' \$(#),#\$\*)+( ' (&%\$( #\$

 

!# ' \$(#),-

**2.4.4 phMRI-responses correlate with dopamine overflow** 

 

CD).

 

&")#+)#%\$-

 -

 

midbrain.

Fig. 6. Hemiparkinsonian nonhuman primates have markedly dimished dopaminergic function. K+ (100 mM)- and amphetamine (250 µM)-evoked DA release was significantly attenuated in the ipsilesional A) putamen (Put) and B) SNc; \*\*\*: P < 0.0001 (paired *t*-test).

the putamen (each measured for a single time point, 30 minutes after stimulus administration) had significant correlations with phMRI responses in the putamen. DA levels in the putamen were also significantly correlated with phMRI responses in the premotor cortex and cingulate gyrus, as well as in the caudate nucleus. Finally, damphetamine-evoked DA release in the SNc was found to have a significant, but negatively correlated relationship with the motor cortex (Fig. 7).

Fig. 7. DA levels in the right SNc correlate with the BOLD responses in the right motor cortex. In animals with lower DA levels in the right SNc, less activation was observed in the right motor cortex

#### **3. Using phMRI to monitor therapeutic effects in parkinsonian monkeys**

There is a great need for the development of noninvasive, highly sensitive, and widely available imaging methods which can potentially be used to longitudinally monitor treatment of PD. We reported the monitoring of glial-cell-line-derived neurotrophic factor (GDNF) induced functional changes of the basal ganglia in hemiparkinsonian monkeys via phMRI measuring the BOLD response to a direct dopamine agonist, APO, (Luan *et al.*, 2008). The effectiveness of GDNF to protect and restore the nigrostriatal dopaminergic system in rodent and nonhuman primate models of PD has been extensively documented (Beck *et al.*,

Developing an MRI-Based Biomarker for Early Diagnosis of Parkinson's Disease 125

to 32 years old, Cass and colleagues (2006) found significant decreases in motor performance, decreases in striatal DA release, and increases in striatal iron levels in rhesus monkeys as they aged from young adulthood. A comprehensive statistical analysis relating age, motor performance, DA release, and iron content indicated that the best predictor of decreases in motor ability, above and beyond levels of performance that could be explained by age alone, was iron accumulation in the striatum. Compared to the young animals, the relaxation rate 1/T2\* used as an indicator of iron content was elevated by 38-43% in all three regions in the middle-aged monkeys (Fig. 9). In the aged animals, iron content was increased by 55%, 61%, and 79% in the caudate, putamen, and nigra, respectively, compared to the young animals (Fig. 9). Iron content in the nigra of the aged animals was also 30% higher than in the middle-aged animals. ROI data for 1/T2 measures are not shown but exhibited a similar dependence on age. Regression analysis extended the group statistics and further confirmed the strong age-associated increase of the MRI relaxation rate 1/T2\* (equivalent to a T2\*-shortening) in each of the three regions of interest (n = 24; p<0.0001). The intercept and rate of increase were 16.537+ 0.598 sec-1/year, 15.728+0.734 sec-1/year, and 19.047+0.791 sec-1/year for the caudate, putamen, and substantia nigra, respectively. This suggests that striatal iron levels may be a biomarker of motor dysfunction in aging; and as

Fig. 9. Using MR imaging of iron content to identify alterations in the aging brain. The use of MR imaging to identify the relative iron content in particular brain structures illustrates the potential usefulness for a non-invasive means of assessing changes in the nigrostriatal system with aging. \*: *P* <0.05; #: *P* <0.01 (one-way ANOVA). (from Cass *et al.*, 2006)

**5. Diffusion Tensor Image (DTI) and dopamine deficiency in rhesus monkeys**  Diffusion tensor imaging (DTI) has been increasingly used in PD related research (Schuff, 2009). DTI can be used to noninvasively investigate and identify white matter (WM) changes associated with PD. DTI is able to obtain quantitative information about fractional anisotropy (FA) and mean diffusivity (MD). A diminished FA is thought to reflect axonal loss and demyelination. A recent DTI study was conducted by our group in normal (n=9) and hemiparkinsonian (n=8) monkeys to explore the MPTP-effects on WM using the DTI parameters of FA and MD. Under general anesthesia, DTI data was obtained on a 3T Siemens

such, can be monitored non-invasively by longitudinal brain MRI scans.

1995; Tomac *et al.*, 1995; Gash *et al.*, 1996; Kordower *et al.*, 2000; Grondin *et al.*, 2002). This trophic factor has also shown promise in Phase I clinical trials for the treatment of PD (Gill *et al.*, 2003; Slevin *et al.*, 2005). Ample evidence supports the idea that GDNF can protect and promote survival of pre-synaptic dopaminergic neurons in the SNc and axons in the striatum (Gash *et al.*, 1996). After testing BOLD responses to APO in their normal state, additional scans were taken with the same dose of APO stimulation after MPTP-induced hemiparkinsonism. Then, the animals were chronically treated with GDNF for 18 weeks by a programmable pump and catheter system. The catheter was surgically implanted into the right putamen and connected to the pump via flexible polyurethane tubing. phMRI scans were taken at both 6 and 18 weeks while they received 22.5µg of GDNF per day (Fig. 8). In addition, behavioral changes were monitored throughout the entire study. The primary finding of this study was that APO-evoked activations in the DA denervated putamen were attenuated by the chronic intraputamenal infusion of GDNF accompanied by improvements of parkinsonian features, movement speed and APO-induced rotation compared to data collected before the chronic GDNF treatment. The results suggest that phMRI methods in combination with administration of a selective DA agonist may be useful for monitoring neurorestorative therapies in PD patients in the future.

Fig. 8. phMRI (BOLD)-responses to APO can be used to monitor GDNF-induced neurorestorative therapeutic effects in rhesus monkeys with MPTP-induced hemiparkinsonisms. phMRI activation reveals differences in dopaminergic activity after GDNF treatment (from Luan *et al.*, 2008).

## **4. Brain iron and motor deficits in rhesus monkeys**

Schuff (2009) notes in a recent review, perhaps the most consistently reported MRI findings in PD have been the detection of signal changes related to excessive iron, most likely related to ferritin, the main iron-storage protein within the brain. Under normal condition, iron is essential for normal metabolism and used in production of DA. Brain iron may also play an essential role in learning and memory (Fretham *et al.*, 2011). Several years ago, we reported a correlation of R2 with total iron concentration in the brains of rhesus monkeys (Hardy *et al.*, 2005). The results show that the transverse relaxation rate R2 = 1/T2 is highly correlated to and varies linearly with iron content. In the study, Hardy and colleagues demonstrated that R2 was highly correlated with the total iron concentration and that the relationship between R2 and tissue iron concentration appeared to depend upon the iron concentration. In another multidiscipline study of brain iron in a large group of rhesus monkeys ranging in age from 4

1995; Tomac *et al.*, 1995; Gash *et al.*, 1996; Kordower *et al.*, 2000; Grondin *et al.*, 2002). This trophic factor has also shown promise in Phase I clinical trials for the treatment of PD (Gill *et al.*, 2003; Slevin *et al.*, 2005). Ample evidence supports the idea that GDNF can protect and promote survival of pre-synaptic dopaminergic neurons in the SNc and axons in the striatum (Gash *et al.*, 1996). After testing BOLD responses to APO in their normal state, additional scans were taken with the same dose of APO stimulation after MPTP-induced hemiparkinsonism. Then, the animals were chronically treated with GDNF for 18 weeks by a programmable pump and catheter system. The catheter was surgically implanted into the right putamen and connected to the pump via flexible polyurethane tubing. phMRI scans were taken at both 6 and 18 weeks while they received 22.5µg of GDNF per day (Fig. 8). In addition, behavioral changes were monitored throughout the entire study. The primary finding of this study was that APO-evoked activations in the DA denervated putamen were attenuated by the chronic intraputamenal infusion of GDNF accompanied by improvements of parkinsonian features, movement speed and APO-induced rotation compared to data collected before the chronic GDNF treatment. The results suggest that phMRI methods in combination with administration of a selective DA agonist may be useful for monitoring

neurorestorative therapies in PD patients in the future.

GDNF treatment (from Luan *et al.*, 2008).

**4. Brain iron and motor deficits in rhesus monkeys** 

Fig. 8. phMRI (BOLD)-responses to APO can be used to monitor GDNF-induced neurorestorative therapeutic effects in rhesus monkeys with MPTP-induced

A. Pre-GDNF B. Post-GDNF

hemiparkinsonisms. phMRI activation reveals differences in dopaminergic activity after

Schuff (2009) notes in a recent review, perhaps the most consistently reported MRI findings in PD have been the detection of signal changes related to excessive iron, most likely related to ferritin, the main iron-storage protein within the brain. Under normal condition, iron is essential for normal metabolism and used in production of DA. Brain iron may also play an essential role in learning and memory (Fretham *et al.*, 2011). Several years ago, we reported a correlation of R2 with total iron concentration in the brains of rhesus monkeys (Hardy *et al.*, 2005). The results show that the transverse relaxation rate R2 = 1/T2 is highly correlated to and varies linearly with iron content. In the study, Hardy and colleagues demonstrated that R2 was highly correlated with the total iron concentration and that the relationship between R2 and tissue iron concentration appeared to depend upon the iron concentration. In another multidiscipline study of brain iron in a large group of rhesus monkeys ranging in age from 4 to 32 years old, Cass and colleagues (2006) found significant decreases in motor performance, decreases in striatal DA release, and increases in striatal iron levels in rhesus monkeys as they aged from young adulthood. A comprehensive statistical analysis relating age, motor performance, DA release, and iron content indicated that the best predictor of decreases in motor ability, above and beyond levels of performance that could be explained by age alone, was iron accumulation in the striatum. Compared to the young animals, the relaxation rate 1/T2\* used as an indicator of iron content was elevated by 38-43% in all three regions in the middle-aged monkeys (Fig. 9). In the aged animals, iron content was increased by 55%, 61%, and 79% in the caudate, putamen, and nigra, respectively, compared to the young animals (Fig. 9). Iron content in the nigra of the aged animals was also 30% higher than in the middle-aged animals. ROI data for 1/T2 measures are not shown but exhibited a similar dependence on age. Regression analysis extended the group statistics and further confirmed the strong age-associated increase of the MRI relaxation rate 1/T2\* (equivalent to a T2\*-shortening) in each of the three regions of interest (n = 24; p<0.0001). The intercept and rate of increase were 16.537+ 0.598 sec-1/year, 15.728+0.734 sec-1/year, and 19.047+0.791 sec-1/year for the caudate, putamen, and substantia nigra, respectively. This suggests that striatal iron levels may be a biomarker of motor dysfunction in aging; and as such, can be monitored non-invasively by longitudinal brain MRI scans.

Fig. 9. Using MR imaging of iron content to identify alterations in the aging brain. The use of MR imaging to identify the relative iron content in particular brain structures illustrates the potential usefulness for a non-invasive means of assessing changes in the nigrostriatal system with aging. \*: *P* <0.05; #: *P* <0.01 (one-way ANOVA). (from Cass *et al.*, 2006)

#### **5. Diffusion Tensor Image (DTI) and dopamine deficiency in rhesus monkeys**

Diffusion tensor imaging (DTI) has been increasingly used in PD related research (Schuff, 2009). DTI can be used to noninvasively investigate and identify white matter (WM) changes associated with PD. DTI is able to obtain quantitative information about fractional anisotropy (FA) and mean diffusivity (MD). A diminished FA is thought to reflect axonal loss and demyelination. A recent DTI study was conducted by our group in normal (n=9) and hemiparkinsonian (n=8) monkeys to explore the MPTP-effects on WM using the DTI parameters of FA and MD. Under general anesthesia, DTI data was obtained on a 3T Siemens

Developing an MRI-Based Biomarker for Early Diagnosis of Parkinson's Disease 127

We thank Drs. Anders Andersen, Peter Hardy, and Richard Grondin for their technical help in data analysis, behavioral evaluation, and discussions. Support provided by USPHS NIH

Andersen, A.H., Zhang, Z., Barber, T., Rayens, W.S., Zhang, J., Grondin, R., Hardy, P.,

Andringa, G., Drukarch, B., Bol, J.G., de Bruin, K., Sorman, K., Habraken, J.B. & Booij, J.

Arthurs, O.J. & Boniface, S. (2002) How well do we understand the neural origins of the

Beck, K.D., Valverde, J., Alexi, T., Poulsen, K., Moffat, B., Vandlen, R.A., Rosenthal, A. &

Brooks, D.J., Frey, K.A., Marek, K.L., Oakes, D., Paty, D., Prentice, R., Shults,

axotomy-induced degeneration in the adult brain. *Nature*, 373, 339-341. Braak, H. & Del Tredici, K. (2008) Cortico-basal ganglia-cortical circuitry in Parkinson's

Gerhardt, G.A. & Gash, D.M. (2002) Functional MRI studies in awake rhesus monkeys: methodological and analytical strategies. *J Neurosci Methods*, 118, 141-152.

(2005) Pinhole SPECT imaging of dopamine transporters correlates with dopamine transporter immunohistochemical analysis in the MPTP mouse model of

Hefti, F. (1995) Mesencephalic dopaminergic neurons protected by GDNF from

C.W. & Stoessl, A.J. (2003) Assessment of neuroimaging techniques as

Fig. 10. Employing MRI methodologies in the clinic for PD.

Parkinson's disease. *Neuroimage*, 26, 1150-1158.

fMRI BOLD signal? *Trends Neurosci*, 25, 27-31.

disease reconsidered. *Exp Neurol*, 212, 226-229.

Brooks, D.J. (2004) Neuroimaging in Parkinson's disease. *NeuroRx*, 1, 243-254.

**7. Acknowledgment**

**8. References** 

grants NS50242, NS39787, and AG13494.

Trio MRI scanner with a custom-built, single channel, receive-only coil, built on a fiberglass frame and used to enhance the received signal. Imaging consisted of single shot (SS), double pulsed gradient spin echo (double-PGSE), diffusion weighted, echo planar imaging (EPI) with a spatial resolution of 1.23×1.23×2.0 mm3. Images were processed and analyzed by using the publicly available image processing software FSL (http://www.fmrib.ox.ac.uk/fsl) (Smith *et al.*, 2004; Smith *et al.*, 2006). All of the processing tools referred to by their FSL acronyms are available for download at the website. First, we observed a WM tract in the vicinity of the basal ganglia (BG) with FA greater (P<0.01, *t*-test) in the aged-matched control animals than MPTP-treated animals in the same structure. Second, we observed multiple WM tracts in the sensory cortex, with FA greater (P<0.05, *t*-test) in the MPTP-treated than untreated side in the same animals. The result from the pilot study supports the idea that high resolution DTI has the potential to distinguish animals with a MPTP-lesioned nigrostriatal system from normal age-matched, healthy controls on an animal-by-animal basis.

## **6. Conclusion and perspectives**

Since a diagnosis of PD still solely depends on the judgment of the clinician, there is an urgent demand for the development of reliable and applicable test systems or biomarkers to provide a level of certainty to the diagnosis. Objective biomarkers of PD are pivotal to tracking the disease progression and confirm the therapeutic effects. Non- or minimallyinvasive imaging techniques provide a unique, real-time opportunity to assess the changes that occur with neurodegenerative diseases. In addition, with the rapidly expanding use of fMRI to provide a dramatically greater understanding of brain function, imaging techniques such as phMRI are only bound to benefit from this new wealth of knowledge.

The advantage of MRI is that MRIs are far more widely available than other imaging modalities and are most commonly used in clinical practice to differentiate idiopathic PD from secondary cause of parkinsonism (Pavese & Brooks, 2009). Recent advancement in high field MRI technology offers even better opportunities for noninvasively, longitudinally, and objectively assessing brain alterations in PD. For example functional and pharmacological MRI has been increasingly employed for preclinical and clinical research of the disease. Ample evidence supports that MRI signals have the potential to be developed as a noninvasive state biomarker in PD. For example, several MRI methodologies such as structural MRI, imaging of brain iron, DTI, functional MRI and pharmacological MRI have provided meaningful insight of brain alteration in PD. That said, we note that while we have gained greater understanding of the changes that occur in disorders of dopaminergic dysfunction with the use of phMRI in the rhesus model of PD, nevertheless the studies are works in progress and ones that still require cautious interpretation because conditions in patients with PD are more complex than in the animal model used in these studies.

In our hands, MRI studies conducted at the University of Kentucky have demonstrated that phMRI-responses to dopaminergic challenges in MPTP-treated monkeys are highly correlated with 1) the severity of parkinsonism, 2) the loss of dopamine neurons and terminals, 3) the decline of dopamine overflow and 4) the functional recovery from GDNF treatment. In addition, results from imaging brain iron suggest that striatal iron levels may constitute a biomarker for motor dysfunction in aged animals with parkinsonism. As shown in Fig. 10, combining various MRI methodologies may be used to screen populations at high risk, to differentiate idiopathic PD from second causes of parkinsonisms, and to monitor progression of the disease and the therapeutic effects.

Fig. 10. Employing MRI methodologies in the clinic for PD.

## **7. Acknowledgment**

We thank Drs. Anders Andersen, Peter Hardy, and Richard Grondin for their technical help in data analysis, behavioral evaluation, and discussions. Support provided by USPHS NIH grants NS50242, NS39787, and AG13494.

## **8. References**

126 Diagnostics and Rehabilitation of Parkinson's Disease

Trio MRI scanner with a custom-built, single channel, receive-only coil, built on a fiberglass frame and used to enhance the received signal. Imaging consisted of single shot (SS), double pulsed gradient spin echo (double-PGSE), diffusion weighted, echo planar imaging (EPI) with a spatial resolution of 1.23×1.23×2.0 mm3. Images were processed and analyzed by using the publicly available image processing software FSL (http://www.fmrib.ox.ac.uk/fsl) (Smith *et al.*, 2004; Smith *et al.*, 2006). All of the processing tools referred to by their FSL acronyms are available for download at the website. First, we observed a WM tract in the vicinity of the basal ganglia (BG) with FA greater (P<0.01, *t*-test) in the aged-matched control animals than MPTP-treated animals in the same structure. Second, we observed multiple WM tracts in the sensory cortex, with FA greater (P<0.05, *t*-test) in the MPTP-treated than untreated side in the same animals. The result from the pilot study supports the idea that high resolution DTI has the potential to distinguish animals with a MPTP-lesioned nigrostriatal system from normal

Since a diagnosis of PD still solely depends on the judgment of the clinician, there is an urgent demand for the development of reliable and applicable test systems or biomarkers to provide a level of certainty to the diagnosis. Objective biomarkers of PD are pivotal to tracking the disease progression and confirm the therapeutic effects. Non- or minimallyinvasive imaging techniques provide a unique, real-time opportunity to assess the changes that occur with neurodegenerative diseases. In addition, with the rapidly expanding use of fMRI to provide a dramatically greater understanding of brain function, imaging techniques

The advantage of MRI is that MRIs are far more widely available than other imaging modalities and are most commonly used in clinical practice to differentiate idiopathic PD from secondary cause of parkinsonism (Pavese & Brooks, 2009). Recent advancement in high field MRI technology offers even better opportunities for noninvasively, longitudinally, and objectively assessing brain alterations in PD. For example functional and pharmacological MRI has been increasingly employed for preclinical and clinical research of the disease. Ample evidence supports that MRI signals have the potential to be developed as a noninvasive state biomarker in PD. For example, several MRI methodologies such as structural MRI, imaging of brain iron, DTI, functional MRI and pharmacological MRI have provided meaningful insight of brain alteration in PD. That said, we note that while we have gained greater understanding of the changes that occur in disorders of dopaminergic dysfunction with the use of phMRI in the rhesus model of PD, nevertheless the studies are works in progress and ones that still require cautious interpretation because conditions in

such as phMRI are only bound to benefit from this new wealth of knowledge.

patients with PD are more complex than in the animal model used in these studies.

progression of the disease and the therapeutic effects.

In our hands, MRI studies conducted at the University of Kentucky have demonstrated that phMRI-responses to dopaminergic challenges in MPTP-treated monkeys are highly correlated with 1) the severity of parkinsonism, 2) the loss of dopamine neurons and terminals, 3) the decline of dopamine overflow and 4) the functional recovery from GDNF treatment. In addition, results from imaging brain iron suggest that striatal iron levels may constitute a biomarker for motor dysfunction in aged animals with parkinsonism. As shown in Fig. 10, combining various MRI methodologies may be used to screen populations at high risk, to differentiate idiopathic PD from second causes of parkinsonisms, and to monitor

age-matched, healthy controls on an animal-by-animal basis.

**6. Conclusion and perspectives** 


Developing an MRI-Based Biomarker for Early Diagnosis of Parkinson's Disease 129

Grondin, R., Zhang, Z., Yi, A., Cass, W.A., Maswood, N., Andersen, A.H., Elsberry, D.D.,

Group, B.D.W. (2001) Biomarkers and surrogate endpoints: Preferred definitions and

Hardy, P.A., Gash, D., Yokel, R., Andersen, A., Ai, Y. & Zhang, Z. (2005) Correlation of R2

Honey, G. & Bullmore, E. (2004) Human pharmacological MRI. *Trends Pharmacol Sci*, 25, 366-374. Hornykiewicz, O. & Kish, S.J. (1987) Biochemical pathophysiology of Parkinson's disease.

Hoshi, H., Kuwabara, H., Leger, G., Cumming, P., Guttman, M. & Gjedde, A. (1993) 6-

Hubble, J.P. (2000) Pre-clinical studies of pramipexole: clinical relevance. *Eur J Neurol*, 7

Jenkins, B.G., Sanchez-Pernaute, R., Brownell, A.L., Chen, Y.C. & Isacson, O. (2004) Mapping

Katzenschlager, R. & Lees, A.J. (2004) Olfaction and Parkinson's syndromes: its role in

Kordower, J.H., Emborg, M.E., Bloch, J., Ma, S.Y., Chu, Y., Leventhal, L., McBride, J., Chen,

Lang, A.E., Gill, S., Patel, N.K., Lozano, A., Nutt, J.G., Penn, R., Brooks, D.J., Hotton, G.,

Langston, J.W. & Ballard, P.A., Jr. (1983) Parkinson's disease in a chemist working with 1 methyl-4-phenyl-1,2,5,6-tetrahydropyridine. *N Engl J Med*, 309, 310. Luan, L., Ding, F., Ai, Y., Andersen, A., Hardy, P., Forman, E., Gerhardt, G.A., Gash, D.M.,

using pharmacological MRI and correlations with PET. *Synapse*, 36, 57-65. Nicklas, W.J., Youngster, S.K., Kindt, M.V. & Heikkila, R.E. (1987) MPTP, MPP+ and

Ovadia, A., Zhang, Z. & Gash, D.M. (1995) Increased susceptibility to MPTP toxicity in

Pavese, N. & Brooks, D.J. (2009) Imaging neurodegeneration in Parkinson's disease. *Biochim* 

middle-aged rhesus monkeys. *Neurobiol Aging*, 16, 931-937.

conceptual framework\*. *Clin Pharmacol Ther*, 69, 89-95.

analytical methods. *J Cereb Blood Flow Metab*, 13, 57-69.

differential diagnosis. *Curr Opin Neurol*, 17, 417-423.

models of Parkinson's disease. *Science*, 290, 767-773.

monkeys. *Brain*, 125, 2191-2201.

*Imaging*, 21, 118-127.

*Adv Neurol*, 45, 19-34.

*Neurosci*, 24, 9553-9560.

disease. *Ann Neurol*, 59, 459-466.

mitochondrial function. *Life Sci*, 40, 721-729.

*Biophys Acta*, 1792, 722-729.

Suppl 1, 15-20.

Klein, M.C., Gerhardt, G.A. & Gash, D.M. (2002) Chronic, controlled GDNF infusion promotes structural and functional recovery in advanced parkinsonian

with total iron concentration in the brains of rhesus monkeys. *J Magn Reson* 

[18F]fluoro-L-dopa metabolism in living human brain: a comparison of six

dopamine function in primates using pharmacologic magnetic resonance imaging. *J* 

E.Y., Palfi, S., Roitberg, B.Z., Brown, W.D., Holden, J.E., Pyzalski, R., Taylor, M.D., Carvey, P., Ling, Z., Trono, D., Hantraye, P., Deglon, N. & Aebischer, P. (2000) Neurodegeneration prevented by lentiviral vector delivery of GDNF in primate

Moro, E., Heywood, P., Brodsky, M.A., Burchiel, K., Kelly, P., Dalvi, A., Scott, B., Stacy, M., Turner, D., Wooten, V.G., Elias, W.J., Laws, E.R., Dhawan, V., Stoessl, A.J., Matcham, J., Coffey, R.J. & Traub, M. (2006) Randomized controlled trial of intraputamenal glial cell line-derived neurotrophic factor infusion in Parkinson

Grondin, R. & Zhang, Z. (2008) Pharmacological MRI (phMRI) monitoring of treatment in hemiparkinsonian rhesus monkeys. *Cell Transplant*, 17, 417-425. Marsden, C.D. & Obeso, J.A. (1994) The functions of the basal ganglia and the paradox of stereotaxic surgery in Parkinson's disease. *Brain*, 117 ( Pt 4), 877-897. Nguyen, T.V., Brownell, A.L., Iris Chen, Y.C., Livni, E., Coyle, J.T., Rosen, B.R., Cavagna, F.

& Jenkins, B.G. (2000) Detection of the effects of dopamine receptor supersensitivity

biomarkers of the progression of Parkinson's disease. *Exp Neurol*, 184 Suppl 1, S68-79.


Brooks, D.J., Ibanez, V., Sawle, G.V., Quinn, N., Lees, A.J., Mathias, C.J., Bannister, R.,

Cass, W.A., Grondin, R., Andersen, A.H., Zhang, Z., Hardy, P.A., Hussey-Andersen, L.K.,

Chin, C.L., Tovcimak, A.E., Hradil, V.P., Seifert, T.R., Hollingsworth, P.R., Chandran, P.,

de la Fuente-Fernandez, R. & Stoessl, A.J. (2002) Parkinson's disease: imaging update. *Curr* 

DeKosky, S.T. & Marek, K. (2003) Looking backward to move forward: early detection of

Ding, F., Luan, L., Ai, Y., Walton, A., Gerhardt, G.A., Gash, D.M., Grondin, R. & Zhang, Z.

Dorsey, E.R., Constantinescu, R., Thompson, J.P., Biglan, K.M., Holloway, R.G., Kieburtz, K.,

Eckert, T. & Eidelberg, D. (2004) The role of functional neuroimaging in the differential diagnosis

Fearnley, J.M. & Lees, A.J. (1991) Ageing and Parkinson's disease: substantia nigra regional

Ferrer, I., Martinez, A., Blanco, R., Dalfó, E. & Carmona, M. (2010) Neuropathology of

Freed, C.R., Greene, P.E., Breeze, R.E., Tsai, W.Y., DuMouchel, W., Kao, R., Dillon, S.,

Fretham, S.J.B., Carlson, E.S. & Georgieff, M.K. (2011) The Role of Iron in Learning and Memory. *Advances in Nutrition: An International Review Journal*, 2, 112-121. Gash, D.M., Zhang, Z., Ovadia, A., Cass, W.A., Yi, A., Simmerman, L., Russell, D., Martin,

supranuclear palsy. *Ann Neurol*, 28, 547-555.

functional MRI. *Magn Reson Imaging*, 14, 469-476.

MRI. *British Journal of Pharmacology*, 153, 367-379.

neurodegenerative disorders. *Science*, 302, 830-834.

2005 through 2030. *Neurology*, 68, 384-386.

selectivity. *Brain*, 114 ( Pt 5), 2283-2301.

in middle-aged rhesus monkeys. *Exp Neurol*, 212, 431-439.

Parkinson disease. *Journal of Neural Transmission*, 1-19.

neurotrophic factor in Parkinson disease. *Nat Med*, 9, 589-595.

*Opin Neurol*, 15, 477-482.

*Engl J Med*, 344, 710-719.

S68-79.

biomarkers of the progression of Parkinson's disease. *Exp Neurol*, 184 Suppl 1,

Marsden, C.D. & Frackowiak, R.S. (1990) Differing patterns of striatal 18F-dopa uptake in Parkinson's disease, multiple system atrophy, and progressive

Rayens, W.S., Gerhardt, G.A. & Gash, D.M. (2006) Iron accumulation in the striatum predicts aging-related decline in motor function in rhesus monkeys. *Neurobiol.Aging*. Chaudhuri, K.R., Healy, D.G. & Schapira, A.H.V. (2006) Non-motor symptoms of Parkinson's disease: diagnosis and management. *The Lancet Neurology*, 5, 235-245. Chen, Q., Andersen, A.H., Zhang, Z., Ovadia, A., Gash, D.M. & Avison, M.J. (1996) Mapping

drug-induced changes in cerebral R2\* by Multiple Gradient Recalled Echo

Zhu, C.Z., Gauvin, D., Pai, M., Wetter, J., Hsieh, G.C., Honore, P., Frost, J.M., Dart, M.J., Meyer, M.D., Yao, B.B., Cox, B.F. & Fox, G.B. (2008) Differential effects of cannabinoid receptor agonists on regional brain activity using pharmacological

(2008) Development of a stable, early stage unilateral model of Parkinson's disease

Marshall, F.J., Ravina, B.M., Schifitto, G., Siderowf, A. & Tanner, C.M. (2007) Projected number of people with Parkinson disease in the most populous nations,

of idiopathic Parkinson's disease and multiple system atrophy. *Clin Auton Res*, 14, 84-91.

sporadic Parkinson disease before the appearance of parkinsonism: preclinical

Winfield, H., Culver, S., Trojanowski, J.Q., Eidelberg, D. & Fahn, S. (2001) Transplantation of embryonic dopamine neurons for severe Parkinson's disease. *N* 

D., Lapchak, P.A., Collins, F., Hoffer, B.J. & Gerhardt, G.A. (1996) Functional recovery in parkinsonian monkeys treated with GDNF. *Nature*, 380, 252-255. Gill, S.S., Patel, N.K., Hotton, G.R., O'Sullivan, K., McCarter, R., Bunnage, M., Brooks, D.J.,

Svendsen, C.N. & Heywood, P. (2003) Direct brain infusion of glial cell line-derived


**6** 

**Neuroimaging** 

Yangho Kim

*Ulsan, Korea* 

 **in Manganese-Induced Parkinsonism** 

*Ulsan University Hospital, University of Ulsan College of Medicine,* 

Over the last 20 years, the impact of imaging on the clinical sciences has been immense. Tremendous progress has been made in medical imaging of the human body since the invention of computed tomography (CT) and magnetic resonance imaging (MRI). Neuroimaging of patients with metal neurotoxicity can be divided into two types: morphological neuroimaging (anatomy-based imaging) including CT and MRI; and functional neuroimaging (physiology-based imaging) such as magnetic resonance spectroscopy (MRS), single-photon emission computed tomography (SPECT), positron emission tomography (PET), diffusion tensor imaging (DTI), and functional MRI (fMRI). Neuroimaging is undergoing a shift from morphological to functional imaging as new technologies are introduced and technical problems associated with the local production of radioisotopes are solved (Lang, 2000; Walker et al., 2004). MRI, PET, and SPECT have been used for 10 years or more to evaluate workers exposed to manganese (Mn), and to examine the neurological consequences of such exposure. Very recently, functional neuroimaging

The objectives of this chapter are (1) to review the use of neuroimaging in Mn-induced parkinsonism, and (2) to discuss recent developments in the functional neuroimaging in

**2. The pallidal MRI T1-signal reflects the target organ dose of Mn exposure**  The Mn ion (Mn2+) has five unpaired electrons in the 3d orbital, which results in a large magnetic moment, resulting in the shortening of proton T1-relaxation time and an increased signal intensity on T1-weighted MRI. Because of this paramagnetic quality of Mn2+, a bilateral symmetrical increase in signal intensity, mainly confined to the globus pallidus and midbrain, can be observed on T1-weighted MRI, but with no concomitant alteration on the

However, Mn-induced high signals on T1-weighted MRI do not correspond to any abnormal findings on brain CT (Park et al., 2003). The characteristic high signal caused by Mn can be differentiated from signals that increase in intensity for other reasons. Thus, high signals from fat, hemoglobin breakdown products, melanomas, neurofibromatosis, and

modalities such as fMRI, MRS, and DTI have been applied to this end.

**1. Introduction** 

Mn-induced parkinsonism.

T2-weighted image (Kim et al., 1999a) (Fig. 1).

*Department of Occupational and Environmental Medicine,* 

Rasmussen Jr, I. (2010) Psychopharmacological MRI. *Acta Neuropsychiatrica*, 22, 38-39.


## **Neuroimaging in Manganese-Induced Parkinsonism**

## Yangho Kim

*Department of Occupational and Environmental Medicine, Ulsan University Hospital, University of Ulsan College of Medicine, Ulsan, Korea* 

## **1. Introduction**

130 Diagnostics and Rehabilitation of Parkinson's Disease

Richardson, J.R., Caudle, W.M., Guillot, T.S., Watson, J.L., Nakamaru-Ogiso, E., Seo, B.B.,

Seibyl, J., Jennings, D., Tabamo, R. & Marek, K. (2005) The role of neuroimaging in the early diagnosis and evaluation of Parkinson's disease. *Minerva Med*, 96, 353-364. Slevin, J.T., Gerhardt, G.A., Smith, C.D., Gash, D.M., Kryscio, R. & Young, B. (2005)

Smith, S.M., Jenkinson, M., Johansen-Berg, H., Rueckert, D., Nichols, T.E., Mackay, C.E.,

Smith, S.M., Jenkinson, M., Woolrich, M.W., Beckmann, C.F., Behrens, T.E., Johansen-Berg,

Snow, B.J., Tooyama, I., McGeer, E.G., Yamada, T., Calne, D.B., Takahashi, H. & Kimura, H.

Snow, B.J., Vingerhoets, F.J., Langston, J.W., Tetrud, J.W., Sossi, V. & Calne, D.B. (2000)

Thiel, C.M. (2009) Neuropharmacological fMRI. *Neuropharmakologisches fMRT*, 40, 233-238. Tomac, A., Lindqvist, E., Lin, L.F., Ogren, S.O., Young, D., Hoffer, B.J. & Olson, L. (1995)

Tracey, I. (2001) Prospects for human pharmacological functional magnetic resonance

Wu, Y., Le, W. & Jankovic, J. (2011) Preclinical Biomarkers of Parkinson Disease. *Arch* 

Zhang, Z., Andersen, A., Grondin, R., Barber, T., Avison, R., Gerhardt, G. & Gash, D. (2001)

Zhang, Z., Andersen, A.H., Ai, Y., Loveland, A., Hardy, P.A., Gerhardt, G.A. & Gash, D.M.

with dopamine cell counts and levels. *Ann Neurol*, 34, 324-330.

parkinsonism. *J Neurol Neurosurg Psychiatry*, 68, 313-316.

imaging (phMRI). *J Clin Pharmacol*, Suppl, 21S-28S.

of awake rhesus monkeys. *Neuroimage*, 14, 1159-1167.

parkinsonian rhesus macaques. *Neuroimage*, 33, 636-643.

Feigin, A., Fahn, S., Guttman, M., Gwinn-Hardy, K., McFarland, H., Innis, R., Katz, R.G., Kieburtz, K., Kish, S.J., Lange, N., Langston, J.W., Marek, K., Morin, L., Moy, C., Murphy, D., Oertel, W.H., Oliver, G., Palesch, Y., Powers, W., Seibyl, J., Sethi, K.D., Shults, C.W., Sheehy, P., Stoessl, A.J. & Holloway, R. (2005) The role of

Sherer, T.B., Greenamyre, J.T., Yagi, T., Matsuno-Yagi, A. & Miller, G.W. (2007) Obligatory role for complex I inhibition in the dopaminergic neurotoxicity of 1 methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP). *Toxicol Sci*, 95, 196-204. Schuff, N. (2009) Potential role of high-field MRI for studies in Parkinson's disease. *Mov* 

Improvement of bilateral motor functions in patients with Parkinson disease through the unilateral intraputaminal infusion of glial cell line-derived

Watkins, K.E., Ciccarelli, O., Cader, M.Z., Matthews, P.M. & Behrens, T.E. (2006) Tract-based spatial statistics: voxelwise analysis of multi-subject diffusion data.

H., Bannister, P.R., De Luca, M., Drobnjak, I., Flitney, D.E., Niazy, R.K., Saunders, J., Vickers, J., Zhang, Y., De Stefano, N., Brady, J.M. & Matthews, P.M. (2004) Advances in functional and structural MR image analysis and implementation as

(1993) Human positron emission tomographic [18F]fluorodopa studies correlate

Pattern of dopaminergic loss in the striatum of humans with MPTP induced

Protection and repair of the nigrostriatal dopaminergic system by GDNF in vivo.

Pharmacological MRI mapping of age-associated changes in basal ganglia circuitry

(2006) Assessing nigrostriatal dysfunctions by pharmacological MRI in

Rasmussen Jr, I. (2010) Psychopharmacological MRI. *Acta Neuropsychiatrica*, 22, 38-39. Ravina, B., Eidelberg, D., Ahlskog, J.E., Albin, R.L., Brooks, D.J., Carbon, M., Dhawan, V.,

radiotracer imaging in Parkinson disease. *Neurology*, 64, 208-215.

*Disord*, 24 Suppl 2, S684-690.

*Neuroimage*, 31, 1487-1505.

*Nature*, 373, 335-339.

*Neurol*, 68, 22-30.

FSL. *Neuroimage*, 23 Suppl 1, S208-219.

neurotrophic factor. *J Neurosurg.*, 102, 216-222.

Over the last 20 years, the impact of imaging on the clinical sciences has been immense. Tremendous progress has been made in medical imaging of the human body since the invention of computed tomography (CT) and magnetic resonance imaging (MRI). Neuroimaging of patients with metal neurotoxicity can be divided into two types: morphological neuroimaging (anatomy-based imaging) including CT and MRI; and functional neuroimaging (physiology-based imaging) such as magnetic resonance spectroscopy (MRS), single-photon emission computed tomography (SPECT), positron emission tomography (PET), diffusion tensor imaging (DTI), and functional MRI (fMRI). Neuroimaging is undergoing a shift from morphological to functional imaging as new technologies are introduced and technical problems associated with the local production of radioisotopes are solved (Lang, 2000; Walker et al., 2004). MRI, PET, and SPECT have been used for 10 years or more to evaluate workers exposed to manganese (Mn), and to examine the neurological consequences of such exposure. Very recently, functional neuroimaging modalities such as fMRI, MRS, and DTI have been applied to this end.

The objectives of this chapter are (1) to review the use of neuroimaging in Mn-induced parkinsonism, and (2) to discuss recent developments in the functional neuroimaging in Mn-induced parkinsonism.

## **2. The pallidal MRI T1-signal reflects the target organ dose of Mn exposure**

The Mn ion (Mn2+) has five unpaired electrons in the 3d orbital, which results in a large magnetic moment, resulting in the shortening of proton T1-relaxation time and an increased signal intensity on T1-weighted MRI. Because of this paramagnetic quality of Mn2+, a bilateral symmetrical increase in signal intensity, mainly confined to the globus pallidus and midbrain, can be observed on T1-weighted MRI, but with no concomitant alteration on the T2-weighted image (Kim et al., 1999a) (Fig. 1).

However, Mn-induced high signals on T1-weighted MRI do not correspond to any abnormal findings on brain CT (Park et al., 2003). The characteristic high signal caused by Mn can be differentiated from signals that increase in intensity for other reasons. Thus, high signals from fat, hemoglobin breakdown products, melanomas, neurofibromatosis, and

Neuroimaging in Manganese-Induced Parkinsonism 133

al., 2003). Mn-induced high signals can occasionally be observed in patients with severe iron-deficiency anemia (Kim et al., 2005). Kim et al. (1999a) showed, for the first time, that the characteristic high T1 signals were also frequently observed in asymptomatic workers exposed to Mn. The cited authors found a high prevalence (41.6%) of increased MRI signals in Mn-exposed workers, and, interestingly, 73.5% of welders showed increased signal intensities compared to none of the non-exposed clerical workers in the same factories. The cited authors found that the increased signal intensities resolved significantly approximately 1 year after Mn exposure ceased (Kim et al., 1999b). The disappearance of high signal abnormalities on MRI following withdrawal of the Mn source has been shown after the cessation of occupational exposure (Nelson et al., 1993), after discontinuation of TPN (Mirowitz et al., 1991), and after liver transplantation in patients with hepatic failure (Choi et al., 2005; Pujol et al., 1993). These findings suggest that increased signal intensities on a T1-weighted image reflect exposure to Mn, but do not necessarily indicate the presence of manganism. This is very important when the similarities and differences between idiopathic Parkinson's disease (IPD) and manganism are considered. Many reports have shown that blood Mn concentration is highly correlated with PI in liver cirrhotics (Hauser et al., 1996; Krieger et al., 1995; Spahr et al., 1996). In Mn-exposed workers, blood Mn concentration was also found to correlate with PI (Chang et al., 2009a; Dietz et al., 1999; Jiang et al., 2007; Kim

A recent study showed that PI was significantly associated with digit symbol test results, digit span backward ratings, scores on the Stroop Word and Stroop Error indices, and Grooved Pegboard (dominant hand) data (Chang et al., 2009a). This means that PI is a good predictor of neurobehavioral performance in welders without clinical manganism. In particular, PI was a better predictor of neurobehavioral performance than was blood Mn

Taken together, the data suggest that PI on MRI may reflect a target organ dose of occupational Mn exposure (Kim, 2006). In addition, Mn in brain has a longer half-life than in blood (Lucchini and Kim, 2009). Thus, PI reflects the cumulative dose better than does blood Mn level. However, the level of signal intensity indicating progression to manganism from Mn exposure remains to be determined. The development of an animal model of manganism would assist in this regard, but the fact that the routes of exposure in humans differ, and that data on non-human species may not be transferable to human situations, are limiting factors. Hence, a prospective study correlating increases in T1 signal intensities with

**3. PET/SPECT as an index of the integrity of the dopaminergic nigrostriatal** 

The dopaminergic nigrostriatal pathway is the primary focus of neurodegeneration in IPD (Brooks et al., 1990; Morrish et al., 1995, 1996). In IPD, dopamine uptake is reduced in the striatum, particularly the posterior putamen. This finding is in accord with the 40-60% loss of dopaminergic cells seen in the nigrostriatal pathway of patients with IPD. In non-human primates and humans intoxicated with Mn, [18F]-dopa (fluorodopa) PET scans are normal (Erikson et al., 1992; Kim et al., 1998; Shinotoh et al., 1995, 1997; Wolters et al., 1989). This supports the view that, in instances of Mn intoxication, the nigrostriatal pathway is relatively well preserved, consistent with many pathological observations showing that Mn-

clinical and neuropsychological findings in Mn-exposed workers is needed.

et al., 1999a).

**pathway** 

levels in such welders.

Fig. 1. T1-weighted MRI scans with or without increased signal intensities. Axial and sagittal sections show increase in signal intensities mainly confined to the globus pallidus, but with no concomitant alteration on T2-weighted image in workers exposed to manganese (Mn) in the upper row. In the lower row, worker without Mn exposure does not show increased signal intensities (Arrow indicates high signal. Left and middle column: T1-weighted image; right column: T2-weighted image.

calcification, can be seen on T1-weighted images. High signals from hemoglobin breakdown products, melanomas, and neurofibromatosis can be differentiated from Mn-induced high signals on the basis of signal site and symmetry. Iron deposits cause a greater shortening of the T2-relaxation time than the T1-relaxation time, resulting in low signal intensity upon T2 weighted imaging, distinct from that of an Mn deposit. Calcification can be easily identified by CT (Ahn et al., 2003). Krieger et al. (1995) coined the term ''pallidus index'' (PI) to quantify Mn accumulation in the globus pallidus, defined as the ratio of the signal intensity in the globus pallidus to that in the subcortical frontal white matter (WM) in axial T1 weighted MRI planes, multiplied by 100. An increase in signal upon T1-weighted imaging was observed during experimental Mn poisoning of non-human primates (Erikson et al., 1992; Newland et al., 1989). Nelson et al. (1993) were the first to report increased signal intensities in a patient with occupational Mn neurointoxication. A similar MRI pattern has also been observed in patients receiving total parenteral nutrition (TPN) by direct intravenous administration (Ejima et al., 1992; Mirowitz et al., 1991) and in patients with portal systemic shunts such as individuals with liver cirrhosis, leading to an inability to clear Mn via biliary excretion (Butterworth et al., 1995; Hauser et al., 1994, 1996; Krieger et al., 1995; Park et al., 2003; Spahr et al., 1996). A high pallidal signal is very frequently observed in patients with established liver cirrhosis, but who lack exposure to Mn (Park et

Fig. 1. T1-weighted MRI scans with or without increased signal intensities. Axial and sagittal sections show increase in signal intensities mainly confined to the globus pallidus, but with no concomitant alteration on T2-weighted image in workers exposed to manganese (Mn) in the upper row. In the lower row, worker without Mn exposure does not show increased signal intensities (Arrow indicates high signal. Left and middle column: T1-weighted image;

calcification, can be seen on T1-weighted images. High signals from hemoglobin breakdown products, melanomas, and neurofibromatosis can be differentiated from Mn-induced high signals on the basis of signal site and symmetry. Iron deposits cause a greater shortening of the T2-relaxation time than the T1-relaxation time, resulting in low signal intensity upon T2 weighted imaging, distinct from that of an Mn deposit. Calcification can be easily identified by CT (Ahn et al., 2003). Krieger et al. (1995) coined the term ''pallidus index'' (PI) to quantify Mn accumulation in the globus pallidus, defined as the ratio of the signal intensity in the globus pallidus to that in the subcortical frontal white matter (WM) in axial T1 weighted MRI planes, multiplied by 100. An increase in signal upon T1-weighted imaging was observed during experimental Mn poisoning of non-human primates (Erikson et al., 1992; Newland et al., 1989). Nelson et al. (1993) were the first to report increased signal intensities in a patient with occupational Mn neurointoxication. A similar MRI pattern has also been observed in patients receiving total parenteral nutrition (TPN) by direct intravenous administration (Ejima et al., 1992; Mirowitz et al., 1991) and in patients with portal systemic shunts such as individuals with liver cirrhosis, leading to an inability to clear Mn via biliary excretion (Butterworth et al., 1995; Hauser et al., 1994, 1996; Krieger et al., 1995; Park et al., 2003; Spahr et al., 1996). A high pallidal signal is very frequently observed in patients with established liver cirrhosis, but who lack exposure to Mn (Park et

right column: T2-weighted image.

al., 2003). Mn-induced high signals can occasionally be observed in patients with severe iron-deficiency anemia (Kim et al., 2005). Kim et al. (1999a) showed, for the first time, that the characteristic high T1 signals were also frequently observed in asymptomatic workers exposed to Mn. The cited authors found a high prevalence (41.6%) of increased MRI signals in Mn-exposed workers, and, interestingly, 73.5% of welders showed increased signal intensities compared to none of the non-exposed clerical workers in the same factories. The cited authors found that the increased signal intensities resolved significantly approximately 1 year after Mn exposure ceased (Kim et al., 1999b). The disappearance of high signal abnormalities on MRI following withdrawal of the Mn source has been shown after the cessation of occupational exposure (Nelson et al., 1993), after discontinuation of TPN (Mirowitz et al., 1991), and after liver transplantation in patients with hepatic failure (Choi et al., 2005; Pujol et al., 1993). These findings suggest that increased signal intensities on a T1-weighted image reflect exposure to Mn, but do not necessarily indicate the presence of manganism. This is very important when the similarities and differences between idiopathic Parkinson's disease (IPD) and manganism are considered. Many reports have shown that blood Mn concentration is highly correlated with PI in liver cirrhotics (Hauser et al., 1996; Krieger et al., 1995; Spahr et al., 1996). In Mn-exposed workers, blood Mn concentration was also found to correlate with PI (Chang et al., 2009a; Dietz et al., 1999; Jiang et al., 2007; Kim et al., 1999a).

A recent study showed that PI was significantly associated with digit symbol test results, digit span backward ratings, scores on the Stroop Word and Stroop Error indices, and Grooved Pegboard (dominant hand) data (Chang et al., 2009a). This means that PI is a good predictor of neurobehavioral performance in welders without clinical manganism. In particular, PI was a better predictor of neurobehavioral performance than was blood Mn levels in such welders.

Taken together, the data suggest that PI on MRI may reflect a target organ dose of occupational Mn exposure (Kim, 2006). In addition, Mn in brain has a longer half-life than in blood (Lucchini and Kim, 2009). Thus, PI reflects the cumulative dose better than does blood Mn level. However, the level of signal intensity indicating progression to manganism from Mn exposure remains to be determined. The development of an animal model of manganism would assist in this regard, but the fact that the routes of exposure in humans differ, and that data on non-human species may not be transferable to human situations, are limiting factors. Hence, a prospective study correlating increases in T1 signal intensities with clinical and neuropsychological findings in Mn-exposed workers is needed.

## **3. PET/SPECT as an index of the integrity of the dopaminergic nigrostriatal pathway**

The dopaminergic nigrostriatal pathway is the primary focus of neurodegeneration in IPD (Brooks et al., 1990; Morrish et al., 1995, 1996). In IPD, dopamine uptake is reduced in the striatum, particularly the posterior putamen. This finding is in accord with the 40-60% loss of dopaminergic cells seen in the nigrostriatal pathway of patients with IPD. In non-human primates and humans intoxicated with Mn, [18F]-dopa (fluorodopa) PET scans are normal (Erikson et al., 1992; Kim et al., 1998; Shinotoh et al., 1995, 1997; Wolters et al., 1989). This supports the view that, in instances of Mn intoxication, the nigrostriatal pathway is relatively well preserved, consistent with many pathological observations showing that Mn-

Neuroimaging in Manganese-Induced Parkinsonism 135

Fig. 2. Differential diagnosis of manganism from parkinson's disease (Kim, 2006)

evaluation for the differential diagnosis of parkinsonism (Ravina et al., 2005).

**4. Recent developments in functional neuroimaging in Mn-induced** 

In vivo proton magnetic resonance spectroscopy ([1H]-MRS) is an image-guided, noninvasive method for monitoring of neurochemical metabolites in the brain (Rosen and Lenkinski, 2007). Currently, [1H]-MRS is the biomedical technique that is most commonly employed to obtain metabolic information to aid in the diagnosis of many neurological diseases, and also allows disease progression to be followed and response to treatment to be evaluated (Ross et al., 2006). Although MRS permits noninvasive, in vivo measurement of brain metabolites, only a few MRS investigations have been performed to date in efforts to assess the neurological effects of heavy metals in the environmental or occupational health. Recently, a few reports have analyzed the impact of lead exposure on brain metabolism in vivo in adults and children (Meng et al., 2005; Trope et al., 2001; Weisskopf, 2007; Weisskopf et al., 2007). However, little is known about the effects of chronic Mn exposure on brain metabolites in vivo. Two reports employed MRS to investigate the potential neurotoxic effects of chronic Mn exposure on the

**parkinsonism** 

**4.1 MRS** 

When a patient exhibits parkinsonian features, MRI is recommended. Observation of bilateral symmetrical increases in signal intensities, mainly confined to the globus pallidus on T1-weighted MRI, in a patient confirms recent CNS exposure to Mn. It should be noted that a negative MRI signal can occur when worker exposure to Mn ceased more than 6–12 months prior to testing. When a patient with a high T1 signal and an Mn exposure history also shows normal uptake by PET/SPECT, primary manganism should be highly suspected. If a patient who has a high T1 signal and an Mn exposure history also shows reduced uptake upon PET/SPECT, the patient should be categorized as suffering from IPD with coincidental Mn exposure. When a patient yielding a high T1 signal upon MRI does not have an Mn exposure history, but shows normal uptake upon PET/SPECT, he/she may be diagnosed with secondary manganism attributable to liver cirrhosis or TPN. If a patient without a high T1 signal shows reduced uptake on PET/SPECT, he/she could possibly have IPD. When a patient without a high T1 signal on MRI shows normal uptake on PET/SPECT, he/she would be under suspicion of a form of secondary parkinsonism other than manganism (Kim 2006). However, neuroimaging should be combined with clinical

induced damage occurs primarily in pathways postsynaptic to the nigrostriatal system. Dopamine transporter (DAT) imaging using (1r)-2b-carboxymethoxy-3b-(4 iodophenyl)tropane (β-CIT), employed as a SPECT ligand, reveals the density of DAT, and therefore explores the integrity of the nigrostriatal dopaminergic system. DAT is a protein located in the presynaptic nerve terminals of this system. β-CIT binds to DAT with high affinity and a low level of nonspecificity (Laruelle et al., 1993). In IPD, [123I]-β-CIT SPECT reveals that specific striatal β-CIT uptake is reduced (Seibyl et al., 1995). Earlier data showed that this method can distinguish IPD patients from normal controls (Jeon et al., 1998a, 1998b). Further, striatal DAT uptake is nearly normal in patients with Mn-induced parkinsonism, but is markedly reduced in IPD patients (Huang et al., 2003). Various ligands that bind to DAT, such as [123I]-β-CIT, [123I]-fluoropropyl-CIT, and [99mTc]-TRODAT-1 have been used in SPECT studies (Huang et al., 2003; Kim et al., 2002). DAT SPECT is more easily accessible and less expensive than is fluorodopa PET. Fluorodopa and DAT uptake values are (nearly) normal in patients with manganism (Huang et al., 2003; Kim et al., 1998; Shinotoh et al., 1997; Wolters et al., 1989), whereas uptake is markedly reduced in IPD patients. However, Guilarte et al. (2008) reported that, in the non-human primate brain, chronic Mn exposure inhibited dopaminergic transmission, leading to motor deficits, in the absence of changes to presynaptic dopaminergic nerve terminals.

Racette et al. (2005) found relatively symmetrical and severely reduced fluorodopa uptake on PET in the posterior putamen of a patient with manganism secondary to liver failure, together with T1 hyperintensities in the basal ganglia on MRI. This is the only reported case of secondary manganism accompanied by abnormal fluorodopa PET findings. However, SPECT data from our secondary manganism patients (Kim et al., 2010) revealed two different patterns of clinical and neuroradiological features. Four of five patients showed atypical parkinsonism, with normal DAT density, which could be clearly differentiated from PD, whereas one patient showed levodopa-responsive parkinsonism with reduced DAT density (classical PD). These findings are remarkably different from those of Racette et al. (2005). PET/SPECT findings in patients with manganism caused by liver failure should be further studied, with respect to both clinical and pathological features. Further, the pathogenesis, clinical characteristics, and neuroimaging might differ between patients with primary and secondary manganism. Liver cirrhosis might confound the symptoms and accelerate the signs of parkinsonism. It is unclear whether secondary manganism caused by liver cirrhosis, for example, differs (from a neuroimaging standpoint) from manganism associated with occupational or environmental exposure to Mn.

Some welders showed clinical features and PET/SPECT findings typical of IPD, with concurrent Mn exposure (Kim et al., 1999b, 2002; Racette et al., 2001). Initially, Kim et al. (1999b) reported that one welder showed IPD with incidental exposure to Mn. However, they subsequently developed the hypothesis that Mn might have been a risk factor for development of IPD, although they could not exclude the possibility that the patient simply suffered from IPD, with coincidental exposure to Mn (Kim et al., 2002). Racette et al. (2001) suggested that welding might be a possible risk factor for development of early-onset IPD. However, it remains unclear whether Mn causes or accelerates IPD. The link between Mn exposure and an increased risk of IPD should be further examined in clinical, pathological, and epidemiological studies focusing on PET/SPECT findings.

Neuroimaging modalities such as MRI and PET/SPECT may be useful for the differential diagnosis of parkinsonism (Calne et al., 1994; Kim, 2006) (Fig. 2).

induced damage occurs primarily in pathways postsynaptic to the nigrostriatal system. Dopamine transporter (DAT) imaging using (1r)-2b-carboxymethoxy-3b-(4 iodophenyl)tropane (β-CIT), employed as a SPECT ligand, reveals the density of DAT, and therefore explores the integrity of the nigrostriatal dopaminergic system. DAT is a protein located in the presynaptic nerve terminals of this system. β-CIT binds to DAT with high affinity and a low level of nonspecificity (Laruelle et al., 1993). In IPD, [123I]-β-CIT SPECT reveals that specific striatal β-CIT uptake is reduced (Seibyl et al., 1995). Earlier data showed that this method can distinguish IPD patients from normal controls (Jeon et al., 1998a, 1998b). Further, striatal DAT uptake is nearly normal in patients with Mn-induced parkinsonism, but is markedly reduced in IPD patients (Huang et al., 2003). Various ligands that bind to DAT, such as [123I]-β-CIT, [123I]-fluoropropyl-CIT, and [99mTc]-TRODAT-1 have been used in SPECT studies (Huang et al., 2003; Kim et al., 2002). DAT SPECT is more easily accessible and less expensive than is fluorodopa PET. Fluorodopa and DAT uptake values are (nearly) normal in patients with manganism (Huang et al., 2003; Kim et al., 1998; Shinotoh et al., 1997; Wolters et al., 1989), whereas uptake is markedly reduced in IPD patients. However, Guilarte et al. (2008) reported that, in the non-human primate brain, chronic Mn exposure inhibited dopaminergic transmission, leading to motor deficits, in the

Racette et al. (2005) found relatively symmetrical and severely reduced fluorodopa uptake on PET in the posterior putamen of a patient with manganism secondary to liver failure, together with T1 hyperintensities in the basal ganglia on MRI. This is the only reported case of secondary manganism accompanied by abnormal fluorodopa PET findings. However, SPECT data from our secondary manganism patients (Kim et al., 2010) revealed two different patterns of clinical and neuroradiological features. Four of five patients showed atypical parkinsonism, with normal DAT density, which could be clearly differentiated from PD, whereas one patient showed levodopa-responsive parkinsonism with reduced DAT density (classical PD). These findings are remarkably different from those of Racette et al. (2005). PET/SPECT findings in patients with manganism caused by liver failure should be further studied, with respect to both clinical and pathological features. Further, the pathogenesis, clinical characteristics, and neuroimaging might differ between patients with primary and secondary manganism. Liver cirrhosis might confound the symptoms and accelerate the signs of parkinsonism. It is unclear whether secondary manganism caused by liver cirrhosis, for example, differs (from a neuroimaging standpoint) from manganism

Some welders showed clinical features and PET/SPECT findings typical of IPD, with concurrent Mn exposure (Kim et al., 1999b, 2002; Racette et al., 2001). Initially, Kim et al. (1999b) reported that one welder showed IPD with incidental exposure to Mn. However, they subsequently developed the hypothesis that Mn might have been a risk factor for development of IPD, although they could not exclude the possibility that the patient simply suffered from IPD, with coincidental exposure to Mn (Kim et al., 2002). Racette et al. (2001) suggested that welding might be a possible risk factor for development of early-onset IPD. However, it remains unclear whether Mn causes or accelerates IPD. The link between Mn exposure and an increased risk of IPD should be further examined in clinical, pathological,

Neuroimaging modalities such as MRI and PET/SPECT may be useful for the differential

absence of changes to presynaptic dopaminergic nerve terminals.

associated with occupational or environmental exposure to Mn.

and epidemiological studies focusing on PET/SPECT findings.

diagnosis of parkinsonism (Calne et al., 1994; Kim, 2006) (Fig. 2).

Fig. 2. Differential diagnosis of manganism from parkinson's disease (Kim, 2006)

When a patient exhibits parkinsonian features, MRI is recommended. Observation of bilateral symmetrical increases in signal intensities, mainly confined to the globus pallidus on T1-weighted MRI, in a patient confirms recent CNS exposure to Mn. It should be noted that a negative MRI signal can occur when worker exposure to Mn ceased more than 6–12 months prior to testing. When a patient with a high T1 signal and an Mn exposure history also shows normal uptake by PET/SPECT, primary manganism should be highly suspected. If a patient who has a high T1 signal and an Mn exposure history also shows reduced uptake upon PET/SPECT, the patient should be categorized as suffering from IPD with coincidental Mn exposure. When a patient yielding a high T1 signal upon MRI does not have an Mn exposure history, but shows normal uptake upon PET/SPECT, he/she may be diagnosed with secondary manganism attributable to liver cirrhosis or TPN. If a patient without a high T1 signal shows reduced uptake on PET/SPECT, he/she could possibly have IPD. When a patient without a high T1 signal on MRI shows normal uptake on PET/SPECT, he/she would be under suspicion of a form of secondary parkinsonism other than manganism (Kim 2006). However, neuroimaging should be combined with clinical evaluation for the differential diagnosis of parkinsonism (Ravina et al., 2005).

### **4. Recent developments in functional neuroimaging in Mn-induced parkinsonism**

#### **4.1 MRS**

In vivo proton magnetic resonance spectroscopy ([1H]-MRS) is an image-guided, noninvasive method for monitoring of neurochemical metabolites in the brain (Rosen and Lenkinski, 2007). Currently, [1H]-MRS is the biomedical technique that is most commonly employed to obtain metabolic information to aid in the diagnosis of many neurological diseases, and also allows disease progression to be followed and response to treatment to be evaluated (Ross et al., 2006). Although MRS permits noninvasive, in vivo measurement of brain metabolites, only a few MRS investigations have been performed to date in efforts to assess the neurological effects of heavy metals in the environmental or occupational health. Recently, a few reports have analyzed the impact of lead exposure on brain metabolism in vivo in adults and children (Meng et al., 2005; Trope et al., 2001; Weisskopf, 2007; Weisskopf et al., 2007). However, little is known about the effects of chronic Mn exposure on brain metabolites in vivo. Two reports employed MRS to investigate the potential neurotoxic effects of chronic Mn exposure on the

Neuroimaging in Manganese-Induced Parkinsonism 137

NAA level, in the ACC and basal ganglia; these changes are considered to be typical metabolic abnormalities associated with HE (Geissler et al., 1997; Laubenberger et al., 1997; Weissenborn & Kolbe, 1998). In the early stages of HE, spectral alterations in mI and/or choline levels have been observed, but without corresponding increases in the Glx concentration (Kreis et al., 1992; Laubenberger et al., 1997; Miese et al., 2006; Naegele et al., 2000; Spahr et al., 2000). Compared with HE patients, welders did not show any abnormal change in Glx metabolism in a study by Chang et al. (2009b). The MRS results in welders are compatible with findings in patients in the early stages of HE. The cited study suggested that the depletion of mI in welders may reflect a possible glial cell effect rather than a neuronal effect, associated with long-term exposure to Mn. More recently, Dydak et al. (2011) used MRS to investigate brain metabolites in the globus pallidus, putamen, thalamus, and frontal cortex of 10 Mn-exposed smelters and 10 age- and gender-matched controls. Additionally, they used the MEGA-PRESS sequence to determine GABA levels in the thalamus. In addition to a significant decrease in the NAA/Cr ratio in the frontal cortex of exposed subjects, a significant increase in GABA level was observed in the thalamus, attributable to Mn exposure. The authors recommended that a combination of PI assessment and measurement of GABA level may provide a powerful, non-invasive biomarker of both Mn exposure and pre-symptomatic Mn neurotoxicity. Further studies using MRS are

The use of fMRI to study neurological diseases has become much more common over the last decade. However, employing fMRI to assess neurotoxicity in humans is a rather novel approach. Chang et al. (2010a) performed the first-ever fMRI experiment, using sequential finger-tapping, to investigate the behavioral significance of additionally recruited brain regions in welders who had experienced chronic Mn exposure. The study population consisted of 42 males, aged 40 years or older, who were current full-time welders, with more than 5 years of welding experience in a factory (Chang et al., 2010a). The control population consisted of 26 age- and gender-matched non-welding production workers from the same factory, who were not exposed to other hazardous materials such as paint. MRI examinations were performed using a 3.0 T whole-body scanner (Signa Excite HD), and blood oxygenation level-dependent (BOLD) contrast data were collected for each participant. T2\*-weighted echo planar imaging was used in fMRI acquisition. In the fingertapping test, each participant was asked to place the thumb on the tip of the index finger, middle finger, ring finger, little finger, ring finger, middle finger, index finger and another finger, in that order, as quickly and precisely as possible. In the cited study, the mean and standard deviation of blood Mn concentrations in welders and control individuals were 1.55 ± 0.45 and 1.15 ± 0.31 µg/dL, respectively. The mean workplace air Mn concentration was 0.14 mg/m3. The welders had an average welding experience of 20.5 years. All welders were shown to be devoid of clinical manganism by neurological examination. Performance on the Grooved Pegboard and finger-tapping tests (right and left hand) were significantly lower among welders than controls. Maximum frequencies, as determined by evaluation of hand pronation/supination, and finger-tapping test results using CAT SYS 2000 (Danish Product Development), were significantly lower among welders than controls. No difference in the results of other rhythmic tests (slow/fast), again using CAT SYS 2000, was evident between

needed to identify brain metabolites in Mn-exposed workers.

**4.2 fMRI** 

the groups.

brain (Guilarte et al., 2006; Kim et al., 2007). Guilarte et al. (2006) assessed the toxic effects of chronic Mn exposure on the levels of brain metabolites in non-human primates. This [1H]-MRS study found a decrease in the N-acetylaspartate/creatine (NAA/Cr) ratio in the parietal cortex and frontal WM at the end of the period of exposure to Mn, relative to baseline, indicating ongoing neuronal degeneration or dysfunction. NAA is known to serve as a neuronal marker (Birken and Oldendorf, 1989). A reduction in NAA levels in the brain can be interpreted as indicating neuronal dysfunction or even neuronal loss (Vion-Dury et al., 1994). Kim et al. (2007) investigated the potential neurotoxic effects of chronic Mn exposure in welders. Using point-resolved spectroscopy (PRESS) at 1.5 T, the cited authors measured the NAA/Cr, choline/creatine (Cho/Cr), and NAA/Cho ratios in the basal ganglia, and found no significant differences between welders and control subjects.

In a recent study, Chang et al., (2009b) sought to determine whether metabolic differences existed between 35 welders chronically exposed to Mn and 20 healthy age-matched control individuals, by measuring brain metabolites using [1H]-MRS. MRI and in vivo single-voxel MRS were performed using the GE 3T MRI system (Signa Excite HD, General Electric Medical Systems, Milwaukee, WI) equipped with an eight-channel RF head coil. The MRS spectra of individual metabolites were analyzed using a Linear Combination Model (Provencher, 1993) running a Linux system. Five brain metabolites—NAA; the Glx complex, including both glutamine (Gln) and glutamate (Glu); total creatine (tCr); total choline (tCho); and myoinositol (mI)—were measured in the anterior cingulate cortex (ACC) and parietal WM. Further, the cited authors investigated correlations between neurochemical changes in the ACC of the brain and neurobehavioral alterations, to assess possible associations between chronic Mn exposure and cognitive deficits (Chang et al., 2009b). The means and standard deviations of blood Mn concentration in welders and controls were found to be 1.53 ± 0.42 and 1.06 ± 0.29 µg/dL, respectively. The mean value of workplace Mn air concentrations was 0.15 mg/m3. The welders had worked for 21.3 ± 7.2 (mean ± SD) years. All welders were shown to be devoid of clinical manganism, by neurological examination. This study on welders using proton-MRS showed that the NAA/tCr, Glx/tCr, and tCho/tCr ratios in both the ACC and parietal WM did not differ significantly between welders and controls. However, the mI levels in the ACC, but not in the parietal WM, were significantly lower in welders compared with control individuals. Further, in the frontal lobe of the brain, the mI/tCr ratio was significantly correlated with verbal memory scores as well as blood Mn concentrations. Kim et al. (2007) found no statistically significant differences in the levels of brain metabolites (NAA and Cho only were measured) between welders and controls. However, although the cited authors used a PRESS sequence with a short echo time, mI levels was not analyzed, unlike in the study of Chang et al. cited above. The results of the latter work agree with those of a previous study (Kim et al., 2007), but a direct comparison of mI levels is not possible. Guilarte et al. (2006) reported a decrease in NAA level in the parietal cortex and frontal WM of the brains of Mn-exposed monkeys. However, when the spectroscopic findings of the work of Chang (2009b) and that of Guilarte et al. (2006) are compared, it is important to consider methodological differences between a human and animal study. MI is known to serve as a cerebral osmoregulator (Strange et al., 1994), and hence may play a role as an intracellular osmolyte. Thus, ml depletion may reflect glial cell swelling associated with long-term exposure to Mn. Previous [1H]-MRS studies on the brains of cirrhotic patients with overt hepatic encephalopathy (HE) often found a large increase in Glx concentration, and depletion of mI, but no change in NAA level, in the ACC and basal ganglia; these changes are considered to be typical metabolic abnormalities associated with HE (Geissler et al., 1997; Laubenberger et al., 1997; Weissenborn & Kolbe, 1998). In the early stages of HE, spectral alterations in mI and/or choline levels have been observed, but without corresponding increases in the Glx concentration (Kreis et al., 1992; Laubenberger et al., 1997; Miese et al., 2006; Naegele et al., 2000; Spahr et al., 2000). Compared with HE patients, welders did not show any abnormal change in Glx metabolism in a study by Chang et al. (2009b). The MRS results in welders are compatible with findings in patients in the early stages of HE. The cited study suggested that the depletion of mI in welders may reflect a possible glial cell effect rather than a neuronal effect, associated with long-term exposure to Mn. More recently, Dydak et al. (2011) used MRS to investigate brain metabolites in the globus pallidus, putamen, thalamus, and frontal cortex of 10 Mn-exposed smelters and 10 age- and gender-matched controls. Additionally, they used the MEGA-PRESS sequence to determine GABA levels in the thalamus. In addition to a significant decrease in the NAA/Cr ratio in the frontal cortex of exposed subjects, a significant increase in GABA level was observed in the thalamus, attributable to Mn exposure. The authors recommended that a combination of PI assessment and measurement of GABA level may provide a powerful, non-invasive biomarker of both Mn exposure and pre-symptomatic Mn neurotoxicity. Further studies using MRS are needed to identify brain metabolites in Mn-exposed workers.

#### **4.2 fMRI**

136 Diagnostics and Rehabilitation of Parkinson's Disease

brain (Guilarte et al., 2006; Kim et al., 2007). Guilarte et al. (2006) assessed the toxic effects of chronic Mn exposure on the levels of brain metabolites in non-human primates. This [1H]-MRS study found a decrease in the N-acetylaspartate/creatine (NAA/Cr) ratio in the parietal cortex and frontal WM at the end of the period of exposure to Mn, relative to baseline, indicating ongoing neuronal degeneration or dysfunction. NAA is known to serve as a neuronal marker (Birken and Oldendorf, 1989). A reduction in NAA levels in the brain can be interpreted as indicating neuronal dysfunction or even neuronal loss (Vion-Dury et al., 1994). Kim et al. (2007) investigated the potential neurotoxic effects of chronic Mn exposure in welders. Using point-resolved spectroscopy (PRESS) at 1.5 T, the cited authors measured the NAA/Cr, choline/creatine (Cho/Cr), and NAA/Cho ratios in the basal ganglia, and found no significant

In a recent study, Chang et al., (2009b) sought to determine whether metabolic differences existed between 35 welders chronically exposed to Mn and 20 healthy age-matched control individuals, by measuring brain metabolites using [1H]-MRS. MRI and in vivo single-voxel MRS were performed using the GE 3T MRI system (Signa Excite HD, General Electric Medical Systems, Milwaukee, WI) equipped with an eight-channel RF head coil. The MRS spectra of individual metabolites were analyzed using a Linear Combination Model (Provencher, 1993) running a Linux system. Five brain metabolites—NAA; the Glx complex, including both glutamine (Gln) and glutamate (Glu); total creatine (tCr); total choline (tCho); and myoinositol (mI)—were measured in the anterior cingulate cortex (ACC) and parietal WM. Further, the cited authors investigated correlations between neurochemical changes in the ACC of the brain and neurobehavioral alterations, to assess possible associations between chronic Mn exposure and cognitive deficits (Chang et al., 2009b). The means and standard deviations of blood Mn concentration in welders and controls were found to be 1.53 ± 0.42 and 1.06 ± 0.29 µg/dL, respectively. The mean value of workplace Mn air concentrations was 0.15 mg/m3. The welders had worked for 21.3 ± 7.2 (mean ± SD) years. All welders were shown to be devoid of clinical manganism, by neurological examination. This study on welders using proton-MRS showed that the NAA/tCr, Glx/tCr, and tCho/tCr ratios in both the ACC and parietal WM did not differ significantly between welders and controls. However, the mI levels in the ACC, but not in the parietal WM, were significantly lower in welders compared with control individuals. Further, in the frontal lobe of the brain, the mI/tCr ratio was significantly correlated with verbal memory scores as well as blood Mn concentrations. Kim et al. (2007) found no statistically significant differences in the levels of brain metabolites (NAA and Cho only were measured) between welders and controls. However, although the cited authors used a PRESS sequence with a short echo time, mI levels was not analyzed, unlike in the study of Chang et al. cited above. The results of the latter work agree with those of a previous study (Kim et al., 2007), but a direct comparison of mI levels is not possible. Guilarte et al. (2006) reported a decrease in NAA level in the parietal cortex and frontal WM of the brains of Mn-exposed monkeys. However, when the spectroscopic findings of the work of Chang (2009b) and that of Guilarte et al. (2006) are compared, it is important to consider methodological differences between a human and animal study. MI is known to serve as a cerebral osmoregulator (Strange et al., 1994), and hence may play a role as an intracellular osmolyte. Thus, ml depletion may reflect glial cell swelling associated with long-term exposure to Mn. Previous [1H]-MRS studies on the brains of cirrhotic patients with overt hepatic encephalopathy (HE) often found a large increase in Glx concentration, and depletion of mI, but no change in

differences between welders and control subjects.

The use of fMRI to study neurological diseases has become much more common over the last decade. However, employing fMRI to assess neurotoxicity in humans is a rather novel approach. Chang et al. (2010a) performed the first-ever fMRI experiment, using sequential finger-tapping, to investigate the behavioral significance of additionally recruited brain regions in welders who had experienced chronic Mn exposure. The study population consisted of 42 males, aged 40 years or older, who were current full-time welders, with more than 5 years of welding experience in a factory (Chang et al., 2010a). The control population consisted of 26 age- and gender-matched non-welding production workers from the same factory, who were not exposed to other hazardous materials such as paint. MRI examinations were performed using a 3.0 T whole-body scanner (Signa Excite HD), and blood oxygenation level-dependent (BOLD) contrast data were collected for each participant. T2\*-weighted echo planar imaging was used in fMRI acquisition. In the fingertapping test, each participant was asked to place the thumb on the tip of the index finger, middle finger, ring finger, little finger, ring finger, middle finger, index finger and another finger, in that order, as quickly and precisely as possible. In the cited study, the mean and standard deviation of blood Mn concentrations in welders and control individuals were 1.55 ± 0.45 and 1.15 ± 0.31 µg/dL, respectively. The mean workplace air Mn concentration was 0.14 mg/m3. The welders had an average welding experience of 20.5 years. All welders were shown to be devoid of clinical manganism by neurological examination. Performance on the Grooved Pegboard and finger-tapping tests (right and left hand) were significantly lower among welders than controls. Maximum frequencies, as determined by evaluation of hand pronation/supination, and finger-tapping test results using CAT SYS 2000 (Danish Product Development), were significantly lower among welders than controls. No difference in the results of other rhythmic tests (slow/fast), again using CAT SYS 2000, was evident between the groups.

Neuroimaging in Manganese-Induced Parkinsonism 139

manganism. The finding of excessive recruitment of the cortical motor network in chronically Mn-exposed group is in line with the emerging concept of use of adaptive neural mechanisms to compensate for latent dysfunction in the basal ganglia (Buhmann et al.,

Chang et al. (2010b) also performed fMRI, combined with two-back memory tests, to assess the neural correlates of Mn-induced memory impairment in response to subclinical dysfunction in the working memory networks of welders exposed to Mn for extended periods of time. The study population consisted of 23 males, aged 40 years or older, who were current full-time welders with more than 5 years of welding experience in a factory. The control population consisted of 21 age- and gender-matched non-welding production workers from the same factory, who were not exposed to other hazardous materials such as paint. The MRI equipment and the fMRI protocol were identical to those used in the report on fMRI data obtained using the finger-tapping task (this work is summarized above). The working memory paradigm consisted of a two-back memory task combined with a "rest" control task. Stimuli were projected onto a viewing screen, attached within the bore of the scanner, and viewed at a distance of approximately 20 cm from the eyes of the participant, after reflection from two mirrors positioned on top of the head coil. In the cited study, Mn exposure status was similar to that of subjects recruited for the fMRI study that employed the finger-tapping task. All welders were shown to be devoid of clinical manganism, by neurological examination. Welders showed significantly lower performance on cognitive neurobehavioral tests, including the Korean Auditory Verbal Learning Test (K-AVLT) (i.e., delayed recall and recognition), the Korean Complex Figure Test (K-CFT) (i.e., copy,

Fig. 4. The activations in fMRI with two-back memory tests from within group analysis in (a) controls and (b) welders (p < 0.05, FDR corrected for multiple comparison). Chang et al. (2010b)

2005).

Fig. 3. Statistical parametric maps (SPM) of sequential finger tapping movement with right hand for control (A) and welder group (B) displayed on 3D SPM template brain. All activation voxels are significant at P<0.00001 FDR corrected for multiple comparison across whole brain. Chang et al. (2010a)

During finger-tapping tasks conducted on welders who were chronically exposed to Mn, significant activation foci were noted in the bilateral primary sensorimotor cortex (SM1), the bilateral supplementary motor area (SMA), the bilateral dorsolateral premotor cortex (dPMC), the bilateral superior parietal cortex, and the bilateral dentate nucleus, when data from movement and rest periods were compared. In contrast, control participants exhibited significant activation of the contralateral (left) SM1 (Fig. 3). Activation of the bilateral SM1, bilateral SMA, bilateral dPMC, bilateral superior parietal cortex, and ipsilateral dentate nucleus was higher in the welding group than in the control group. No region showed significantly more activation in controls compared to welders. PI correlated with activation observed in the contralateral SM1, in terms of finger-tapping test data from the left hand. The fMRI variables correlated with motor behavior. Grooved Pegboard performance (right hand) correlated with activation, as seen also in ipsilateral and contralateral SMA data obtained during finger-tapping with the right hand. Left-hand finger-tapping data collected during the first 10 sec of the task significantly correlated with activation of the ipsilateral and contralateral SMA when finger-tapping was evaluated on the left side. Bilateral SM1 hyperactivity may reflect motor re-organization in the brains of Mn-exposed welders, which might compensate for existing subclinical motor deficits. It seems likely that the mechanisms regulating sensorimotor control (i.e., systems operative from the basal ganglial output to the cortical sensorimotor regions, via the thalamus) may compensate for abnormalities in the basal ganglia and thereby prevent the appearance of symptoms in presymptomatic welders. In addition, hyperactivity of the SMA suggests that it is more difficult for welders (compared to controls) to perform a simple sequential finger-tapping task; thus, more SMA activity is recruited via the basal ganglial-thalamo-cortical loop, which allows for successful performance of the sequential finger-tapping task. However, these findings do not agree with those reported for patients with IPD. Functional neuroimaging of participants performing tasks requiring motor selection and initiation showed that the SMA was hypoactivated in patients with IPD, compared to normal participants (Sabatini et al., 2000). In summary, the collective findings suggest that, when relatively simple tasks are set, fMRI may uncover evidence of compromised brain functioning in patients with subclinical

Fig. 3. Statistical parametric maps (SPM) of sequential finger tapping movement with right hand for control (A) and welder group (B) displayed on 3D SPM template brain. All activation voxels are significant at P<0.00001 FDR corrected for multiple comparison across

During finger-tapping tasks conducted on welders who were chronically exposed to Mn, significant activation foci were noted in the bilateral primary sensorimotor cortex (SM1), the bilateral supplementary motor area (SMA), the bilateral dorsolateral premotor cortex (dPMC), the bilateral superior parietal cortex, and the bilateral dentate nucleus, when data from movement and rest periods were compared. In contrast, control participants exhibited significant activation of the contralateral (left) SM1 (Fig. 3). Activation of the bilateral SM1, bilateral SMA, bilateral dPMC, bilateral superior parietal cortex, and ipsilateral dentate nucleus was higher in the welding group than in the control group. No region showed significantly more activation in controls compared to welders. PI correlated with activation observed in the contralateral SM1, in terms of finger-tapping test data from the left hand. The fMRI variables correlated with motor behavior. Grooved Pegboard performance (right hand) correlated with activation, as seen also in ipsilateral and contralateral SMA data obtained during finger-tapping with the right hand. Left-hand finger-tapping data collected during the first 10 sec of the task significantly correlated with activation of the ipsilateral and contralateral SMA when finger-tapping was evaluated on the left side. Bilateral SM1 hyperactivity may reflect motor re-organization in the brains of Mn-exposed welders, which might compensate for existing subclinical motor deficits. It seems likely that the mechanisms regulating sensorimotor control (i.e., systems operative from the basal ganglial output to the cortical sensorimotor regions, via the thalamus) may compensate for abnormalities in the basal ganglia and thereby prevent the appearance of symptoms in presymptomatic welders. In addition, hyperactivity of the SMA suggests that it is more difficult for welders (compared to controls) to perform a simple sequential finger-tapping task; thus, more SMA activity is recruited via the basal ganglial-thalamo-cortical loop, which allows for successful performance of the sequential finger-tapping task. However, these findings do not agree with those reported for patients with IPD. Functional neuroimaging of participants performing tasks requiring motor selection and initiation showed that the SMA was hypoactivated in patients with IPD, compared to normal participants (Sabatini et al., 2000). In summary, the collective findings suggest that, when relatively simple tasks are set, fMRI may uncover evidence of compromised brain functioning in patients with subclinical

whole brain. Chang et al. (2010a)

manganism. The finding of excessive recruitment of the cortical motor network in chronically Mn-exposed group is in line with the emerging concept of use of adaptive neural mechanisms to compensate for latent dysfunction in the basal ganglia (Buhmann et al., 2005).

Chang et al. (2010b) also performed fMRI, combined with two-back memory tests, to assess the neural correlates of Mn-induced memory impairment in response to subclinical dysfunction in the working memory networks of welders exposed to Mn for extended periods of time. The study population consisted of 23 males, aged 40 years or older, who were current full-time welders with more than 5 years of welding experience in a factory. The control population consisted of 21 age- and gender-matched non-welding production workers from the same factory, who were not exposed to other hazardous materials such as paint. The MRI equipment and the fMRI protocol were identical to those used in the report on fMRI data obtained using the finger-tapping task (this work is summarized above). The working memory paradigm consisted of a two-back memory task combined with a "rest" control task. Stimuli were projected onto a viewing screen, attached within the bore of the scanner, and viewed at a distance of approximately 20 cm from the eyes of the participant, after reflection from two mirrors positioned on top of the head coil. In the cited study, Mn exposure status was similar to that of subjects recruited for the fMRI study that employed the finger-tapping task. All welders were shown to be devoid of clinical manganism, by neurological examination. Welders showed significantly lower performance on cognitive neurobehavioral tests, including the Korean Auditory Verbal Learning Test (K-AVLT) (i.e., delayed recall and recognition), the Korean Complex Figure Test (K-CFT) (i.e., copy,

Fig. 4. The activations in fMRI with two-back memory tests from within group analysis in (a) controls and (b) welders (p < 0.05, FDR corrected for multiple comparison). Chang et al. (2010b)

Neuroimaging in Manganese-Induced Parkinsonism 141

The mean workplace Mn air concentration was 0.15 mg/m3. The welders had an average welding experience of 20.6 years. All welders were shown to be devoid of clinical manganism by neurological examination. Welders showed significantly lower performances in all of the digit symbol, digit span, Stroop, Grooved Pegboard, and finger-tapping tests, compared to controls. Further, the results of the digit symbol, digit span, and Stroop tests were significantly associated with PI and blood Mn level after controlling for age, educational level, smoking status, and alcohol consumption. In addition, relationships between dependent measures and PI were stronger than those seen when blood Mn was used as an independent variable. Direct comparisons between welders and controls using investigator-independent Statistical Parametric Mapping (SPM) voxel-wise analysis of DTI metrics revealed a reduction in FA in the genu, body, and splenium of the corpus callosum (CC), and the frontal WM, in Mn-exposed welders. PI showed a statistically significant correlation with FA in the genu (left), body, and splenium of the CC. Blood Mn levels showed statistically significant correlations with FA in the genu (left) and body of the CC, and in the frontal WM. Further, marked increases in RD, but negligible changes in AD, were evident in the genu, body, and splenium of the CC, and the frontal WM. PI was significantly correlated with RD in the body of the CC. However, the blood Mn level did not show a statistically significant correlation with RD. All of these findings suggested that microstructural changes in the CC and the frontal WM result from a compromised radial directionality of fibers in such areas, primarily caused by demyelination. As the digit span (forward) test is more likely to measure attention and immediate recall, and the digit span (backward) test more specifically explores working memory, the statistically significant positive correlation between FA and digit span performance score (forward) suggests that the reduced FA in the frontal WM is in part responsible for the impaired attention of welders. The Stroop word and color/word tests are often used to measure executive function. Therefore, correlations between FA in the frontal WM, and the Stroop word and color/word test scores, suggest that poor performance on executive functioning, as measured using the Stroop word test (information processing) and the color/word test

(executive function), are closely associated with lower FA values in the frontal WM.

Mn exposure needs to be established in further work.

**5. Conclusion** 

In summary, correlation of DTI matrices with motor and cognitive neurobehavioral performance indices suggested that the observed microstructural abnormalities were associated with subtle motor and cognitive differences between welders and controls. This was the first study to use DTI to examine Mn-exposed workers (Kim et al., 2011). However, the functional significance of reduced frontal WM integrity evident in welders with chronic

Neuroimaging is undergoing a shift from morphological to functional approaches as new technologies are gradually introduced. For morphological neuroimaging reflecting Mn exposure, PI on T1-weighted MRI data exploring target organ dosages of Mn reflects the cumulative Mn dose better than does assessment of blood Mn. For use in functional neuroimaging exploring Mn exposure, fluorodopa-PET/DAT SPECT serves as an index of the integrity of the dopaminergic nigrostriatal pathway, and is useful to differentiate between manganism and IPD. Recently, proton MRS has been used to identify brain metabolites in Mn-exposed workers. Chang et al. (1999b) suggested that subclinical neurologic effects attributable to long-term Mn exposure are associated with possible glial

immediate recall, and delayed recall), digit span tests (both forward and backward), and the Stroop tests, compared to controls. Chronic Mn exposure caused increased brain activity in working memory networks during the two-back verbal working memory task.

The cited authors observed activation of the inferior frontal cortex, the basal ganglia (including the putamen), and the bilateral cerebellum, as well as activation of the common memory-related network of frontal and parietal cortical areas including the premotor cortex, the middle frontal cortex, the inferior and superior frontal cortex, the inferior and superior parietal cortex, the precuneus, and the cuneus, in welders exposed to Mn (Fig. 4). Betweengroup analysis revealed increased brain activity in the left (contralateral) SM1, the right inferior parietal cortex, the anterior and posterior cingulated cortex, the bilateral inferior frontal cortex, and the basal ganglia of welders, compared to controls, during the memory task. No region was significantly more activated in controls compared to welders. After controlling for age and educational level, the percentage change in activation of the parietal cortex was associated with K-AVLT (i.e., delayed recall and recognition). The percentage change in activation of the inferior frontal cortex was significantly associated with scores on the Stroop color and error indices. The percentage change in activation of the ACC was significantly associated with K-AVLT (i.e., recognition) and digit span (i.e., forward) test results.

The basal ganglial-thalamo-cortical circuitry was originally viewed as almost exclusively involved in control of movement. However, these structures are now considered to be essential for non-motor function (DeLong & Wichmann, 2009). Considering that the basal ganglia are the brain regions that receive most Mn deposits, a speculative explanation of the higher basal ganglial activity in welders is that, if performance is to be matched to that of normal subjects, an increased recruitment of basal ganglial cells is required in welders to compensate for a diminished working memory capacity. Together, the fMRI findings indicate that welders might need to recruit more neural resources to the working memory network, to compensate for subtle working memory deficits and alterations in working memory processes, if they are to perform tasks at the same level as is possible by healthy control individuals.

#### **4.3 DTI**

DTI is a unique method used to characterize WM micro-integrity, and relies on the principle that water diffusion is highly anisotropic in brain WM structures (Beaulieu, 2002). Thus, DTI reveals the orientation of WM tracts in vivo, and yields indices of microstructural integrity by quantification of the directionality of water diffusion (Le Bihanetal, 2001; Moseley et al., 1990). Although a few previous studies have explored the neurotoxicity associated with exposure to heavy metals such as Hg (Kinoshita et al., 1999) and Mn (McKinney et al., 2004) using diffusion-weighted image (DWI), no report on DTI-detected alteration of microscopic integrity within the WM of subjects experiencing chronic Mn exposure has appeared. Kim et al. (2011) used DTI to investigate whether welders exposed to Mn exhibited differences in WM integrity, compared to control subjects. MRI examinations were performed using a 3.0 T whole body scanner (Signa Excite HD). Fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity (RD) were measured on a voxel-wise basis in 30 male welders exposed to Mn and in 19 age- and gender-matched control subjects (Kim et al., 2011). In the cited study, the means and standard deviations of blood Mn concentration in welders and control individuals were 1.52 ± 0.47 µg/dL and 1.17 ± 0.33 µg/dL, respectively.

immediate recall, and delayed recall), digit span tests (both forward and backward), and the Stroop tests, compared to controls. Chronic Mn exposure caused increased brain activity in

The cited authors observed activation of the inferior frontal cortex, the basal ganglia (including the putamen), and the bilateral cerebellum, as well as activation of the common memory-related network of frontal and parietal cortical areas including the premotor cortex, the middle frontal cortex, the inferior and superior frontal cortex, the inferior and superior parietal cortex, the precuneus, and the cuneus, in welders exposed to Mn (Fig. 4). Betweengroup analysis revealed increased brain activity in the left (contralateral) SM1, the right inferior parietal cortex, the anterior and posterior cingulated cortex, the bilateral inferior frontal cortex, and the basal ganglia of welders, compared to controls, during the memory task. No region was significantly more activated in controls compared to welders. After controlling for age and educational level, the percentage change in activation of the parietal cortex was associated with K-AVLT (i.e., delayed recall and recognition). The percentage change in activation of the inferior frontal cortex was significantly associated with scores on the Stroop color and error indices. The percentage change in activation of the ACC was significantly associated with K-AVLT (i.e., recognition) and digit span (i.e., forward) test

The basal ganglial-thalamo-cortical circuitry was originally viewed as almost exclusively involved in control of movement. However, these structures are now considered to be essential for non-motor function (DeLong & Wichmann, 2009). Considering that the basal ganglia are the brain regions that receive most Mn deposits, a speculative explanation of the higher basal ganglial activity in welders is that, if performance is to be matched to that of normal subjects, an increased recruitment of basal ganglial cells is required in welders to compensate for a diminished working memory capacity. Together, the fMRI findings indicate that welders might need to recruit more neural resources to the working memory network, to compensate for subtle working memory deficits and alterations in working memory processes, if they are to perform tasks at the same level as is possible by healthy

DTI is a unique method used to characterize WM micro-integrity, and relies on the principle that water diffusion is highly anisotropic in brain WM structures (Beaulieu, 2002). Thus, DTI reveals the orientation of WM tracts in vivo, and yields indices of microstructural integrity by quantification of the directionality of water diffusion (Le Bihanetal, 2001; Moseley et al., 1990). Although a few previous studies have explored the neurotoxicity associated with exposure to heavy metals such as Hg (Kinoshita et al., 1999) and Mn (McKinney et al., 2004) using diffusion-weighted image (DWI), no report on DTI-detected alteration of microscopic integrity within the WM of subjects experiencing chronic Mn exposure has appeared. Kim et al. (2011) used DTI to investigate whether welders exposed to Mn exhibited differences in WM integrity, compared to control subjects. MRI examinations were performed using a 3.0 T whole body scanner (Signa Excite HD). Fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity (RD) were measured on a voxel-wise basis in 30 male welders exposed to Mn and in 19 age- and gender-matched control subjects (Kim et al., 2011). In the cited study, the means and standard deviations of blood Mn concentration in welders and control individuals were 1.52 ± 0.47 µg/dL and 1.17 ± 0.33 µg/dL, respectively.

working memory networks during the two-back verbal working memory task.

results.

control individuals.

**4.3 DTI** 

The mean workplace Mn air concentration was 0.15 mg/m3. The welders had an average welding experience of 20.6 years. All welders were shown to be devoid of clinical manganism by neurological examination. Welders showed significantly lower performances in all of the digit symbol, digit span, Stroop, Grooved Pegboard, and finger-tapping tests, compared to controls. Further, the results of the digit symbol, digit span, and Stroop tests were significantly associated with PI and blood Mn level after controlling for age, educational level, smoking status, and alcohol consumption. In addition, relationships between dependent measures and PI were stronger than those seen when blood Mn was used as an independent variable. Direct comparisons between welders and controls using investigator-independent Statistical Parametric Mapping (SPM) voxel-wise analysis of DTI metrics revealed a reduction in FA in the genu, body, and splenium of the corpus callosum (CC), and the frontal WM, in Mn-exposed welders. PI showed a statistically significant correlation with FA in the genu (left), body, and splenium of the CC. Blood Mn levels showed statistically significant correlations with FA in the genu (left) and body of the CC, and in the frontal WM. Further, marked increases in RD, but negligible changes in AD, were evident in the genu, body, and splenium of the CC, and the frontal WM. PI was significantly correlated with RD in the body of the CC. However, the blood Mn level did not show a statistically significant correlation with RD. All of these findings suggested that microstructural changes in the CC and the frontal WM result from a compromised radial directionality of fibers in such areas, primarily caused by demyelination. As the digit span (forward) test is more likely to measure attention and immediate recall, and the digit span (backward) test more specifically explores working memory, the statistically significant positive correlation between FA and digit span performance score (forward) suggests that the reduced FA in the frontal WM is in part responsible for the impaired attention of welders. The Stroop word and color/word tests are often used to measure executive function. Therefore, correlations between FA in the frontal WM, and the Stroop word and color/word test scores, suggest that poor performance on executive functioning, as measured using the Stroop word test (information processing) and the color/word test (executive function), are closely associated with lower FA values in the frontal WM.

In summary, correlation of DTI matrices with motor and cognitive neurobehavioral performance indices suggested that the observed microstructural abnormalities were associated with subtle motor and cognitive differences between welders and controls. This was the first study to use DTI to examine Mn-exposed workers (Kim et al., 2011). However, the functional significance of reduced frontal WM integrity evident in welders with chronic Mn exposure needs to be established in further work.

#### **5. Conclusion**

Neuroimaging is undergoing a shift from morphological to functional approaches as new technologies are gradually introduced. For morphological neuroimaging reflecting Mn exposure, PI on T1-weighted MRI data exploring target organ dosages of Mn reflects the cumulative Mn dose better than does assessment of blood Mn. For use in functional neuroimaging exploring Mn exposure, fluorodopa-PET/DAT SPECT serves as an index of the integrity of the dopaminergic nigrostriatal pathway, and is useful to differentiate between manganism and IPD. Recently, proton MRS has been used to identify brain metabolites in Mn-exposed workers. Chang et al. (1999b) suggested that subclinical neurologic effects attributable to long-term Mn exposure are associated with possible glial

Neuroimaging in Manganese-Induced Parkinsonism 143

Chang, Y., Song, H-J., Lee, J-J., Seo, J.H., Kim, J-H., Lee, H.J., Kim, H.J., Ahn, J-H., Park, S-J.,

Chang, Y., Lee, J.J., Seo, J.H., Song, H.J., Kim, J.H., Bae, S.J., Ahn, J.H., Park, S.J., Jeong, K.S.,

Choi, Y., Park, J.K., Park, N.H., Shin, J.W., Yoo, C-I., Lee, C.R., Lee, H., Kim, H.K., Kim, S-R.,

DeLong, M. & Wichmann, T. (2009) Update on models of basal ganglia function and dysfunction. *Parkinsonism Relat Disord* Suppliment 3., pp. 237–240, ISSN 1353-8020 Dietz, M.C., Ihrig, A., Bader, M., Wradzillo, W. & Triebig, G. (1999) Effects on the nervous

Dydak, U., Jiang, Y.M., Long, L.L., Zhu, H.,Chen, J., Li, W.M., Edden, R.A., Hu, S., Fu, X.,

Erikson, H., Tedroff, J., Thuomas, K.A., Aquilonius, S.M., Hartvig, P., Fasth, K.J., Bjurling, P.,

Geissler, A., Lock, G., Frund, R., Held, P., Hollerbach, S., Andus, T., Schölmerich, J.,

Guilarte, T.R., McGlothan, J.L., Degaonkar, M., Chen, M.K., Barker, P.B., Syversen, T. &

*Perspect* Vol.119, No.2, (February 2011), pp. 219–224,, ISSN 0091-6765 Ejima, A., Imamur, T., Nakamura, S., Saito, H., Matsumoto, K. & Momono, S. (1992)

0161-813X

2010), pp. 809-815, ISSN 1351-0711

occupational and environmental health.

(February 1992), pp. 426, ISSN 0140-6736

1285, ISSN 1053-8119

ISSN 1341-9145

ISSN 0340-5761

358, ISSN 1096-6080

9139

Neurochemical changes in welders revealed by proton magnetic resonance spectroscopy. *Neurotoxicology* Vol.30, No.6, (November 2009), pp. 950-957, ISSN

Kwon, J.H., Jeong, K.S., Jung, D-K. & Kim, Y. (2010a) Neuroplastic changes within the brains of manganese-exposed welders: Recruiting additional neural resources for successful motor performance. *Occup Environ Med* Vol.67, No.12, (December

Kwon. Y.J., Kim, S.H. & Kim, Y. (2010b) Altered working memory process in the manganese-exposed brain. *Neuroimage* Vol.53, No.4, (December 2010), pp. 1279-

Jung, T-H., Park, J., Yoon, C.S. & Kim, Y. (2005) Whole blood and red blood cell manganese reflected signal intensities of T1-weighted MRI better than plasma manganese in liver cirrhotics. *J Occup Health* Vol.47, No.1, (January 2005), pp. 68–73,

system of manganese exposed workers in a dry cell battery factory. In: Proceedings of the seventh international symposium on neurobehavioral methods and effects in

Long, Z., Mo, X.A., Meier, D., Harezlak, J., Aschner, M., Murdoch, J.B. & Zheng, W. (2011) In vivo measurement of brain GABA concentrations by magnetic resonance spectroscopy in smelters occupationally exposed to manganese. *Environ Health* 

Manganese intoxication during total parental nutrition. *Lancet* Vol.339, No.8790,

Långström, B., Hedström, K.G. & Heilbronn, E. (1992) Manganese induced brain lesions in Macaca fasciculularis as revealed by positron emission tomography and magnetic resonance imaging. *Arch Toxicol* Vol.66, No.6. (June 1992), pp. 403–407,

Feuerbach, S. & Holstege, A. (1997) Cerebral abnormalities in patients with cirrhosis detected by proton magnetic resonance spectroscopy and magnetic resonance imaging. *Hepatology* Vol.25, No.1, (January 1997), pp. 48–54, ISSN 0270-

Schneider, J,S. (2006) Evidence for cortical dysfunction and widespread manganese accumulation in the nonhuman primate brain following chronic manganese exposure: a 1H-MRS and MRI study. *Toxicol Sci* Vol.94, No.2 (December 2006):351–

cell effect rather than neuronal deficits. The use of fMRI, combined with motor tasks, has suggested that cortical hyperactivity may reflect motor re-organization in the brains of Mnexposed welders, to compensate for subclinical motor deficits. When cognitive tasks are set, fMRI findings indicate that welders might need to recruit more neural resources to the working memory network to compensate for subtle subclinical working memory deficits. Therefore, fMRI is useful to detect subclinical cortical deficits in subjects who have experienced chronic exposure to Mn. DTI revealed microstructural deficits in WM integrity in welders exposed to Mn. Thus, functional neuroimaging can evaluate both subclinical WM integrity and cortical function in those exposed to Mn. Such neuroimaging combined with neurobehavioral performance evaluation shows promise in the elucidation of the pathophysiology of Mn neurotoxicity.

### **6. References**


cell effect rather than neuronal deficits. The use of fMRI, combined with motor tasks, has suggested that cortical hyperactivity may reflect motor re-organization in the brains of Mnexposed welders, to compensate for subclinical motor deficits. When cognitive tasks are set, fMRI findings indicate that welders might need to recruit more neural resources to the working memory network to compensate for subtle subclinical working memory deficits. Therefore, fMRI is useful to detect subclinical cortical deficits in subjects who have experienced chronic exposure to Mn. DTI revealed microstructural deficits in WM integrity in welders exposed to Mn. Thus, functional neuroimaging can evaluate both subclinical WM integrity and cortical function in those exposed to Mn. Such neuroimaging combined with neurobehavioral performance evaluation shows promise in the elucidation of the

Ahn, J., Yoo, C-I., Lee, C.R., Lee, J.H., Lee, H., Park, J.K., Sakai, T., Yoon, C.S. & Kim, Y.

*Neurotoxicology* Vol.24, No.6, (December 2003), pp. 835-838, ISSN 0161-813X Beaulieu, C. (2002). The basis of anisotropic diffusion imaging in the nervous system. *NMR Biomed* Vol.15, No.7-8, (November-December 2002), pp. 435–455. ISSN 0952-3480 Birken, D.L., & Oldendorf, W.H. (1989). N-Acetyl-L-aspartic acid: a literature review of a

Brooks, D.J., Salmon, E.P., Mathias, C.J., Quinn, N., Leenders, K.L., Bannister. R., Marsden,

Buhmann, C., Binkofski, F., Klein, C., Büchel, C., van Eimeren, T., Erdmann, C., Hedrich, K.,

Butterworth, R.F., Spahr, L., Fontaine, S. & Layrargues, G.P. (1995) Manganese toxicity,

Calne, D.B., Chu, N.S., Huang, C.C., Lu, C.S. & Olanow W. (1994) Manganism and

Chang, Y., Kim, Y., Woo, S-T., Song, H-J., Kim, S.H., Lee, H., Kwon, Y.J., Ahn, J.H., Park, S-J.,

Chang, Y., Woo, S-T., Lee, J-J., Song, H-J., Lee, H.J., Yoo, D-S., Kim, S.H., Lee, H., Kwon, Y.J.,

*Rev* Vol.13, No.1, (Spring 1989), pp. 23–31, ISSN 0149-7634

No.4, (December 1995), pp. 259–267, ISSN 0885-7490

(September 1994), pp. 1583–1586, ISSN 0028-3878

(2003). Calcification mimicking manganese-induced increased signal intensities in T1-weighted MR images in a patient taking herbal medicine: case report.

compound prominent in 1H-NMR spectroscopic studies of brain. *Neurosci Biobehav* 

C.D. & Frackowiakm R.S. (1990). The relationship between locomotor disability, autonomic dysfunction, and the integrity of the striatal dopaminergic system in patients with multiple system atrophy, pure autonomic failure and Parkinson's disease, studied with PET. *Brain* Vol.113, No.5, (October 1990), pp. 1539–1552, ISSN

Kasten, M., Hagenah, J., Deuschl, G., Pramstaller, P.P. & Siebner, H.R. (2005) Motor reorganization in asymptomatic carriers of a single mutant Parkin allele: a human model for presymptomatic parkinsonism. *Brain* Vol.128, No.10, (October 2005), pp.

dopaminergic dysfunction and hepatic encephalopathy. *Metab Brain Dis* Vol.10,

idiopathic parkinsonism: similarities and differences. *Neurology* Vol.44, No.9

Chung, I-S. & Jeong, K.S. (2009) High signal intensity on magnetic resonance imaging is a better predictor of neurobehavioral performances than blood manganese in asymptomatic welders. *Neurotoxicology* Vol.30, No.4 (July 2009), pp.

Ahn, H.J., Ahn, J.H., Park, S-J., Weon, Y.C., Chung, I-S., Jeong, K.S. & Kim, Y. (2009)

pathophysiology of Mn neurotoxicity.

**6. References** 

0006-8950

2281-2290, ISSN 0006-8950

555-563, ISSN 0161-813X

Neurochemical changes in welders revealed by proton magnetic resonance spectroscopy. *Neurotoxicology* Vol.30, No.6, (November 2009), pp. 950-957, ISSN 0161-813X


Neuroimaging in Manganese-Induced Parkinsonism 145

Kim, Y., Kim, J., Ito, K., Lim, H-S., Cheong, H-K., Kim, J.Y., Shin, Y.C., Kim, K.S. & Moon, Y

Kim, Y., Kim, J.M., Kim, J., Yoo, C., Lee, C.R., Lee, J.H., Kim, H.K., Yang, S.O., Park, J.,

Kim, Y., Park, J.K., Choi, Y., Yoo, C-I., Lee, C.R., Lee, H., Lee, J-H., Kim, S-R., Jung, T-H.,

Kinoshita, Y., Ohnishi, A., Kohshi, K. & Yokota, A. (1999) Apparent diffusion coefficient on

Kreis, R., Ross, B.D., Farrow, N.A. & Ackerman, Z. (1992) Metabolic disorders of the brain in

Krieger, D., Krieger, S., Jansen, O., Gass, P., Theilmann, L. & Lichtnecker, H. (1995)

Lang, C.J. (2000) The use of neuroimaging techniques for clinical detection of neurotoxicity:

Laubenberger, J., Haussinger, D., Bayer, S., Gufler, H., Hennig, J. & Langer, M. (1997) Proton

Laruelle, M., Baldwin, R.M., Malison, R.T., Zea-Ponce, Y., Zoghbi, S.S., al-Tikriti, M.,

Le Bihan, D., Mangin, J.F., Poupon, C., Clark, C.A., Pappata, S., Molko, N. & Chabriat, H.

Lucchini, R. & Kim, Y. (2009) Manganese, Health Effects. In Vojtisek, M and Prakash, R Eds. Metals and Neurotoxicity. Society for Science and Environment. 81-92, India. McKinney, A.M., Filice, R.W., Teksam, M., Casey, S., Truwit, C., Clark, H.B., Woon, C. &

*Mov Disord* Vol.17, No.3, (May 2002) pp. 568–575, ISSN 0885-3185

(January 2011), pp. 100-109, ISSN 0161-813X

Vol.182, No.1, (January 1992), pp. 19–27, ISSN 0033-8419

Vol.13, No.4, (April 1993), pp. 295–309, ISSN 0887-4476

No.4, (April 2001), pp. 534–546, ISSN 1090-7807

(May 1999) pp. 348–354, ISSN 0013-9351

1995) pp. 270–274, ISSN 0140-6736

1616 ISSN 0016-5085

249–252, ISSN 0161-813X

(1999b) Idiopathic parkinsonism with superimposed manganese exposure: utility of positron emission tomography. *Neurotoxicology* Vol.20, No.2-3, (April 1999), pp.

Chung, H.K., Lee, D.S. & Jeon, B. (2002) Dopamine transporter density is decreased in parkinsonian patients with a history of manganese exposure; what does it mean?

Yoon, C.S. & Park J-H. (2005) Blood manganese concentration is elevated in iron deficiency anemia patients, whereas globus pallidus signal intensity is minimally affected. *Neurotoxicology* Vol.26, No.1, (January 2005), pp. 107-111, ISSN 0161-813X Kim, Y., Jeong, K.S., Song, H-J., Lee, J-J., Seo, J-H., Kim, G-C., Lee, H.J., Kim, H.J., Ahn, J-H.,

Park, S-J., Kim, S.H., Kwon, Y.J. & Chang, Y. (2011) Altered white matter microstructural integrity revealed by voxel-wise analysis of diffusion tensor imaging in welders with manganese exposure. *Neurotoxicology* Vol.32, No.1,

rat brain and nerves intoxicated with methylmercury. *Environ Res* Vol.80, No.4,

chronic hepatic encephalopathy detected with H-1 MR spectroscopy. *Radiology*

Manganese and chronic hepatic encephalopathy. *Lancet* Vol.346, No.8970, (July

a review. *Neurotoxicology* Vol.21, No.5, (October 2000), pp. 847-855, ISSN 0161-813X

magnetic resonance spectroscopy of the brain in symptomatic and asymptomatic patients with liver cirrhosis. *Gastroenterology* Vol.112, No.5, (May 1997), pp. 1610–

Sybirska, E.H., Zimmermann, R.C., Wisniewski, G. & Neumeyer, J.L. (1993) SPECT imaging of dopamine and serotonin transporters with [123I]beta-CIT SPECT: pharmacological characterization of brain uptake in nonhuman primates. *Synapse*

(2001) Diffusion tensor imaging: concepts and applications. J Magn Reson Vol.13,

Liu, H.Y. (2004) Diffusion abnormalities of the globi pallidi in manganese neurotoxicity. *Neuroradiology* Vol.46, No.4 (April 2004), pp. 291–295, ISSN 0028-3940


Guilarte, T.R., Burton, N.C., McGlothan, J.L., Vernia, T., Zhou, Y., Alexander, M., Pham, L.,

*Neurochem* Vol.107, No.5, (December 2008), pp. 1236–1247, ISSN 0022-3042 Hauser, R.A., Zesiewicz, T.A., Rosemurgy, A.S., Martinez, C. & Olanow, C.W. (1994)

Hauser, R.A., Zesiewicz, T.A., Martinez, C., Rosemurgy, A.S. & Olanow, C.W. (1996) Blood

Huang, C.C., Weng, Y.H., Lu, C.S., Chu, N.S. & Yen, T.C. (2003) Dopamine transporter

Jeon, B., Jeong, J.M., Park, S.S., Kim, J.M., Chang, Y.S., Song, H.C., Kim, K.M., Yoon, K.Y.,

Jeon, B., Kim, J.M., Jeong, J.M., Kim, K.M., Chang, Y.S., Lee, D.S. & Lee. M.C. (1998b)

Jiang, Y.M., Zheng, W., Long, L.L, Zhao, W.J., Li, X.R., Mo, X.A., Lu, J., Fu, X., Li, W., Liu, S.,

Kim, E.A., Cheong, H.K., Choi, D.S., Sakong, J., Ryoo, J.W., Park, I. & Kang. D.M. (2007)

Kim, Y. (2006) Neuroimaging in manganism. *Neurotoxicology* Vol.27, No.3, (May 2006), pp.

Kim, J.M., Kim, J.S., Jeong, S.H., Kim, Y.K., Kim, S.E., Kim, S.H. & Kim, Y. (2010)

*Neurotoxicology*. Vol.31, No.4, (August 2010), pp. 351-355, ISSN 0161-813X Kim, J.W., Kim, Y., Cheong, H.K. & Ito, K. (1998) Manganese induced parkinsonism: a case report. *J Korean Med Sci* Vol.13, No.4, (August 1998), pp. 437–439, ISSN 1011-8934 Kim, Y., Kim, K.S., Yang, J.S., Park, I.J., Kim, E., Jin, Y., Kwon, K.R., Chang, K.H., Kim, J.W.,

*Psychiatry* Vol.65, No.1, (July 1998), pp. 60–64, ISSN 0022-3050

Vol.28, No.2, (March 2007), pp. 276–283, ISSN 0161-813X

(December 1994), pp. 871–875, ISSN 0364-5134

2003), pp. 1335–1339, ISSN 0340-5354

pp. 126–135, ISSN 0161-813X

369–372, ISSN 0161-813X

901–907, ISSN 0161-813X

1671

5134

Griswold, M., Wong, D.F., Syversen, T. & Schneider, J.S. (2008) Impairment of nigrostriatal dopamine neurotransmission by manganese is mediated by presynaptic mechanism(s): implications to manganese-induced parkinsonism. *J* 

Manganese intoxication and chronic liver failure. *Ann Neurol* Vol.36, No.6

manganese correlates with brain magnetic resonance imaging changes in patients with liver disease. *Can J Neurol Sci* Vol.23, No.2, (May 1996), pp. 95–98, ISSN 0317-

binding in chronic manganese intoxication. *J Neurol* Vol.250, No.11. (November

Lee, M.C. & Lee, S.B. (1998a) Dopamine transporter density measured by [123I]beta-CIT single-photon emission computed tomography is normal in doparesponsive dystonia. *Ann Neurol* Vol.43, No.6, (June 1998), pp. 792–800, ISSN 0364-

Dopamine transporter imaging with [123I]beta-CIT demonstrates presynaptic nigrostriatal dopaminergic damage in Wilson's disease. *J Neurol Neurosurg* 

Long, Q., Huang. J. & Pira, E. (2007) Brain magnetic resonance imaging and manganese concentrations in red blood cells of smelting workers: search for biomarkers of manganese exposure. *Neurotoxicology* Vol.28, No.1, (January 2007).

Effect of occupational manganese exposure on the central nervous system of welders: 1H magnetic resonance spectroscopy and MRI findings. *Neurotoxicology*

Dopaminergic neuronal integrity in parkinsonism associated with liver cirrhosis.

Park, S,H,, Lim,. H.S., Cheong, H.K., Shin, Y.C., Park, J. & Moon, Y. (1999a) Increase in signal intensities on T1-weighted magnetic resonance images in asymptomatic manganese-exposed workers. *Neurotoxicology* Vol.20, No.6, (December 1999), pp.


Neuroimaging in Manganese-Induced Parkinsonism 147

Ravina, B., Eidelberg, D., Ahlskog, J.E., Albin, R.L., Brooks, D.J., Carbon, M., Dhawan, V.,

Rosen, Y, & Lenkinski, R.E. (2007) Recent advances in magnetic resonance

Ross, A.J., Sachdev, P.S., Wen, W., Brodaty, H., Joscelyne, A. & Lorentz, L.M. (2006)

Sabatini, U., Boulanouar, K., Fabre, N., Martin, F., Carel, C., Colonnese, C., Bozzao, L., Berry,

Seibyl, J.P., Marek, K.L., Quinlan, D., Sheff, K., Zoghbi, S., Zea-Ponce, Y., Baldwin, R.M.,

Shinotoh, H., Snow, B.J., Hewitt, K.A., Pate, B.D., Doudet, D., Nugent, R., Perl, D.P.,

Spahr, L., Butterworth, R.F., Fontaine, S., Bui, L., Therrien, G., Milette, P.C., Lebrun, L.H.,

Spahr, L., Vingerhoets, F., Lazeyras, F., Delavelle, J., DuPasquier, R., Giostra, E., Mentha, G.,

Strange, K., Emma, F., Paredes, A. & Morrison R. (1994) Osmoregulatory changes in myo-

Trope, I., Lopez-Villegas, D., Cecil, K.M. & Lenkinski, R.E. (2001) Exposure to lead appears

No.2, (February 2000), pp. 394-403, ISSN 0006-8950

Vol.48, No.4, (April 1997), pp. 1053–1056, ISSN 0028-3878

Vol.12, No.1, (September 1994), pp. 35–43, ISSN 0894-1491

1995), pp. 589–598, ISSN 0364-5134

1996), pp. 1116–1120, ISSN 0270-9139

2001), pp.1437–1442, ISSN 0031-4005

pp. 208–215, ISSN 0028-3878

1933-7213

510X

5085

Feigin, A., Fahn, S., Guttman, M., Gwinn-Hardy, K., McFarland, H., Innis, R., Katz, R.G., Kieburtz, K., Kish, S.J., Lange, N., Langston, J.W., Marek, K., Morin, L., Moy, C., Murphy, D., Oertel, W.H., Oliver, G, Palesch, Y., Powers, W., Seibyl, J., Sethi, K.D., Shults, C.W., Sheehy, P., Stoessl, A.J. & Holloway, R. (2005) The role of radiotracer imaging in Parkinson's disease. *Neurology* Vol.64, No.2, (January 2005),

neurospectroscopy. *Neurotherapeutics* Vol.4, No.3, (July 2007), pp. 330–345, ISSN

Prediction of cognitive decline after stroke using proton magnetic resonance spectroscopy. *J Neurol Sci* Vol.251, No.1–2, (December 2006), pp. 62–69, ISSN 0022-

I., Montastruc, J.L., Chollet, F. & Rascol, O. (2000) Cortical motor reorganization in akinetic patients with Parkinson's disease: a functional MRI study. *Brain* Vol.123,

Fussell, B., Smith, E.O., Charney, D.S. & van Dyck, C. (1995) Decreased singlephoton emission computed tomographic [123I]beta-CIT striatal uptake correlates with symptom severity in Parkinson's disease. *Ann Neurol* Vol.38, No.4, (October

Olanow, W. & Calne, D.B. (1995) MRI and PET studies of manganese-intoxicated monkeys. *Neurology* Vol.45, No.6, (June 1995), pp. 1199–1204, ISSN 0028-3878 Shinotoh, H., Snow, B.J., Chu, N.S., Huang, C.C., Lu, C.S., Lee, C., Takahashi, H. & Calne,

DB. (1997) Presynaptic and postsynaptic striatal dopaminergic function in patients with manganese intoxication: a position emission tomography study. *Neurology*

Zayed, J., Leblanc, A. & Pomier-Layrargues, G. (1996) Increased blood manganese in cirrhotic patients: relationship to pallidal magnetic resonance signal hyperintensity and neurological symptoms. *Hepatology* Vol.24, No.5, (November

Terrier, F. & Hadengue, A. (2000) Magnetic resonance imaging and proton spectroscopic alterations correlate with parkinsonian signs in patients with cirrhosis. *Gastroenterology* Vol.119, No.3, (September 2000), pp. 774–781, ISSN 0016-

inositol content and Na+/myo-inositol cotransport in rat cortical astrocytes. *Glia*

to selectively alter metabolism of cortical gray matter. *Pediatrics* Vol.107, No.6, (June


Meng, X.M., Zhu, D.M., Ruan, D.Y., She, J.Q. & Luo, L. (2005) Effects of chronic lead

Miese, F., Kircheis, G., Wittsack, H.J., Wenserski, F., Hemker, J., Mödder, U., Häussinger, D.

Mirowitz, S.A., Westrich, T.J. & Hirsch, J.D. (1991) Hyperintense basal ganglia on T1-

Morrish, P.K., Sawle, G.V. & Brooks, D.J. (1995) Clinical and [18F]dopa PET findings in early

Morrish, P.K., Sawle, G.V. & Brooks, D.J. (1996) An [18F]dopa-PET and clinical study of the

Moseley, M.E., Cohen, Y. & Kucharczyk, J. (1990) Diffusion-weighted MR imaging of

Naegele, T., Grodd, W., Viebahn, R., Seeger, U., Klose, U., Seitz, D., Kaiser, S., Mader, I.,

Nelson, K., Golnick, J., Korn, T. & Angle, C. (1993) Manganese encephalopathy: utility of

Newland, M.C., Ceckler, T.L., Kordower, J.H. & Weiss, B. (1989) Visualizing manganese in

Park, N.H., Park, J.K., Choi, Y., Yoo, C-I., Lee, C.R., Lee, H., Kim, H.K., Kim, S-R., Jung, T.H.,

*Neurotoxicology* Vol.24, No.6, (December 2003), pp. 909–915, ISSN 0161-813X Pujol, A., Pujol, J., Graus, F., Rimola, A., Peri, J., Mercader, J.M., García-Pagan, J.C., Bosch, J.,

Racette, B.A., McGee-Minnich, L., Moerlein, S.M., Mink, J.W., Videen, T.O. & Perlmutter, J.S.

Vol.64, No.9, (May 2005), pp. 1644–1647, ISSN 0028-3878

No.1, (October, 1991), pp. 117–120, ISSN 0033-8419

(August 1990), pp.439–446, ISSN 0033-8419

No.3, (December 1989), pp. 251–258, ISSN 0014-4886

(January 1993), pp. 65–69, ISSN 0028-3878

2000), pp. 683–691, ISSN 0033-8419

1026, ISSN 0195-6108

591, ISSN 0006-8950

513, ISSN 0007-1072

pp. 597–600, ISSN 0364-5134

exposure on 1H MRS of hippocampus and frontal lobes in children. *Neurology*

& Cohnen, M. (2006) 1H-MR spectroscopy, magnetization transfer, and diffusionweighted imaging in alcoholic and nonalcoholic patients with cirrhosis with hepatic encephalopathy. *AJNR Am J Neuroradiol* Vol.27, No.5, (May 2006):1019–

weighted MR images in patients receiving parenteral nutrition. *Radiology* Vol.181,

Parkinson's disease. *J Neurol Neurosurg Psychiatry* Vol.59, No.6, (December 1995),

rate of progression in Parkinson's disease. *Brain* Vol.119, No.2 (April 1996), pp. 585–

anisotropic water diffusion in cat central nervous system. *Radiology* Vol.176, No.2,

Mayer, J., Lauchart, W., Gregor, M. & Voigt, K. (2000) MR imaging and (1)H spectroscopy of brain metabolites in hepatic encephalopathy: time-course of renormalization after liver transplantation. *Radiology* Vol.216, No.3, (September

early magnetic resonance imaging. *Br J Ind Med* Vol.50, No.6, (June 1993), pp. 510–

the primate basal ganglia with magnetic resonance imaging. *Exp Neurol* Vol.106,

Park, J., Yoon, C.S. & Kim, Y (2003) Whole blood manganese correlates with high signal intensities on T1-weighted MRI in patients with liver cirrhosis.

Rodés, J. & Tolosa, E. (1993) Hyperintense globus pallidus on T1-weighted MRI in cirrhotic patients is associated with severity of liver failure. *Neurology* Vol.43, No.1,

(2001) Welding-related parkinsonism: clinical features, treatment, and pathophysiology. *Neurology* Vol.56, No.1, (January 2001), pp. 8–13, ISSN 0028-3878 Racette, B.A., Antenor, J.A., McGee-Minnich, L., Moerlein, S.M., Videen, T.O., Kotagal, V. &

Perlmutter, J.S. (2005) [18F]FDOPA PET and clinical features in parkinsonism due to manganism. *Mov Disord* Vol.20, No.4, (April 2005), pp. 492–496, ISSN 0885-3185


**7** 

*Italy* 

**Minor and Trace Elements in Cerebrospinal** 

**Fluid of Parkinson's Patients – Suggestions** 

**After a Critical Review of the Analytical Data** 

Patients suffering from neurodegenerative diseases are known to present, in comparison to controls, variations on the contents of minor and trace elements in body tissues and fluids. For individuals affected by Parkinson's disease (PD), some findings, regarding brain and serum, are cited hereafter. BRAIN. Various brain areas were characterized for trace element levels and some alterations were observed in patients. Higher concentrations of aluminum were determined by Yasui et al. (1992) in different sites; increased levels of copper were detected by Riederer et al. (1989) in raphe plus reticular formation, whereas diminished amounts were found in substantia nigra by Rajput et al. (1985) and Dexter et al. (1989). Dexter et al. (1991) and Griffits et al. (1999) observed an iron enrichment in substantia nigra; regarding to zinc, Dexter et al. (1989) found more elevated amounts in a few areas, while Riederer et al. (1989) noticed lower contents in raphe formation. Variations of aluminum, copper, iron and zinc levels in definite brain sites of PD patients were reviewed by Speziali & Orvini (2003). SERUM. For PD patients, trace element changes were observed also in serum. Several studies were carried out at the Italian Istituto Superiore di Sanità (ISS) by Bocca et al. (2004, 2006), Forte et al. (2004, 2005), Alimonti et al. (2007a). A decreasing trend for aluminum was observed by Bocca et al. (2004) and Forte et al. (2004), as well as by Hedge et al. (2004) and Pande et al. (2005). Copper resulted elevated in these last two works and in a paper by Mindadse & Tschikowani (1967); in other investigations, by Bocca et al. (2006), Forte et al. (2004) and Tan et al. (2007), copper resulted diminished. Hedge et al. (2004), Pande et al. (2005), Forte et al. (2005), Alimonti et al. (2007a) detected lower iron concentrations, whereas Tan et al. (2007) reported a higher amount. In the case of zinc, a slight increase was noticed by Tan et al. (2007), whereas Hedge et al. (2004), Pande et al. (2005), Forte et al. (2005), Alimonti et al. (2007a) observed a significant

decrease. A lower mercury content was found by Gellein et al. (2008).

From this survey, it emerges that disagreeing findings for the same element are quite frequent. In the case of brain, we can suppose that the discrepancies among the various trials are related to the different areas examined. The less expected controversial findings for serum stimulated us to examine the up to date knowledge about the CSF of PD subjects. We have already published a short review on this topic (Speziali & Di Casa, 2009). In this

**1. Introduction**

Margherita Speziali1 and Michela Di Casa2 *1CNR-IENI (Institute for Energetics and Interphases), Department of Pavia, University of Pavia, Pavia, 2 Department of Chemistry, University of Pavia, Pavia,* 


## **Minor and Trace Elements in Cerebrospinal Fluid of Parkinson's Patients – Suggestions After a Critical Review of the Analytical Data**

Margherita Speziali1 and Michela Di Casa2

*1CNR-IENI (Institute for Energetics and Interphases), Department of Pavia, University of Pavia, Pavia, 2 Department of Chemistry, University of Pavia, Pavia, Italy* 

## **1. Introduction**

148 Diagnostics and Rehabilitation of Parkinson's Disease

Vion-Dury, J., Meyerhoff, D.J., Cozzone, P.J. & Weiner MW. (1994) What might be the

Walker, R..C, Purnell, G.L., Jones-Jackson, L.B., Thomas, K.L., Brito, J.A. & Ferris EJ. (2004)

*Neurotoxicology* Vol.25, No.4, (June 2004), pp. 533-542, ISSN 0161-813X Weissenborn, K. & Kolbe, H. (1998) The basal ganglia and portal-systemic encephalopathy. *Metab Brain Dis* Vol.13, No.4, (December 1998), pp. 261–272, ISSN 0885-7490 Weisskopf, M.G. Magnetic resonance spectroscopy and environmental toxicant exposure. (2007) *Ann N Y Acad Sci* Vol.1097, (February 2007), pp. 179–182, ISSN 0077-8923 Weisskopf, M.G., Hu, H., Sparrow, D., Lenkinski, R.E. & Wright, R.O. (2007) Proton

(April 2007), pp. 519–523, ISSN 0091-6765

0340-5354

5134

impact on neurology of the analysis of brain metabolism by in vivo magnetic resonance spectroscopy? *J Neurol* Vol.241, No.6, (May 1994), pp. 354–371, ISSN

Introduction to PET imaging with emphasis on biomedical research.

magnetic resonance spectroscopic evidence of glial effects of cumulative lead exposure in the adult human hippocampus. *Environ Health Perspect* Vol.115, No.4,

T.J., Li, D. & Calne, D.B. (1989) Positron emission tomography in manganese intoxication. *Ann Neurol* Vol.26, No.5, (November 1989), pp. 647–651, ISSN 0364-

Wolters, E.C., Huang, C.C., Clark, C., Peppard, R.F., Okada, J., Chu, N.S., Adam, M.J., Ruth,

Patients suffering from neurodegenerative diseases are known to present, in comparison to controls, variations on the contents of minor and trace elements in body tissues and fluids. For individuals affected by Parkinson's disease (PD), some findings, regarding brain and serum, are cited hereafter. BRAIN. Various brain areas were characterized for trace element levels and some alterations were observed in patients. Higher concentrations of aluminum were determined by Yasui et al. (1992) in different sites; increased levels of copper were detected by Riederer et al. (1989) in raphe plus reticular formation, whereas diminished amounts were found in substantia nigra by Rajput et al. (1985) and Dexter et al. (1989). Dexter et al. (1991) and Griffits et al. (1999) observed an iron enrichment in substantia nigra; regarding to zinc, Dexter et al. (1989) found more elevated amounts in a few areas, while Riederer et al. (1989) noticed lower contents in raphe formation. Variations of aluminum, copper, iron and zinc levels in definite brain sites of PD patients were reviewed by Speziali & Orvini (2003). SERUM. For PD patients, trace element changes were observed also in serum. Several studies were carried out at the Italian Istituto Superiore di Sanità (ISS) by Bocca et al. (2004, 2006), Forte et al. (2004, 2005), Alimonti et al. (2007a). A decreasing trend for aluminum was observed by Bocca et al. (2004) and Forte et al. (2004), as well as by Hedge et al. (2004) and Pande et al. (2005). Copper resulted elevated in these last two works and in a paper by Mindadse & Tschikowani (1967); in other investigations, by Bocca et al. (2006), Forte et al. (2004) and Tan et al. (2007), copper resulted diminished. Hedge et al. (2004), Pande et al. (2005), Forte et al. (2005), Alimonti et al. (2007a) detected lower iron concentrations, whereas Tan et al. (2007) reported a higher amount. In the case of zinc, a slight increase was noticed by Tan et al. (2007), whereas Hedge et al. (2004), Pande et al. (2005), Forte et al. (2005), Alimonti et al. (2007a) observed a significant decrease. A lower mercury content was found by Gellein et al. (2008).

From this survey, it emerges that disagreeing findings for the same element are quite frequent. In the case of brain, we can suppose that the discrepancies among the various trials are related to the different areas examined. The less expected controversial findings for serum stimulated us to examine the up to date knowledge about the CSF of PD subjects. We have already published a short review on this topic (Speziali & Di Casa, 2009). In this

Minor and Trace Elements in Cerebrospinal

24.696 27.911

21.229 20.913

21.868

23.693

median: 22.665

median: 22.365

**Subjects Mean SD N. of subjects** 

153.6 75.1 141.5\* 183.6

> 150 170

72.7 29.4

> 37 47 50

30.2 14.6

5.03

14.6

75.9 181 28.3 266.4

> 210 170

73.3 33.0

237 345 397

45.0 28.2

35.5 median: 36.8 28.2 median: 24.4

Table 3. Iron in Controls and Patients (µg/L)

these two values are incoherent

**ment Subjects Mean SD N. of subjects** 

26.956 median: 25.579 27.312 median: 27.301 1.997 4.964

> 5.515 3.385

3.160 2.024

3.509 4.263

Table 2. Calcium and magnesium in Controls and Patients (mg/L)

22 (20 M + 2 F) 11 (10 M + 1 F) 6 5

37 (16 M + 21 F) 37 (14 M + 23 F)

13 (6 M + 7 F) 26 (24 M + 2 F)

21 (13 M + 8 F) 17 (10 M + 7 F) 19 (13 M + 6 F)

18 (10 M + 8 F) 91 (64 M + 27 F)

20 (17 M + 3 F)

42 (36 M + 6 F)

**Ele-**

Ca <sup>C</sup> PD

 Mg <sup>C</sup> PD

C PD tot PD untreated PD treated

C PD

C PD

C PD (On) PD (On/Off)

C PD

C PD C PD

C PD

Fluid of Parkinson's Patients – Suggestions After a Critical Review of the Analytical Data 151

13 (6 M + 7 F) 26 (24 M + 2 F)

20 (17 M + 3 F) 42 (36 M + 6 F)

13 (6 M + 7 F) 26 (24 M + 2 F)

20 (17 M + 3 F) 42 (36 M + 6 F)

**(Gender) Age,** *y* **Technique Signif-**

ICP-AES

ICP-AES

Et-AAS

Et-AAS

ICP-AES

Et-AAS

ICP-AES

ICP-AES

63.8 13.7 64.9 10.8

66.2 14.7 64.5 10.7

63.8 13.7 64.9 10.8

66.2 14.7 64.5 10.7

**(Gender) Age,** *y* **(Range) Technique Signif-**

age-matched 64.9 (49-78) 63.1 (49-78) 67 (59-77)

62.4 17.8 65.7 8.8

63.8 13.7 64.9 10.8

63.8 13.8 65.5 9.7

66.2 14.7

64.5 10.7

**icance References** 

2004

NS Alimonti et al., 2007b

2004

NS Alimonti et al., 2007b

**icance References** 

1992

al., 1998

Gazzaniga et al.,

Jiménez-Jiménez et

Forte et al., 2004

Qureshi et al., 2005 and 2006

Bocca et al., 2006

Alimonti et al., 2007b

NS NS NS

NS

NS

S S

S

S

ICP-AES NS Forte et al.,

ICP-AES NS Forte et al.,

Chapter we present in a series of tables, for the first time, all the original values retrieved, along with several parameters that can influence the results. Here we discuss more extensively the role of all the factors affecting the results, which are the parameters reported in the tables along with the criteria for the enrollment of subjects, the analytical procedures and the statistical tests used. Finally, we propose with wider completeness several suggestions useful for possible future studies.


Table 1. Captions for the tables

## **2. Aim**

We performed an investigation on the minor and trace element amounts, available in the literature, regarding the CSF of PD patients and paired controls. Our purpose was to obtain a comprehensive picture of the element concentrations and to verify possible imbalances in the CSF of the diseased individuals.

## **3. Data presentation**

We considered only studies where: a) both patients and controls were examined in the same investigation; b) the concentration values determined were reported as numbers; c) statistical tests were employed to verify the significance of potential changes of element amounts in the CSF of patients. The scientific publications were retrieved through the data bank Medline along with the Personal Alert Service of Thomson Reuters, Philadelphia, PA. From the bibliographies of the recruited papers further references were derived. The concentration data we recruited in the literature were published from 1987 to 2008. Values of Al, Ba, Be, Bi, Ca, Cd, Co, Cr, Cu, Fe, Hg, Li, Mg, Mn, Mo, Ni, Pb, Sb, Se, Si, Sn, Sr, Tl, V, W, Zn and Zr were found. In Tab. 1 we set out the captions useful for all our tables. In Tab. 2 - 11 we report the mean concentration values, along with the standard deviations; we also show several parameters affecting the results: number, gender and age of the subjects enrolled, analytical technique employed, significance of possible differences between concentration values for patients and controls. The simultaneous availability of all these factors allows scientists to evaluate immediately the reliability of each trial findings. From the tables, indications can be also deducted on the possibility (or not) to compare directly the results of different trials; increasing or decreasing element trends in patients are evident right away. Finally, in Tab. 12 we sum up some indications recorded in other publications of interest.

Chapter we present in a series of tables, for the first time, all the original values retrieved, along with several parameters that can influence the results. Here we discuss more extensively the role of all the factors affecting the results, which are the parameters reported in the tables along with the criteria for the enrollment of subjects, the analytical procedures and the statistical tests used. Finally, we propose with wider completeness several

> Et-AAS = Electrothermal Atomic Absorption Spectrometry ICP-AES = Inductively Coupled Plasma - Atomic Emission Spectometry DCP-AES = Direct Current Plasma -AES SF-ICP-MS = Sector Field – Inductively Coupled Plasma – Mass

Spectrometry

difference

We performed an investigation on the minor and trace element amounts, available in the literature, regarding the CSF of PD patients and paired controls. Our purpose was to obtain a comprehensive picture of the element concentrations and to verify possible imbalances in

We considered only studies where: a) both patients and controls were examined in the same investigation; b) the concentration values determined were reported as numbers; c) statistical tests were employed to verify the significance of potential changes of element amounts in the CSF of patients. The scientific publications were retrieved through the data bank Medline along with the Personal Alert Service of Thomson Reuters, Philadelphia, PA. From the bibliographies of the recruited papers further references were derived. The concentration data we recruited in the literature were published from 1987 to 2008. Values of Al, Ba, Be, Bi, Ca, Cd, Co, Cr, Cu, Fe, Hg, Li, Mg, Mn, Mo, Ni, Pb, Sb, Se, Si, Sn, Sr, Tl, V, W, Zn and Zr were found. In Tab. 1 we set out the captions useful for all our tables. In Tab. 2 - 11 we report the mean concentration values, along with the standard deviations; we also show several parameters affecting the results: number, gender and age of the subjects enrolled, analytical technique employed, significance of possible differences between concentration values for patients and controls. The simultaneous availability of all these factors allows scientists to evaluate immediately the reliability of each trial findings. From the tables, indications can be also deducted on the possibility (or not) to compare directly the results of different trials; increasing or decreasing element trends in patients are evident right away. Finally, in Tab. 12 we sum up some indications recorded in other publications of interest.

S = significant or highly significant

NS = non-significant difference

suggestions useful for possible future studies.

PD (On) = PD with positive response to the

PD (On/Off) = PD without positive response

PDCN = PD cognitively normal patients

PD = Parkinson's disease patients

to the therapy

PDD = PD demented patients SD = Standard Deviation

Table 1. Captions for the tables

the CSF of the diseased individuals.

**3. Data presentation** 

C = Controls

M = male F = female

**2. Aim**

therapy


Table 2. Calcium and magnesium in Controls and Patients (mg/L)


these two values are incoherent

Table 3. Iron in Controls and Patients (µg/L)

Minor and Trace Elements in Cerebrospinal

19.2 (range: 10 – 33) 18.7 (range: 7 – 30)

64.9 67.7 63.2 67.0

109.1 104.9

22.5 23.7

132 119 109

19.6 17.4 10.0

21.9 median: 2.2 19.4 median: 17.0

C PD

C PD tot PD untreated PD treated

C PD

C PD

C PD (On) PD (On/Off)

C PD

C PDCN PDD

**4.1.2 Age** 

**Subjects Mean SD N. of subjects** 

5.8 6.3

14.4 19.9 11.5 18.5

88.2 86.3

4.76 10.5

17 18 19

4.77 7.97

1.3 4.3 1.1

and treated (with levodopa) patients was lower than in controls.

Table 4. Copper in Controls and Patients (µg/L)

Fluid of Parkinson's Patients – Suggestions After a Critical Review of the Analytical Data 153

**Age,** *y* 

age-matched 64.9 (49 – 78) 63.1 (49 – 78) 67 (59 – 77)

62.4 17.8 65.7 8.8

63.8 13.7 64.9 10.8

66.2 14.7 64.5 10.7

85.2 1.0 85.3 1.4 78.6 2.2

**(Range) Technique Signif-**

ICP-AES

DCP-AES NS Belliveau et al.,

NS NS

NS

ICP-AES NS Alimonti et al.,

NS NS

Et-AAS NS Jiménez-Jiménez

Et-AAS NS

Et-AAS NS

**icance References** 

1990

1992

et al., 1998

NS Forte et al., 2004

2007b

Qureshi et al., 2005 and 2006

Sparks et al., 2008

Gazzaniga et al.,

**(Gender)** 

22 (20 M + 2 F) 11 (10 M + 1 F)

37 (16 M + 21 F) 37 (14 M + 23 F)

13 (6 M + 7 F) 26 (24 M + 2 F)

21 (13 M + 8 F) 17 (10 M + 7 F) 19 (13 M + 6 F)

20 (17 M + 3 F) 42 (36 M + 6 F)

many authors, as Gazzaniga et al. (1992), Qureshi et al. (2005, 2006), Aguilar et al. (1998), Jimenez-Jimenez et al. (1998), Bocca et al. (2004, 2006), Forte et al. (2004), Alimonti et al. (2007b) along with Campanella et al. (1973) and Takahashi et al. (1994). Aguilar et al. (1998) carried out studies about the influence of antiparkinsonian treatment with various drugs on Se and Cr levels; in the entire group of PD patients, Se showed a non significant increase compared to controls, but the elevation attained the significance when only patients not treated with levodopa were considered. This interesting observation is just recorded in the article text. Jimenez-Jimenez et al. (1998) studied the effects of the same drugs on the concentrations of Fe, Cu, Zn and Mn; they did not observe any significant influence. Gazzaniga et al. (1992), confronting long-term levodopa treated and untreated patients, did not find any significant differences in the amounts of Cu, Fe and Mn. Qureshi et al. (2005, 2006) determined the amounts of Cu, Fe, Se and Zn in patients treated with levodopa, who were divided into two groups (PD On and PD On/Off), depending on the positive or negative response to the therapy. Fe and Se were found to be markedly higher than in controls in both kinds of patients; Zn resulted instead significantly reduced in both groups. Bocca et al. (2006) evaluated that the type of therapy did not influence the concentrations of all the elements studied (Al, Ca, Cu, Fe, Mg, Mn, Si, Zn). Alimonti et al. (2007b) observed that diverse drugs did not affect the concentration of Fe; on the contrary, they influenced the amounts of the elements which resulted significantly different between controls and patients (Co, Cr, Pb, Si, Sn). Takahashi et al. (1994 - see Tab. 12) found that the Mg concentration in both untreated

The subject age is known to influence the amounts of some elements in tissues and fluids. In serum, it has been documented for Cu by Ghayour-Mobarhan et al. (2005) and Kouremenou-Dona et al. (2006); for Se by Lopes et al. (2004). In brain, Markesbery et al. (1984), Ongkana et al. (2010) and Tohno et al. (2010) found age-related changes for several

32 12 5

21 16

6 5

## **4. Considerations on factors influencing the results**

As already pointed out in the **Introduction**, many factors affect the results, conditioning then the comparability among the findings of the various studies. We describe now in details the influence of each factor as it emerged in the examined publications.

## **4.1 Criteria for subject enrollment**

## **4.1.1 Health conditions**

In the subject enrollment, the criteria for inclusion/exclusion are fundamental and should absolutely include health conditions along with age and gender. Ideally, exposures due to environmental pollution or occupational activities and diet should also be considered.

In the reviewed papers, only a few teams give full details of the selection criteria used. The researchers of the Italian Istituto Superiore di Sanità (ISS), Bocca, Forte, Alimonti and coworkers, along with the Spanish scientists (Jiménez-Jiménez et al. 1998 and Aguilar et al. 1998) are among the few authors who describe extensively the criteria employed. The diseases affecting the individuals recruited as patients or controls are often not precisely described, mainly in the less recent works. In some publications, incongruities appear within the text or among the text and the data presented in the tables. It is not always clear whether the patients were affected by comorbidities. Due to the fact that severe illnesses of heart, liver, kidney and also tumors are known to affect the levels of trace elements in human fluids and tissues, an exhaustive health report is also necessary in the case of controls. For these subjects, very heterogeneous situations have been observed. In Bourrier-Guerin et al. (1985) - see Tab. 12 - controls were not enrolled at all. Mindadse & Tschikowani (1967) - see Tab. 12 - employed blood donors. In Qureshi et al. (2005, 2006) control individuals, simply defined "healthy", were affected by tension headache, ischemic cerebrovascular disease or polyneuropathy. The Spanish group selected "healthy" subjects with suspected subarachnoid hemorrhage or pseudotumor cerebri, oculomotor palsies, etc. The scientists of the ISS enrolled individuals not suffering from any central neurological disease. Kjellin (1967a and 1967b - see Tab. 12) assumed psychoneurotic outpatients as representative of the "normal" condition. It is evident that, in the diverse investigations, were enrolled as controls subjects in really different health conditions. It is worth considering that, differently from blood, the samples of CSF are not easily available; therefore, the control specimens are mostly withdrawn from subjects undergoing lumbar puncture for clinical analyses. In the case of patients, the differences among the groups enrolled in the diverse studies are amplified by some clinical variables, as duration and severity of the disease along with medical treatments and possible comorbidities. Regarding duration and severity of the disease, in the trial by Aguilar et al. (1998) Se and Cr levels showed no correlation with age at onset and duration of the illness. On the other hand, Alimonti et al. (2007b) detected in patients a negative association of Cr amount and severity and duration of the illness; in the same study, Pb appeared to be negatively related to the severity of the disorder, while Sn resulted to be negatively associated with the duration of the disease. The authors also found that age at onset did not affect the concentration of Fe and of the other elements that resulted significantly different between controls and patients (Co, Cr, Pb, Si, Sn). Bocca et al. (2006) observed that duration and severity of the disease appeared not to be correlated with Al, Ca, Cu, Fe, Mn, Si and Zn amounts; on the other hand, Mg level decreased with the duration and severity of the illness. Concerning medical treatments, the therapies followed by patients are described by


Table 4. Copper in Controls and Patients (µg/L)

many authors, as Gazzaniga et al. (1992), Qureshi et al. (2005, 2006), Aguilar et al. (1998), Jimenez-Jimenez et al. (1998), Bocca et al. (2004, 2006), Forte et al. (2004), Alimonti et al. (2007b) along with Campanella et al. (1973) and Takahashi et al. (1994). Aguilar et al. (1998) carried out studies about the influence of antiparkinsonian treatment with various drugs on Se and Cr levels; in the entire group of PD patients, Se showed a non significant increase compared to controls, but the elevation attained the significance when only patients not treated with levodopa were considered. This interesting observation is just recorded in the article text. Jimenez-Jimenez et al. (1998) studied the effects of the same drugs on the concentrations of Fe, Cu, Zn and Mn; they did not observe any significant influence. Gazzaniga et al. (1992), confronting long-term levodopa treated and untreated patients, did not find any significant differences in the amounts of Cu, Fe and Mn. Qureshi et al. (2005, 2006) determined the amounts of Cu, Fe, Se and Zn in patients treated with levodopa, who were divided into two groups (PD On and PD On/Off), depending on the positive or negative response to the therapy. Fe and Se were found to be markedly higher than in controls in both kinds of patients; Zn resulted instead significantly reduced in both groups. Bocca et al. (2006) evaluated that the type of therapy did not influence the concentrations of all the elements studied (Al, Ca, Cu, Fe, Mg, Mn, Si, Zn). Alimonti et al. (2007b) observed that diverse drugs did not affect the concentration of Fe; on the contrary, they influenced the amounts of the elements which resulted significantly different between controls and patients (Co, Cr, Pb, Si, Sn). Takahashi et al. (1994 - see Tab. 12) found that the Mg concentration in both untreated and treated (with levodopa) patients was lower than in controls.

#### **4.1.2 Age**

152 Diagnostics and Rehabilitation of Parkinson's Disease

As already pointed out in the **Introduction**, many factors affect the results, conditioning then the comparability among the findings of the various studies. We describe now in

In the subject enrollment, the criteria for inclusion/exclusion are fundamental and should absolutely include health conditions along with age and gender. Ideally, exposures due to environmental pollution or occupational activities and diet should also be considered. In the reviewed papers, only a few teams give full details of the selection criteria used. The researchers of the Italian Istituto Superiore di Sanità (ISS), Bocca, Forte, Alimonti and coworkers, along with the Spanish scientists (Jiménez-Jiménez et al. 1998 and Aguilar et al. 1998) are among the few authors who describe extensively the criteria employed. The diseases affecting the individuals recruited as patients or controls are often not precisely described, mainly in the less recent works. In some publications, incongruities appear within the text or among the text and the data presented in the tables. It is not always clear whether the patients were affected by comorbidities. Due to the fact that severe illnesses of heart, liver, kidney and also tumors are known to affect the levels of trace elements in human fluids and tissues, an exhaustive health report is also necessary in the case of controls. For these subjects, very heterogeneous situations have been observed. In Bourrier-Guerin et al. (1985) - see Tab. 12 - controls were not enrolled at all. Mindadse & Tschikowani (1967) - see Tab. 12 - employed blood donors. In Qureshi et al. (2005, 2006) control individuals, simply defined "healthy", were affected by tension headache, ischemic cerebrovascular disease or polyneuropathy. The Spanish group selected "healthy" subjects with suspected subarachnoid hemorrhage or pseudotumor cerebri, oculomotor palsies, etc. The scientists of the ISS enrolled individuals not suffering from any central neurological disease. Kjellin (1967a and 1967b - see Tab. 12) assumed psychoneurotic outpatients as representative of the "normal" condition. It is evident that, in the diverse investigations, were enrolled as controls subjects in really different health conditions. It is worth considering that, differently from blood, the samples of CSF are not easily available; therefore, the control specimens are mostly withdrawn from subjects undergoing lumbar puncture for clinical analyses. In the case of patients, the differences among the groups enrolled in the diverse studies are amplified by some clinical variables, as duration and severity of the disease along with medical treatments and possible comorbidities. Regarding duration and severity of the disease, in the trial by Aguilar et al. (1998) Se and Cr levels showed no correlation with age at onset and duration of the illness. On the other hand, Alimonti et al. (2007b) detected in patients a negative association of Cr amount and severity and duration of the illness; in the same study, Pb appeared to be negatively related to the severity of the disorder, while Sn resulted to be negatively associated with the duration of the disease. The authors also found that age at onset did not affect the concentration of Fe and of the other elements that resulted significantly different between controls and patients (Co, Cr, Pb, Si, Sn). Bocca et al. (2006) observed that duration and severity of the disease appeared not to be correlated with Al, Ca, Cu, Fe, Mn, Si and Zn amounts; on the other hand, Mg level decreased with the duration and severity of the illness. Concerning medical treatments, the therapies followed by patients are described by

details the influence of each factor as it emerged in the examined publications.

**4. Considerations on factors influencing the results** 

**4.1 Criteria for subject enrollment** 

**4.1.1 Health conditions** 

The subject age is known to influence the amounts of some elements in tissues and fluids. In serum, it has been documented for Cu by Ghayour-Mobarhan et al. (2005) and Kouremenou-Dona et al. (2006); for Se by Lopes et al. (2004). In brain, Markesbery et al. (1984), Ongkana et al. (2010) and Tohno et al. (2010) found age-related changes for several

Minor and Trace Elements in Cerebrospinal

14.6 14.5

13.5 17.9

14.2 19.7 22.7

**ment Subjects Mean SD N. of subjects** 

1.39 median: 1.47 0.60 median: 0.54

1.28 median: 1.39 0.65 median: 0.55

**ment Subjects Mean SD N. of subjects** 

1.06 median: 1.0 0.42 median: 0.30

0.91 median: 0.84 0.46 median: 0.43

95.0 median: 96.3 58.4 median: 52.3

**4.1.4 Number of subjects examined** 

For gender difference, see text (section 4.1.3 Gender)

Table 8. Lead and silicon in Controls and Patients (µg/L)

105 66.9

95.0 92.5 58.4 63.9 28.9

6.3 7.4

0.64 0.47

0.59 0.46

8.2 12.3

1.8 1.9 2.1

Table 7. Chromium and selenium in Controls and Patients (µg/L)

0.34 0.38

0.36 0.24

39.3 49.7

38.0 44.3 44.8 46.5 13.7

38.3 44.8

**Ele-**

Cr C PD

 C PD

 C PD

Se C PD

C

**Ele-**

Pb C PD

 C PD

Si C PD

 C " PD " "

 C PD

PD (On) PD (On/Off)

Fluid of Parkinson's Patients – Suggestions After a Critical Review of the Analytical Data 155

43 (19 M + 24 F) 28 (11 M + 17 F )

13 (6 M + 7 F) 26 (24 M + 2 F)

20 (17 M + 3 F) 42 (36 M + 6 F)

43 (19 M + 24 F) 28 (11 M + 17 F )

21 (13 M + 8 F) 17 (10 M + 7 F) 19 (13 M + 6 F)

13 (6 M + 7 F) 26 (24 M + 2 F)

20 (17 M + 3 F) 42 (36 M + 6 F)

13 (6 M + 7 F) 26 (24 M + 2 F)

18 (10 M + 8 F)

91 (64 M + 27 F)

20 (17 M + 3 F) 42 (36 M + 6 F)

In the reviewed papers, the authors usually publish the total number of controls and patients, and even the numbers of males and females; however, they frequently do not report the information actually needed, that is the number of individuals really tested for each element. In our review, we observed that Be, Cd, Hg, and V were determined in two investigations by the team of ISS (Bocca et al. 2004 and Alimonti et al. 2007b). In the previous one, where a lower number of individuals was considered, the element decrements in patients were evaluated as significant; in the second trial, where more subjects were enrolled, the variations came out to be not significant. Fe resulted decreased in patients at the limits of significance (p = 0.052) in a trial carried out by Forte et al. (2004); in two successive investigations by the same team (Bocca et al. 2006 and Alimonti et al. 2007b), with a higher number of individuals, Fe was found to be significantly reduced. The control and patient groups of successive trials by the same authors probably included the corresponding groups already examined in the previous ones; the disagreeing findings could be due to the

1 F

1 M 1 F

**(Gender) Age,** *y* **Technique Signif-**

Et-AAS

SF-ICP-MS

SF-ICP-MS

Et-AAS

Et-AAS

65.2 13.0 65.5 9.1

63.8 13.7 64.9 10.8

66.2 14.7 64.5 10.7

65.2 13.0 65.5 9.1

**(Gender) Age,** *y* **Technique Signif-**

ICP-AES

63.8 13.7 64.9 10.8

66.2 14.7 64.5 10.7

63.8 13.7 64.9 10.8

63.8 13.8 65.5 9.7

66.2 14.7 64.5 10.7

**icance References** 

NS Aguilar et al., 1998

<sup>S</sup> Bocca et al., 2004

<sup>S</sup> Alimonti et al., 2007b

NS Aguilar et al., 1998

**icance References** 

2004

2007b

2004

2007b

Bocca et al., 2006

Qureshi et al., 2006

S S

SF-ICP-MS S Bocca et al.,

ICP-AES S Forte et al.,

S S S

ICP-AES S Alimonti et al.,

SF-ICP-MS S Alimonti et al.,


Table 5. Zinc in Controls and Patients (µg/L)


data converted from nmol/L

Table 6. Manganese in Controls and Patients (µg/L)

elements. All authors who studied CSF and published element concentration values also for patients reported the mean age of each subject group; Gazzaniga et al. (1992) specified also the age range. Regarding element changes with age, Aguilar et al. (1998) found that Se and Cr levels were not correlated with the age of PD patients. Bocca et al. (2006) found no Zn changes in patients (no data given); in controls, they observed a significant Zn increment in subjects elder than 70 years in comparison with younger individuals, but these differences disappeared in patients.

#### **4.1.3 Gender**

This parameter also influences trace element levels. For changes of Se, Cu and Zn in serum, see Lopes et al. (2004) and Ghayour-Mobarhan et al. (2005). For Zn variations in brain, see Ongkana et al. (2010). Regarding CSF, Bocca et al. (2006) noticed lesser Fe amounts in PD males than in females, whereas the opposite was found in controls. They also report that Si concentration resulted significantly lower in patients than in controls and that in PD females it was two-times lower than in males. This remarkable observation could come out because the authors calculated distinct values, not published, for the two genders.


Table 7. Chromium and selenium in Controls and Patients (µg/L)


For gender difference, see text (section 4.1.3 Gender)

Table 8. Lead and silicon in Controls and Patients (µg/L)

## **4.1.4 Number of subjects examined**

154 Diagnostics and Rehabilitation of Parkinson's Disease

37 (16 M + 21 F) 37 (14 M + 23 F)

13 (6 M + 7 F) 26 (24 M + 2 F)

21 (13 M + 8 F) 17 (10 M + 7 F) 19 (13 M + 6 F)

20 (17 M + 3 F) 42 (36 M + 6 F)

**(Gender)** 

22 (20 M + 2 F) 11 (10 M + 1 F)

37 (16 M + 21 F) 37 (14 M + 23 F)

13 (6 M + 7 F) 26 (24 M + 2 F)

18 (10 M + 8 F) 91 (64 M + 27 F)

20 (17 M + 3 F) 42 (36 M + 6 F)

elements. All authors who studied CSF and published element concentration values also for patients reported the mean age of each subject group; Gazzaniga et al. (1992) specified also the age range. Regarding element changes with age, Aguilar et al. (1998) found that Se and Cr levels were not correlated with the age of PD patients. Bocca et al. (2006) found no Zn changes in patients (no data given); in controls, they observed a significant Zn increment in subjects elder than 70 years in comparison with younger individuals, but these differences

This parameter also influences trace element levels. For changes of Se, Cu and Zn in serum, see Lopes et al. (2004) and Ghayour-Mobarhan et al. (2005). For Zn variations in brain, see Ongkana et al. (2010). Regarding CSF, Bocca et al. (2006) noticed lesser Fe amounts in PD males than in females, whereas the opposite was found in controls. They also report that Si concentration resulted significantly lower in patients than in controls and that in PD females it was two-times lower than in males. This remarkable observation could come out because

the authors calculated distinct values, not published, for the two genders.

**(Gender) Age,** *y* **Technique Signif-**

Et-AAS

Et-AAS

ICP-AES

**(Range) Technique Signif-**

SF-ICP-MS

Et-AAS NS

62.4 17.8 65.7 8.8

63.8 13.7 64.9 10.8

66.2 14.7 64.5 10.7

**Age,** *y* 

age-matched 64.9 (49 - 78) 63.1 (49 - 78) 67 (59 - 77)

62.4 17.8 65.7 8.8

63.8 13.7 64.9 10.8

63.8 13.8 65.5 9.7

66.2 14.7 64.5 10.7

**icance References** 

ICP-AES NS Forte et al., 2004

S S

Et-AAS NS Pall et al., 1987

NS NS

NS

Et-AAS NS Jiménez-Jiménez

SF-ICP-MS NS Bocca et al., 2006

SF-ICP-MS NS Alimonti et al.,

<sup>S</sup> Jiménez-Jiménez et al., 1998

> Qureshi et al., 2005 and 2006

NS Alimonti et al., 2007b

**icance References** 

1992

2004

2007b

et al., 1998

Gazzaniga et al.,

Bocca et al., 2004 and Forte et al.,

**Subjects Mean SD N. of subjects** 

140 60

8.85 10.5

31 19 11

11.4 9.01

0.34 0.36

1.8 3.9 1.3 2.4

0.76 0.98

0.36 0.43

0.39 0.42

0.39 0.42

Table 6. Manganese in Controls and Patients (µg/L)

29 9

6 5

C PD

C PD

C PD (On) PD (On/Off)

C PD

C PD

C PD tot PD untreated PD treated

C PD

C PD

C PD

C PD 170 100

32.9 27.3

161 117 96

> 0.97 0.96

5.7 5.4 6.0 5.4

0.88 1.20

0.95 0.69

data converted from nmol/L

disappeared in patients.

**4.1.3 Gender** 

0.85 median: 0.91 0.63 median: 0.54

0.95 median: 1.02 0.69 median: 0.58

32.3 median: 33.5 27.7 median: 27.8

Table 5. Zinc in Controls and Patients (µg/L)

**Subjects Mean SD N. of subjects** 

In the reviewed papers, the authors usually publish the total number of controls and patients, and even the numbers of males and females; however, they frequently do not report the information actually needed, that is the number of individuals really tested for each element. In our review, we observed that Be, Cd, Hg, and V were determined in two investigations by the team of ISS (Bocca et al. 2004 and Alimonti et al. 2007b). In the previous one, where a lower number of individuals was considered, the element decrements in patients were evaluated as significant; in the second trial, where more subjects were enrolled, the variations came out to be not significant. Fe resulted decreased in patients at the limits of significance (p = 0.052) in a trial carried out by Forte et al. (2004); in two successive investigations by the same team (Bocca et al. 2006 and Alimonti et al. 2007b), with a higher number of individuals, Fe was found to be significantly reduced. The control and patient groups of successive trials by the same authors probably included the corresponding groups already examined in the previous ones; the disagreeing findings could be due to the

Minor and Trace Elements in Cerebrospinal

PD

PD

PD

PD

PD

PD

PD

PD

PD

PD

PD

N. of subjects (Gender)

C 20 (17 M + 3 F) PD 42 (36 M + 6 F)

Al <sup>C</sup>

Ba <sup>C</sup>

Bi <sup>C</sup>

Li <sup>C</sup>

Mo <sup>C</sup>

Sb <sup>C</sup>

Sn <sup>C</sup>

Sr <sup>C</sup>

Tl <sup>C</sup>

<sup>W</sup><sup>C</sup>

Zr <sup>C</sup>

al., 2007b)

**4.3 Statistical tests**

Fluid of Parkinson's Patients – Suggestions After a Critical Review of the Analytical Data 157

to previous publications. The techniques used for the most studied elements were principally electrothermal atomic absorption spectrometry (Et-AAS) and inductively coupled plasma atomic emission spectrometry (ICP-AES). For some elements, sector field inductively coupled plasma mass spectrometry (SF-ICP-MS) was also employed by the team of ISS. All these analytical techniques are widely used for trace element determination in human samples.

Element Subjects Mean SD Median Significance

0.51 1.03

0.15 0.13

0.07 0.05

0.13 0.53

0.27 0.17

0.02 0.04

0.07 0.11

8.69 8.66

0.01 0.007

0.02 0.02

0.05 0.03

SF-ICP-MS

2.72

0.31

0.09

0.52

0.43

0.09

0.31

27.7

0.01

0.04

0.06

Age, *y*

66.2 14.7 64.5 10.7

2.05 NS

0.24 NS

0.06 NS

0.65 NS

0.27 NS

0.07 NS

0.24 S

22.6 NS

0.01 NS

0.03 NS

0.04 NS

2.64 2.15

0.35 0.26

0.10 0.08

0.52 0.82

0.45 0.33

0.08 0.06

0.32 0.26

30.0 24.6

0.01 0.01

0.04 0.03

0.06 0.04

Technique:

Table 11. Other trace elements in Controls and Patients (µg/L); (modified from Alimonti et

In this kind of studies, statistical tests of various types are required for diverse appraisals. Within each study and for each element considered, tests are applied at first to evaluate whether the concentrations found for controls and patients are significantly different or not. In the same trial, other tests can reveal non negligible dissimilarities among the control and patient groups, regarding one or more factors affecting the results. When a significant


Table 9. Cobalt and nickel in Controls and Patients (µg/L)


Table 10. Berillium, cadmium, mercury and vanadium in Controls and Patients (µg/L)

different number of the considered subjects. In the case of Co, Cr, Pb and Si, the outcomes for significance are always the same when the number of subjects, in both control and patient groups, is either lower or higher. We wonder whether the changes in patients of these elements are so marked that result noticeable in every case. As a general consideration, it is obvious that the higher is the number of the subjects examined, the higher is the representativeness of the results obtained.

#### **4.2 Analytical procedures**

When determining elements at trace levels, the entire analytical process is critical. Sampling and storage should be carried out in an appropriate way to minimize contamination and losses, following the recognized requirements in the field. The chemical treatments needed by each method should be as standardized as possible. The analytical technique employed must assure high sensitivity and good reproducibility.

In the reviewed studies, the preanalytical steps were described with more or less details; the techniques employed were cited by all authors, except Sparks et al. (2008). A careful description of the method is generally available in the most recent papers, that sometimes refer to previous publications. The techniques used for the most studied elements were principally electrothermal atomic absorption spectrometry (Et-AAS) and inductively coupled plasma atomic emission spectrometry (ICP-AES). For some elements, sector field inductively coupled plasma mass spectrometry (SF-ICP-MS) was also employed by the team of ISS. All these analytical techniques are widely used for trace element determination in human samples.


Table 11. Other trace elements in Controls and Patients (µg/L); (modified from Alimonti et al., 2007b)

## **4.3 Statistical tests**

156 Diagnostics and Rehabilitation of Parkinson's Disease

13 (6 M + 7 F) 26 (24 M + 2 F)

20 (17 M + 3 F) 42 (36 M + 6 F)

13 (6 M + 7 F) 26 (24 M + 2 F)

20 (17 M + 3 F) 42 (36 M + 6 F)

13 (6 M + 7 F) 26 (24 M + 2 F)

20 (17 M + 3 F) 42 (36 M + 6 F)

13 (6 M + 7 F) 26 (24 M + 2 F)

20 (17 M + 3 F) 42 (36 M + 6 F)

13 (6 M + 7 F) 26 (24 M + 2 F)

20 (17 M + 3 F) 42 (36 M + 6 F)

13 (6 M + 7 F) 26 (24 M + 2 F)

20 (17 M + 3 F) 42 (36 M + 6 F)

Table 10. Berillium, cadmium, mercury and vanadium in Controls and Patients (µg/L)

different number of the considered subjects. In the case of Co, Cr, Pb and Si, the outcomes for significance are always the same when the number of subjects, in both control and patient groups, is either lower or higher. We wonder whether the changes in patients of these elements are so marked that result noticeable in every case. As a general consideration, it is obvious that the higher is the number of the subjects examined, the higher is the

When determining elements at trace levels, the entire analytical process is critical. Sampling and storage should be carried out in an appropriate way to minimize contamination and losses, following the recognized requirements in the field. The chemical treatments needed by each method should be as standardized as possible. The analytical technique employed

In the reviewed studies, the preanalytical steps were described with more or less details; the techniques employed were cited by all authors, except Sparks et al. (2008). A careful description of the method is generally available in the most recent papers, that sometimes refer

**(Gender) Age,** *y* **Technique Signif-**

63.8 13.7 64.9 10.8

66.2 14.7 64.5 10.7

63.8 13.7 64.9 10.8

66.2 14.7 64.5 10.7

**(Gender) Age,** *y* **Technique Signif-**

63.8 13.7 64.9 10.8

66.2 14.7 64.5 10.7

63.8 13.7 64.9 10.8

66.2 14.7 64.5 10.7

63.8 13.7 64.9 10.8

66.2 14.7 64.5 10.7

63.8 13.7 64.9 10.8

66.2 14.7 64.5 10.7 **icance References** 

2004

2007b

2004

2007b

SF-ICP-MS S Bocca et al.,

SF-ICP-MS NS Bocca et al.,

SF-ICP-MS S Alimonti et al.,

SF-ICP-MS NS Alimonti et al.,

**icance References** 

2007b

2007b

2007b

2007b

SF-ICP-MS <sup>S</sup> Bocca et al., 2004

SF-ICP-MS NS Alimonti et al.,

SF-ICP-MS <sup>S</sup> Bocca et al., 2004

SF-ICP-MS NS Alimonti et al.,

SF-ICP-MS <sup>S</sup> Bocca et al., 2004

SF-ICP-MS NS Alimonti et al.,

SF-ICP-MS <sup>S</sup> Bocca et al., 2004

SF-ICP-MS NS Alimonti et al.,

**ment Subjects Mean SD N. of subjects** 

0.04 0.04

0.05 0.09

1.39 5.61

3.33 3.61

Table 9. Cobalt and nickel in Controls and Patients (µg/L)

0.33 0.13

0.37 0.21

0.02 0.01

0.03 0.02

0.50 0.32

0.46 0.32

0.06 0.03

0.03 0.03

0.15 median: 0.16 0.04 median: 0.03

0.13 median: 0.13 0.09 median: 0.06

8.01 median: 7.54 4.37 median: 1.07

5.40 median: 6.44 3.34 median: 1.53

**ment Subjects Mean SD N. of subjects**

0.87 median: 0.85 0.44 median: 0.44

0.70 median: 0.55 0.56 median: 0.54

0.06 median: 0.06 0.03 median: 0.03

0.05 median: 0.05 0.04 median: 0.4

1.20 median: 1.19 0.67 median: 0.74

1.05 median: 0.85 0.73 median: 0.81

0.12 median: 0.11 0.07 median: 0.08

0.09 median: 0.10 0.07 median: 0.07

representativeness of the results obtained.

must assure high sensitivity and good reproducibility.

**4.2 Analytical procedures**

**Ele-**

Co C PD

 C PD

Ni C PD

 C PD

**Ele-**

Be C PD

 C PD

Cd C PD

 C PD

Hg C PD

 C PD

V C PD

 C PD

> In this kind of studies, statistical tests of various types are required for diverse appraisals. Within each study and for each element considered, tests are applied at first to evaluate whether the concentrations found for controls and patients are significantly different or not. In the same trial, other tests can reveal non negligible dissimilarities among the control and patient groups, regarding one or more factors affecting the results. When a significant

Minor and Trace Elements in Cerebrospinal

values was wider than for controls.

was probably in both cases a male of 65 y.

suffering from parkinsonism.

however not reported.

Fluid of Parkinson's Patients – Suggestions After a Critical Review of the Analytical Data 159

**Bourrier-Guerin et al.** 1985 report values for 13 elements in 70 patients (34 M and 36 F) affected by different neurodegenerative diseases; patients were grouped all together. Si

**Campanella et al**. 1973 enrolled 18 individuals (5 controls; 7 untreated patients; 6 patients treated with dopaminergic drugs), age 39 y, no gender given. They published the Cu mean amount found for each subject. For both patients groups, the range of the mean

**Kjellin** 1967a and 1967b reported Cu and Fe amounts in the CSF of a female patient (69 y)

Cu and Fe resulted respectively higher and lesser in comparison to a unique control, who

**Mindadse & Tschikowani** 1967 found that Au amount in PD patients was 66 µg/g, about the double than in controls. The Au concentration in controls (blood donors) is

**Pall et al.** 1987 found in patients (24) with untreated, idiopathic PD, a higher Cu

did not observe a difference between patients (26) and controls (33).

the amounts of Cd, Fe, Mn and Zn did not change.

zinc concentration than in controls (2).

Table 12. Additional information

**6. Conclusion**

concentration than in controls (34) with various other neurological diseases. For Fe, they

**Pan et al.** 1997 observed that Cu increased significantly in PD patients; on the other hand,

**Takahashi et al.** 1994 evaluated Br, Cu, Fe, Se, Zn and Mg levels in 25 controls and 20 PD patients (13 untreated and 7 treated with L-dopa). The mean Mg concentration in both

**Woodbury et al**. 1968 determined in one PD patient a Mg amount overlapping the mean value found for controls (11). Always in one patient, these authors determined a higher

Regarding the matrix CSF, the first remark we make is that the element concentration values available in the literature are non numerous, probably due to the rareness of the

treated and untreated parkinsonians was found to be lower than in controls.

and Zn resulted to be significantly higher in men than in women.

discrepancy is disclosed, the comparison between the mean concentration values for controls and those for patients results rather inappropriate.

A crucial point, worth of a close investigation by the scientists of the field, is to assess at what extent the outcomes for significance of the various tests applied are the same. When comparing the results of different investigations, the diversity of the statistical tests applied in each one causes an amplification of the general inhomogeneity.

In the reviewed papers, the statistical tests used to verify the difference between the results for controls and patients are generally indicated. Some authors checked also possible differences, for one or more variability factors, among the various subject groups; their information is therefore more accurate.

## **5. Summary of the retrieved data**

The retrieved data are non numerous, being the withdrawal of the fluid unpleasant; the control samples are taken from individuals undergoing lumbar puncture for neurological exams. Some values have been found for Cu, Fe, Mn and Zn, whereas only few results have been retrieved for Cr and Si. As far as other elements are concerned, the data are absolutely scarce or determined only once, mainly by the scientists of ISS.

Examining the collected values, regarding **copper** - see Tab. 4 - no significant variations for patients as compared to controls were found in trials performed by diverse teams; nevertheless, in other papers (not showing analytical data for Cu), Pall et al. (1987) and Pan et al. (1997) - see Tab. 12 - assess to have observed a remarkable elevation. In the case of **manganese** too – see Tab. 6 - no changes were observed in the different investigations; of note, the levels determined by Gazzaniga et al. (1992) are higher than those found by the other author groups. Concerning **calcium** and **magnesium** - see Tab. 2 - no significant alterations are reported; however, Takahashi et al. (1994) - see Tab. 12 - assess to have found a lesser Mg level in patients. As for **zinc** - see Tab. 5 - Forte et al. (2004) and Alimonti et al. (2007b) observed in PD a slight diminution, which in the trials by Jiménez- Jiménez et al. (1998) and Qureshi et al. (2005 and 2006) attained the significance. Aguilar et al (1998) found for PD subjects a non significant **selenium** increment - see Tab. 7; a significant elevation resulted instead in all the patients, with both positive and negative response to the therapy, enrolled by Qureshi et al. (2006). **Lead** - see Tab. 8 - was found to be significantly reduced in patients by the team of ISS (Bocca et al. 2004 and Alimonti et al. 2007b), that obtained the same finding also for **silicon** (Forte et al. 2004, Bocca et al. 2006, Alimonti et al. 2007b) - see Table 8. In the case of **iron** - see Tab. 3 - the most interesting element for PD, discordant results were unfortunately recruited. A significant depletion was found by Bocca et al. 2006 and Alimonti et al. 2007b; for a detailed description, see the paragraph 4.1.4. An elevation, also significant, was seen by Qureshi et al. (2005 and 2006); other scientists as Gazzaniga et al. (1992) and Jiménez- Jiménez et al. (1998) did not observe noticeable variations. The values determined by the ISS team appear to be remarkably lower than those published by the other groups. Dealing with **chromium** - see Tab. 7 - Aguilar et al. (1998) found similar amounts in the CSF of patients and controls; differently, Bocca et al. (2004) and Alimonti et al. (2007b) obtained much lower values and noticed a significant decrement in patients. **Al**, **Ba**, **Be, Bi**, **Co**, **Li**, **Mo**, **Ni**, **Sb**, **Sn, Sr**, **Tl**, **V, W** and **Zr** were determined only by the scientists of ISS - see Tab. 9, 10, 11. No variations were observed, except significant decreases of Co and Sn. Regarding the results for Be and V, see the paragraph 4.1.4, where are described also the findings for **Cd** and **Hg**; the values for these last four elements are shown in Tab. 10.

**Bourrier-Guerin et al.** 1985 report values for 13 elements in 70 patients (34 M and 36 F) affected by different neurodegenerative diseases; patients were grouped all together. Si and Zn resulted to be significantly higher in men than in women.

**Campanella et al**. 1973 enrolled 18 individuals (5 controls; 7 untreated patients; 6 patients treated with dopaminergic drugs), age 39 y, no gender given. They published the Cu mean amount found for each subject. For both patients groups, the range of the mean values was wider than for controls.

**Kjellin** 1967a and 1967b reported Cu and Fe amounts in the CSF of a female patient (69 y) suffering from parkinsonism.

Cu and Fe resulted respectively higher and lesser in comparison to a unique control, who was probably in both cases a male of 65 y.

**Mindadse & Tschikowani** 1967 found that Au amount in PD patients was 66 µg/g, about the double than in controls. The Au concentration in controls (blood donors) is however not reported.

**Pall et al.** 1987 found in patients (24) with untreated, idiopathic PD, a higher Cu concentration than in controls (34) with various other neurological diseases. For Fe, they did not observe a difference between patients (26) and controls (33).

**Pan et al.** 1997 observed that Cu increased significantly in PD patients; on the other hand, the amounts of Cd, Fe, Mn and Zn did not change.

**Takahashi et al.** 1994 evaluated Br, Cu, Fe, Se, Zn and Mg levels in 25 controls and 20 PD patients (13 untreated and 7 treated with L-dopa). The mean Mg concentration in both treated and untreated parkinsonians was found to be lower than in controls.

**Woodbury et al**. 1968 determined in one PD patient a Mg amount overlapping the mean value found for controls (11). Always in one patient, these authors determined a higher zinc concentration than in controls (2).

Table 12. Additional information

### **6. Conclusion**

158 Diagnostics and Rehabilitation of Parkinson's Disease

discrepancy is disclosed, the comparison between the mean concentration values for

A crucial point, worth of a close investigation by the scientists of the field, is to assess at what extent the outcomes for significance of the various tests applied are the same. When comparing the results of different investigations, the diversity of the statistical tests applied

In the reviewed papers, the statistical tests used to verify the difference between the results for controls and patients are generally indicated. Some authors checked also possible differences, for one or more variability factors, among the various subject groups; their

The retrieved data are non numerous, being the withdrawal of the fluid unpleasant; the control samples are taken from individuals undergoing lumbar puncture for neurological exams. Some values have been found for Cu, Fe, Mn and Zn, whereas only few results have been retrieved for Cr and Si. As far as other elements are concerned, the data are absolutely

Examining the collected values, regarding **copper** - see Tab. 4 - no significant variations for patients as compared to controls were found in trials performed by diverse teams; nevertheless, in other papers (not showing analytical data for Cu), Pall et al. (1987) and Pan et al. (1997) - see Tab. 12 - assess to have observed a remarkable elevation. In the case of **manganese** too – see Tab. 6 - no changes were observed in the different investigations; of note, the levels determined by Gazzaniga et al. (1992) are higher than those found by the other author groups. Concerning **calcium** and **magnesium** - see Tab. 2 - no significant alterations are reported; however, Takahashi et al. (1994) - see Tab. 12 - assess to have found a lesser Mg level in patients. As for **zinc** - see Tab. 5 - Forte et al. (2004) and Alimonti et al. (2007b) observed in PD a slight diminution, which in the trials by Jiménez- Jiménez et al. (1998) and Qureshi et al. (2005 and 2006) attained the significance. Aguilar et al (1998) found for PD subjects a non significant **selenium** increment - see Tab. 7; a significant elevation resulted instead in all the patients, with both positive and negative response to the therapy, enrolled by Qureshi et al. (2006). **Lead** - see Tab. 8 - was found to be significantly reduced in patients by the team of ISS (Bocca et al. 2004 and Alimonti et al. 2007b), that obtained the same finding also for **silicon** (Forte et al. 2004, Bocca et al. 2006, Alimonti et al. 2007b) - see Table 8. In the case of **iron** - see Tab. 3 - the most interesting element for PD, discordant results were unfortunately recruited. A significant depletion was found by Bocca et al. 2006 and Alimonti et al. 2007b; for a detailed description, see the paragraph 4.1.4. An elevation, also significant, was seen by Qureshi et al. (2005 and 2006); other scientists as Gazzaniga et al. (1992) and Jiménez- Jiménez et al. (1998) did not observe noticeable variations. The values determined by the ISS team appear to be remarkably lower than those published by the other groups. Dealing with **chromium** - see Tab. 7 - Aguilar et al. (1998) found similar amounts in the CSF of patients and controls; differently, Bocca et al. (2004) and Alimonti et al. (2007b) obtained much lower values and noticed a significant decrement in patients. **Al**, **Ba**, **Be, Bi**, **Co**, **Li**, **Mo**, **Ni**, **Sb**, **Sn, Sr**, **Tl**, **V, W** and **Zr** were determined only by the scientists of ISS - see Tab. 9, 10, 11. No variations were observed, except significant decreases of Co and Sn. Regarding the results for Be and V, see the paragraph 4.1.4, where are described also the findings for **Cd**

controls and those for patients results rather inappropriate.

information is therefore more accurate.

**5. Summary of the retrieved data**

in each one causes an amplification of the general inhomogeneity.

scarce or determined only once, mainly by the scientists of ISS.

and **Hg**; the values for these last four elements are shown in Tab. 10.

Regarding the matrix CSF, the first remark we make is that the element concentration values available in the literature are non numerous, probably due to the rareness of the

Minor and Trace Elements in Cerebrospinal

pp. 1245-1251

*Neurotoxicol.,* Vol.28, pp. 450-456

*Med. Biol*., Vol.21, pp. 234-241

*J. Neurol. Sci*., Vol.248, pp. 23-30

*Neurol*. (Napoli), Vol.28, pp. 1-34

Vol.2, pp. 88-91

pp. 1031-1040

(Eds.) 89-91, John Libbey Eurotext, London-Paris

**7. References** 

Fluid of Parkinson's Patients – Suggestions After a Critical Review of the Analytical Data 161

Aguilar, M.V.; Jimenez-Jimenez, F.J.; Molina, J.A.; Meseguer, I.; Mateos-Vega, C.J.;

Alimonti, A.; Ristori, G.; Giubilei F.; Stazi, M.A.; Pino, A.; Visconti, A.; Brescianini, S.; Sepe

Alimonti, A.; Bocca, B.; Pino, A.; Ruggieri, F.; Forte, G. & Sancesario, G. (2007b). Elemental

Belliveau, J.F.; Friedman, J.H.; O'Leary, G.P. Jr. & Guarrera, D. (1990). Evaluation of

Bocca, B.; Alimonti, A.; Petrucci, F.; Violante, N.; Sancesario, G.; Forte, G. & Senofonte, O.

Bocca, B.; Alimonti, A.; Senofonte, O.; Pino, A.; Violante, N.; Petrucci, F.; Sancesario, G. &

Bourrier-Guerin, L.; Mauras, Y.; Truelle, J.L. & Allain, P. (1985). CSF and plasma

Campanella, G.; Carrieri, P.; Romito, D. & Pasqual-Marsettin, E. (1973). Ferro, transferrina,

Dexter, D.T.; Wells, F.R.; Lees, A.J.; Agid, F.; Agid, Y.; Jenner, P. & Marsde, C.D. (1989).

Dexter, D.T.; Carayon, A.; Javoy-Agid, F.; Agid, Y.; Wells, F.R.; Daniel, S.E.; Lees, A.J.;

Forte, G.; Bocca, B.; Senofonte, O.; Petrucci, F.; Brusa, L.; Stanzione, P.; Zannino, S.;

in Parkinson's disease. *J. Neurochem*., Vol.52, pp. 1830-1836

affecting the basal ganglia. *Brain,* Vol.114, pp. 1953-1975

Parkinson's disease. *Spectrochim. Acta Part B*, Vol.59, pp. 559-566

Gonzalez-Munoz, M.J.; de Bustos, F.; Gomez-Escalonilla, C.; Ortí-Pareja, M.; Zurdo, M. & Martinez-Para, M.C. (1998). Cerebrospinal fluid selenium and chromium levels in patients with Parkinson's disease. *J. Neural Transm*., Vol.105,

Monti, M.; Forte, G.; Stanzione, P.; Bocca, B.; Bomboi, G.; D'Ippolito, C.; Annibali, V.; Salvetti, M. & Sancesario, G. (2007a). Serum chemical elements and oxidative status in Alzheimer's disease, Parkinson disease and multiple sclerosis.

profile of cerebrospinal fluid in patients with Parkinson's disease. *J. Trace Elem.* 

increased copper levels in the cerebrospinal-fluid of Parkinson's patients, In: *Metal Ions in Biology and Medicine,* P. Collery, L.A. Poirier, M. Manfait & J.C. Etienne,

(2004). Quantification of trace elements by sector field inductively coupled plasma mass spectrometry in urine, serum, blood and cerebrospinal fluid of patients with

Forte, G. (2006). Metal changes in CSF and peripheral compartments of PD patients.

concentrations of 13 elements in various neurological disorders. *Trace Elem. Med*.,

rame e ceruloplasmina del siero e del liquor nelle malattie extrapiramidali e nelle miopatie primitive. [Iron, transferrin, copper and ceruloplasmin of the serum and cerebrospinal fluid in extrapyramidal diseases and primary myopathies]. *Acta* 

Increased nigral iron content and alterations in other metal ions occurring in brain

Jenner, P. & Marsden, C.D. (1991). Alterations in the levels of iron, ferritin and other trace metals in Parkinson's disease and other neurodegenerative diseases

Violante, N.; Alimonti, A. & Sancesario, G. (2004). Trace and major elements in whole blood, serum, CSF and urine of patients with PD. *J. Neural Trans*., Vol.111,

samples. Among the recruited papers, the range of values was recorded only in that by Belliveau et al. (1990). Knowing the ranges for controls and patients would allow to establish a range of normalcy for each element and, as a consequence, to individuate in patients a possible shift towards elevation or diminution. Examining the retrieved data, it is evident that for some elements the results obtained by the various research groups are of different levels. For Cu, the values published by the different teams vary from less than two decades to more than a hundred of µg/L. For Fe and Zn, the scientists of ISS determined concentrations much lesser than the other teams. For Cr, Aguilar et al. (1998) found values an order of magnitude higher than those reported by the team of ISS. The discrepancies regarding the element levels are difficult to explain. The mean values retrieved have often very large standard deviations. In the case of Be, Cd, Cr Hg, Se, Si, V, the SDs are sometimes as high as the half of the mean. A similar situation resulted for Mn and Cu in the study by Jiménez-Jiménez et al. (1998). Dealing with Fe, Jiménez-Jiménez et al. (1998) and Forte et al. (2004) detected SD values very close to the mean. SDs close to the mean were also found for Co and Pb by the researchers of ISS; they report, for Ni, SDs even higher. The large SDs can be due to the individual variability and/or to the low number of the subjects enrolled; they are not surprising also when the element concentration level is very low (a few µg/L or less). In the case of Cr and Se, and mostly in that of Fe and Si, high SDs are less expected. Obviously, they make it really difficult to evaluate the significance of the difference among the results.

In this review, we have verified the influence, on the results, of number, age, gender of the subjects enrolled; health conditions (with regard also to clinical variables as duration and severity of the disease and pharmacological therapies) were demonstrated to be other influencing factors. The importance of adequate analytical procedures and statistical tests has been previously described (see the respective paragraphs).

At this point, we can suggest that, in a trial, attention should be paid to match, as far as possible, age and health conditions of the subjects belonging to the same group; this is more difficult to obtain in the case of patients. Concerning gender, separate male and female groups could reveal possible unexpected information. A similar number of individuals in the various groups should be enrolled; anyway, we are aware that, in the clinical practice, the scarceness of the CSF control samples and the prevalent number of male PD patients (Alimonti et al. 2007b) make these requirements not always achievable. In addition, all the previously mentioned factors should be not too different when confronting the results of the various studies, to allow a proper comparison. It is evident that this is a truly unattainable task.

In our opinion, a real upgrading in this field could actually be achieved if many specific indications were recorded in the single studies. Regarding every subject enrolled, information as age, gender, health condition, lifestyle and environmental exposure should be clearly reported; for each individual, also the results obtained for every element should be published. A detailed description of the various steps of sampling and analytical procedures should also be given; the single steps should be performed according to the indications most recently standardized.

Following all these suggestions, a database useful for diverse kind of investigations would be obtained; retrospective studies as meta-analyses, based on single factors affecting the results, could be derived; even findings not detectable at the moment could arise.

## **7. References**

160 Diagnostics and Rehabilitation of Parkinson's Disease

samples. Among the recruited papers, the range of values was recorded only in that by Belliveau et al. (1990). Knowing the ranges for controls and patients would allow to establish a range of normalcy for each element and, as a consequence, to individuate in patients a possible shift towards elevation or diminution. Examining the retrieved data, it is evident that for some elements the results obtained by the various research groups are of different levels. For Cu, the values published by the different teams vary from less than two decades to more than a hundred of µg/L. For Fe and Zn, the scientists of ISS determined concentrations much lesser than the other teams. For Cr, Aguilar et al. (1998) found values an order of magnitude higher than those reported by the team of ISS. The discrepancies regarding the element levels are difficult to explain. The mean values retrieved have often very large standard deviations. In the case of Be, Cd, Cr Hg, Se, Si, V, the SDs are sometimes as high as the half of the mean. A similar situation resulted for Mn and Cu in the study by Jiménez-Jiménez et al. (1998). Dealing with Fe, Jiménez-Jiménez et al. (1998) and Forte et al. (2004) detected SD values very close to the mean. SDs close to the mean were also found for Co and Pb by the researchers of ISS; they report, for Ni, SDs even higher. The large SDs can be due to the individual variability and/or to the low number of the subjects enrolled; they are not surprising also when the element concentration level is very low (a few µg/L or less). In the case of Cr and Se, and mostly in that of Fe and Si, high SDs are less expected. Obviously, they make it really difficult to evaluate the significance of the

In this review, we have verified the influence, on the results, of number, age, gender of the subjects enrolled; health conditions (with regard also to clinical variables as duration and severity of the disease and pharmacological therapies) were demonstrated to be other influencing factors. The importance of adequate analytical procedures and statistical tests

At this point, we can suggest that, in a trial, attention should be paid to match, as far as possible, age and health conditions of the subjects belonging to the same group; this is more difficult to obtain in the case of patients. Concerning gender, separate male and female groups could reveal possible unexpected information. A similar number of individuals in the various groups should be enrolled; anyway, we are aware that, in the clinical practice, the scarceness of the CSF control samples and the prevalent number of male PD patients (Alimonti et al. 2007b) make these requirements not always achievable. In addition, all the previously mentioned factors should be not too different when confronting the results of the various studies, to allow a proper comparison. It is evident that this is a truly

In our opinion, a real upgrading in this field could actually be achieved if many specific indications were recorded in the single studies. Regarding every subject enrolled, information as age, gender, health condition, lifestyle and environmental exposure should be clearly reported; for each individual, also the results obtained for every element should be published. A detailed description of the various steps of sampling and analytical procedures should also be given; the single steps should be performed according to the

Following all these suggestions, a database useful for diverse kind of investigations would be obtained; retrospective studies as meta-analyses, based on single factors affecting the

results, could be derived; even findings not detectable at the moment could arise.

has been previously described (see the respective paragraphs).

difference among the results.

unattainable task.

indications most recently standardized.


Minor and Trace Elements in Cerebrospinal

Vol.2, pp. 238-241

Vol.24, pp. 426-430

871

705

Fluid of Parkinson's Patients – Suggestions After a Critical Review of the Analytical Data 163

Ongkana, N.; Tohno, S.; Tohno, Y.; Suwannahoy, P.; Mahakkanukrauh, P.; Azuma, C. &

Pall, H.S.; Williams, A.; Blake, D.R.; Lunec, J.; Gutteridge, J.M.; Hall, M. & Taylor, A. (1987).

Pan, B.Y.; Cheng, Q.L.; He, Z.X. & Su, C.C. (1997). Transition metals in serum and CSF of patients with Parkinson's disease. *Mov. Disord*., Vol.12 (Suppl.), p. 33 Pande, M.B.; Nagabhushan, P.; Hegde, M.L.; Rao, T.S. & Rao, K.S. (2005). An algorithmic

Qureshi, G.A.; Qureshi, A.A.; Memon, S.A.; Sarwar, M. & Parvez, S.H. (2005). The role of

Qureshi, G.A.; Qureshi, A.A.; Memon, S.A. & Parvez, S.H. (2006). Impact of selenium, iron,

Rajput, A.H.; Uitti, R.J.; Rozdilsky, B. & Yuen, W.K. (1985). Distribution of metals in Parkinson's disease and control brains. *Neurol*., Volo.35 (Suppl. 1), p. 224. Riederer, P.; Sofic, E.; Rausch, W.D.; Schmidt, B.; Reynolds, G.P.; Jellinger, K. & Youdim,

Sparks, D.L.; Ziolkowski, C.; Connor, D.; Beach, T.; Adler, C. & Sabbagh, M. (2008). Copper

Speziali, M. & Orvini, E. (2003). Metals distribution and regionalization in the brain, In:

Speziali, M. & Di Casa, M. (2009) Copper, iron, zinc and other element concentrations in

Takahashi, S.; Takahashi, J.; Osawa, N.; Abe, T.; Yonezawa, H.; Sera, K. & Tohgi, H. (1994).

Tan, X.; Luo, Y.; Pan, J.; Huang, B. & Wang P.Q. (2007). Serum Cu, Fe, Mn, and Zn levels and

Tohno, Y.; Tohno, S.; Ongkana, N.; Suwannahoy, P.; Azuma, C.; Minami, T. &

Woodbury, J.; Lyons, K.; Carretta, R.; Hahn, A. & Sullivan, J.F. (1968). Cerebrospinal fluid

Parkinson's disease. *Parkinsonism Rel. Dis.* Vol.13 (Suppl.), p. S134

Parkinson disease. *Comput. Biol. Med*., Vol.35, pp. 475-493

parkinsonian brains. *J. Neurochem*., Vol.52, pp. 515-520

therapy. *Biogenic Amines*, Vol.19, pp. 257-267

*Transm*., Vol.71, (Suppl.), pp. 229-236

Publishing Co., Singapore-New Jersey

*Elem. Res*. *,*Vol.138, pp. 42-52.

data. *Trace Elem. Electrol.*, Vol.26, pp. 171-176

Minami, T.( 2010). Age-related changes of elements in the anterior commissures and the relationships among their elements. *Biol. Trace Elem. Res.,* Vol.135, pp.86-97

Raised cerebrospinal-fluid copper concentration in Parkinson's disease. *Lancet*,

approach to understand trace elemental homeostasis in serum samples of

iron, copper and zinc and their effects in on/off Parkinson's patients on L-dopa

copper and zinc in on/off Parkinson's patients on L-dopa therapy. *J. Neural* 

M.B. (1989). Transition metals, ferritin, glutathione and ascorbic acid in

and cognition in Alzheimer's disease and Parkinson's disease. *Cell Biol. Toxicol*.,

*Metal Ions and Neurodegenerative Disorders*, P. Zatta (Ed.) 15-65,World Scientific

cerebrospinal fluid of Parkinson's disease patients – considerations on literature

Trace elements analysis of serum and cerebrospinal fluid with PIXE - Effect of age and changes in parkinsonian patients. *Nippon Ronen Igakkai Zasshi*, Vol.31, pp. 865-

Mahakkanukrauh, P. (2010). Age-related changes of elements and relationships among elements in human hippocampus, dentate gyrus, and fornix. *Biol. Trace* 

and serum levels of magnesium, zinc, and calcium in man. *Neurol*., Vol.18, pp. 700-


Forte, G.; Alimonti, A.; Pino, A.; Stanzione, P.; Brescianini, S.; Brusa, L.; Sancesario, G.;

Gazzaniga, G.C.; Ferraro, B.; Camerlingo, M.; Casto, L.; Viscardi, M. & Mamoli, A. (1992). A

Gellein, K.; Syversen, T.; Steinnes, E.; Nilsen, T.I.; Dahl, O.P.; Mitrovic, S.; Duraj, D. & Flaten,

Ghayour-Mobarhan, M.; Taylor, A.; New, S.A.; Lamb, D.J. & Ferns, G.A. (2005).

Griffiths, P.D.; Dobson, B.R.; Jones, G.R. & Clarke, D.T. (1999). Iron in the basal ganglia in

Hegde, M.L.; Shanmugavelu, P.; Vengamma, B.; Rao, T.S.; Menon, R.B.; Rao, R.V. & Rao,

Jimenez-Jimenez, F.J.; Molina, J.A.; Aguilar, M.V.; Meseguer, I.; Mateos-Vega, C.J.;

patients with Parkinson's disease. *J. Neural Transm*., Vol.105, pp. 497-505 Kjellin, K.G. (1967). Trace elements in the cerebrospinal fluid, In: *Nuclear Activation* 

Kjellin, K.G. (1967). The CSF iron in patients with neurological diseases. *Acta Neurol. Scand*.,

Kouremenou-Dona, E.; Dona, A.; Papoutsis, J. & Spiliopoulou, C. (2006). Copper and zinc

Lopes, P.A.; Santos, M.C.; Vicente, L.; Rodrigues, M.O.; Pavão, M.L.; Nève, J. & Viegas-

Markesbery, W.R.; Ehmann, W.D.; Alauddin, M. & Hossain, T.I. (1984). Brain trace element

Mindadse, A.A. & Tschikowani, T.I. (1967). Über die Verteilung von Spurenelementen

concentrations in aging. *Neurobiol. Aging*, Vol.5, pp. 19-28

*Gesundheitsw*., Vol.22, pp. 1746-1748

structure and cryo-electron microscopy. *Brain*, Vol.122, pp. 667-673

Parkinson's disease. *Ann. Ist. Super. Sanità*, Vol.41, pp. 189-195

*Neurol. Sci*., Vol.13, pp. 239-243

*Res*., Vol.1219, pp. 111-115

*Biochem*., Vol.42, pp. 364-375

163-171

91/3

81

1-17

Vol.43, pp. 299-313

Violante, N. & Bocca, B. (2005). Metals and oxidative stress in patients with

case control study of CSF copper, iron and manganese in Parkinson disease. *Ital. J.* 

T.P. (2008). Trace elements in serum from patients with Parkinson's disease - A prospective case-control study - The Nord-Trøndelag Health Study (HUNT). *Brain* 

Determinants of serum copper, zinc and selenium in healthy subjects. *Ann. Clin.* 

Parkinson's disease. An in vitro study using extended X-ray absorption fine

K.S. (2004). Serum trace element levels and the complexity of inter-element relations in patients with Parkinson's disease. *J. Trace Elem. Med. Biol*., Vol.18, pp.

Gonzalez-Munoz, M.J.; de Bustos, F.; Martínez-Salio, A.; Ortí-Pareja, M.; Zurdo, M. & Martinez-Para, M.C. (1998). Cerebrospinal fluid levels of transition metals in

*Techniques in The Life Sciences*, A. Ericson, (Ed.), 517-532, IAEA Proc. Series SM-

concentrations in serum of healthy Greek adults. *Sci. Total Environ*., Vol.359, pp. 76-

Crespo, A.M. (2004). Trace element status (Se, Cu, Zn) in healthy Portuguese subjects of Lisbon population: a reference study. *Biol. Trace Elem. Res*., Vol.101, pp.

(Mangan, Kupfer, Zink und Gold) im Serum und Liquor bei Epilepsie und Parkinsonsyndrom [On the distribution of trace elements (manganese, copper, zinc, and gold) in serum and cerebro-spinal fluid in epilepsy and parkinsonism]. *Dtsch* 


**Language Processing in Parkinson's** 

One of the major pathophysiological features in Parkinson's disease, from now on referred to as PD, is the loss of dopaminergic neurons in the substantia nigra, which in turn results in dysfunction of the cortico-striato-cortical circuits (Bartels & Leenders, 2009). In PD the components of the cortico-striato-cortical circuits are not in an optimal interaction, leading to insufficient engagement of for example the frontal and prefrontal lobes. Motor symptoms of tremor, bradykinesia, and rigidity are the clinical hallmark of PD (Wolters & Bosboom, 2007), however, non-motor symptoms are often present (Dubois & Pillon, 1995, 1997). In particular cognitive impairments in the domain of executive functioning have frequently been observed, both in late and also in very early stages of PD (Muslimovic et al., 2005). The term 'executive functioning' is used as a blanket term referring to a set of abilities that allow individuals to achieve goal-oriented behavior. These aspects of behavior can be regarded as top-down processes, in contrast to bottom-up processes that only represent stimulus-driven processing. Strauss et al. (2006) defined executive functioning as a collection of processes that are responsible for guiding, directing, and managing cognitive, emotional and behavioral functions, particularly during active, novel problem solving. As PD progresses, more severe cognitive impairments or dementia can occur (Aarsland et al., 2003). The dementia in PD exhibits normal or only slightly decreased performance in gnosis and praxis functions, and is typically characterized by a progressive dysexecutive syndrome with

In addition, it has repeatedly been shown that language functions in PD patients with dementia are affected. Demented PD patients show reduced verbal fluency, poor confrontation naming abilities, decreased word list generation, and difficulties in wordfinding (Dubois & Pillon, 1997; Pahwa et al., 1998). However, prior to dementia, PD patients also evidence subtle language impairments. The question whether the language system itself is impaired, as for example in aphasia, or whether language performance is disrupted because of non-linguistic executive function disorders in PD is still unanswered. We assume that, intact executive functioning is a prerequisite for normal language functioning. Therefore, language processing deficits in PD will always be associated with executive function deficits. Under this view, the language faculty is not considered to be totally modular in nature, but thought to depend on other cognitive functions, since, for example, comprehending a sentence demands that a listener flexibly guides his/her attention to relevant linguistic information, maintains information in working memory during the

disturbed memory functions and attention (Dubois & Pillon, 1997).

**1. Introduction** 

**Disease Patients Without Dementia** 

Katrien Colman and Roelien Bastiaanse

*University of Groningen* 

*The Netherlands* 

Yasui, M.; Kihira, T. & Ota, K. (1992). Calcium, magnesium and aluminum concentrations in Parkinson's disease. *Neurotoxicol*., Vol.13, pp. 593-600. **8** 

## **Language Processing in Parkinson's Disease Patients Without Dementia**

Katrien Colman and Roelien Bastiaanse *University of Groningen The Netherlands* 

## **1. Introduction**

164 Diagnostics and Rehabilitation of Parkinson's Disease

Yasui, M.; Kihira, T. & Ota, K. (1992). Calcium, magnesium and aluminum concentrations in

One of the major pathophysiological features in Parkinson's disease, from now on referred to as PD, is the loss of dopaminergic neurons in the substantia nigra, which in turn results in dysfunction of the cortico-striato-cortical circuits (Bartels & Leenders, 2009). In PD the components of the cortico-striato-cortical circuits are not in an optimal interaction, leading to insufficient engagement of for example the frontal and prefrontal lobes. Motor symptoms of tremor, bradykinesia, and rigidity are the clinical hallmark of PD (Wolters & Bosboom, 2007), however, non-motor symptoms are often present (Dubois & Pillon, 1995, 1997). In particular cognitive impairments in the domain of executive functioning have frequently been observed, both in late and also in very early stages of PD (Muslimovic et al., 2005). The term 'executive functioning' is used as a blanket term referring to a set of abilities that allow individuals to achieve goal-oriented behavior. These aspects of behavior can be regarded as top-down processes, in contrast to bottom-up processes that only represent stimulus-driven processing. Strauss et al. (2006) defined executive functioning as a collection of processes that are responsible for guiding, directing, and managing cognitive, emotional and behavioral functions, particularly during active, novel problem solving. As PD progresses, more severe cognitive impairments or dementia can occur (Aarsland et al., 2003). The dementia in PD exhibits normal or only slightly decreased performance in gnosis and praxis functions, and is typically characterized by a progressive dysexecutive syndrome with disturbed memory functions and attention (Dubois & Pillon, 1997).

In addition, it has repeatedly been shown that language functions in PD patients with dementia are affected. Demented PD patients show reduced verbal fluency, poor confrontation naming abilities, decreased word list generation, and difficulties in wordfinding (Dubois & Pillon, 1997; Pahwa et al., 1998). However, prior to dementia, PD patients also evidence subtle language impairments. The question whether the language system itself is impaired, as for example in aphasia, or whether language performance is disrupted because of non-linguistic executive function disorders in PD is still unanswered. We assume that, intact executive functioning is a prerequisite for normal language functioning. Therefore, language processing deficits in PD will always be associated with executive function deficits. Under this view, the language faculty is not considered to be totally modular in nature, but thought to depend on other cognitive functions, since, for example, comprehending a sentence demands that a listener flexibly guides his/her attention to relevant linguistic information, maintains information in working memory during the

Language Processing in Parkinson's Disease Patients Without Dementia 167

two steps, the Formulator translates this pre-verbal message into a linguistic structure. In a first step, the Grammatical Encoder must access lemma information1 from the mental lexicon (i.e., declarative knowledge) and activate syntactic building procedures stored in the Grammatical Encoder (i.e., procedural knowledge). Based on the properties of the message, the Grammatical Encoder will assign grammatical functions to the words and build a phrasal representation (e.g., verb phrases or noun phrases), specifying the hierarchical relation between syntactic constituents and their linear order. In a second step, the Phonological Encoder fills in the word forms in the structure that was generated by the Grammatical Encoder. It then constructs a phonetic plan, which is transformed into a spoken utterance by the Articulator. Formulation is "a largely automatic process" (Levelt, 1989, p. 21), implying that lexical retrieval and syntactic planning during production do not rely much on executive functions. However, declarative and procedural memory are both not disconnected from executive functions. For example, during the course of syntactic structure building the selected lemmas from declarative memory need to be maintained and

On the right-hand side in Fig. 1, the Speech Comprehension System is depicted. During comprehension, a spoken utterance is mapped to a phonetic string by the Audition component, from which the Speech Comprehension System computes parsed speech, a representation of the input in terms of phonological, morphological, syntactic, and semantic composition. This representation is further processed by the Conceptualizer. Sentence parsing during comprehension is constrained by working memory capacity (Caplan &

Speakers inspect their overt and covert speech for errors, thereby allowing themselves to inhibit and repair erroneous utterances. As Levelt (1989, p. 13) says, "a speaker is his own listener". Levelt localizes the central Monitor in the Conceptualizer (see Fig. 1). Very much simplified, Levelt's framework proposes that during language production the speaker monitors production through the Comprehension module. This proposal is known as the 'perceptual loop theory of speech monitoring', and claims that a speaker's phonetic plan is processed by the Speech Comprehension System during speech production, which allows the speaker to compare the comprehension of what he is about to say ('the internal loop') to what he originally intended to express. Speakers are also hypothesized to listen to their own overt speech, giving them another chance to detect errors ('the external loop'). In that case, they use the Audition component to analyze their own speech. Both feedback loops will reach the Monitor located in the Conceptualizer, which checks whether the parsed speech matches the intended speech. Upon error detection, the Monitor signals the speech production system to interrupt speech and to plan a repair process. The Monitor in Levelt's framework has been described as being a central, conscious process that oversees end products of speech production (Postma, 2000). Analogously to the monitoring system in speech production, Van Herten (2006), Van Herten et al. (2006), Vissers (2008), and Van de Meerendonk et al. (2009) proposed a monitoring process during comprehension inspired by the conflict monitoring theory of Botvinick et al. (2001). In the same line, Kuperberg (2007) suggested a monitoring process embedded in her non-syntactocentric, dynamic model of

1 The lemma of a word contains the semantic and syntactic information, necessary for the construction of the syntactic structure of the sentence. A lemma is still very abstract and distinct from the word

Waters, 1999; Just & Carpenter, 1992; Just et al., 1996; Waters & Caplan, 1996).

updated by executive functions until the process is terminated.

language processing.

forms, that are stored at a different level in the Lexicon

incremental development of the sentence interpretation and inhibits prepotent or incorrect parsing. This raises the question which aspect(s) of executive functioning are most important for language comprehension.

The studies described in this review reported on PD patients' production and comprehension in several languages (English, French, German, Greek and Dutch). From the literature it is clear that PD disrupts the processes involved in both language production and language comprehension.

In the present chapter, we will use Levelt's framework for sentence processing (1983, 1989) to clarify production and comprehension of spoken language. This includes implementation of the distinction between controlled and automatic cognitive processing. Figure 1 depicts Levelt's "Blueprint for the speaker" and shows the complex architecture of the various processes involved in speech production and comprehension.

Fig. 1. Blueprint for the speaker (Adapted from Levelt, 1989)

In this figure, the boxes represent processing components and the circle as well as the ellipse represent knowledge stores. The framework consists of two subsystems, one for production and one for comprehension. The Production System is further divided into a Conceptualizer, a Formulator and an Articulator. When a speaker produces speech, he starts with an idea that he intends to communicate in the Conceptualizer. Conceptualizing demands working memory (Levelt, 1989), since during this stage an intention needs to be conceived and relevant information needs to be retrieved from long-term memory and ordered while keeping track of the discourse. In short, the Conceptualizer provides an interface between thought and language and produces a pre-verbal message. Then, using

incremental development of the sentence interpretation and inhibits prepotent or incorrect parsing. This raises the question which aspect(s) of executive functioning are most

The studies described in this review reported on PD patients' production and comprehension in several languages (English, French, German, Greek and Dutch). From the literature it is clear that PD disrupts the processes involved in both language production and

In the present chapter, we will use Levelt's framework for sentence processing (1983, 1989) to clarify production and comprehension of spoken language. This includes implementation of the distinction between controlled and automatic cognitive processing. Figure 1 depicts Levelt's "Blueprint for the speaker" and shows the complex architecture of the various

important for language comprehension.

processes involved in speech production and comprehension.

Fig. 1. Blueprint for the speaker (Adapted from Levelt, 1989)

In this figure, the boxes represent processing components and the circle as well as the ellipse represent knowledge stores. The framework consists of two subsystems, one for production and one for comprehension. The Production System is further divided into a Conceptualizer, a Formulator and an Articulator. When a speaker produces speech, he starts with an idea that he intends to communicate in the Conceptualizer. Conceptualizing demands working memory (Levelt, 1989), since during this stage an intention needs to be conceived and relevant information needs to be retrieved from long-term memory and ordered while keeping track of the discourse. In short, the Conceptualizer provides an interface between thought and language and produces a pre-verbal message. Then, using

language comprehension.

two steps, the Formulator translates this pre-verbal message into a linguistic structure. In a first step, the Grammatical Encoder must access lemma information1 from the mental lexicon (i.e., declarative knowledge) and activate syntactic building procedures stored in the Grammatical Encoder (i.e., procedural knowledge). Based on the properties of the message, the Grammatical Encoder will assign grammatical functions to the words and build a phrasal representation (e.g., verb phrases or noun phrases), specifying the hierarchical relation between syntactic constituents and their linear order. In a second step, the Phonological Encoder fills in the word forms in the structure that was generated by the Grammatical Encoder. It then constructs a phonetic plan, which is transformed into a spoken utterance by the Articulator. Formulation is "a largely automatic process" (Levelt, 1989, p. 21), implying that lexical retrieval and syntactic planning during production do not rely much on executive functions. However, declarative and procedural memory are both not disconnected from executive functions. For example, during the course of syntactic structure building the selected lemmas from declarative memory need to be maintained and updated by executive functions until the process is terminated.

On the right-hand side in Fig. 1, the Speech Comprehension System is depicted. During comprehension, a spoken utterance is mapped to a phonetic string by the Audition component, from which the Speech Comprehension System computes parsed speech, a representation of the input in terms of phonological, morphological, syntactic, and semantic composition. This representation is further processed by the Conceptualizer. Sentence parsing during comprehension is constrained by working memory capacity (Caplan & Waters, 1999; Just & Carpenter, 1992; Just et al., 1996; Waters & Caplan, 1996).

Speakers inspect their overt and covert speech for errors, thereby allowing themselves to inhibit and repair erroneous utterances. As Levelt (1989, p. 13) says, "a speaker is his own listener". Levelt localizes the central Monitor in the Conceptualizer (see Fig. 1). Very much simplified, Levelt's framework proposes that during language production the speaker monitors production through the Comprehension module. This proposal is known as the 'perceptual loop theory of speech monitoring', and claims that a speaker's phonetic plan is processed by the Speech Comprehension System during speech production, which allows the speaker to compare the comprehension of what he is about to say ('the internal loop') to what he originally intended to express. Speakers are also hypothesized to listen to their own overt speech, giving them another chance to detect errors ('the external loop'). In that case, they use the Audition component to analyze their own speech. Both feedback loops will reach the Monitor located in the Conceptualizer, which checks whether the parsed speech matches the intended speech. Upon error detection, the Monitor signals the speech production system to interrupt speech and to plan a repair process. The Monitor in Levelt's framework has been described as being a central, conscious process that oversees end products of speech production (Postma, 2000). Analogously to the monitoring system in speech production, Van Herten (2006), Van Herten et al. (2006), Vissers (2008), and Van de Meerendonk et al. (2009) proposed a monitoring process during comprehension inspired by the conflict monitoring theory of Botvinick et al. (2001). In the same line, Kuperberg (2007) suggested a monitoring process embedded in her non-syntactocentric, dynamic model of language processing.

<sup>1</sup> The lemma of a word contains the semantic and syntactic information, necessary for the construction of the syntactic structure of the sentence. A lemma is still very abstract and distinct from the word forms, that are stored at a different level in the Lexicon

Language Processing in Parkinson's Disease Patients Without Dementia 169

other cognitive deficits. This conversational speech analysis showed that changes in language production in PD reflect concomitant cognitive and motor speech impairments, rather than being a pure language deficit. Ellis et al. (2006, see also Ellis, 2006 and Ellis & Rosenbek, 2007) analyzed narrative discourse in individuals with PD and in healthy control speakers. According to Ellis et al. (2006) the analysis of narrative discourse is as a method to differentially characterize expressive language form versus use2. They concluded that patients with mild to moderate PD demonstrate deficits in language use while maintaining

Earlier, McNamara et al. (1992) suggested that mildly to moderately impaired PD patients have a reduced capacity to simultaneously speak and monitor one's own speech resulting in self-monitoring impairments during narrative discourse. To test overt speech monitoring in narrative discourse of patients with PD, they used the procedures of the Cookie Theft picture description task of the Boston Diagnostic Aphasia Examination (BDAE, Goodglass & Kaplan, 1972). The number and the distribution of uncorrected errors and two repair types were tallied. The results showed that PD patients made three times more errors than the age-matched control speakers and used both repair strategies, but relatively less often than the control speakers. According to the authors, this significant unawareness of speech errors is related to attentional dysfunctioning in PD. They furthermore suggested that PD patients display reduced sensitivity to context, which may complicate their language comprehension. In order to explicitly evaluate PD patients' pragmatic skills3, McNamara and Durso (2003) used a formal pragmatic communication skills protocol (Prutting & Kirchner, 1987). The pragmatic communication skills were also rated on the basis of the assessment of (self-)awareness of the problem by individual PD patients and their spouses. It was concluded that PD patients were significantly impaired on measures of pragmatic communication abilities and were less aware of their communication problems. In line with Levelt's framework (1989, see Fig. 1) it is concluded that PD patients have a problem in their monitoring system and, thus, are not aware of their errors or, in other words, do not detect

the mismatch when comparing their intentions and the actual speech output.

the grammar of the language, while language use is akin to the area of pragmatics.

In 1997, Ullman and colleagues obtained evidence for a role of the basal ganglia in morphosyntactic production. Ullman et al. (1997) reported the results of a sentence completion task, which required the participants to read aloud randomly ordered sentence pairs and to fill in a past tensed verb. The authors found a correlation between right-side hypokinesia and the impaired production of rule-generated (regular) past tense forms in PD. The authors concluded that PD leads to the suppression of both motor activity and grammatical rule application. In essence, Ullman et al. (1997) and Ullman (2001) proposed that the frontal basal ganglia system, which is damaged in PD, constitutes the procedural memory system that regulates grammar (Grammatical Encoder in Fig. 1) and that the

2 In defining 'what is language', Bloom and Lahey (1978) divided language into three different, but overlapping aspects: content, form and use. In brief, language content includes factors such as semantics, including word knowledge and world knowledge, and vocabulary. Language form refers to

3 Pragmatic skills involve the ability to use and interpret verbal and nonverbal language appropriately within the social context in which communication occurs, requiring a degree of inference and

**2.2 Verb production in sentence context** 

interpretation (Perkins, 2005).

spared language form.

This chapter presents an overview of the extensive and still growing literature examining the underlying mechanisms of the subtle language impairments in non-demented PD patients. The connectivity of the basal ganglia with especially the frontal cortical regions explains why language processing is a vulnerable cognitive function in the course of PD. We start with reviewing what is known about language production deficits in non-demented patients with PD, followed by a summary of the receptive language deficits in PD. This review will not be limited to deficits at the sentence level, but will also consider deficits at the word and discourse level. Over the years, a variety of methodologies have been used, and recently functional imaging in PD patients has begun to add information to the neural instantiation of the patients' language impairments. Studying language processing in PD allows researchers to analyze the effects of poorly functioning, yet still engaged corticostriato-cortical circuitry during language performance. Some of the studies reviewed in this chapter aimed at examining language processing in PD, to ultimately define the role of the basal ganglia in language processing (e.g., Ullman et al., 1997; Friederici et al., 2003; Grossman et al., 2003; Kotz et al., 2003). In the final section of this chapter, advice for communication guidelines that would guarantee a better quality of life for patients suffering from PD is given. The chapter will be concluded with suggestions for future research on language processing in PD.

## **2. Language production in PD**

#### **2.1 Spontaneous speech**

Spontaneous speech in PD patients is often characterized by hypokinetic dysarthria and hypophonia, joined in the term 'dysarthrophonia' (Ackermann & Ziegler, 1989). Some PD patients in the advanced disease stage produce repetitions of speech, which are also labeled as stuttering, speech iterations, or palilalia (Benke et al., 2000). The major complaints reported by PD patients are not as much related to the acoustic, perceptual and physiological changes to their speech, but are related to the effect of these changes on communication overall, their view of themselves and the detrimental effects of the effort required to overcome physical and mental limitations (Miller et al., 2006). Also, PD patients' prosody, facial expression and gestures are abnormal, probably because these are influenced by the cardinal motor impairment.

One of the focuses of this review on language processing is grammatical effects in the spontaneous speech of PD patients, which were first reported by Illes et al. (1988). The sentences produced by the moderately impaired PD patients were syntactically simple. The pattern may reflect an adaptive, compensatory mechanism to reduce speech-motor difficulty, or may actually be evidence of a language impairment intrinsic to the disease process. Illes and colleagues (Illes, 1989; Illes et al., 1988) favored the adaptation hypothesis, stating that as the severity of the disease and, hence, the dysarthria increases, PD patients adapt to or compensate for their motor speech difficulties. Using a verbal picture description task, Murray (2000) observed compromised grammar and informativeness of spoken language in PD patients. Furthermore, a relationship between syntactic changes in production and concomitant cognitive changes was found. While analyzing conversational speech, Murray and Lenz (2001) found that patients with greater cognitive deficits and dysarthria performed more poorly on syntactic measures than patients with either more intact cognitive abilities or more intelligible speech. They suggested that PD patients show syntax limitations in production, but only under certain task requirements or related to

This chapter presents an overview of the extensive and still growing literature examining the underlying mechanisms of the subtle language impairments in non-demented PD patients. The connectivity of the basal ganglia with especially the frontal cortical regions explains why language processing is a vulnerable cognitive function in the course of PD. We start with reviewing what is known about language production deficits in non-demented patients with PD, followed by a summary of the receptive language deficits in PD. This review will not be limited to deficits at the sentence level, but will also consider deficits at the word and discourse level. Over the years, a variety of methodologies have been used, and recently functional imaging in PD patients has begun to add information to the neural instantiation of the patients' language impairments. Studying language processing in PD allows researchers to analyze the effects of poorly functioning, yet still engaged corticostriato-cortical circuitry during language performance. Some of the studies reviewed in this chapter aimed at examining language processing in PD, to ultimately define the role of the basal ganglia in language processing (e.g., Ullman et al., 1997; Friederici et al., 2003; Grossman et al., 2003; Kotz et al., 2003). In the final section of this chapter, advice for communication guidelines that would guarantee a better quality of life for patients suffering from PD is given. The chapter will be concluded with suggestions for future research on

Spontaneous speech in PD patients is often characterized by hypokinetic dysarthria and hypophonia, joined in the term 'dysarthrophonia' (Ackermann & Ziegler, 1989). Some PD patients in the advanced disease stage produce repetitions of speech, which are also labeled as stuttering, speech iterations, or palilalia (Benke et al., 2000). The major complaints reported by PD patients are not as much related to the acoustic, perceptual and physiological changes to their speech, but are related to the effect of these changes on communication overall, their view of themselves and the detrimental effects of the effort required to overcome physical and mental limitations (Miller et al., 2006). Also, PD patients' prosody, facial expression and gestures are abnormal, probably because these are influenced

One of the focuses of this review on language processing is grammatical effects in the spontaneous speech of PD patients, which were first reported by Illes et al. (1988). The sentences produced by the moderately impaired PD patients were syntactically simple. The pattern may reflect an adaptive, compensatory mechanism to reduce speech-motor difficulty, or may actually be evidence of a language impairment intrinsic to the disease process. Illes and colleagues (Illes, 1989; Illes et al., 1988) favored the adaptation hypothesis, stating that as the severity of the disease and, hence, the dysarthria increases, PD patients adapt to or compensate for their motor speech difficulties. Using a verbal picture description task, Murray (2000) observed compromised grammar and informativeness of spoken language in PD patients. Furthermore, a relationship between syntactic changes in production and concomitant cognitive changes was found. While analyzing conversational speech, Murray and Lenz (2001) found that patients with greater cognitive deficits and dysarthria performed more poorly on syntactic measures than patients with either more intact cognitive abilities or more intelligible speech. They suggested that PD patients show syntax limitations in production, but only under certain task requirements or related to

language processing in PD.

**2.1 Spontaneous speech** 

**2. Language production in PD** 

by the cardinal motor impairment.

other cognitive deficits. This conversational speech analysis showed that changes in language production in PD reflect concomitant cognitive and motor speech impairments, rather than being a pure language deficit. Ellis et al. (2006, see also Ellis, 2006 and Ellis & Rosenbek, 2007) analyzed narrative discourse in individuals with PD and in healthy control speakers. According to Ellis et al. (2006) the analysis of narrative discourse is as a method to differentially characterize expressive language form versus use2. They concluded that patients with mild to moderate PD demonstrate deficits in language use while maintaining spared language form.

Earlier, McNamara et al. (1992) suggested that mildly to moderately impaired PD patients have a reduced capacity to simultaneously speak and monitor one's own speech resulting in self-monitoring impairments during narrative discourse. To test overt speech monitoring in narrative discourse of patients with PD, they used the procedures of the Cookie Theft picture description task of the Boston Diagnostic Aphasia Examination (BDAE, Goodglass & Kaplan, 1972). The number and the distribution of uncorrected errors and two repair types were tallied. The results showed that PD patients made three times more errors than the age-matched control speakers and used both repair strategies, but relatively less often than the control speakers. According to the authors, this significant unawareness of speech errors is related to attentional dysfunctioning in PD. They furthermore suggested that PD patients display reduced sensitivity to context, which may complicate their language comprehension. In order to explicitly evaluate PD patients' pragmatic skills3, McNamara and Durso (2003) used a formal pragmatic communication skills protocol (Prutting & Kirchner, 1987). The pragmatic communication skills were also rated on the basis of the assessment of (self-)awareness of the problem by individual PD patients and their spouses. It was concluded that PD patients were significantly impaired on measures of pragmatic communication abilities and were less aware of their communication problems. In line with Levelt's framework (1989, see Fig. 1) it is concluded that PD patients have a problem in their monitoring system and, thus, are not aware of their errors or, in other words, do not detect the mismatch when comparing their intentions and the actual speech output.

#### **2.2 Verb production in sentence context**

In 1997, Ullman and colleagues obtained evidence for a role of the basal ganglia in morphosyntactic production. Ullman et al. (1997) reported the results of a sentence completion task, which required the participants to read aloud randomly ordered sentence pairs and to fill in a past tensed verb. The authors found a correlation between right-side hypokinesia and the impaired production of rule-generated (regular) past tense forms in PD. The authors concluded that PD leads to the suppression of both motor activity and grammatical rule application. In essence, Ullman et al. (1997) and Ullman (2001) proposed that the frontal basal ganglia system, which is damaged in PD, constitutes the procedural memory system that regulates grammar (Grammatical Encoder in Fig. 1) and that the

 2 In defining 'what is language', Bloom and Lahey (1978) divided language into three different, but overlapping aspects: content, form and use. In brief, language content includes factors such as semantics, including word knowledge and world knowledge, and vocabulary. Language form refers to the grammar of the language, while language use is akin to the area of pragmatics.

<sup>3</sup> Pragmatic skills involve the ability to use and interpret verbal and nonverbal language appropriately within the social context in which communication occurs, requiring a degree of inference and interpretation (Perkins, 2005).

Language Processing in Parkinson's Disease Patients Without Dementia 171

working memory capacity compared to healthy speakers, verb production was associated with working memory in PD patients. In healthy speakers, the production of verbs is a rather automatic language processing task, which is confirmed by the fact that no association was found between verb production and working memory in healthy controls. Automatic behavior is thought to be mediated by the basal ganglia (Saling & Phillips, 2007). Since PD is characterized by a dopaminergic dysfunction of the basal ganglia, we assume that PD patients cannot produce verbs in a rather automatic way as well as healthy speakers and, therefore, they need to rely more on their working memory, which we consider to be a

The tests of word fluency that were employed in the studies that will be discussed in the next paragraphs all test, apart from semantic memory, aspects of executive functioning. In a standard word fluency task the subjects are asked to name as many words as possible within a given semantic category (known as semantic or category fluency) or starting with a certain letter (known as phonemic or letter fluency) during a restricted time period. During an alternating fluency task, subjects have to generate words alternately using two fluency probes, which could either be from the same domain (i.e., letter-letter or category-category) or from different domains (i.e., category-letter). In a standard fluency task, planning abilities

are evaluated, while in an alternative fluency task, set shifting abilities are evaluated.

Impairments in non-demented PD patients have been reported in both semantic and phonemic fluency, but the most consistent finding is impaired performance in semantic fluency (e.g., Flowers et al., 1995; Grossman, Carvell et al., 1992, Grossman et al., 1993; Gurd

Henry and Crawford (2004) did a meta-analysis of 68 studies published between 1983 and 2002 which included more than 4600 PD participants. One of the aims was to find out if the word fluency deficit associated with PD predominantly reflects executive dysfunction, or problems with semantic memory, which is related to declarative memory. The outcome of the analysis was that, although PD was associated with deficits upon tests of phonemic and semantic fluency for studies that assessed both measures, the semantic fluency deficit was significantly larger than the phonemic fluency deficit. Moreover, since the confrontation naming deficit for the Boston Naming Test (BNT; Kaplan et al., 1983), a measure that imposes only minimal demands upon cognitive speed and effortful retrieval, was equivalent in magnitude to the deficits of these two types of fluency, Henry and Crawford concluded that PD is associated with a particular deficit in semantic memory. However, tests of alternating fluency were associated with slightly larger deficits than standard measures of fluency, which supports evidence for a specific deficit in cognitive set-shifting (Henry & Crawford, 2004). Some PD patients evidenced impairments in semantic knowledge, which correlate with their executive dysfunctions (Portin et al. 2000). The exact underlying nature

Interestingly, Auriacombe et al. (1993) examined the traditional semantic and phonemic fluency tasks, but also examined fluency performance in the non-verbal modality (i.e., design fluency and category drawing task). They found that PD patients' performance on the non-verbal fluency task was comparable to healthy speakers, and confirmed the discrepancy between relatively intact phonemic fluency and impaired semantic fluency. It is not necessary to retrieve a word form during category drawing, since knowledge of the concept underlying a target superordinate (i.e., vegetable) and the exemplars that contribute

compensatory mechanism.

**2.3 Single word production tasks** 

& Ward, 1989; Van Spaendonck et al., 1996).

of the semantic deficits has yet to be determined.

mental lexicon depends on declarative memory (see Fig. 1), embedded in the temporal lobe, which is largely intact in PD. Set in Levelt's framework (Fig. 1), it is proposed that PD patients have a deficit in grammatical encoding. As a result, PD patients are not able to produce the past tense form of regular verbs.

In the following years, the vast majority of studies on verbal morphosyntactic production in PD focused on testing the Declarative-Procedural hypothesis of Ullman et al. (1997), but the PD data of the Ullman study could not be replicated (Almor et al., 2002; Longworth et al., 2003; Longworth et al., 2005; Penke et al., 2005; Terzi et al., 2005). Longworth et al. (2005) found a tendency in English-speaking PD patients (among other patients with striatal damage) to perseverate on the cue (i.e., verb stem) rather than to produce past tense verbs as requested. Longworth et al. (2005) argued against an isolated grammatical deficit in PD and suggested that the striatum plays a general (i.e., not specific to language), inhibitory role in the later, controlled stages of language comprehension and production. The deficits in PD may reflect impairment of inhibition of competing alternatives during the later controlled processes involved in both comprehension and production (Longworth et al., 2005). Related is our evidence for executive dysfunctions being correlated to deficits in verb production in sentence context (Colman et al., 2009). Contrary to the findings of Ullman et al. (1997), but consistent with the findings of Longworth et al. (2005), no influence of regularity on verb production in sentence context was detected in the Dutch-speaking PD patients. In a study on verb production in sentence context, we showed that a deficit with regular inflection is not a characteristic for Dutch-speaking PD patients (Colman et al., 2009). We furthermore suggested that because of failing automaticity, PD patients relied more on the cortically represented executive functions. Unfortunately, due to the disturbed intimate relation between the basal ganglia and the frontal cortex, these executive functions are also dysfunctional. We manipulated the grammatical features of the test sentences, in order to simultaneously test verb retrieval and sentence integration processes in a group of PD patients compared to a control group consisting of age and education matched healthy participants. All subjects were assessed on both verb production in sentence context as well as on cognitive functions relevant for sentence processing. The verb production performances of the PD patients were correlated to their scores on executive function tasks. Analyses of PD patients' performance revealed that they have set-switching deficits and decreased sustained visual attention. The performance on verb production of PD patients was associated with the set-switching deficits, suggesting that PD patients who show poor set-switching have more difficulties with verb production. Many verb tense errors were made in sentences targeting the present tense. In our verb production task participants were instructed to inflect the verb in the past tense only in the presence of a temporal adverb referring to the past (e.g., 'yesterday') and in the present tense if the adverbial time phrase was absent. It is therefore suggested that the test materials and associated instructions provoked the tense errors. Due to the absence of a temporal adverb, PD patients were unable to switch to the present tense and showed 'stuck-in-set perseverations' which were evoked by the previous sentences. Evidence for self-monitoring deficits has earlier been reported by McNamara et al. (1992). While monitoring their performance, PD patients seemed to forget the instruction, especially in the longer subordinate sentences where working memory was challenged more than in the short main clauses. Hence, in Colman et al. (2009) set-switching impairments played a major role in performing the task assessing verb production in sentence context. These set-switching impairments reduce PD patients' performance seriously. Although the PD patients in our study did not show a decreased working memory capacity compared to healthy speakers, verb production was associated with working memory in PD patients. In healthy speakers, the production of verbs is a rather automatic language processing task, which is confirmed by the fact that no association was found between verb production and working memory in healthy controls. Automatic behavior is thought to be mediated by the basal ganglia (Saling & Phillips, 2007). Since PD is characterized by a dopaminergic dysfunction of the basal ganglia, we assume that PD patients cannot produce verbs in a rather automatic way as well as healthy speakers and, therefore, they need to rely more on their working memory, which we consider to be a compensatory mechanism.

#### **2.3 Single word production tasks**

170 Diagnostics and Rehabilitation of Parkinson's Disease

mental lexicon depends on declarative memory (see Fig. 1), embedded in the temporal lobe, which is largely intact in PD. Set in Levelt's framework (Fig. 1), it is proposed that PD patients have a deficit in grammatical encoding. As a result, PD patients are not able to

In the following years, the vast majority of studies on verbal morphosyntactic production in PD focused on testing the Declarative-Procedural hypothesis of Ullman et al. (1997), but the PD data of the Ullman study could not be replicated (Almor et al., 2002; Longworth et al., 2003; Longworth et al., 2005; Penke et al., 2005; Terzi et al., 2005). Longworth et al. (2005) found a tendency in English-speaking PD patients (among other patients with striatal damage) to perseverate on the cue (i.e., verb stem) rather than to produce past tense verbs as requested. Longworth et al. (2005) argued against an isolated grammatical deficit in PD and suggested that the striatum plays a general (i.e., not specific to language), inhibitory role in the later, controlled stages of language comprehension and production. The deficits in PD may reflect impairment of inhibition of competing alternatives during the later controlled processes involved in both comprehension and production (Longworth et al., 2005). Related is our evidence for executive dysfunctions being correlated to deficits in verb production in sentence context (Colman et al., 2009). Contrary to the findings of Ullman et al. (1997), but consistent with the findings of Longworth et al. (2005), no influence of regularity on verb production in sentence context was detected in the Dutch-speaking PD patients. In a study on verb production in sentence context, we showed that a deficit with regular inflection is not a characteristic for Dutch-speaking PD patients (Colman et al., 2009). We furthermore suggested that because of failing automaticity, PD patients relied more on the cortically represented executive functions. Unfortunately, due to the disturbed intimate relation between the basal ganglia and the frontal cortex, these executive functions are also dysfunctional. We manipulated the grammatical features of the test sentences, in order to simultaneously test verb retrieval and sentence integration processes in a group of PD patients compared to a control group consisting of age and education matched healthy participants. All subjects were assessed on both verb production in sentence context as well as on cognitive functions relevant for sentence processing. The verb production performances of the PD patients were correlated to their scores on executive function tasks. Analyses of PD patients' performance revealed that they have set-switching deficits and decreased sustained visual attention. The performance on verb production of PD patients was associated with the set-switching deficits, suggesting that PD patients who show poor set-switching have more difficulties with verb production. Many verb tense errors were made in sentences targeting the present tense. In our verb production task participants were instructed to inflect the verb in the past tense only in the presence of a temporal adverb referring to the past (e.g., 'yesterday') and in the present tense if the adverbial time phrase was absent. It is therefore suggested that the test materials and associated instructions provoked the tense errors. Due to the absence of a temporal adverb, PD patients were unable to switch to the present tense and showed 'stuck-in-set perseverations' which were evoked by the previous sentences. Evidence for self-monitoring deficits has earlier been reported by McNamara et al. (1992). While monitoring their performance, PD patients seemed to forget the instruction, especially in the longer subordinate sentences where working memory was challenged more than in the short main clauses. Hence, in Colman et al. (2009) set-switching impairments played a major role in performing the task assessing verb production in sentence context. These set-switching impairments reduce PD patients' performance seriously. Although the PD patients in our study did not show a decreased

produce the past tense form of regular verbs.

The tests of word fluency that were employed in the studies that will be discussed in the next paragraphs all test, apart from semantic memory, aspects of executive functioning. In a standard word fluency task the subjects are asked to name as many words as possible within a given semantic category (known as semantic or category fluency) or starting with a certain letter (known as phonemic or letter fluency) during a restricted time period. During an alternating fluency task, subjects have to generate words alternately using two fluency probes, which could either be from the same domain (i.e., letter-letter or category-category) or from different domains (i.e., category-letter). In a standard fluency task, planning abilities are evaluated, while in an alternative fluency task, set shifting abilities are evaluated.

Impairments in non-demented PD patients have been reported in both semantic and phonemic fluency, but the most consistent finding is impaired performance in semantic fluency (e.g., Flowers et al., 1995; Grossman, Carvell et al., 1992, Grossman et al., 1993; Gurd & Ward, 1989; Van Spaendonck et al., 1996).

Henry and Crawford (2004) did a meta-analysis of 68 studies published between 1983 and 2002 which included more than 4600 PD participants. One of the aims was to find out if the word fluency deficit associated with PD predominantly reflects executive dysfunction, or problems with semantic memory, which is related to declarative memory. The outcome of the analysis was that, although PD was associated with deficits upon tests of phonemic and semantic fluency for studies that assessed both measures, the semantic fluency deficit was significantly larger than the phonemic fluency deficit. Moreover, since the confrontation naming deficit for the Boston Naming Test (BNT; Kaplan et al., 1983), a measure that imposes only minimal demands upon cognitive speed and effortful retrieval, was equivalent in magnitude to the deficits of these two types of fluency, Henry and Crawford concluded that PD is associated with a particular deficit in semantic memory. However, tests of alternating fluency were associated with slightly larger deficits than standard measures of fluency, which supports evidence for a specific deficit in cognitive set-shifting (Henry & Crawford, 2004). Some PD patients evidenced impairments in semantic knowledge, which correlate with their executive dysfunctions (Portin et al. 2000). The exact underlying nature of the semantic deficits has yet to be determined.

Interestingly, Auriacombe et al. (1993) examined the traditional semantic and phonemic fluency tasks, but also examined fluency performance in the non-verbal modality (i.e., design fluency and category drawing task). They found that PD patients' performance on the non-verbal fluency task was comparable to healthy speakers, and confirmed the discrepancy between relatively intact phonemic fluency and impaired semantic fluency. It is not necessary to retrieve a word form during category drawing, since knowledge of the concept underlying a target superordinate (i.e., vegetable) and the exemplars that contribute

Language Processing in Parkinson's Disease Patients Without Dementia 173

One explanation for the discrepancy between verb and noun retrieval is that verb retrieval is more demanding than noun retrieval in terms of executive functioning (e.g., Péran et al., 2003; Piatt et al., 1999a and 1999b). The idea is that retrieving the name of an object elicits a more automatic lexical retrieval response than retrieval of the action name, which demands a more controlled retrieval. In other words, impaired action naming is seen as a result of executive function impairment. According to Levelt's framework (see Fig. 1), the lemmas contain information about word meaning, and word class. The lemmas of verbs additionally contain information on thematic roles, argument structure, and subcategorisation frame. Comparable to what was found for individuals with Broca's aphasia (Bastiaanse & Van Zonneveld, 2004), we suggest that for PD patients verbs are more difficult to produce than nouns, because verb lemmas contain simply more grammatical information than noun

An alternative hypothesis for the discrepancy between verb and noun retrieval is that the link between representation of action words and representation of motor acts per se in the human motor and premotor cortex is damaged, leading to verb retrieval problems. The existence of a similar verb-naming deficit in other motor disorders, such as corticobasal degeneration (CBD) and progressive supranuclear palsy (PSP) (Cotelli et al., 2006), has provided a major argument for the idea that semantic mechanisms concerning the verb are grounded in the motor system of the brain. To test whether the motor system comes into play during the processing of verbs, Boulenger et al. (2008) compared lexical decision latencies for action verbs and concrete nouns of non-demented PD patients (off and on dopaminergic medication) using a masked priming paradigm. Priming effects for action verbs, but not for concrete nouns, were nearly absent in PD patients off treatment, confirming that processing lexico-semantic information of action words depends on the integrity of the motor system. As a follow up to their earlier French verb generation task, Péran et al. (2009) explored the relationship between the motor deficit in PD patients and brain activation in noun and verb generation tasks conducting a functional neuroimaging study. Although they did not find differences between the brain activity during the production of object-related action words and of object names, they did observe a clear relationship between brain activity and the severity of the motor deficit (as assessed by the Unified Parkinson's Disease Rating Scale (UPDRS), Fahn et al., 1987) in PD. This relation was particularly found during generation of action verbs in response to manipulable biological objects, in the pre- and post-central gyri bilaterally, left frontal operculum, left supplementary motor area and right superior temporal cortex. The impairment in the motor cortico-striato-cortical circuits in PD may result in the recruitment of a wider cortical network designed to alleviate the disturbed motor representations during the demanding

In the following section, receptive language functions in PD, with a particular focus on

From the early nineties of the last century, off-line tasks such as sentence-to-picture matching and grammaticality judgment have revealed that comprehension of complex syntactic structures (i.e., non-canonical structures such as passives) is vulnerable in

generation of action verbs in response to manipulable objects.

sentence comprehension of non-canonical sentences, will be discussed.

**3. Receptive language functions in PD** 

**3.1 Comprehension of non-canonical sentences** 

lemmas.

to a superordinate is sufficient. To check the hypothesis that PD patients are impaired in the retrieval of semantic information, Auriacombe et al. (1993) also administered a supraspan verbal learning task. A large proportion of the PD patients showed difficulties with free recall, but these patients were accurate at recognition, which is consistent with a retrieval deficit, and not an impairment of semantic memory itself. PD patients thus have difficulties retrieving the phonological form that is the label of an exemplar (Levelt et al., 1991).

In addition, in PD, action naming is often found to be more impaired than object naming (Bertella et al., 2002; Cotelli et al., 2007), a phenomenon also observed in agrammatic/Broca's aphasic patients. Related to this, Signorini and Volpato (2006) found that PD patients were impaired on an action fluency task but not on semantic and phonemic fluency tasks. However, analysis of spontaneous speech production of PD patients did not show the expected discrepancy between nouns and verbs, which supports the hypothesis that it is not the representation of verbs, but rather the utilization of the verb emerging under specific task demands that is troublesome (Pignatti et al., 2006). Moreover, verb fluency scores also discriminate between demented PD patients and non-demented PD patients and healthy elderly control participants, whereas tests of letter or category verbal fluency do not (Piatt et al., 1999a and 1999b). Piatt et al. (1999a, 1999b) concluded that verb fluency was particularly sensitive to the fronto-striatal pathophysiology of PD patients with dementia. According to these authors, verb fluency reflects the underlying integrity of frontal lobe circuitry, and problems on verbal fluency tasks could therefore indicate deficits in executive functioning.

Péran et al. (2003) developed a French word generation task that requires a semantic and grammar driven selection of single words over a limited time period. Compared to healthy control participants, non-demented PD patients made more grammatical errors in the noun-verb-generation task than in the verb-noun-generation task. Péran et al. (2003) suggested that this discrepancy was due to the combined effect of impaired set switching and a specific grammatical impairment in verb production. The authors assume that in the verb-noun task, the impact of impaired switching is compensated by the easier noun production, whereas in the noun-verb task both the switching and production of the verb were dysfunctional.

However, the argument that PD specifically affects verb processing was contradicted in a recent word generation study in PD conducted by Crescentini et al. (2008). Behavioral tasks already showed before that the Reaction Times (RTs) and accuracy of word generation both depend on the number of possible responses (response selection) and on the strength of association between cues and responses (associative strength) (Cheng & Martin, 2005; Martin & Cheng, 2006; Thompson-Schill & Botvinick, 2006). Based on these findings, Crescentini et al. (2008) controlled the response selection demands and association strength of the verb and the noun stimuli during a word generation task. The critical condition for PD patients was the one with a weak association between the stimulus and the response as opposed to the grammatical class. Crescentini et al. (2008) suggested that the verb generation problem in PD is caused by the fact that nouns are typically more associated with other nouns than with verbs in the semantic network. During the noun-verb condition, PD patients seem to have problems with both switching to task-relevant representations (i.e., verbs) and with inhibiting the task-irrelevant and more strongly activated options (i.e., nouns). Based on these findings, the authors proposed a non-language-specific involvement of the basal ganglia in the controlled rather than the routine semantic processes required during lexical retrieval.

to a superordinate is sufficient. To check the hypothesis that PD patients are impaired in the retrieval of semantic information, Auriacombe et al. (1993) also administered a supraspan verbal learning task. A large proportion of the PD patients showed difficulties with free recall, but these patients were accurate at recognition, which is consistent with a retrieval deficit, and not an impairment of semantic memory itself. PD patients thus have difficulties

In addition, in PD, action naming is often found to be more impaired than object naming (Bertella et al., 2002; Cotelli et al., 2007), a phenomenon also observed in agrammatic/Broca's aphasic patients. Related to this, Signorini and Volpato (2006) found that PD patients were impaired on an action fluency task but not on semantic and phonemic fluency tasks. However, analysis of spontaneous speech production of PD patients did not show the expected discrepancy between nouns and verbs, which supports the hypothesis that it is not the representation of verbs, but rather the utilization of the verb emerging under specific task demands that is troublesome (Pignatti et al., 2006). Moreover, verb fluency scores also discriminate between demented PD patients and non-demented PD patients and healthy elderly control participants, whereas tests of letter or category verbal fluency do not (Piatt et al., 1999a and 1999b). Piatt et al. (1999a, 1999b) concluded that verb fluency was particularly sensitive to the fronto-striatal pathophysiology of PD patients with dementia. According to these authors, verb fluency reflects the underlying integrity of frontal lobe circuitry, and problems on verbal fluency tasks could therefore indicate deficits

Péran et al. (2003) developed a French word generation task that requires a semantic and grammar driven selection of single words over a limited time period. Compared to healthy control participants, non-demented PD patients made more grammatical errors in the noun-verb-generation task than in the verb-noun-generation task. Péran et al. (2003) suggested that this discrepancy was due to the combined effect of impaired set switching and a specific grammatical impairment in verb production. The authors assume that in the verb-noun task, the impact of impaired switching is compensated by the easier noun production, whereas in the noun-verb task both the switching and production of the verb

However, the argument that PD specifically affects verb processing was contradicted in a recent word generation study in PD conducted by Crescentini et al. (2008). Behavioral tasks already showed before that the Reaction Times (RTs) and accuracy of word generation both depend on the number of possible responses (response selection) and on the strength of association between cues and responses (associative strength) (Cheng & Martin, 2005; Martin & Cheng, 2006; Thompson-Schill & Botvinick, 2006). Based on these findings, Crescentini et al. (2008) controlled the response selection demands and association strength of the verb and the noun stimuli during a word generation task. The critical condition for PD patients was the one with a weak association between the stimulus and the response as opposed to the grammatical class. Crescentini et al. (2008) suggested that the verb generation problem in PD is caused by the fact that nouns are typically more associated with other nouns than with verbs in the semantic network. During the noun-verb condition, PD patients seem to have problems with both switching to task-relevant representations (i.e., verbs) and with inhibiting the task-irrelevant and more strongly activated options (i.e., nouns). Based on these findings, the authors proposed a non-language-specific involvement of the basal ganglia in the controlled rather than the routine semantic processes required

retrieving the phonological form that is the label of an exemplar (Levelt et al., 1991).

in executive functioning.

were dysfunctional.

during lexical retrieval.

One explanation for the discrepancy between verb and noun retrieval is that verb retrieval is more demanding than noun retrieval in terms of executive functioning (e.g., Péran et al., 2003; Piatt et al., 1999a and 1999b). The idea is that retrieving the name of an object elicits a more automatic lexical retrieval response than retrieval of the action name, which demands a more controlled retrieval. In other words, impaired action naming is seen as a result of executive function impairment. According to Levelt's framework (see Fig. 1), the lemmas contain information about word meaning, and word class. The lemmas of verbs additionally contain information on thematic roles, argument structure, and subcategorisation frame. Comparable to what was found for individuals with Broca's aphasia (Bastiaanse & Van Zonneveld, 2004), we suggest that for PD patients verbs are more difficult to produce than nouns, because verb lemmas contain simply more grammatical information than noun lemmas.

An alternative hypothesis for the discrepancy between verb and noun retrieval is that the link between representation of action words and representation of motor acts per se in the human motor and premotor cortex is damaged, leading to verb retrieval problems. The existence of a similar verb-naming deficit in other motor disorders, such as corticobasal degeneration (CBD) and progressive supranuclear palsy (PSP) (Cotelli et al., 2006), has provided a major argument for the idea that semantic mechanisms concerning the verb are grounded in the motor system of the brain. To test whether the motor system comes into play during the processing of verbs, Boulenger et al. (2008) compared lexical decision latencies for action verbs and concrete nouns of non-demented PD patients (off and on dopaminergic medication) using a masked priming paradigm. Priming effects for action verbs, but not for concrete nouns, were nearly absent in PD patients off treatment, confirming that processing lexico-semantic information of action words depends on the integrity of the motor system. As a follow up to their earlier French verb generation task, Péran et al. (2009) explored the relationship between the motor deficit in PD patients and brain activation in noun and verb generation tasks conducting a functional neuroimaging study. Although they did not find differences between the brain activity during the production of object-related action words and of object names, they did observe a clear relationship between brain activity and the severity of the motor deficit (as assessed by the Unified Parkinson's Disease Rating Scale (UPDRS), Fahn et al., 1987) in PD. This relation was particularly found during generation of action verbs in response to manipulable biological objects, in the pre- and post-central gyri bilaterally, left frontal operculum, left supplementary motor area and right superior temporal cortex. The impairment in the motor cortico-striato-cortical circuits in PD may result in the recruitment of a wider cortical network designed to alleviate the disturbed motor representations during the demanding generation of action verbs in response to manipulable objects.

#### **3. Receptive language functions in PD**

In the following section, receptive language functions in PD, with a particular focus on sentence comprehension of non-canonical sentences, will be discussed.

#### **3.1 Comprehension of non-canonical sentences**

From the early nineties of the last century, off-line tasks such as sentence-to-picture matching and grammaticality judgment have revealed that comprehension of complex syntactic structures (i.e., non-canonical structures such as passives) is vulnerable in

Language Processing in Parkinson's Disease Patients Without Dementia 175

violations of grammatical agreements during on-line word detection. The comprehension impairment on the traditional measure in PD was argued to be related to impairments in inhibition and planning, emphasizing the important influence of task requirements on

In the same year another study by Grossman and colleagues was published, using a different on-line methodology, that is, a list priming task (Grossman, Zurif et al., 2002). Those PD patients who had problems comprehending sentences with a non-canonical structure when measured off-line (e.g., "The boy that the girl chased was friendly") showed delayed lexical retrieval during the priming task. This was reported earlier for Broca's

The Grossman group gained additional information on the connection between slowed lexical activation and sentence comprehension deficits in PD by applying the same word detection methodology as before, but by using a different violation type. Based on previous observation of PD patients' difficulty detecting phonetic errors in grammatical morphemes (Grossman, Carvell et al., 1992), the researchers tested phonetic errors in free grammatical morphemes and words as violation type (Lee et al., 2003). PD patients were insensitive to phonetic errors in free grammatical morphemes and showed a slowed sensitivity to words located in the non-canonical sentences. This delayed sensitivity was correlated with the measure of planning, which was seen as evidence for the fundamental contribution of executive functions to sentence comprehension. Lee et al. (2003) concluded that sentence comprehension impairments are due to limitations in specific executive resources such as attention to grammatical morphemes and delayed lexical retrieval of words, rather than

Hochstadt et al. (2006) conducted the first off-line study that also tested the interrelationship between the distinct executive functions. The authors concluded that limits on sequencing and/or verbal working memory (i.e., executive component and articulatory rehearsal) are responsible for the sentence comprehension deficits in PD. Later, Hochstadt (2009) used eye-tracking to minimize the extraneous executive demands during off-line sentence-picture matching. Some of the PD patients in this study showed difficulties comprehending passive sentences and they looked toward a distractor picture before giving a response. One of the proposed explanations by Hochstadt (2009) for the errors in passive sentences is the exaggerated agent-first bias pointing to a reliance on heuristics to compensate for impaired syntax processing. However, this explanation did not hold for passives in general, since there was no evidence that the bias differed between patients with

high and low error rates in final passive trials as compared to center passive trials.

To further explore the hypothesis that executive dysfunctions are involved in the comprehension deficits of passive sentences in PD (Lieberman et al., 1992), we recently tested Dutch-speaking PD patients on the comprehension of sentences that were varied for phrase structure complexity and sentence length (Colman, 2011) to see whether there was a relation between the processing of the sentences and relevant executive function deficits. In general, the PD patients showed slightly poorer sentence comprehension compared to healthy control participants. However, the difficulties encountered by PD patients were not limited to one specific grammatical aspect. Decreased set-switching, inhibition, and working memory abilities were all associated with comprehending non-canonical passive sentences, rather than one specific executive function being primarily associated with the comprehension difficulties. Deficits in sustained visual attention appear to underlie PD patients' overall comprehension performance, possibly due to the demands of the picture-

sentence comprehension in PD.

being a pure linguistic deficit.

aphasic patients (Swinney et al., 1996; Zurif et al., 1993).

individuals with PD (see Grossman, 1999 and Murray, 2008 for an extensive review). Sentences are defined as syntactically complex when the thematic roles are not in base (or canonical) word order and therefore require extra grammatical operations. The following examples are given of an active (a) and a passive (b) sentence in Dutch. Important to note is that base word order in Dutch is Subject-Object-Verb (SOV).


In passive constructions, the *grammatical roles* are in base order. In sentence (b) the subject ('the apples') precedes the finite verb ('are') which precedes the prepositional phrase ('by the children'). However, the *thematic (semantic) roles* are not in their base position. The theme ('apples'), precedes the finite verb, whereas the agent ('children') follows the finite verb.

Lieberman et al. (1990) were among the first to find a comprehension deficit that could not be attributed to compensatory motor strategies, which had been claimed to be responsible for the sentence production deficits in PD till then (Illes et al., 1988; Illes, 1989). Lieberman et al. (1990) attributed the sentence comprehension errors in PD to "some deterioration of the patient's ability to make use of the syntactic 'rules' involved in English" (1990, p. 364). Similarly, researchers have attributed the sentence comprehension deficit to an impairment of some aspects of grammatical processing as such (Cohen et al., 1994; Natsopoulos et al., 1991, 1993). However, according to Lieberman et al. (1990), the cognitive impairments and syntactic comprehension deficits in PD have a common physiological basis; they are both caused by disruption of the cortico-striato-cortical circuits. Lieberman et al. (1990, 1992) do not regard grammatical processing and executive functions as separate mechanisms. They take the position that syntax comprehension is achieved by the operations of non-domainspecific executive functions over language-specific knowledge. Consistent with this view, some researchers claim that it is not syntax itself, but rather the interaction with executive dysfunction that might reflect the sentence comprehension deficits in PD (see for example Colman, 2011; Colman et al., 2006; Geyer & Grossman, 1994; Grossman, Carvell et al., 1992; Hochstadt et al., 2006, 2009; Kemmerer, 1999; Lieberman et al., 1990, 1992). In addition, some researchers reported deficits in lexical-semantic processing during sentence comprehension. For example, Angwin et al. (2005) reported a general semantic processing deficit, but also reported that PD patients with comprehension deficits for non-canonical sentences showed a delayed time course of semantic activation. This finding added evidence to the proposal that slowed information processing is one of the causes of the sentence processing deficits in patients with PD (Grossman, Zurif et al., 2002; Lee et al., 2003).

Grossman, Lee et al. (2002) administered both a traditional off-line sentence processing task and an on-line word detection task to the same PD patients. Subjects were instructed to press a button as soon as they heard the target word in an auditorily presented sentence. Half of the sentences contained a grammatical agreement violation (e.g., subject-verb agreement violation) prior to the target word. In healthy persons, responses to the target word were slowed down when they immediately followed a morphosyntactic error. The offline measure of sentence comprehension required subjects to answer a simple question about a semantically unconstrained sentence. In addition to the language tasks, a battery of executive function tests was also run. Off-line, PD patients were significantly impaired on non-canonical sentences and their comprehension was correlated with the executive measures. However, PD patients and healthy control participants were equally sensitive to

individuals with PD (see Grossman, 1999 and Murray, 2008 for an extensive review). Sentences are defined as syntactically complex when the thematic roles are not in base (or canonical) word order and therefore require extra grammatical operations. The following examples are given of an active (a) and a passive (b) sentence in Dutch. Important to note is

In passive constructions, the *grammatical roles* are in base order. In sentence (b) the subject ('the apples') precedes the finite verb ('are') which precedes the prepositional phrase ('by the children'). However, the *thematic (semantic) roles* are not in their base position. The theme ('apples'), precedes the finite verb, whereas the agent ('children') follows the finite verb. Lieberman et al. (1990) were among the first to find a comprehension deficit that could not be attributed to compensatory motor strategies, which had been claimed to be responsible for the sentence production deficits in PD till then (Illes et al., 1988; Illes, 1989). Lieberman et al. (1990) attributed the sentence comprehension errors in PD to "some deterioration of the patient's ability to make use of the syntactic 'rules' involved in English" (1990, p. 364). Similarly, researchers have attributed the sentence comprehension deficit to an impairment of some aspects of grammatical processing as such (Cohen et al., 1994; Natsopoulos et al., 1991, 1993). However, according to Lieberman et al. (1990), the cognitive impairments and syntactic comprehension deficits in PD have a common physiological basis; they are both caused by disruption of the cortico-striato-cortical circuits. Lieberman et al. (1990, 1992) do not regard grammatical processing and executive functions as separate mechanisms. They take the position that syntax comprehension is achieved by the operations of non-domainspecific executive functions over language-specific knowledge. Consistent with this view, some researchers claim that it is not syntax itself, but rather the interaction with executive dysfunction that might reflect the sentence comprehension deficits in PD (see for example Colman, 2011; Colman et al., 2006; Geyer & Grossman, 1994; Grossman, Carvell et al., 1992; Hochstadt et al., 2006, 2009; Kemmerer, 1999; Lieberman et al., 1990, 1992). In addition, some researchers reported deficits in lexical-semantic processing during sentence comprehension. For example, Angwin et al. (2005) reported a general semantic processing deficit, but also reported that PD patients with comprehension deficits for non-canonical sentences showed a delayed time course of semantic activation. This finding added evidence to the proposal that slowed information processing is one of the causes of the sentence processing deficits in

Grossman, Lee et al. (2002) administered both a traditional off-line sentence processing task and an on-line word detection task to the same PD patients. Subjects were instructed to press a button as soon as they heard the target word in an auditorily presented sentence. Half of the sentences contained a grammatical agreement violation (e.g., subject-verb agreement violation) prior to the target word. In healthy persons, responses to the target word were slowed down when they immediately followed a morphosyntactic error. The offline measure of sentence comprehension required subjects to answer a simple question about a semantically unconstrained sentence. In addition to the language tasks, a battery of executive function tests was also run. Off-line, PD patients were significantly impaired on non-canonical sentences and their comprehension was correlated with the executive measures. However, PD patients and healthy control participants were equally sensitive to

that base word order in Dutch is Subject-Object-Verb (SOV).

patients with PD (Grossman, Zurif et al., 2002; Lee et al., 2003).

b. De appelsi worden door de kinderen ti geplukt 'The apples are by the children picked'

a. De kinderen plukken de appels 'The children pick the apples'

violations of grammatical agreements during on-line word detection. The comprehension impairment on the traditional measure in PD was argued to be related to impairments in inhibition and planning, emphasizing the important influence of task requirements on sentence comprehension in PD.

In the same year another study by Grossman and colleagues was published, using a different on-line methodology, that is, a list priming task (Grossman, Zurif et al., 2002). Those PD patients who had problems comprehending sentences with a non-canonical structure when measured off-line (e.g., "The boy that the girl chased was friendly") showed delayed lexical retrieval during the priming task. This was reported earlier for Broca's aphasic patients (Swinney et al., 1996; Zurif et al., 1993).

The Grossman group gained additional information on the connection between slowed lexical activation and sentence comprehension deficits in PD by applying the same word detection methodology as before, but by using a different violation type. Based on previous observation of PD patients' difficulty detecting phonetic errors in grammatical morphemes (Grossman, Carvell et al., 1992), the researchers tested phonetic errors in free grammatical morphemes and words as violation type (Lee et al., 2003). PD patients were insensitive to phonetic errors in free grammatical morphemes and showed a slowed sensitivity to words located in the non-canonical sentences. This delayed sensitivity was correlated with the measure of planning, which was seen as evidence for the fundamental contribution of executive functions to sentence comprehension. Lee et al. (2003) concluded that sentence comprehension impairments are due to limitations in specific executive resources such as attention to grammatical morphemes and delayed lexical retrieval of words, rather than being a pure linguistic deficit.

Hochstadt et al. (2006) conducted the first off-line study that also tested the interrelationship between the distinct executive functions. The authors concluded that limits on sequencing and/or verbal working memory (i.e., executive component and articulatory rehearsal) are responsible for the sentence comprehension deficits in PD. Later, Hochstadt (2009) used eye-tracking to minimize the extraneous executive demands during off-line sentence-picture matching. Some of the PD patients in this study showed difficulties comprehending passive sentences and they looked toward a distractor picture before giving a response. One of the proposed explanations by Hochstadt (2009) for the errors in passive sentences is the exaggerated agent-first bias pointing to a reliance on heuristics to compensate for impaired syntax processing. However, this explanation did not hold for passives in general, since there was no evidence that the bias differed between patients with high and low error rates in final passive trials as compared to center passive trials.

To further explore the hypothesis that executive dysfunctions are involved in the comprehension deficits of passive sentences in PD (Lieberman et al., 1992), we recently tested Dutch-speaking PD patients on the comprehension of sentences that were varied for phrase structure complexity and sentence length (Colman, 2011) to see whether there was a relation between the processing of the sentences and relevant executive function deficits. In general, the PD patients showed slightly poorer sentence comprehension compared to healthy control participants. However, the difficulties encountered by PD patients were not limited to one specific grammatical aspect. Decreased set-switching, inhibition, and working memory abilities were all associated with comprehending non-canonical passive sentences, rather than one specific executive function being primarily associated with the comprehension difficulties. Deficits in sustained visual attention appear to underlie PD patients' overall comprehension performance, possibly due to the demands of the picture-

Language Processing in Parkinson's Disease Patients Without Dementia 177

participants when dealing with non-canonical passive sentences. First, Carpenter et al. (1994) hypothesized that working memory load is directly related to sentence complexity. As mentioned before, higher sentence complexity is related to the non-canonical order of roles (such as in passives). All in all, non-canonical sentences impose a higher demand on working memory than canonical sentences (King & Just, 1991). The PD patients possibly relied more on their intact working memory allocated in the prefrontal cortex to compensate for their difficulties to process the non-canonical passive sentences (for a review see Wager & Smith, 2003). Secondly, examining ambiguity resolution, Stowe et al. (2004) reported a similar left medial prefrontal area as in our study, which they linked to supporting higherlevel semantic processes involved in evaluation of plausibility. Finally, it is suggested that the exclusive activation of the prefrontal cortex in PD patients for passive sentences reflects a lexical-semantic strategy for dealing with word order information, which was probably

Semantic priming tasks are a straightforward measure for the evaluation of lexical and semantic processes in patients with PD. In healthy participants, RTs to the target word are faster if the prime and the target are semantically related (doctor-NURSE) as compared to when the prime and target are not related (doctor-FLOWER) (see Neely, 1991 for extensive review on priming tasks). Copland (2003) found that PD patients are unable to suppress the infrequent meaning of homophones (bank-RIVER) and proposes therefore that the selective attentional engagement of the semantic network is impaired. Thus, PD compromises the controlled aspects of semantic processing rather than the automatic processes. During sentence comprehension tasks, lexical-semantic processing has been found to be abnormal in PD patients as well (Angwin et al., 2005). Angwin et al. (2004, 2006) also found that semantic processing deficits in PD are related to striatal dopamine deficiency since automatic semantic activation was compromised in PD patients when off medication . Spicer et al. (1994) were the first to evidence a unique increased semantic priming effect in PD patients as compared to the normal control subjects, which they called 'hyperpriming'. This hyperpriming was suggested to be caused by slowness in the unrelated prime-target conditions. Spicer et al. (1994) suggested two possible levels of the deficit, either the prelexical level or the post-lexical level. Somewhat later, the same research group (McDonald et al., 1996) revised their theory and concluded that PD patients show poor performance whenever the task requires switching between response sets or different semantic categories. However, rather than hyperpriming reflecting a switching problem between semantically unrelated words, Mari-Beffa et al. (2005) suggested that a lack of lexicalsemantic inhibitory control in participants with PD is responsible for it. This idea was confirmed by Castner er al. (2007), who furthermore concluded that subthalamic nucleus stimulation restored these inhibitory processes. Consequently, it is concluded that the basal ganglia are involved in both the automatic and controlled aspects of semantic priming and

not always a guarantee for successful comprehension.

thus support both the involved facilitation and inhibition processes.

Using receptive tasks, the existence of a specific verb processing deficit in PD was found. Grossman et al. (1994) reported impaired verb learning. They taught PD patients and healthy age-matched controls the grammatical and semantic information of a new verb ('to wamble'). The semantic and grammatical information of the new verb was probed by sentence judgment

**3.3 Verb processing** 

**3.2 Lexical and semantic processing** 

sentence matching task. Generally, our study confirms that the language faculty is not independent from executive functioning.

Several studies using Positron Emission Tomography (PET) and functional Magnetic Resonance Imaging (fMRI) have investigated the pattern of brain activation during sentence processing in non-brain-damaged individuals. However, only a few imaging studies have investigated the underlying neural activity during sentence processing in PD patients. In an fMRI study, Grossman et al. (2003) found striatal activation in exclusively the brains of healthy senior volunteers for long sentences, relative to short sentences. Moreover, PD patients engaged significantly more brain regions associated with working memory than healthy participants to achieve the same level of comprehension accuracy as the control subjects. According to Grossman et al. (2003) the striatum contributes to cognitive resources such as working memory and information-processing speed. PD patients' sentence comprehension difficulties have been ascribed to their limited striatal recruitment, which causes an interruption of a large scale network important for cognitive resources that can interfere with sentence processing (Grossman et al., 2003).

Using Event Related Potential (ERP) studies, Friederici and colleagues have demonstrated that degeneration of the basal ganglia due to PD influences language-related ERP components dramatically and correlates with different aspects of language processing during comprehension (for an overview see Kutas & Van Petten, 1994; Osterhout & Holcomb, 1995) . In a study by Kotz et al. (2002) and by Friederici et al. (2003), the PD patients included showed an intact ELAN (reflecting highly automatic first-pass parsing processes), but a strongly reduced P600. The P600 is an ERP component that is controlled by attention and is explained as indicating secondary syntactic processes such as reanalysis and repair (Friederici & Mecklinger, 1996), or as reflecting syntactic integration processes in general (Kaan et al., 2000). According to Friederici et al. (2003), the alteration in the P600 reflected distortions of the late controlled syntactic integration processes in PD. This reduction in amplitude points to a failure in the activation of the generators of this ERP component in PD patients. The reduction in PD patients' P600 amplitude points to a lack of integrity of the cortico-striato-cortical circuits responsible for the P600 generation. The patient studies by Friederici and colleagues suggest that the frontal cortex and the basal ganglia are differently involved in sentence processing or are active during different stages of auditory sentence processing. The left frontal cortex and the left anterior temporal cortex both contribute to the early automatic processing underlying the (E)LAN, whereas the left basal ganglia contribute to the late controlled syntactic integration processes underlying the P600. The difficulties with syntactic integration processes as described by Friederici et al. suggest that the language system itself is disrupted in PD patients.

In a recent fMRI study, we evaluated the patterns of activation during the comprehension of sentences in which canonicity and grammaticality were manipulated in fifteen patients with PD compared to fifteen healthy older adults (Colman, 2011). Here we focus on the activation patterns related to the processing of the passives by the PD patients and healthy control participants. Our intergroup analysis contradicted the expectation of compensatory cortical activation (Grossman et al., 2003). However, PD patients showed significant increased activation for passive versus active sentences in the left medial/superior frontal gyrus compared to healthy control participants. Three possible explanations for the activation in this frontal area during the processing of passive sentences are suggested. PD patients may rely on working memory, lexical semantics or higher-level semantic processes involved in evaluation of plausibility to compensate for the lack of activation seen in the healthy control participants when dealing with non-canonical passive sentences. First, Carpenter et al. (1994) hypothesized that working memory load is directly related to sentence complexity. As mentioned before, higher sentence complexity is related to the non-canonical order of roles (such as in passives). All in all, non-canonical sentences impose a higher demand on working memory than canonical sentences (King & Just, 1991). The PD patients possibly relied more on their intact working memory allocated in the prefrontal cortex to compensate for their difficulties to process the non-canonical passive sentences (for a review see Wager & Smith, 2003). Secondly, examining ambiguity resolution, Stowe et al. (2004) reported a similar left medial prefrontal area as in our study, which they linked to supporting higherlevel semantic processes involved in evaluation of plausibility. Finally, it is suggested that the exclusive activation of the prefrontal cortex in PD patients for passive sentences reflects a lexical-semantic strategy for dealing with word order information, which was probably not always a guarantee for successful comprehension.

### **3.2 Lexical and semantic processing**

176 Diagnostics and Rehabilitation of Parkinson's Disease

sentence matching task. Generally, our study confirms that the language faculty is not

Several studies using Positron Emission Tomography (PET) and functional Magnetic Resonance Imaging (fMRI) have investigated the pattern of brain activation during sentence processing in non-brain-damaged individuals. However, only a few imaging studies have investigated the underlying neural activity during sentence processing in PD patients. In an fMRI study, Grossman et al. (2003) found striatal activation in exclusively the brains of healthy senior volunteers for long sentences, relative to short sentences. Moreover, PD patients engaged significantly more brain regions associated with working memory than healthy participants to achieve the same level of comprehension accuracy as the control subjects. According to Grossman et al. (2003) the striatum contributes to cognitive resources such as working memory and information-processing speed. PD patients' sentence comprehension difficulties have been ascribed to their limited striatal recruitment, which causes an interruption of a large scale network important for cognitive resources that can

Using Event Related Potential (ERP) studies, Friederici and colleagues have demonstrated that degeneration of the basal ganglia due to PD influences language-related ERP components dramatically and correlates with different aspects of language processing during comprehension (for an overview see Kutas & Van Petten, 1994; Osterhout & Holcomb, 1995) . In a study by Kotz et al. (2002) and by Friederici et al. (2003), the PD patients included showed an intact ELAN (reflecting highly automatic first-pass parsing processes), but a strongly reduced P600. The P600 is an ERP component that is controlled by attention and is explained as indicating secondary syntactic processes such as reanalysis and repair (Friederici & Mecklinger, 1996), or as reflecting syntactic integration processes in general (Kaan et al., 2000). According to Friederici et al. (2003), the alteration in the P600 reflected distortions of the late controlled syntactic integration processes in PD. This reduction in amplitude points to a failure in the activation of the generators of this ERP component in PD patients. The reduction in PD patients' P600 amplitude points to a lack of integrity of the cortico-striato-cortical circuits responsible for the P600 generation. The patient studies by Friederici and colleagues suggest that the frontal cortex and the basal ganglia are differently involved in sentence processing or are active during different stages of auditory sentence processing. The left frontal cortex and the left anterior temporal cortex both contribute to the early automatic processing underlying the (E)LAN, whereas the left basal ganglia contribute to the late controlled syntactic integration processes underlying the P600. The difficulties with syntactic integration processes as described by Friederici et al.

In a recent fMRI study, we evaluated the patterns of activation during the comprehension of sentences in which canonicity and grammaticality were manipulated in fifteen patients with PD compared to fifteen healthy older adults (Colman, 2011). Here we focus on the activation patterns related to the processing of the passives by the PD patients and healthy control participants. Our intergroup analysis contradicted the expectation of compensatory cortical activation (Grossman et al., 2003). However, PD patients showed significant increased activation for passive versus active sentences in the left medial/superior frontal gyrus compared to healthy control participants. Three possible explanations for the activation in this frontal area during the processing of passive sentences are suggested. PD patients may rely on working memory, lexical semantics or higher-level semantic processes involved in evaluation of plausibility to compensate for the lack of activation seen in the healthy control

independent from executive functioning.

interfere with sentence processing (Grossman et al., 2003).

suggest that the language system itself is disrupted in PD patients.

Semantic priming tasks are a straightforward measure for the evaluation of lexical and semantic processes in patients with PD. In healthy participants, RTs to the target word are faster if the prime and the target are semantically related (doctor-NURSE) as compared to when the prime and target are not related (doctor-FLOWER) (see Neely, 1991 for extensive review on priming tasks). Copland (2003) found that PD patients are unable to suppress the infrequent meaning of homophones (bank-RIVER) and proposes therefore that the selective attentional engagement of the semantic network is impaired. Thus, PD compromises the controlled aspects of semantic processing rather than the automatic processes. During sentence comprehension tasks, lexical-semantic processing has been found to be abnormal in PD patients as well (Angwin et al., 2005). Angwin et al. (2004, 2006) also found that semantic processing deficits in PD are related to striatal dopamine deficiency since automatic semantic activation was compromised in PD patients when off medication .

Spicer et al. (1994) were the first to evidence a unique increased semantic priming effect in PD patients as compared to the normal control subjects, which they called 'hyperpriming'. This hyperpriming was suggested to be caused by slowness in the unrelated prime-target conditions. Spicer et al. (1994) suggested two possible levels of the deficit, either the prelexical level or the post-lexical level. Somewhat later, the same research group (McDonald et al., 1996) revised their theory and concluded that PD patients show poor performance whenever the task requires switching between response sets or different semantic categories. However, rather than hyperpriming reflecting a switching problem between semantically unrelated words, Mari-Beffa et al. (2005) suggested that a lack of lexicalsemantic inhibitory control in participants with PD is responsible for it. This idea was confirmed by Castner er al. (2007), who furthermore concluded that subthalamic nucleus stimulation restored these inhibitory processes. Consequently, it is concluded that the basal ganglia are involved in both the automatic and controlled aspects of semantic priming and thus support both the involved facilitation and inhibition processes.

#### **3.3 Verb processing**

Using receptive tasks, the existence of a specific verb processing deficit in PD was found. Grossman et al. (1994) reported impaired verb learning. They taught PD patients and healthy age-matched controls the grammatical and semantic information of a new verb ('to wamble'). The semantic and grammatical information of the new verb was probed by sentence judgment

Language Processing in Parkinson's Disease Patients Without Dementia 179

society. Miller et al. (2006) investigated the impact of particularly 'speech and voice' deficits on the life of the individual with PD and their family. To this purpose, a group of PD patients was interviewed to explore the onset of speech changes, their impact and patients' strategies to manage these changes. In general, the changes in PD patients' speech and voice had an effect on the overall communication, roles and relationships of those confronted with the disease. It was shown that alterations in speech do not need to be severe to have a significant impact. However, in addition to the speech and voice problems some of the interviewees reported difficulties with word retrieval, sentence formulation and comprehension. This suggests the necessity to refer all newly diagnosed PD patients to speech and language therapy. According to us, this preventive therapy will not only serve articulation and intelligibility abilities, but should also focus on the assessment and remediation of language problems. From our review it is clear that some PD patients suffer from unawareness of the extent of their communicative problems. During social conversations, deficit in the monitoring system influenced turn taking abilities and topic maintenance (McNamara et al., 2003). This unawareness or self-monitoring deficit can prevent the development of adaptive coping strategies, provokes feelings of frustration and might lead to complete withdrawal from communication. PD patients can profit from insights in their language disorders, for example it can help them to use effective compensation strategies or to simply inform the other speech partner of the impact of their disease on communication. From our clinical experience it is clear that patients and their caregivers are often surprised to hear that not only motor symptoms, but also language processing can be affected in the course of the disease. This review on language problems in PD may help in bringing the topic under the attention of those confronted with the disease, meaning that professionals need more up-to-date information. Up-to-date knowledge on the language problems on the part of the patients' environment will facilitate successful communication and, thus, support good family relations. Hence, including routine screening for cognitive decline and language problems early in the disease, in addition to supplying information on PD patients' language problems to caregivers and professionals could keep the PD patients from becoming socially isolated. Examples of communication advice for caregivers could be to simplify and avoid redundancy of information. Speech and language therapy must provide information in tune with the patient's individual limitations and whishes towards language and speech, which in turn can facilitate patients and their environment to implement coping strategies when communicative contexts are arising. In addition, we expect that intensive training of cognitive functions and strategies in PD patients will positively influence processing in the language domain. In the near future a therapy effectiveness study will be developed, which will remediate language problems in

combination with executive function deficits in PD.

Medication with levodopa is well known to improve the motor symptoms. However, the effects on cognitive functions are more complex: both positive as well as negative effects have been observed. According to Cools (2006), these contrasting effects of levodopa are due to the spatio-temporal progression of dopamine (DA) depletion in PD. PD starts in the dorsal striatum (tail of the Caudate nucleus) and progresses to the ventral striatum (head of the Caudate nucleus). Levodopa in early stages of the disease may improve cognitive functions of the dorsal striatum while simultaneously 'over-dosing' functions of the ventral

**5. Suggestions for future research** 

and picture classification. Significant impairment in recalling some aspects of the new verb was seen in 55% of the PD patients. These patients demonstrated a language-sensitive deficit in "appreciating grammatical information represented in the new verb" (Grossman et al., 1994, p. 413). However, a small number of PD patients responded randomly to probes of all information about the new verb, which suggests a memory impairment in these patients. More recently, Whiting et al. (2005) evaluated verb and context processing in PD by using a selfpaced stop making sense task. The participants had to pace themselves through a sentence that was preceded by a context, which made the thematic role of the verb plausible or implausible. They found that PD patients were impaired in thematic role mapping, which was consistent with previous findings of Geyer and Grossman from 1994. Whiting et al. (2005) proposed that PD participants in their study processed sentences "on a more superficial level" than control subjects and concluded that the PD patients' performance was caused by both global discourse comprehension difficulties and impaired working memory.

## **3.4 Perceptive pragmatic language abilities**

In daily life, healthy individuals interpret the intended meaning of language appropriate to the social context. Another line of research in receptive language functions has been focusing on the pragmatic language skills of PD patients. Pragmatic language use entails the ability to interpret nonliteral elements of language such as metaphors, proverbs, idioms, etc. Berg et al. (2003) conducted a survey of pragmatic language abilities and reported that PD patients exhibit impairments in making inferences,, comprehending metaphors and lexical ambiguities. The study by Whiting et al. (2005) showed that PD patients were less accurate than the control participants in using previously encountered discourse antecedents when deciding that a sentence stopped making sense. This is in line with the finding of Grossman, Crino et al. (1992) in which PD participants displayed an impaired ability to answer questions about previously encountered discourse elements compared to control participants. In addition, patients with PD have these problems also when resolving lexical ambiguities (Copland et al., 2001).

Monetta and Pell (2007) investigated how PD patients process metaphors using a timed property verification task (by Gernsbacher et al., 2001) compared to healthy control participants. The impact of PD on metaphor comprehension varied as a function of working memory ability, meaning that PD patients with a reduced working memory capacity were impaired in the comprehension of metaphors, whereas PD participants at a similar stage of disease but without working memory difficulties performed as good as the healthy control participants (Monetta & Pell, 2007). In a follow-up, similar resutls were found for inference generation (Monetta et al., 2008) and irony comprehension (Monetta et al., 2009). McKinlay et al. (2009) related pragmatic language skills to cognitive functions and suggested that processing speed was a stronger determiner of pragmatic language performance than working memory.

Research relating the pragmatic language problems of PD to their executive function deficits might be influenced by the research investigating morphosyntactic processes during sentence comprehension in PD patients.

## **4. Impact of language processing deficits on the daily life of PD patients**

The subtle deficits in language comprehension and production in PD will lead to communication problems that may result in decreased socialization and participation in

and picture classification. Significant impairment in recalling some aspects of the new verb was seen in 55% of the PD patients. These patients demonstrated a language-sensitive deficit in "appreciating grammatical information represented in the new verb" (Grossman et al., 1994, p. 413). However, a small number of PD patients responded randomly to probes of all information about the new verb, which suggests a memory impairment in these patients. More recently, Whiting et al. (2005) evaluated verb and context processing in PD by using a selfpaced stop making sense task. The participants had to pace themselves through a sentence that was preceded by a context, which made the thematic role of the verb plausible or implausible. They found that PD patients were impaired in thematic role mapping, which was consistent with previous findings of Geyer and Grossman from 1994. Whiting et al. (2005) proposed that PD participants in their study processed sentences "on a more superficial level" than control subjects and concluded that the PD patients' performance was caused by both global discourse

In daily life, healthy individuals interpret the intended meaning of language appropriate to the social context. Another line of research in receptive language functions has been focusing on the pragmatic language skills of PD patients. Pragmatic language use entails the ability to interpret nonliteral elements of language such as metaphors, proverbs, idioms, etc. Berg et al. (2003) conducted a survey of pragmatic language abilities and reported that PD patients exhibit impairments in making inferences,, comprehending metaphors and lexical ambiguities. The study by Whiting et al. (2005) showed that PD patients were less accurate than the control participants in using previously encountered discourse antecedents when deciding that a sentence stopped making sense. This is in line with the finding of Grossman, Crino et al. (1992) in which PD participants displayed an impaired ability to answer questions about previously encountered discourse elements compared to control participants. In addition, patients with PD have these problems also when resolving lexical

Monetta and Pell (2007) investigated how PD patients process metaphors using a timed property verification task (by Gernsbacher et al., 2001) compared to healthy control participants. The impact of PD on metaphor comprehension varied as a function of working memory ability, meaning that PD patients with a reduced working memory capacity were impaired in the comprehension of metaphors, whereas PD participants at a similar stage of disease but without working memory difficulties performed as good as the healthy control participants (Monetta & Pell, 2007). In a follow-up, similar resutls were found for inference generation (Monetta et al., 2008) and irony comprehension (Monetta et al., 2009). McKinlay et al. (2009) related pragmatic language skills to cognitive functions and suggested that processing speed was a stronger determiner of pragmatic language performance than

Research relating the pragmatic language problems of PD to their executive function deficits might be influenced by the research investigating morphosyntactic processes during

The subtle deficits in language comprehension and production in PD will lead to communication problems that may result in decreased socialization and participation in

**4. Impact of language processing deficits on the daily life of PD patients** 

comprehension difficulties and impaired working memory.

**3.4 Perceptive pragmatic language abilities** 

ambiguities (Copland et al., 2001).

sentence comprehension in PD patients.

working memory.

society. Miller et al. (2006) investigated the impact of particularly 'speech and voice' deficits on the life of the individual with PD and their family. To this purpose, a group of PD patients was interviewed to explore the onset of speech changes, their impact and patients' strategies to manage these changes. In general, the changes in PD patients' speech and voice had an effect on the overall communication, roles and relationships of those confronted with the disease. It was shown that alterations in speech do not need to be severe to have a significant impact. However, in addition to the speech and voice problems some of the interviewees reported difficulties with word retrieval, sentence formulation and comprehension. This suggests the necessity to refer all newly diagnosed PD patients to speech and language therapy. According to us, this preventive therapy will not only serve articulation and intelligibility abilities, but should also focus on the assessment and remediation of language problems. From our review it is clear that some PD patients suffer from unawareness of the extent of their communicative problems. During social conversations, deficit in the monitoring system influenced turn taking abilities and topic maintenance (McNamara et al., 2003). This unawareness or self-monitoring deficit can prevent the development of adaptive coping strategies, provokes feelings of frustration and might lead to complete withdrawal from communication. PD patients can profit from insights in their language disorders, for example it can help them to use effective compensation strategies or to simply inform the other speech partner of the impact of their disease on communication. From our clinical experience it is clear that patients and their caregivers are often surprised to hear that not only motor symptoms, but also language processing can be affected in the course of the disease. This review on language problems in PD may help in bringing the topic under the attention of those confronted with the disease, meaning that professionals need more up-to-date information. Up-to-date knowledge on the language problems on the part of the patients' environment will facilitate successful communication and, thus, support good family relations. Hence, including routine screening for cognitive decline and language problems early in the disease, in addition to supplying information on PD patients' language problems to caregivers and professionals could keep the PD patients from becoming socially isolated. Examples of communication advice for caregivers could be to simplify and avoid redundancy of information. Speech and language therapy must provide information in tune with the patient's individual limitations and whishes towards language and speech, which in turn can facilitate patients and their environment to implement coping strategies when communicative contexts are arising. In addition, we expect that intensive training of cognitive functions and strategies in PD patients will positively influence processing in the language domain. In the near future a therapy effectiveness study will be developed, which will remediate language problems in combination with executive function deficits in PD.

## **5. Suggestions for future research**

Medication with levodopa is well known to improve the motor symptoms. However, the effects on cognitive functions are more complex: both positive as well as negative effects have been observed. According to Cools (2006), these contrasting effects of levodopa are due to the spatio-temporal progression of dopamine (DA) depletion in PD. PD starts in the dorsal striatum (tail of the Caudate nucleus) and progresses to the ventral striatum (head of the Caudate nucleus). Levodopa in early stages of the disease may improve cognitive functions of the dorsal striatum while simultaneously 'over-dosing' functions of the ventral

Language Processing in Parkinson's Disease Patients Without Dementia 181

PSP is a neurodegenerative disease characterized by defects in the vertical ocular gaze, bulbar dysfunction, increased frequency of falling, and akinetic-rigid features. In addition, cognitive impairments, in particular executive dysfunctions associated with alterations

CBD is characterized by slowly progressing, unilateral Parkinsonism with dystonia or myoclonus, unresponsiveness to Levodopa, and limb apraxia. Patients with CBD often demonstrate impairments in visuospatial processing and visuoconstruction (Tang-Wai et al., 2003) in combination with acalculia, dysexecutive symptoms and aphasia (McMonagle et al.,

Thus far, only a few studies have investigated language processing in patients with atypical Parkinson syndromes, such as MSA (Apostolova et al., 2006), PSP and CBD (Josephs &

This review highlights that the progressive degeneration of the cortico-striato-cortical circuits due to PD disturbs executive functioning and, thus contributes to deficits in language production and comprehension. One of the major conclusions based on this review is the importance of evaluating both the executive functions and modalities of language processing (i.e., comprehension and production) in patients with PD. This is not only crucial for our understanding of PD and for the relationship between languages and executive functions, but it is also particularly useful for efficiently identifying the needs for

More research still needs to be done to illuminate further the impact of PD on language processing. Future research needs focus on the clinical implementation of evidence-based communication guidelines in order to guarantee a better quality of life for patients suffering

This chapter was written on the basis of Katrien Colman's PhD dissertation. Roelien Bastiaanse was the principal supervisor of the PhD project. This research project was funded

Aarsland, D., Andersen, K., Larsen, J. P., & Lolk, A. (2003). Prevalence and characteristics of

Ackermann, H., & Ziegler, W. (1989). [Dysarthrophonia of parkinson syndrome]. *Fortschritte* 

Almor, A., Kempler, D., Andersen, E. S., Macdonald, M. C., Hayes, U. L., & Hintiryan, H. (2002).

parkinson patients. *Brain and Language, 83*(1), 149-151, ISSN-L 0093-934X. Angwin, A. J., Chenery, H. J., Copland, D. A., Murdoch, B. E., & Silburn, P. A. (2004). The

dementia in parkinson disease: An 8-year prospective study. *Archives of Neurology,* 

The production of regularly and irregularly inflected nouns and verbs in alzheimer and

time course of semantic activation in parkinson's disease. *Brain and Language, 91*(1),

by the Stichting Internationaal Parkinson Fonds (Hoofddorp, the Netherlands).

*Der Neurologie-Psychiatrie, 57*(4), 149-160, ISSN 0720-4299.

*60* (3), 387-392, ISSN-L 0003-9942.

145-146, ISSN-L 0093-934X.

within the frontostriatal circuitry, occur (Millar et al., 2006).

Duffy, 2008; McMonagle et al., 2006).

2006).

**6. Conclusion** 

direct intervention

**8. References** 

**7. Acknowledgements** 

from PD.

striatum. This effect of over-dosing is related to the base level of DA in underlying corticostriato-cortical circuitry and the task instructions. Therefore, in future research, to control for the influence of dopaminergic medication on cognitive processing, we suggest conducting experiments in the practically defined 'off state'. This is typically following an overnight fast from the patient's anti-Parkinson medications. More positive results are expected in this 'off state', but we also expect more influence of other factors such as frustrations with task performance and tremors and rigidity making testing in the MRI scanner impossible. Ultimately, conducting experiments in drug naïve 'de novo' patients is preferred, but clinically these patients are not always willing to participate in research. Apart from the important effects of medication on language processing, the variables of disease duration and age of onset of PD should be taken in consideration in future research.

Future studies on the influence of set-shifting and working memory on sentence processing in PD can benefit from the use of better-controlled and better-understood methods than the clinical accepted neuropsychological tests which were used in the studies reported above. For example, reading span tasks have been used as tests of working memory because they require active manipulation of information and concurrent item retention (Just & Carpenter, 1992). However, reading span tasks rely on many of the same processes as reading comprehension tasks (Engle et al., 1992), which makes it difficult to draw any strong conclusions in terms of the mediating value of working memory for exactly that language process.

In the near future, the nature of the connectivity between the inferior frontal gyrus and the basal ganglia can be further explored. A functional connectivity analysis can provide functional evidence for a basal ganglia-frontal cortical network during the comprehension of sentences in which the variables of canonicity and grammaticality are crossed. However, it is generally known that in fMRI, temporal resolution is inferior and that it cannot index neural activity that is specifically time-locked to the critical word itself. The temporal coarseness of the fMRI method probably blurred the linguistic processes. Simultaneous ERP/fMRI may allow improved localization of neural generators as well as enhanced temporal resolution of BOLD activation foci. Functional connectivity analysis can be used to examine the degree of collaboration between language-specific cortical areas and the basal ganglia, when processing violated compared to non-violated sentences. In on-line behavioral tasks, the impact of executive functions necessary for syntactic processes per se and the executive functions necessary for the task can potentially be disentangled. Therefore, a valuable technique for obtaining on-line data from sentence-picture matching is the eye-tracking method as suggested by Hochstadt (2009).

Finally, as evidenced by this review, there exists an extensive amount of literature on language processing in PD, but language processing in other motor syndromes has received little attention. The existence of a similar verb naming deficit in other movement disorders, such as CBD and PSP (Cotelli et al., 2006) has provided a major argument for the theory that semantic characteristics of the verb are grounded in the motor system of the brain. It will be interesting to test verb production related to cognitive functions in the following movement disorders:

Multiple system atrophy (MSA) is an adult-onset, sporadic, progressive neurodegenerative disease characterized by varying severity of Parkinsonian features, cerebellar ataxia, autonomic failure, urogenital dysfunction, and corticospinal disorders (Gilman et al., 2008). MSA is also accompanied by cognitive impairments associated with dysfunctional corticostriato-cortical circuits (Herting et al., 2007).

PSP is a neurodegenerative disease characterized by defects in the vertical ocular gaze, bulbar dysfunction, increased frequency of falling, and akinetic-rigid features. In addition, cognitive impairments, in particular executive dysfunctions associated with alterations within the frontostriatal circuitry, occur (Millar et al., 2006).

CBD is characterized by slowly progressing, unilateral Parkinsonism with dystonia or myoclonus, unresponsiveness to Levodopa, and limb apraxia. Patients with CBD often demonstrate impairments in visuospatial processing and visuoconstruction (Tang-Wai et al., 2003) in combination with acalculia, dysexecutive symptoms and aphasia (McMonagle et al., 2006).

Thus far, only a few studies have investigated language processing in patients with atypical Parkinson syndromes, such as MSA (Apostolova et al., 2006), PSP and CBD (Josephs & Duffy, 2008; McMonagle et al., 2006).

## **6. Conclusion**

180 Diagnostics and Rehabilitation of Parkinson's Disease

striatum. This effect of over-dosing is related to the base level of DA in underlying corticostriato-cortical circuitry and the task instructions. Therefore, in future research, to control for the influence of dopaminergic medication on cognitive processing, we suggest conducting experiments in the practically defined 'off state'. This is typically following an overnight fast from the patient's anti-Parkinson medications. More positive results are expected in this 'off state', but we also expect more influence of other factors such as frustrations with task performance and tremors and rigidity making testing in the MRI scanner impossible. Ultimately, conducting experiments in drug naïve 'de novo' patients is preferred, but clinically these patients are not always willing to participate in research. Apart from the important effects of medication on language processing, the variables of disease duration

Future studies on the influence of set-shifting and working memory on sentence processing in PD can benefit from the use of better-controlled and better-understood methods than the clinical accepted neuropsychological tests which were used in the studies reported above. For example, reading span tasks have been used as tests of working memory because they require active manipulation of information and concurrent item retention (Just & Carpenter, 1992). However, reading span tasks rely on many of the same processes as reading comprehension tasks (Engle et al., 1992), which makes it difficult to draw any strong conclusions in terms of the mediating value of working memory for exactly that language

In the near future, the nature of the connectivity between the inferior frontal gyrus and the basal ganglia can be further explored. A functional connectivity analysis can provide functional evidence for a basal ganglia-frontal cortical network during the comprehension of sentences in which the variables of canonicity and grammaticality are crossed. However, it is generally known that in fMRI, temporal resolution is inferior and that it cannot index neural activity that is specifically time-locked to the critical word itself. The temporal coarseness of the fMRI method probably blurred the linguistic processes. Simultaneous ERP/fMRI may allow improved localization of neural generators as well as enhanced temporal resolution of BOLD activation foci. Functional connectivity analysis can be used to examine the degree of collaboration between language-specific cortical areas and the basal ganglia, when processing violated compared to non-violated sentences. In on-line behavioral tasks, the impact of executive functions necessary for syntactic processes per se and the executive functions necessary for the task can potentially be disentangled. Therefore, a valuable technique for obtaining on-line data from sentence-picture matching is

Finally, as evidenced by this review, there exists an extensive amount of literature on language processing in PD, but language processing in other motor syndromes has received little attention. The existence of a similar verb naming deficit in other movement disorders, such as CBD and PSP (Cotelli et al., 2006) has provided a major argument for the theory that semantic characteristics of the verb are grounded in the motor system of the brain. It will be interesting to test verb production related to cognitive functions in the following movement

Multiple system atrophy (MSA) is an adult-onset, sporadic, progressive neurodegenerative disease characterized by varying severity of Parkinsonian features, cerebellar ataxia, autonomic failure, urogenital dysfunction, and corticospinal disorders (Gilman et al., 2008). MSA is also accompanied by cognitive impairments associated with dysfunctional cortico-

and age of onset of PD should be taken in consideration in future research.

the eye-tracking method as suggested by Hochstadt (2009).

striato-cortical circuits (Herting et al., 2007).

process.

disorders:

This review highlights that the progressive degeneration of the cortico-striato-cortical circuits due to PD disturbs executive functioning and, thus contributes to deficits in language production and comprehension. One of the major conclusions based on this review is the importance of evaluating both the executive functions and modalities of language processing (i.e., comprehension and production) in patients with PD. This is not only crucial for our understanding of PD and for the relationship between languages and executive functions, but it is also particularly useful for efficiently identifying the needs for direct intervention

More research still needs to be done to illuminate further the impact of PD on language processing. Future research needs focus on the clinical implementation of evidence-based communication guidelines in order to guarantee a better quality of life for patients suffering from PD.

## **7. Acknowledgements**

This chapter was written on the basis of Katrien Colman's PhD dissertation. Roelien Bastiaanse was the principal supervisor of the PhD project. This research project was funded by the Stichting Internationaal Parkinson Fonds (Hoofddorp, the Netherlands).

## **8. References**


Language Processing in Parkinson's Disease Patients Without Dementia 183

Colman, K. S. F. (2011). *Behavioral and neuroimaging studies on language processing in Dutch* 

Cools, R. (2006). Dopaminergic modulation of cognitive function-implications for L-DOPA

Copland, D. A., Chenery, H. J., & Murdoch, B. E. (2001). Discourse priming of homophones

Copland, D. A. (2003). The basal ganglia and semantic engagement: Potential insights from

Cotelli, M., Borroni, B., Manenti, R., Alberici, A., Calabria, M., Agosti, C., . . . Cappa, S. F.

Cotelli, M., Borroni, B., Manenti, R., Zanetti, M., Arévalo, A., Cappa, S. F., & Padovani, A.

Crescentini, C., Mondolo, F., Biasutti, E., & Shallice, T. (2008). Supervisory and routine

Dubois, B., & Pillon, B. (1995). Do cognitive changes of parkinson's disease result from

Dubois, B., & Pillon, B. (1997). Cognitive deficits in parkinson's disease. *Journal of Neurology,* 

Ellis, C. (2006). *The contribution of the basal ganglia to expressive language performance.* 

Ellis, C., & Rosenbek, J. C. (2007). The basal ganglia and expressive language. A review and directions for research. *Communicative Disorders Review, 1*(1), 1-15, ISSN-L 1933-2831. Engle, R. W., Cantor, J., & Carullo, J. J. (1992). Individual differences in working memory

Fahn, S., & Elton, R. (1987). Unified parkinson's disease rating scale. In: Fahn, S., Marsden,

*Volume 2*, (153-163), Florham Park, NJ. Macmillan Health Care Information. Flowers, K. A., Robertson, C., & Sheridan, M. R. (1995). Some characteristics of word fluency in parkinson's disease. *Journal of Neurolinguistics, 9*(1), 33-46, ISSN-L 0911-6044.

*Learning, Memory, and Cognition, 18*(5), 972-992, ISSN-L 0278-7393.

*European Journal of Neurology, 14*(6), 632-637, ISSN-L 1351-5101.

disease. *Neuropsychologia, 46*(2), 434-447, ISSN-L 0028-3932.

Groningen, The Netherlands, ISBN (digital version) 9789036747530. Colman, K. S. F., Koerts, J., van Beilen, M., Leenders, K. L., Post, W. J., & Bastiaanse, R.

parkinson's disease. *Cortex, 45*(8), 930-942, ISSN-L 0010-9452.

ISSN-L 0149-7634.

ISSN-L 0894-4105.

L 0303-6995.

*244*(1), 2-8, ISSN-L 0340-5354.

897-898, ISSN-L 0885-3185.

556, ISSN-L 1380-3395.

*9*(7), 1041-1052, ISSN-L 1355-6177.

*speakers with parkinson's disease.* (Unpublished PhD thesis). University of Groningen,

(2009). The impact of executive functions on verb production in patients with

treatment in parkinson's disease. *Neuroscience and Biobehavioral Reviews, 30*(1), 1-23,

in individuals with dominant nonthalamic subcortical lesions, cortical lesions and parkinson's disease. *Journal of Clinical and Experimental Neuropsychology, 23*(4), 538-

semantic priming in individuals with subcortical vascular lesions, parkinson's disease, and cortical lesions. *Journal of the International Neuropsychological Society,* 

(2006). Action and object naming in frontotemporal dementia, progressive supranuclear palsy, and corticobasal degeneration. *Neuropsychology, 20*(5), 558-565,

(2007). Action and object naming in parkinson's disease without dementia.

processes in noun and verb generation in nondemented patients with parkinson's

dopamine depletion? *Journal of Neural Transmission. Supplementum, 45,* 27-34, ISSN-

(Unpublished PhD thesis), University of Florida, Gainesville, ISBN 9780542351716. Ellis, C., okun, M. S., Gonzalez-Rothi, I. J., Crosson, B., Rogalski, Y., & Rosenbek, J. C. (2006).

Expressive language after PD: Deficits in use but not form. *Movement Disorders, 21*,

and comprehension: A test of four hypotheses. *Journal of Experimental Psychology.* 

C. D., Calne, D. B., Goldstein, M. (Eds.), *Recent Developments in Parkinson's Disease,* 


Angwin, A. J., Chenery, H. J., Copland, D. A., Murdoch, B. E., & Silburn, P. A. (2005).

parkinson's disease. *Cognitive Brain Research, 25*(1), 78-89, ISSN-L 0926-6410. Angwin, A. J., Copland, D. A., Chenery, H. J., Murdoch, B. E., & Silburn, P. A. (2006). The

Auriacombe, S., Grossman, M., Carvell, S., Gollomp, S., Stern, M. B., & Hurtig, H. I. (1993).

Bartels, A. L., & Leenders, K. L. (2009). Parkinson's disease: The syndrome, the pathogenesis

Bastiaanse, R., & van Zonneveld, R. (2004). Broca's aphasia, verbs and the mental lexicon.

Benke, T., Hohenstein, C., Poewe, W., & Butterworth, B. (2000). Repetitive speech

Berg, E., Bjornram, C., Hartelius, L., Laakso, K., & Johnels, B. (2003). High-level language

Bertella, L., Albani, G., Greco, E., Priano, L., Mauro, A., Marchi, S., . . . Semenza, C. (2002).

Bloom, L., & Lahey, M. (1978). *Language development and language disorders*. New York: John

Botvinick, M. M., Braver, T. S., Barch, D. M., Carter, C. S., & Cohen, J. D. (2001). Conflict

Boulenger, V., Mechtouff, L., Thobois, S., Broussolle, E., Jeannerod, M., & Nazir, T. A. (2008).

Caplan, D., & Waters, G. S. (1999). Verbal working memory and sentence comprehension.

Castner, J. E., Copland, D. A., Silburn, P. A., Coyne, T. J., Sinclair, F., & Chenery, H. J. (2007).

Colman, K., Koerts, J., van Beilen, M., Leenders, K. L., & Bastiaanse, R. (2006). The role of

concrete nouns. *Neuropsychologia, 46*(2), 743-756, ISSN-L 0028-3932.

*The Behavioral and Brain Sciences, 22*(1), 77-94, ISSN-L 0140-525X.

and pathophysiology. *Cortex, 45*(8), 915-921, ISSN-L 0010-9452.

*Brain and Language, 90*(1-3), 198-202, ISSN-L 0093-934X.

*Psychiatry, 69*(3), 319-324, ISSN-L 0022-3050.

Wiley & Sons, ISBN 0-471-04438-5.

(3), 166-175, ISSN-L 0894-878X.

726-727, ISSN-L 0028-3878.

L 0894-4105.

ISSN-L 0269-9206.

ISSN-L 0278-2626.

0033-295X.

0093-934X.

Summation of semantic priming and complex sentence comprehension in

influence of dopamine on semantic activation in parkinson's disease: Evidence from a multipriming task. *Neuropsychology, 20*(3), 299-306, ISSN-L 0894-4105. Apostolova, L. G., Klement, I., Bronstein, Y., Vinters, H. V., & Cummings, J. L. (2006).

Multiple system atrophy presenting with language impairment. *Neurology, 67*(4),

Verbal fluency deficits in parkinson's disease. *Neuropsychology, 7*(2), 182-192, ISSN-

phenomena in parkinson's disease. *Journal of Neurology, Neurosurgery, and* 

difficulties in parkinson's disease. *Clinical Linguistics and Phonetics, 17*(1), 63-80,

Noun verb dissociation in parkinson's disease. *Brain and Cognition, 48*(2-3), 277-280,

monitoring and cognitive control. *Psychological Review, 108*(3), 624-652, ISSN-L

Word processing in parkinson's disease is impaired for action verbs but not for

Lexical-semantic inhibitory mechanisms in parkinson's disease as a function of subthalamic stimulation. *Neuropsychologia, 45*(14), 3167-3177, ISSN-L 0028-3932. Cheng, Y., & Martin, R. (2005). Selection demands vs. association strength in the verb generation task. *Brain and Language, 95*(1), 193-194, ISSN-L 0093-934X. Cohen, H., Bouchard, S., Scherzer, P., & Whitaker, H. (1994). Language and verbal reasoning

in parkinson's disease. *Neuropsychiatry, Neuropsychology, & Behavioral Neurology,* 7

cognitive mechanisms in sentence comprehension in dutch speaking parkinson's disease patients: Preliminary data. *Brain and Language, 99*(1-2), 109-110, ISSN-L


Language Processing in Parkinson's Disease Patients Without Dementia 185

Hochstadt, J., Nakano, H., Lieberman, P., & Friedman, J. (2006). The roles of sequencing and

Hochstadt, J. (2009). Set-shifting and the on-line processing of relative clauses in parkinson's

Illes, J. (1989). Neurolinguistic features of spontaneous language production dissociate three

Illes, J., Metter, E. J., Hanson, W. R., & Iritani, S. (1988). Language production in parkinson's

Josephs, K. A., & Duffy, J. R. (2008). Apraxia of speech and nonfluent aphasia: A new clinical

Just, M. A., & Carpenter, P. A. (1992). A capacity theory of comprehension: Individual differences in working memory. *Psychological Review, 99*(1), 122-149, ISSN-L 0033-295X. Just, M. A., Carpenter, P. A., & Keller, T. A. (1996). The capacity theory of comprehension:

Kaan, E., Harris, A., Gibson, E., & Holcomb, P. (2000). The P600 as an index of syntactic

Kaplan, E., Goodglass, H., & Weintraub, S. (1983). *The Boston naming test*. Philadelphia: Lea

Kemmerer, D. (1999). Impaired comprehension of raising-to-subject constructions in parkinson's disease, *Brain and Language, 66*(3), 311-328, ISSN-L 0093-934X. Kotz, S. A., Frisch, S., von Cramon, D. Y., & Friederici, A. D. (2003). Syntactic language

*International Neuropsychological Society, 9*(7), 1053-1060, ISSN-L 1355-6177. Kotz, S. A., Frisch, S., Werheid, K., Hein, G., von Cramon, D. Y., & Friederici, A. D. (2002).

Kuperberg, G. R. (2007). Neural mechanisms of language comprehension: Challenges to

Kutas, M., & Van Petten, C. K. (1994). Psycholinguistics electrified: Event-related brain

Lee, C., Grossman, M., Morris, J., Stern, M. B., & Hurtig, H. I. (2003). Attentional resource

Levelt, W. J. M. (1983). Monitoring and self-repair in speech. *Cognition, 14*(1), 41-104, ISSN-L

Levelt, W. J. M. (1989). *Speaking: From intention to articulation*. Cambridge, MA: MIT Press,

*Brain and Language, 97*(3), 243-257, ISSN-L 0093-934X.

*Brain and Language, 37*(4), 628-642, ISSN-L 0093-934X.

*Opinion in Neurology, 21*(6), 688-692, ISSN-L 1080-8248.

*Brain and Language, 83*(1), 69-70, ISSN-L 0093-934X.

syntax. *Brain Research, 1146*, 23-49, ISSN-L 0006-8993.

(83-143). San Diego: Academic Press, ISBN 0122808908.

disease. *Brain and Language, 85*(3), 347-356, ISSN-L 0093-934X.

0010-9452.

ISSN-L 0093-934X.

ISSN-L 0033-295X.

0010-0277.

ISBN 0262121379.

& Febiger, ISBN 9780812101003.

verbal working memory in sentence comprehension deficits in parkinson's disease.

disease: Results from a novel eye-tracking method. *Cortex, 45*(8), 991-1011, ISSN-L

forms of neurodegenerative disease: Alzheimer's, huntington's, and parkinson's.

disease: Acoustic and linguistic considerations. *Brain and Language, 33*(1), 146-160,

marker for corticobasal degeneration and progressive supranuclear palsy. *Current* 

New frontiers of evidence and arguments. *Psychological Review, 103*(4), 773-780,

integration difficulty. *Language and Cognitive Processes, 15*(2), 159-201, ISSN-L 0169-0965.

processing: ERP lesion data on the role of the basal ganglia. *Journal of the* 

The role of the basal ganglia in syntactic language processing: Event-related potential evidence from different patient populations and syntactic paradigms.

potential investigations. In M. A. Gernsbacher (Ed.), *Handbook of psycholinguistics*

and processing speed limitations during sentence processing in parkinson's


Friederici, A. D., Kotz, S. A., Werheid, K., Hein, G., & von Cramon, D. Y. (2003). Syntactic

Friederici, A. D., & Mecklinger, A. (1996). Syntactic parsing as revealed by brain responses:

Gernsbacher, M. A., Keysar, B., Robertson, R. R. W., & Werner, N. K. (2001). The role of

Geyer, H. L., & Grossman, M. (1994). Investigating the basis for the sentence comprehension

Gilman, S., Wenning, G. K., Low, P. A., Brooks, D. J., Mathias, C. J., Trojanowski, J. Q., . . .

Goodglass, H., & Kaplan, E. (1972). *The assessment of aphasia and related disorders*.

Grossman, M. (1999). Sentence processing in parkinson's disease. *Brain and Cognition, 40*(2),

Grossman, M., Carvell, S., Gollomp, S., Stern, M. B., Reivich, M., Morrison, D., . Hurtig, H. I.

Grossman, M., Carvell, S., Stern, M. B., Gollomp, S., & Hurtig, H. I. (1992). Sentence

Grossman, M., Cooke, A., DeVita, C., Lee, C., Alsop, D., Detre, J., . Hurtig, H. I. (2003).

Grossman, M., Stern, M. B., Gollomp, S., Vernon, G., & Hurtig, H. I. (1994). Verb learning in parkinson's disease. *Neuropsychology, 8*(3), 413-423, ISSN-L 0894-4105. Grossman, M., Zurif, E., Lee, C., Prather, P., Kalmanson, J., Stern, M. B., & Hurtig, H. I.

Gurd, J. M., & Ward, C. D. (1989). Retrieval from semantic and letter-initial categories in patients with parkinson's disease. *Neuropsychologia, 27*(5), 743-746, ISSN-L 0028-3932.

Henry, J. D., & Crawford, J. R. (2004). Verbal fluency deficits in parkinson's disease: A meta-

Herting, B., Beuthien-Baumann, B., Pöttrich, K., Donix, M., Triemer, A., Lampe, J. B., . . .

disease: An fMRI study. *Neurology, 60*(5), 775-781, ISSN-L 0028-3878. Grossman, M., Lee, C., Morris, J., Stern, M. B., & Hurtig, H. I. (2002). Assessing resource

disease. *Neuropsychology, 16*(2), 174-181, ISSN-L 0894-4105.

system atrophy. *Neurology, 71*(9), 670-676, ISSN-L 0028-3878.

142, ISSN-L 0894-4105.

*25*(1), 157-176, ISSN-L 0090-6905.

387-413, ISSN-L 0278-2626.

1530-8898.

1355-6177.

*Language, 45*(3), 433-450, ISSN-L 0749-596X.

Philadelphia: Lea & Febiger, ISBN 0812103572.

*Language, 42*(4), 347-384, ISSN-L 0093-934X.

*80*(3), 603-616, ISSN-L 0093-934X.

comprehension in parkinson'd disease: Investigating early automatic and late integrational processes using event-related potentials. *Neuropsychology, 17*(1), 133-

First-pass and second-pass parsing processes. *Journal of Psycholinguistic Research,* 

suppression and enhancement in understanding metaphors. *Journal of Memory and* 

deficit in parkinson's disease. *Journal of Neurolinguistics, 8*(3), 191-205, ISSN-L 0911-6044.

Vidailhet, M. (2008). Second consensus statement on the diagnosis of multiple

(1993). Cognitive and physiological substrates of impaired sentence processing in parkinson's disease. *Journal of Cognitive Neuroscience, 5*(4), 480-498, 1530-8898, ISSN

comprehension in parkinson's disease: The role of attention and memory. *Brain and* 

Grammatical and resource components of sentence processing in parkinson's

demands during sentence processing in parkinson's disease. *Brain and Language,* 

(2002). Information processing speed and sentence comprehension in parkinson's

analysis. *Journal of the International Neuropsychological Society, 10*(4), 608-622, ISSN-L

Holthoff, V. A. (2007). Prefrontal cortex dysfunction and depression in atypical parkinsonian syndromes. *Movement Disorders, 22*(4), 490-497, ISSN-L 0885-3185.


Language Processing in Parkinson's Disease Patients Without Dementia 187

Murray, L. L. (2000). Spoken language production in huntington's and parkinson's diseases.

Murray, L. L. (2008). Language and parkinson's disease. *Annual Review of Applied Linguistics,* 

Murray, L. L., & Lenz, L. P. (2001). Productive syntax abilities in huntington's and parkinson's diseases. *Brain and Cognition, 46*(1-2), 213-219, ISSN-L 0278-2626. Muslimović, D., Post, B., Speelman, J. D., & Schmand, B. (2005). Cognitive profile of patients with newly diagnosed parkinson disease. *Neurology, 65*(8), 1239-1245, ISSN-L 0028-3878. Natsopoulos, D., Grouios, G., Bostantzopoulou, S., Mentenopoulos, G., Katsarou, Z., &

Natsopoulos, D., Katsarou, Z., Bostantzopoulou, S., Grouios, G., Mentenopoulos, G., &

Osterhout, L., & Holcomb, P. (1995). Event related potentials and language comprehension.

Pahwa, R., Paolo, A., Troster, A., & Koller, W. (1998). Cognitive impairment in parkinson's disease. *European Journal of Neurology, 5*(5), 431-441, ISSN-L 1351-5101. Péran, P., Rascol, O., Demonet, J. F., Celsis, P., Nespoulous, J. L., Dubois, B., & Cardebat, D.

Péran, P., Cardebat, D., Cherubini, A., Piras, F., Luccichenti, G., Peppe, A., . . . Sabatini, U.

Perkins, M. R. (2005). Pragmatic ability and disability as emergent phenomena. *Clinical* 

Piatt, A. L., Fields, J. A., Paolo, A. M., Koller, W. C., & Tröster, A. I. (1999a). Lexical, semantic,

*of Clinical and Experimental Neuropsychology, 21*(4), 435-443, ISSN-L 1380-3395. Piatt, A. L., Fields, J. A., Paolo, A. M., & Tröster, A. I. (1999b). Action (verb naming) fluency

Pignatti, R., Ceriani, F., Bertella, L., Mori, I., & Semenza, C. (2006). Naming abilities in

Portin, R., Laatu, S., Revonsuo, A., & Rinne, U. K. (2000). Impairment of semantic

Postma, A. (2000). Detection of errors during speech production: A review of speech

Prutting, C. A., & Kirchner, D. M. (1987). A clinical appraisal of the pragmatic aspects of

Saling, L. L., & Phillips, J. G. (2007). Automatic behaviour: Efficient not mindless. *Brain* 

monitoring models. *Cognition, 77*(2), 97-132, ISSN-L 0010-0277.

*Research Bulletin, 73*(1-3), 1-20, ISSN-L 0361-9230.

parkinsonian patients. *Cortex, 27*(2), 255-268, ISSN-L 0010-9452.

disease. *Movement Disorders, 18*(2), 150-156, ISSN-L 0885-3185.

*Linguistics and Phonetics, 19*(5), 367-377, ISSN-L 0269-9206.

*Neuropsychologia, 37*(13), 1499-1503, ISSN-L 0028-3932.

*99*(1-2), 113-114, ISSN-L 0093-934X.

study. *Cortex, 45*(8), 960-971, ISSN-L 0010-9452.

*28*(-1), 113-127, ISSN-L 0267-1905.

951-964, ISSN-L 0028-3932.

0198521359.

0003-9942.

*Journal of Speech, Language, and Hearing Research, 43*(6), 1350-1366, ISSN-L 1092-4388.

Logothetis, J. (1993). Algorithmic and heuristic strategies in comprehension of complement clauses by patients with parkinson's disease. *Neuropsychologia, 31*(9),

Logothetis, J. (1991). Strategies in comprehension of relative clauses by

In M. D. Rugg, & M. G. H. Coles (Eds.), *Electrophysiology of mind: Event-related brain potentials and cognition* (171-215). New York: Oxford University Press, ISBN

(2003). Deficit of verb generation in nondemented patients with parkinson's

(2009). Object naming and action-verb generation in parkinson's disease: A fMRI

and action verbal fluency in parkinson's disease with and without dementia. *Journal* 

as an executive function measure: Convergent and divergent evidence of validity.

spontaneous speech in Parkinson's and Alzheimer's disease. *Brain and Language,* 

knowledge in parkinson disease. *Archives of Neurology, 57*(9), 1338-1343, ISSN-L

language. *The Journal of Speech and Hearing Disorders, 52*(2), 105-119, ISSN-L 0022-4677.


Levelt, W. J. M., Schriefers, H., Vorberg, D., Meyer, A. S., Pechmann, T., & Havinga, J. (1991).

Lieberman, P., Friedman, J., & Feldman, L. S. (1990). Syntax comprehension deficits in

Lieberman, P., Kako, E., Friedman, J., Tajchman, G., Feldman, L. S., & Jiminez, E. B. (1992).

Longworth, C. E., Keenan, S. E., Barker, R. A., Marslen-Wilson, W. D., & Tyler, L. K. (2005).

Martin, R. C., & Cheng, Y. (2006). Selection demands versus association strength in the verb generation task. *Psychonomic Bulletin and Review, 13*(3), 396-401, ISSN-L 1069-9384. McDonald, C., Brown, G. G., & Gorell, J. M. (1996). Impaired set-shifting in parkinson's

McKinlay, A., Dalrymple-Alford, J. C., Grace, R. C., & Roger, D. (2009). The effect of

McMonagle, P., Blair, M., & Kertesz, A. (2006). Corticobasal degeneration and progressive

McNamara, P., & Durso, R. (2003). Pragmatic communication skills in patients with parkinson's disease. *Brain and Language, 84*(3), 414-423, ISSN-L 0093-934X. McNamara, P., Obler, L. K., Au, R., Durso, R., & Albert, M. L. (1992). Speech monitoring

Millar, D., Griffiths, P., Zermansky, A. J., & Burn, D. J. (2006). Characterizing behavioral and

Miller, N., Noble, E., Jones, D., & Burn, D. J. (2006). Life with communication changes in parkinson's disease. *Age and Ageing, 35*(3), 235-239, ISSN-L 0002-0729. Monetta, L., Grindrod, C. M., & Pell, M. D. (2008). Effects of working memory capacity on

disease. *Journal of Neurolinguistics, 21*(5), 400-417, ISSN-L 0911-6044. Monetta, L., Grindrod, C. M., & Pell, M. D. (2009). Irony comprehension and theory of mind deficits in patients with parkinson's disease. *Cortex, 45*(8), 972-981, ISSN-L 0010-9452. Monetta, L., & Pell, M. D. (2007). Effects of verbal working memory deficits on metaphor

degenerative conditions. *Brain, 128*(3), 584-596, Online ISSN 1460-2156. Longworth, C. E., Tyler, L. K., & Marslen-Wilson, W. D. (2003). Language deficits and basal ganglia lesions: The past tense. *Brain and Language, 87*(1), 7-8, ISSN-L 0093-934X. Mari-Beffa, P., Hayes, A. E., Machado, L., & Hindle, J. V. (2005). Lack of inhibition in

*Psychological Review, 98*(1), 122-142, ISSN-L 0033-295X.

disease. *Brain and Language, 43*(2), 169-189, ISSN-L 0093-934X.

*Experimental Neuropsychology, 18*(6), 793-809, ISSN-L 1380-3395.

*Psychology, 21*(2), 330-346, ISSN-L 0954-1446.

*Language, 42*(1), 38-51, ISSN-L 0093-934X.

*Disorders, 21*(2), 199-207, ISSN-L 0885-3185.

89, ISSN-L 0093-934X.

aphasia. *Neurology, 67*(8), 1444-1451, ISSN-L 0028-3878.

L 0022-3018.

638-646, ISSN-L 0028-3932.

The time course of lexical access in speech production: A study of picture naming.

parkinson's disease. *The Journal of Nervous and Mental Disease, 178*(6), 360-365, ISSN-

Speech production, syntax comprehension, and cognitive deficits in parkinson's

The basal ganglia and rule-governed language use: Evidence from vascular and

parkinson's disease: Evidence from a lexical decision task. *Neuropsychologia, 43*(4),

disease: New evidence from a lexical decision task. *Journal of Clinical and* 

attentional set-shifting, working memory, and processing speed on pragmatic language functioning in parkinson's disease. *European Journal of Cognitive* 

skills in alzheimer's disease, parkinson's disease, and normal aging. *Brain and* 

cognitive dysexecutive changes in progressive supranuclear palsy. *Movement* 

inference generation during story comprehension in adults with parkinson's

comprehension in patients with parkinson's disease. *Brain and Language, 101*(1), 80-


**Part 2** 

**Novel Methods to Evaluate the Symptoms** 

 **in Parkinson's Disease** 


## **Part 2**

## **Novel Methods to Evaluate the Symptoms in Parkinson's Disease**

188 Diagnostics and Rehabilitation of Parkinson's Disease

Signorini, M., & Volpato, C. (2006). Action fluency in parkinson's disease: A follow-up

Spicer, K. B., Brown, G. G., & Gorell, J. M. (1994). Lexical decision in parkinson disease: Lack

Strauss, E., Sherman, E. M. S., & Spreen, O. (2006). *A compendium of neuropsychological tests:* 

Swinney, D., Zurif, E. B., Prather, P., & Love, T. (1996). Neurological distribution of

Tang-Wai, D. F., Josephs, K. A., Boeve, B. F., Dickson, D. W., Parisi, J. E., & Petersen, R. C.

Ullman, M. T. (2001). A neurocognitive perspective on language: The declarative/procedural model. *Nature Reviews Neuroscience, 2*(10), 717-726, ISSN-L 1471-003X. Ullman, M. T., Corkin, S., Coppola, M., Hickok, G., Growdon, J. H., Koroshetz, W. J., & Pinker,

van Spaendonck, K. P. M., Berger, H. J. C., Horstink, M. W. I. M., Buytenhuijs, E. L., & Cools,

van, Herten M. (2006). *Executive control of sentence perception: An electrophysiological* 

van, Herten M., Chwilla, D. J., & Kolk, H. H. (2006). When heuristics clash with parsing

Vissers, C. T. W. M. (2008). *Monitoring in language perception: An electrophysiological* 

Waters, G. S., & Caplan, D. (1996). The measurement of verbal working memory capacity

Whiting, E., Copland, D., & Angwin, A. (2005). Verb and context processing in parkinson's disease. *Journal of Neurolinguistics, 18*(3), 259-276, ISSN-L 0911-6044. Wolters, E. C., & Bosboom, J. L. W. (2007). Parkinson's disease. In E. C. Wolters, T. van Laar

Zurif, E., Swinney, D., Prather, P., Solomon, J., & Bushell, C. (1993). An on-line analysis of

visuospatial dysfunction. *Neurology, 61*(8), 1134-1135, ISSN-L 0028-3878. Thompson-Schill, S. L., & Botvinick, M. M. (2006). Resolving conflict: A response to martin

of evidence for generalized bradyphrenia. *Journal of Clinical and Experimental* 

*Administration, norms and commentary*. New York: Oxford University Press, ISBN

processing resources underlying language comprehension. *Journal of Cognitive* 

(2003). Pathologically confirmed corticobasal degeneration presenting with

and cheng (2006). *Psychonomic Bulletin and Review, 13*(3), 402-408, ISSN-L 1069-9384.

S. (1997). A neural dissociation within language: Evidence that the mental dictionary is part of declarative memory, and that grammatical rules are processed by the procedural system. *Journal of Cognitive Neuroscience, 9*(2), 266-276, ISSN 1530-8898. van de Meerendonk, N., Kolk, H. H. J., Chwilla, D. J., & Vissers, C. T. W. M. (2009).

Monitoring in language perception *Language and Linguistics Compass, 3*(5), 1211-

A. R. (1996). Executive functions and disease characteristics in parkinson's disease.

*investigation.* (Unpublished PhD thesis). Radbout Universiteit Nijmegen, the

routines: ERP evidence for conflict monitoring in sentence perception. *Journal of* 

*investigation.* (Unpublished PhD thesis). Radbout Universiteit Nijmegen, the

and its relation to reading comprehension. *The Quarterly Journal of Experimental* 

& H. W. Berendse (Eds.), *Parkinsonism and related disorders*. Amsterdam: VU

syntactic processing in broca's and wernicke's aphasia. *Brain and Language, 45*(3),

study. *Movement Disorders, 21*(4), 467-472, ISSN-L 0885-3185.

*Neuropsychology, 16*(3), 457-471, ISSN-L 1380-3395.

*Neuropsychologia, 34*(7), 617-626, ISSN-L 0028-3932.

*Cognitive Neuroscience, 18*(7), 1181-1197, ISSN 1530-8898.

*Neuroscience, 8*, 174-184, ISSN 1530-8898.

9780195159578.

1224, ISSN-L 1749-818X.

Netherlands, ISBN 9090208984.

Netherlands, ISBN 9789090230061.

*Psychology. A, 49*(1), 51-75, ISSN-L 0272-4987.

University Press, ISBN 9789086591503.

448-464, ISSN-L 0093-934X.

**9** 

*Japan* 

**Novel Methods to Evaluate** 

*2Osaka University Graduate School of Medicine,* 

*1Toneyama National Hospital,* 

**– Rigidity and Finger Tapping** 

**Symptoms in Parkinson's Disease** 

Takuyuki Endo1\*, Masaru Yokoe2\*, Harutoshi Fujimura1 and Saburo Sakoda1†

Parkinsonian symptoms such as tremor, rigidity, akinesia, and postural instability are perceived subjectively, and therefore understanding the degree of the symptoms varies depending on the neurologist. Sensing technologies and computer science have advanced and can now detect neurological symptoms and the detected data can be analyzed by software and described in a similar manner to how neurologists perceive those symptoms. This chapter discusses two popular neurological examinations in Parkinson's disease (PD); one is rigidity, which is representative of passive movement, and the other is finger tapping,

Rigidity, a well known symptom of PD, is defined as increased muscle tone that is elicited when an examiner moves the patient's limbs, neck, or trunk, and this increased resistance to passive movement is equal in all directions (Fahn & Przedborski 2005). Many researchers have analyzed rigidity by applying biomedical engineering principles and electrophysiological techniques (Fung et al. 2000, Prochazka et al. 1997, Teravainen et al.

Finger tapping, one of The Unified Parkinson's Disease Rating Scale (UPDRS) items, is commonly used in daily neurological examinations. Its evaluation includes velocity,

To evaluate rigidity and finger tapping, it is necessary to sense muscle tone and finger movement. We have previously developed novel methods to evaluate rigidity and finger tapping (Endo et al. 2009, Kandori et al., 2004). In this chapter, we showed the usefulness of

We evaluated the effects of deep brain stimulation (DBS) of the subthalamic nucleus (STN) on rigidity and finger tapping using our measuring materials. The preceded study of the

**2. Evaluating the effects of deep brain stimulation on rigidity and finger** 

1989). However, we do not know exactly what we feel in muscle tone in PD.

amplitude, and rhythm. However, observation of these is subjective.

**1. Introduction**

**tapping** 

equally contributed authors †corresponding author

 \*

which is representative of active movement.

these systems as objective markers of treatment.

## **Novel Methods to Evaluate Symptoms in Parkinson's Disease – Rigidity and Finger Tapping**

Takuyuki Endo1\*, Masaru Yokoe2\*, Harutoshi Fujimura1 and Saburo Sakoda1† *1Toneyama National Hospital, 2Osaka University Graduate School of Medicine, Japan* 

## **1. Introduction**

Parkinsonian symptoms such as tremor, rigidity, akinesia, and postural instability are perceived subjectively, and therefore understanding the degree of the symptoms varies depending on the neurologist. Sensing technologies and computer science have advanced and can now detect neurological symptoms and the detected data can be analyzed by software and described in a similar manner to how neurologists perceive those symptoms. This chapter discusses two popular neurological examinations in Parkinson's disease (PD); one is rigidity, which is representative of passive movement, and the other is finger tapping, which is representative of active movement.

Rigidity, a well known symptom of PD, is defined as increased muscle tone that is elicited when an examiner moves the patient's limbs, neck, or trunk, and this increased resistance to passive movement is equal in all directions (Fahn & Przedborski 2005). Many researchers have analyzed rigidity by applying biomedical engineering principles and electrophysiological techniques (Fung et al. 2000, Prochazka et al. 1997, Teravainen et al. 1989). However, we do not know exactly what we feel in muscle tone in PD.

Finger tapping, one of The Unified Parkinson's Disease Rating Scale (UPDRS) items, is commonly used in daily neurological examinations. Its evaluation includes velocity, amplitude, and rhythm. However, observation of these is subjective.

To evaluate rigidity and finger tapping, it is necessary to sense muscle tone and finger movement. We have previously developed novel methods to evaluate rigidity and finger tapping (Endo et al. 2009, Kandori et al., 2004). In this chapter, we showed the usefulness of these systems as objective markers of treatment.

## **2. Evaluating the effects of deep brain stimulation on rigidity and finger tapping**

We evaluated the effects of deep brain stimulation (DBS) of the subthalamic nucleus (STN) on rigidity and finger tapping using our measuring materials. The preceded study of the

<sup>\*</sup> equally contributed authors

<sup>†</sup>corresponding author

Novel Methods to Evaluate Symptoms in Parkinson's Disease – Rigidity and Finger Tapping 193

Figure 1 shows a schematic diagram of the muscle tonus measurement system. Details of the device were described in a previous report (Endo et al. 2009). Briefly, elbow joint torque was estimated using the force along the Z-axis and the longitudinal length of the forearm. The elbow joint angle was calculated from the signal generated by the gyroscope. The EMG activity was recorded from surface electrodes attached to the *biceps brachii* and *triceps brachii*.

Fig. 1. Schematic diagram of the muscle tonus measurement system

The basic method for sensing finger tap movement has been described previously (Kandori et al. 2003, Shima et al. 2008). The finger-tapping measurement system used in this study is shown in Figure 2. A magnetic sensor consisting of two coils is used to measure fingertapping movement. The coil voltage depending on the distance between the two coils enables estimation of the distance between two fingertips. We calculated the rhythm,

Each subject with DBS-on state or DBS-off state was instructed to relax in a sitting position; the examiner applied the measuring device to the wrist joint of the subject and practiced passive flexion and extension movements at the elbow joint. The measurement of DBS-off state started at 1 min after DBS was turned off. The measurement was made by repeating the four phases of movement as described in a previous report (Endo et al. 2009): (1) holding the elbow at maximum extension for at least 3 s (Fig. 3A), (2) passive flexion for 2 s, (3) holding the elbow at maximum flexion for at least 3 s (Fig. 3B), and (4) passive extension for 2 s (ramp-and-hold). This measurement was repeated twice for each of the left and right upper limbs and the resulting values were averaged on each side independently. Two

measurements each for left and right upper limbs were obtained per subject.

**2.2.2 Finger tapping measurement system**

**2.3.1 Protocols for measuring rigidity**

**2.3 Protocols** 

amplitude, and velocity of the finger-tapping movement.

**2.2 Sensing methods** 

**2.2.1 Muscle tonus measurement device**

effects of STN-DBS revealed that rigidity responded immediately upon tuning DBS, while improvement of finger tapping needed longer time to manifest after tuning DBS. Thus, we analyzed Parkinsonian rigidity by comparing the DBS on state to the DBS off state and finger tapping by comparing pre-operation DBS to post-operation DBS in this study.

## **2.1 Subjects**

Five patients in whom PD was diagnosed according to British Brain Bank clinical criteria (Gibb & Lees 1988) and who received STN-DBS were included in this study. Clinical details of patients with PD who participated in rigidity analysis are shown in Table 1, and those in finger tapping are shown in Table 2. Prior to measurement, patients with PD were assessed using the UPDRS Part III. In this examination, rigidity was scored using a five-point scale (0 = no rigidity, 1 = slight or detectable only when activated, 2 = mild to moderate, 3 = marked, and 4 = severe), and finger-tapping was also scored using the five-point scale (0 = normal; 1 = mild slowing and/or reduction in amplitude; 2 = moderately impaired, definite and early fatiguing, may have occasional arrests in movement; 3 = severely impaired, frequent hesitation in initiating movements or arrests in ongoing movement; and 4 = can barely perform the task). This study was approved by the Institutional Review Board of Osaka University Hospital and written informed consent was obtained from all subjects.




Table 2. Clinical details of patients who participated in finger tapping analysis

## **2.2 Sensing methods**

192 Diagnostics and Rehabilitation of Parkinson's Disease

effects of STN-DBS revealed that rigidity responded immediately upon tuning DBS, while improvement of finger tapping needed longer time to manifest after tuning DBS. Thus, we analyzed Parkinsonian rigidity by comparing the DBS on state to the DBS off state and

Five patients in whom PD was diagnosed according to British Brain Bank clinical criteria (Gibb & Lees 1988) and who received STN-DBS were included in this study. Clinical details of patients with PD who participated in rigidity analysis are shown in Table 1, and those in finger tapping are shown in Table 2. Prior to measurement, patients with PD were assessed using the UPDRS Part III. In this examination, rigidity was scored using a five-point scale (0 = no rigidity, 1 = slight or detectable only when activated, 2 = mild to moderate, 3 = marked, and 4 = severe), and finger-tapping was also scored using the five-point scale (0 = normal; 1 = mild slowing and/or reduction in amplitude; 2 = moderately impaired, definite and early fatiguing, may have occasional arrests in movement; 3 = severely impaired, frequent hesitation in initiating movements or arrests in ongoing movement; and 4 = can barely perform the task). This study was approved by the Institutional Review Board of Osaka

finger tapping by comparing pre-operation DBS to post-operation DBS in this study.

University Hospital and written informed consent was obtained from all subjects.

Duration after DBS

pd1 73 M 5 One month 28 1/1(\*on/off) 1/1(\*on/off) pd2 70 F 13 One month 8 1/2 1/1 pd3 60 F 11 One year 59 2/3 2/2 pd4 63 F 18 6 years 40 1/2 1/1 pd5 72 F 29 5 years 29 1/1 1/1

Table 1. Clinical details of patients who participated in rigidity analysis. \* on/off; DBS-on/off

PD1 67 M Right Left 3 49 29 3/2 2/1 PD2 69 F Right Right 1 26 8 1/2 1/1 PD3 69 M Right Right 2 24 17 1/2 1/1 PD4 62 F Right Right 1 40 20 2/2 1/1 PD5 73 F Right Left 1 34 29 2/1 1/1

**Evaluation interval between pre and post (month)** 

**Initially affected site** 

Table 2. Clinical details of patients who participated in finger tapping analysis

Part III (\*on)

UPDRS score

Rigidity (Left)

**UPDRS Part**Ⅲ **Finger-tapping score** 

> **Post (L/R)**

**(L/R)** 

Rigidity (Right)

**UPDRS PartIII** 

**Pre Post Pre** 

Disease duration (y)

**2.1 Subjects** 

Age

**Age** 

(y) Sex

**(y) Sex Handed** 

### **2.2.1 Muscle tonus measurement device**

Figure 1 shows a schematic diagram of the muscle tonus measurement system. Details of the device were described in a previous report (Endo et al. 2009). Briefly, elbow joint torque was estimated using the force along the Z-axis and the longitudinal length of the forearm. The elbow joint angle was calculated from the signal generated by the gyroscope. The EMG activity was recorded from surface electrodes attached to the *biceps brachii* and *triceps brachii*.

Fig. 1. Schematic diagram of the muscle tonus measurement system

## **2.2.2 Finger tapping measurement system**

The basic method for sensing finger tap movement has been described previously (Kandori et al. 2003, Shima et al. 2008). The finger-tapping measurement system used in this study is shown in Figure 2. A magnetic sensor consisting of two coils is used to measure fingertapping movement. The coil voltage depending on the distance between the two coils enables estimation of the distance between two fingertips. We calculated the rhythm, amplitude, and velocity of the finger-tapping movement.

#### **2.3 Protocols**

## **2.3.1 Protocols for measuring rigidity**

Each subject with DBS-on state or DBS-off state was instructed to relax in a sitting position; the examiner applied the measuring device to the wrist joint of the subject and practiced passive flexion and extension movements at the elbow joint. The measurement of DBS-off state started at 1 min after DBS was turned off. The measurement was made by repeating the four phases of movement as described in a previous report (Endo et al. 2009): (1) holding the elbow at maximum extension for at least 3 s (Fig. 3A), (2) passive flexion for 2 s, (3) holding the elbow at maximum flexion for at least 3 s (Fig. 3B), and (4) passive extension for 2 s (ramp-and-hold). This measurement was repeated twice for each of the left and right upper limbs and the resulting values were averaged on each side independently. Two measurements each for left and right upper limbs were obtained per subject.

Novel Methods to Evaluate Symptoms in Parkinson's Disease – Rigidity and Finger Tapping 195

Fig. 3. Measuring protocol. A: holding the elbow at maximum extension. B: holding the

elbow at maximum flexion

Fig. 2. Finger-tapping measurement system

Figure 4A and Figure 5A shows the typical longitudinal data extracted from the right upper limb of patient pd3 in Table 1 with a UPDRS rigidity score of 2/3 (DBS-on/off). Figure 4A represents the DBS-off state and Figure 5A represents the DBS-on state. Torque-angle characteristics in passive flexion and passive extension are also shown in Figure 4B (DBS-off state) and Figure 5B (DBS-on state).

## **2.3.2 Protocols for measuring finger tapping**

Five patients with PD were evaluated 1 week before and 3 to 5 months after surgery. The magnetic sensors were worn on the subject's index finger and thumb. The subject practiced the finger tapping movement for about 10 s. The subject was asked to execute the finger tapping movements as quickly and widely as possible for 15 s. The finger-tapping wave of patient PD1 before and after intervention is shown in Figure 6.

## **2.4 Data analysis**

## **2.4.1 Data analysis for rigidity**

The resulting data were analyzed by extracting features from elbow joint torque-angle characteristics during passive flexion and extension as shown in Figure 7. The features used here were elastic coefficients in extension and flexion and the sum of the differences of averaged torque values. These were calculated as follows: for the elastic coefficients, the slopes of the regression lines for both flexion and extension were estimated based on the torque-angle data. The data from the start point to the last maximal extension phase were used to calculate the elastic coefficient, which included four to five cycles. At this time, torque values were adjusted for gravity using the mass of the forearms and hands as estimated from the subject's body weight (de Leva 1996). For the sum of the differences of averaged torque values, first we averaged the flexion torque values across four trials at a certain joint angle and also averaged the extension torque values similarly. Then, the differences of the averaged torque values at 30°, 60°, and 90° were calculated and the resulting values were summed.

These three features, that is, the elastic coefficients in extension and flexion and the sum of the differences of averaged torque values, were normalized using the mass of the subject's body weight, because these are dependent on the subject's muscle mass.

Figure 4A and Figure 5A shows the typical longitudinal data extracted from the right upper limb of patient pd3 in Table 1 with a UPDRS rigidity score of 2/3 (DBS-on/off). Figure 4A represents the DBS-off state and Figure 5A represents the DBS-on state. Torque-angle characteristics in passive flexion and passive extension are also shown in Figure 4B (DBS-off

Five patients with PD were evaluated 1 week before and 3 to 5 months after surgery. The magnetic sensors were worn on the subject's index finger and thumb. The subject practiced the finger tapping movement for about 10 s. The subject was asked to execute the finger tapping movements as quickly and widely as possible for 15 s. The finger-tapping wave of

The resulting data were analyzed by extracting features from elbow joint torque-angle characteristics during passive flexion and extension as shown in Figure 7. The features used here were elastic coefficients in extension and flexion and the sum of the differences of averaged torque values. These were calculated as follows: for the elastic coefficients, the slopes of the regression lines for both flexion and extension were estimated based on the torque-angle data. The data from the start point to the last maximal extension phase were used to calculate the elastic coefficient, which included four to five cycles. At this time, torque values were adjusted for gravity using the mass of the forearms and hands as estimated from the subject's body weight (de Leva 1996). For the sum of the differences of averaged torque values, first we averaged the flexion torque values across four trials at a certain joint angle and also averaged the extension torque values similarly. Then, the differences of the averaged torque values at 30°, 60°, and 90° were calculated and the

These three features, that is, the elastic coefficients in extension and flexion and the sum of the differences of averaged torque values, were normalized using the mass of the subject's

body weight, because these are dependent on the subject's muscle mass.

Fig. 2. Finger-tapping measurement system

**2.3.2 Protocols for measuring finger tapping**

patient PD1 before and after intervention is shown in Figure 6.

state) and Figure 5B (DBS-on state).

**2.4 Data analysis** 

**2.4.1 Data analysis for rigidity** 

resulting values were summed.

Fig. 3. Measuring protocol. A: holding the elbow at maximum extension. B: holding the elbow at maximum flexion

Novel Methods to Evaluate Symptoms in Parkinson's Disease – Rigidity and Finger Tapping 197

Fig. 5. Typical longitudinal data (A) and torque-angle characteristics (B) in passive flexion and passive extension (DBS-on state) obtained from the right upper limb of patient pd3 with

UPDRS rigidity score 2.

Fig. 4. Typical longitudinal data (A) and torque-angle characteristics (B) in passive flexion and passive extension (DBS-off state) obtained from the right upper limb of patient pd3 with UPDRS rigidity score 3.

Fig. 4. Typical longitudinal data (A) and torque-angle characteristics (B) in passive flexion and passive extension (DBS-off state) obtained from the right upper limb of patient pd3 with

UPDRS rigidity score 3.

Fig. 5. Typical longitudinal data (A) and torque-angle characteristics (B) in passive flexion and passive extension (DBS-on state) obtained from the right upper limb of patient pd3 with UPDRS rigidity score 2.

Novel Methods to Evaluate Symptoms in Parkinson's Disease – Rigidity and Finger Tapping 199

We statistically analyzed five parameters of repetitive index finger-to-thumb oppositions for 15 seconds (Fig. 8). A single finger-tapping interval (FTI) was defined as the interval between the onset of a finger tap and the onset of the next finger tap. We measured the following: the maximum opening velocity (MoV) in a single finger-tapping movement; the maximum closing velocity (McV) in a single finger-tapping movement; the maximum amplitude (MA) during a single finger-tapping movement; and the standard deviation (SD) of FTI, the index of rhythm as the variation of finger-tapping coordination. The mean MA, MoV, and McV for 15 s were

calculated. The frequency was the number of finger taps in 15 s (NFT).

Fig. 8. Measured amplitude and calculated velocity in finger-tapping movement. (a)

Using the data obtained from both left and right upper limbs of five patients with PD, 10 data sets on muscle tonus were available for final analysis. The effects of STN-DBS on three parameters are shown in Figures 9, 10, and 11. Age-matched normal values of elastic coefficients in extension and flexion and the sum of the differences of averaged torque values from 20 control subjects were 1.0[N\*m/rad\*kg], 1.0[N\*m/rad\*kg], and 1.0[N\*m/kg],

In the arms with a UPDRS rigidity score 2 or 3 in the DBS-off state, DBS-on improved their scores. Figures of elastic coefficients in extension and flexion and the sum of the differences of averaged torque values in this muscle tonus system supported UPDRS rigidity score improvement. In addition, these three parameters also showed improvement even in arms where the UPDRS rigidity scores did not improve in the DBS-on state. This result indicates

Measured amplitude, (b) Calculated velocity.

**2.5 Results**  *Rigidity* 

respectively.

**2.4.2 Data analysis for finger tapping** 

Fig. 6. The finger-tapping wave of patient PD1 before and after intervention

Fig. 7. Extracting features from torque-angle characteristics. Elastic coefficients in flexion and extension were calculated by estimating the slopes of the regression lines for both phases. Differences in the averaged torque values were calculated at 30°, 60°, and 90°.

## **2.4.2 Data analysis for finger tapping**

198 Diagnostics and Rehabilitation of Parkinson's Disease

Amplitude

[m/s]

[mm]

Velocity

1

0

100


100

Amplitude

[mm]

Velocity

[m/s]

1

0


0 15

Fig. 7. Extracting features from torque-angle characteristics. Elastic coefficients in flexion and extension were calculated by estimating the slopes of the regression lines for both phases. Differences in the averaged torque values were calculated at 30°, 60°, and 90°.

Fig. 6. The finger-tapping wave of patient PD1 before and after intervention

Post-STN-DBS

[sec]

[sec]

Pos show corr tapp

Preo perfo was

0 15

Pre-STN-DBS

We statistically analyzed five parameters of repetitive index finger-to-thumb oppositions for 15 seconds (Fig. 8). A single finger-tapping interval (FTI) was defined as the interval between the onset of a finger tap and the onset of the next finger tap. We measured the following: the maximum opening velocity (MoV) in a single finger-tapping movement; the maximum closing velocity (McV) in a single finger-tapping movement; the maximum amplitude (MA) during a single finger-tapping movement; and the standard deviation (SD) of FTI, the index of rhythm as the variation of finger-tapping coordination. The mean MA, MoV, and McV for 15 s were calculated. The frequency was the number of finger taps in 15 s (NFT).

Fig. 8. Measured amplitude and calculated velocity in finger-tapping movement. (a) Measured amplitude, (b) Calculated velocity.

## **2.5 Results**

### *Rigidity*

Using the data obtained from both left and right upper limbs of five patients with PD, 10 data sets on muscle tonus were available for final analysis. The effects of STN-DBS on three parameters are shown in Figures 9, 10, and 11. Age-matched normal values of elastic coefficients in extension and flexion and the sum of the differences of averaged torque values from 20 control subjects were 1.0[N\*m/rad\*kg], 1.0[N\*m/rad\*kg], and 1.0[N\*m/kg], respectively.

In the arms with a UPDRS rigidity score 2 or 3 in the DBS-off state, DBS-on improved their scores. Figures of elastic coefficients in extension and flexion and the sum of the differences of averaged torque values in this muscle tonus system supported UPDRS rigidity score improvement. In addition, these three parameters also showed improvement even in arms where the UPDRS rigidity scores did not improve in the DBS-on state. This result indicates

Novel Methods to Evaluate Symptoms in Parkinson's Disease – Rigidity and Finger Tapping 201

Fig. 11. Effects of deep brain stimulation on the sum of the differences of averaged torque

As shown in Table 2, improvement in UPDRS finger-tapping score after DBS was observed in PD1, PD2, and PD4. The finger-tapping wave of PD1 before and after intervention is shown in Figure 6. Irregular and disordered finger tapping changed to a smooth and correct performance after DBS. This system allows examiners to understand improvement at first sight. In the parameter analysis of finger-tapping movement, all patients with PD showed significant improvement after DBS in three parameters: mean of MoV, mean of McV, and mean of MA. However, it was not necessarily the case that STN-DBS improved the SD of FTI (Fig. 12). In summary, MoV, McV, and MA in PD1, PD2, and PD4 apparently improved,

We succeeded in showing the effects of DBS on rigidity and finger-tapping movement quantitatively using these instruments. The severity of symptoms obtained by these systems would not show much difference among examiners. Because neurologists could grasp subtle changes after not only DBS but also an increase in drug dose such as dopamine receptor agonists, these instruments would indicate treatment efficacy to neurologists before patients

In the present analysis, rigidity was quantified by "work", in which the average work was done by the torque motor over one cycle (Shapiro et al. 2007). However, the concept of "work" views the flexion and extension movements as a single system, and strictly speaking, it is different from the sum of the differences of averaged torque values that we extracted. If one repeats sinusoidal flexion and extension movements as a measurement protocol, most features could not be properly evaluated at each phase because the stretch

reflex has greater impact when the flexion phase is switched to the extension phase.

values. The filled area (less than 1.0[N\*m/kg]) represents the normal region.

suggesting these are possible treatment markers.

realized the improvement in their symptoms.

*Finger tapping* 

**3. Conclusion** 

that this muscle tonus measuring system is sensitive, objective, and precise. On the other hand, in arms with a UPDRS rigidity score of 1, which is a subtle change in muscle tonus, apparent improvement was not detected using this system. The difference of averaged torque values is the most sensitive among the three parameters.

Fig. 9. Effects of deep brain stimulation on the elastic coefficient in flexion. The filled area (less than 1.0[N\*m/rad\*kg]) represents the normal region.

Fig. 10. Effects of deep brain stimulation on the elastic coefficient in extension. The filled area (less than 1.0[N\*m/rad\*kg]) represents the normal region.

Fig. 11. Effects of deep brain stimulation on the sum of the differences of averaged torque values. The filled area (less than 1.0[N\*m/kg]) represents the normal region.

#### *Finger tapping*

200 Diagnostics and Rehabilitation of Parkinson's Disease

that this muscle tonus measuring system is sensitive, objective, and precise. On the other hand, in arms with a UPDRS rigidity score of 1, which is a subtle change in muscle tonus, apparent improvement was not detected using this system. The difference of averaged

Fig. 9. Effects of deep brain stimulation on the elastic coefficient in flexion. The filled area

Fig. 10. Effects of deep brain stimulation on the elastic coefficient in extension. The filled

area (less than 1.0[N\*m/rad\*kg]) represents the normal region.

torque values is the most sensitive among the three parameters.

(less than 1.0[N\*m/rad\*kg]) represents the normal region.

As shown in Table 2, improvement in UPDRS finger-tapping score after DBS was observed in PD1, PD2, and PD4. The finger-tapping wave of PD1 before and after intervention is shown in Figure 6. Irregular and disordered finger tapping changed to a smooth and correct performance after DBS. This system allows examiners to understand improvement at first sight. In the parameter analysis of finger-tapping movement, all patients with PD showed significant improvement after DBS in three parameters: mean of MoV, mean of McV, and mean of MA. However, it was not necessarily the case that STN-DBS improved the SD of FTI (Fig. 12). In summary, MoV, McV, and MA in PD1, PD2, and PD4 apparently improved, suggesting these are possible treatment markers.

## **3. Conclusion**

We succeeded in showing the effects of DBS on rigidity and finger-tapping movement quantitatively using these instruments. The severity of symptoms obtained by these systems would not show much difference among examiners. Because neurologists could grasp subtle changes after not only DBS but also an increase in drug dose such as dopamine receptor agonists, these instruments would indicate treatment efficacy to neurologists before patients realized the improvement in their symptoms.

In the present analysis, rigidity was quantified by "work", in which the average work was done by the torque motor over one cycle (Shapiro et al. 2007). However, the concept of "work" views the flexion and extension movements as a single system, and strictly speaking, it is different from the sum of the differences of averaged torque values that we extracted. If one repeats sinusoidal flexion and extension movements as a measurement protocol, most features could not be properly evaluated at each phase because the stretch reflex has greater impact when the flexion phase is switched to the extension phase.

Novel Methods to Evaluate Symptoms in Parkinson's Disease – Rigidity and Finger Tapping 203

Fig. 13. Prototype of compact muscle tonus measurement system

Fig. 12. Differences in five finger tapping parameters between before and after STN-DBS (A)SD of FTI (B)Mean of MA (C)Mean of MoV (D)Mean of McV (E)NFT

Fig. 12. Differences in five finger tapping parameters between before and after STN-DBS

(A)SD of FTI (B)Mean of MA (C)Mean of MoV (D)Mean of McV (E)NFT

Fig. 13. Prototype of compact muscle tonus measurement system

Novel Methods to Evaluate Symptoms in Parkinson's Disease – Rigidity and Finger Tapping 205

We previously reported that the muscle activity index in the static phase (EMG index) obtained for *biceps brachii* muscles, elastic coefficients, and sum of the differences of averaged torque values correlated well with the UPDRS score. Recently, we found that the EMG index is a good marker to distinguish a UPDRS rigidity score of 1 from the normal control (unpublished data). Because the elastic coefficients and the sum of the differences of averaged torque values seemed to be simple and better indicators of drug efficacy than the EMG index (unpublished data), we decided to use elastic coefficients and the sum of the differences of averaged torque values in this study. Rigidity is a clinical sign that gets worse

In finger tapping, we previously reported fourteen parameters of finger-tapping movement and a radar chart showed obvious differences in most of these parameters between normal controls and patients with PD (Yokoe et al. 2009). Principal component analysis showed that these parameters could be classified into three components: (1) mean of both amplitude and velocity, (2) number of finger tappings and mean FTI, and (3) SD of FTI. The first (velocityand amplitude-related parameters) and third (rhythm-related parameters) components contributed to the discrimination of PD from normal controls. Regarding which component reflects treatment efficacy, parameters in the first component, including mean of MoV, mean of McV, and mean of MA, are good markers. The second component, including the number of finger tappings, does not reflect treatment efficacy. The third component, including the SD of FTI, depends on the patient. The left hand of PD1 showed improvement, although the right hand of PD2 worsened. However, both fingers moved faster and larger after DBS (Fig. 12). These results indicate that DBS works on the first component parameters rather than

These novel systems for testing muscle tonus and finger-tapping, which are compact, simple, and efficient, are very useful for daily neurological examinations. The muscle tonus measurement system was recently established, as shown in Figure 13 (product of PI System Co. Ltd, http://www.pis.co.jp), and the finger-tapping measurement system recently came

These sensing systems identify rigidity or spasticity and the nature of abnormal finger tapping in PD and show Parkinsonian symptoms as a system error in software of repetitive

We thank Dr. Kenzo Akazawa (Osaka Institute of Technology, Department of Biomedical Engineering), Dr. Ryuhei Okuno (Setsunan University, Department of Electrical and Electronic Engineering), Dr. Toshio Tsuji (Hiroshima University, Faculty of Engineering), Dr. Akihiko Kandori (Hitachi Co. Ltd., Advanced Research Laboratory) and Dr. Kei Fukada (Osaka General Medical Center) for their assistance during the design of this study. This study was supported by the Program for Promotion of Fundamental Studies in Health

de Leva, P. (1996). Adjustments to Zatsiorsky-Seluyanov's segment inertia parameters.

*Journal of Biomechanics* Vol.29, No.9, (September1996), pp.1223-1230, ISSN 0022-0949

on the market in Japan from Hitachi Co. Ltd. as shown in Figure 14.

Sciences of the National Institute of Biomedical Innovation (NIBIO).

immediately after DBS and therefore, this system is suitable for the tuning of DBS.

those of the third component.

movement.

**4. Acknowledgment**

**5. References** 

Fig. 14. New finger tapping analysis system by Hitachi Co. Ltd. (Hitachi Computer Peripherals Co. Ltd., Tokyo branch, 1-11-1, Ohmorikita, Ohta-ku, Tokyo, Zip.143-0016, JAPAN, TEL: +81-3-5753-6870, FAX: +81-3-5753-6872)

We previously reported that the muscle activity index in the static phase (EMG index) obtained for *biceps brachii* muscles, elastic coefficients, and sum of the differences of averaged torque values correlated well with the UPDRS score. Recently, we found that the EMG index is a good marker to distinguish a UPDRS rigidity score of 1 from the normal control (unpublished data). Because the elastic coefficients and the sum of the differences of averaged torque values seemed to be simple and better indicators of drug efficacy than the EMG index (unpublished data), we decided to use elastic coefficients and the sum of the differences of averaged torque values in this study. Rigidity is a clinical sign that gets worse immediately after DBS and therefore, this system is suitable for the tuning of DBS.

In finger tapping, we previously reported fourteen parameters of finger-tapping movement and a radar chart showed obvious differences in most of these parameters between normal controls and patients with PD (Yokoe et al. 2009). Principal component analysis showed that these parameters could be classified into three components: (1) mean of both amplitude and velocity, (2) number of finger tappings and mean FTI, and (3) SD of FTI. The first (velocityand amplitude-related parameters) and third (rhythm-related parameters) components contributed to the discrimination of PD from normal controls. Regarding which component reflects treatment efficacy, parameters in the first component, including mean of MoV, mean of McV, and mean of MA, are good markers. The second component, including the number of finger tappings, does not reflect treatment efficacy. The third component, including the SD of FTI, depends on the patient. The left hand of PD1 showed improvement, although the right hand of PD2 worsened. However, both fingers moved faster and larger after DBS (Fig. 12). These results indicate that DBS works on the first component parameters rather than those of the third component.

These novel systems for testing muscle tonus and finger-tapping, which are compact, simple, and efficient, are very useful for daily neurological examinations. The muscle tonus measurement system was recently established, as shown in Figure 13 (product of PI System Co. Ltd, http://www.pis.co.jp), and the finger-tapping measurement system recently came on the market in Japan from Hitachi Co. Ltd. as shown in Figure 14.

These sensing systems identify rigidity or spasticity and the nature of abnormal finger tapping in PD and show Parkinsonian symptoms as a system error in software of repetitive movement.

## **4. Acknowledgment**

204 Diagnostics and Rehabilitation of Parkinson's Disease

Fig. 14. New finger tapping analysis system by Hitachi Co. Ltd. (Hitachi Computer Peripherals Co. Ltd., Tokyo branch, 1-11-1, Ohmorikita, Ohta-ku, Tokyo, Zip.143-0016,

JAPAN, TEL: +81-3-5753-6870, FAX: +81-3-5753-6872)

We thank Dr. Kenzo Akazawa (Osaka Institute of Technology, Department of Biomedical Engineering), Dr. Ryuhei Okuno (Setsunan University, Department of Electrical and Electronic Engineering), Dr. Toshio Tsuji (Hiroshima University, Faculty of Engineering), Dr. Akihiko Kandori (Hitachi Co. Ltd., Advanced Research Laboratory) and Dr. Kei Fukada (Osaka General Medical Center) for their assistance during the design of this study. This study was supported by the Program for Promotion of Fundamental Studies in Health Sciences of the National Institute of Biomedical Innovation (NIBIO).

## **5. References**

de Leva, P. (1996). Adjustments to Zatsiorsky-Seluyanov's segment inertia parameters. *Journal of Biomechanics* Vol.29, No.9, (September1996), pp.1223-1230, ISSN 0022-0949

**10** 

**Relevance of Aerodynamic Evaluation** 

Sarr Mamadou Moustapha1, Ghio Alain2, Espesser Robert2, Teston Bernard2, Dramé Moustapha3 and Viallet François2,4

*4Service de Neurologie du Centre Hospitalier du Pays d'Aix- Aix-en-Provence* 

Parkinsonian dysarthria is generally known under the name of hypokinetic dysarthria. Dysarthria, according to Darley et al (1969), is characterized by all speech disorders related to disturbances of muscular control of the speech organs, whose origin is a central or peripheral nervous system injury. So we must understand by dysarthria all failures related to either different levels of speech production (respiratory, phonatory, articulatory and even prosodic). Parkinsonian dysarthria, meanwhile, is mainly based on rigidity and hypokinesia. That's why it is considered as « hypokinetic » (Darley et al., 1975; Gentil et al., 1995). This term refers not only to reduction of articulatory movements but also to decreasing of speech prosody modulation described as monotonic (Viallet & Teston, 2007). Parkinsonian dysarthria arises, like other signs of Parkinson's disease, the depletion of dopamine in charge of phonatory incompetence by muscular hypokinesia. It is a major handicap factor that may compromise in long-term oral communication of the patient, as worsening over the course of the disease, responding less well to treatment and thereby posing additional difficulties in support. So we thought to better assess this dysarthria in order to gain a better understanding and improve management. This assessment can be done by perceptual analysis. She could also be done by various instrumental methods (acoustic and physiological) focusing on one of the speech production levels mentioned above. Such studies are numerous in literature and we will report some examples in this chapter. What is more rare in literature is assessment of parkinsonian dysarthria in study combined several levels as might allow, for example, the dual approach appealing to physiology of speech production with firstly an aerodynamic component related to pneumophonic coordination (respiratory and phonatory levels) and, secondly, an acoustic component in relation to phonoarticulatory coordination (phonatory and articulatory levels). Through this chapter we want show that it is possible to assess appropriately parkinsonian dysarthria by using aerodynamic parameters that combine respiratory and phonatory levels, so such an experiment that we report in this chapter after

**1. Introduction** 

having reviewed main methods of evaluation.

**in Parkinsonian Dysarthria** 

*2Laboratoire Parole et Langage-Aix-en-Provence* 

*1UFR Santé- Université deThiès* 

*3Université de Reims* 

*1Sénégal 2,3,4France* 


## **Relevance of Aerodynamic Evaluation in Parkinsonian Dysarthria**

Sarr Mamadou Moustapha1, Ghio Alain2, Espesser Robert2, Teston Bernard2, Dramé Moustapha3 and Viallet François2,4 *1UFR Santé- Université deThiès 2Laboratoire Parole et Langage-Aix-en-Provence 3Université de Reims 4Service de Neurologie du Centre Hospitalier du Pays d'Aix- Aix-en-Provence 1Sénégal 2,3,4France* 

### **1. Introduction**

206 Diagnostics and Rehabilitation of Parkinson's Disease

Endo, T.; Okuno, R.; Yokoe, M.; Akazawa, K. & Sakoda, S. (2009). A Novel Method for

Fahn, S.M.C.; Calne, D. & Goldstein, M. (1987). UPDRS Development Comittee. Unified Parkinson's disease rating scale, In: *Recent Developments in Parkinson*

Fahn, S. & Przedborski S. (2005). Parkinsonism, In: *Merritt's Neurology, eleventh edition,* Lewis P. Rawland, pp. 828-846, Lippincott Williams & Wilkins, ISBN 0-7817-5311-2, USA Fung, V.S.; Burne, J.A. & Morris, J.G.(2000). Objective quantification of resting and activated

Gibb, W.R. & Lees, A.J. (1988). The relevance of the Lewy body to the pathogenesis of

Kandori, A.; Yokoe, M.; Sakoda, S.; Abe, K.; Miyashita, T.; Oe, H.; Naritomi, H.; Ogata, K. &

Prochazka, A.; Bennett, D.J.; Stephens, M.J.; Patric, S.K.; Sears-Duru, R.; Roberts, T. &

Shapiro, M. B.; Vaillancourt, D. E.; Sturman, M. M.; Metman, L. V.; Bakay, R. A. E.; &

Teravainen, H.; Tsui, J.K.; Mak, E. & Calne, D.B. (1989). Optimal indices for testing

Yokoe, M.; Okuno, R.; Hamasaki, T.; Kurachi, Y.; Akazawa, K. & Sakoda, S. (2009). Opening

*ENGINEERING,* Vol.15, No.2, (June 2007), pp. 173-181, ISSN 1534-4320 Shima, K.; Tsui, T.; Kan, E.; Kandori, A.; Yokoe, M. & Sakoda, S. (2008). Measurement and

No.15, (November 2009), pp. 2218-2224, ISSN 0885-3185

Macmillan Healthcare Innformation, ISBN: 0-8816-7132-0, USA

*Disorders,* Vol.15, No.1, (January 2000), pp.48-55, ISSN 0885-3185

*Disorders,* Vol.12, No.1, (January 1997), pp.24-32, ISSN 0885-3185

Vol.51, No6, (June 1988), pp745-752, ISSN 0022-3050

pp.253-260, ISSN 0306-4552

pp.5628-5631, ISBN 978-1-4244-1814-5

(May1989), pp.180-183, ISSN:0317-1671

1353-8020

Systematic Analysis of Rigidity in Parkinson's Disease. *Movement Disorders,* Vol.24,

parkinsonian rigidity: a comparison of angular impulse and work scores. *Movement* 

idiopathic Parkinson's disease.*Journal of Neurology, Neurosurgery and Psychiatry,*

Tsukada, K. (2004). Quantitative magnetic detection of finger movements in patients with Parkinson's disease, *Neuroscience Research*, Vol.49, No.2 (June 2004),

Jhamandas, J.H. (1997). Measurement of rigidity in Parkinson's disease. *Movement* 

Corcos, D. M.(2007). Effects of STN DBS on Rigidity in Parkinson's Disease. *IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION* 

evaluation of finger tapping movements using magnetic sensors. *Conference Proceedings of IEEE Engineering in Medicine and Biology Society,* (August 2008),

parkinsonian rigidity. *Canadian Journal of Neurological Scieince,* Vol.16, No.2,

velocity, a novel parameter, for finger tapping test in patients with Parkinson's disease, *Parkinsonism & Related Disorders*, Vol.15, No.6, (July 2009), pp.440-444, ISSN

'

*s Disease,*

Parkinsonian dysarthria is generally known under the name of hypokinetic dysarthria. Dysarthria, according to Darley et al (1969), is characterized by all speech disorders related to disturbances of muscular control of the speech organs, whose origin is a central or peripheral nervous system injury. So we must understand by dysarthria all failures related to either different levels of speech production (respiratory, phonatory, articulatory and even prosodic). Parkinsonian dysarthria, meanwhile, is mainly based on rigidity and hypokinesia. That's why it is considered as « hypokinetic » (Darley et al., 1975; Gentil et al., 1995). This term refers not only to reduction of articulatory movements but also to decreasing of speech prosody modulation described as monotonic (Viallet & Teston, 2007). Parkinsonian dysarthria arises, like other signs of Parkinson's disease, the depletion of dopamine in charge of phonatory incompetence by muscular hypokinesia. It is a major handicap factor that may compromise in long-term oral communication of the patient, as worsening over the course of the disease, responding less well to treatment and thereby posing additional difficulties in support. So we thought to better assess this dysarthria in order to gain a better understanding and improve management. This assessment can be done by perceptual analysis. She could also be done by various instrumental methods (acoustic and physiological) focusing on one of the speech production levels mentioned above. Such studies are numerous in literature and we will report some examples in this chapter. What is more rare in literature is assessment of parkinsonian dysarthria in study combined several levels as might allow, for example, the dual approach appealing to physiology of speech production with firstly an aerodynamic component related to pneumophonic coordination (respiratory and phonatory levels) and, secondly, an acoustic component in relation to phonoarticulatory coordination (phonatory and articulatory levels). Through this chapter we want show that it is possible to assess appropriately parkinsonian dysarthria by using aerodynamic parameters that combine respiratory and phonatory levels, so such an experiment that we report in this chapter after having reviewed main methods of evaluation.

Relevance of Aerodynamic Evaluation in Parkinsonian Dysarthria 209

PD patients while F0 average value for control subject apparied in age and sex was 143.2 Hz. As well Canter (1963) found F0 average values of 129 Hz for patients and 106 Hz for normal subjects. F0 increased with disease severity (Metter & Hanson, 1986). Nevertheless, other studies have clearly demonstrated a F0 average reduction (Jankowski et al., 2004; Sanabria et al. 2001; Viallet et al., 2002,). It is therefore logical to agree on a certain diversity of trends in F0 that can be either lowered or increased or unchanged. F0 trends diversity could be due to biases related to patient age, gender, disease duration, variability of performance interand intra-individual as well as heterogeneity of measurement or evaluation methods. Regarding F0 variability in sentences production, it is reported much lower values in PD patients than normal subjects. Thus Canter (1963) noted frequency variations between 0.15 and 0.59 octaves for PD patients against 0.60 and 1.64 octaves for normal subjects. This limited variability observed in PD may be related to laryngeal rigidity that induces insufficient contraction including lack of crico-thyroid muscle which is mainly responsible for F0 increase. In sustained vowel task, there is disclosed an increase in F0 variability from cycle to cycle (Jitter) in patients, indicating an alteration of pneumophonic control stability

Regarding the vocal intensity, the results of perceptual analysis and acoustic measurements are not always consistent. For example, Fox and Ramig (1997) reported that the sonorous volume of PD patients was significantly lower than control subjects, around 2 to 4 decibels during speech production or other speech tasks such as sustained vowel. This result demonstrates clearly the hypophonic caracter of parkinsonian dysarthria. The results of other acoustic studies showed no significant differences between PD patients and normal subjects (Canter, 1963; Metter & Hanson, 1986). The alteration or no of the sonorous volume rather depend on the degree of severity of illness (Ludlow & Bassich, 1984). Despite these mixed results, however, there would be leaning towards a small reduction of mean intensity which falls within the phonatory incompetence associated with the subglottic pressure decreasing. The shimmer is for intensity what the jitter is for frequency, and it reflects intensity variability of sound vibration from cycle to cycle. A shimmer increasing in the task of sustained vowel has been reported in Parkinson's disease (PD) patients compared with control subjects, indicating an alteration of laryngeal stability control (Jimenez et al., 1997). These findings suggest that a reflex part of speech production control appears to be intact,

contrary to the dysfunction of voluntary control directly induced by the disease.

with a reduction of signal/noise ratio (Viallet & Teston, 2007).

Acoustic measurements during sustained vowel confirmed the perceptual abnormalities of timbre (blown, frayed or tremulous character) in addition to showing F0 and intensity increasing variability from cycle to cycle, changes longer term due mainly to the tremor,

The speed of speech of PD patients is highly variable from one subject to another (Darley et al, 1975). Some studies showed no significant difference between parkinsonian and normal subjects (Ackermann & Ziegler, 1991; Ludlow et al. 1987). Other studies have reported a

(Jankowski et al, 2004).

**3.3 Abnormalities of vocal timbre** 

**3.4 Speed of speech**

**3.2 Intensity** 

## **2. Perceptual analysis**

Perceptual analysis is subject to a large degree of subjectivity and inter or intra individual differences. However it can capture all functions involved in speech production system and is main foundation of parkinsonian dysarthria evaluation. On perceptual side, major disorder of parkinsonian speech is dysprosody. Prosody is defined as using of three vocal parameters (pitch, intensity and duration) which variations contribute to emotional and linguistic information. Parkinsonian voice is often described as low, monotonous, altered in timbre, too slow with hoarse character and difficult starting (Hartelius & Svensson, 1994). In addition articulations'problems were reported including a certain loss of identity of phonemes, the most suitable example being realization of plosives (/ t /, / d /) as fricatives (/ s /, / z /) due to insufficient closure of vocal tract (Robert & Spezza, 2005). These disorders can occur very early during disease's course, perhaps as early as the clinical onset of it even at the presymptomatic stage (Harel et al., 2004). Dysphonia is first manifestation that appears early. It is secondarily complemented by articulatory disorders and airflow dysfunctions (Ho et al. 1998; Logeman et al., 1978). However articulatory disorders and airflow dysfunctions alter intelligibility more than dysphonia. Chronological order of disorders appearing suggests abnormalities progression down to up of the vocal tract during disease's course. Disorders begin at laryngeal level and end with bilabial constriction via lingual and palate constriction also. In all cases, perceptual marks of parkinsonian dysarthria were well reported by Selby (1968). Points of emphasis disappear, voice volume decreases, while consonants pronunciation is deteriorating and sentence ends in a whisper. At clinical onset of parkinsonian dysarthria, voice is low, monotonous (no variation in height). After, progressive worsening of dysarthria leads to inaudible and unintelligible diction. In some cases general slowness of movement is also reflected in speech rate. In others cases patients talk quickly, tangle words and sometimes carry words acceleration until sentence ending, imitating feast walking. Perceptual disturbances of Parkinsonian speech could also be summarized by identifying two clusters. On the one hand, a main cluster of prosodic insufficiency that combines monotony of pitch and intensity, accent reduction, quick acceleration, variable flow and consonants imprecision. On the other hand, an accessory cluster of phonatory incompetence that is related to voice disturbances.

Despite large amount of information it provides, perceptual analysis must be supplemented by more objective methods of assessment.

## **3. Acoustic analysis**

Instrumental methods are generally limited in their analysis field. Despite this limitation, they allow, from quantified data, complex functions evaluation and objective comparisons between patients and normal subjects.

On acoustic side, perceptual impressions physical basis of Parkinsonian dysarthria have been studied by measuring several parameters.

### **3.1 Fundamental frequency**

Measurements of voice fundamental frequency (F0) reported mixed results. However, most studies concluded that a F0 average increase in PD patients during sustained vowel, text reading or spontaneous speech (Flint et al. 1992; Hertrich & Ackermann, 1993; Robert & Spezza, 2005). For example, Ludlow and Bassich (1984) found a F0 average of 165.8 Hz for PD patients while F0 average value for control subject apparied in age and sex was 143.2 Hz. As well Canter (1963) found F0 average values of 129 Hz for patients and 106 Hz for normal subjects. F0 increased with disease severity (Metter & Hanson, 1986). Nevertheless, other studies have clearly demonstrated a F0 average reduction (Jankowski et al., 2004; Sanabria et al. 2001; Viallet et al., 2002,). It is therefore logical to agree on a certain diversity of trends in F0 that can be either lowered or increased or unchanged. F0 trends diversity could be due to biases related to patient age, gender, disease duration, variability of performance interand intra-individual as well as heterogeneity of measurement or evaluation methods. Regarding F0 variability in sentences production, it is reported much lower values in PD patients than normal subjects. Thus Canter (1963) noted frequency variations between 0.15 and 0.59 octaves for PD patients against 0.60 and 1.64 octaves for normal subjects. This limited variability observed in PD may be related to laryngeal rigidity that induces insufficient contraction including lack of crico-thyroid muscle which is mainly responsible for F0 increase. In sustained vowel task, there is disclosed an increase in F0 variability from cycle to cycle (Jitter) in patients, indicating an alteration of pneumophonic control stability (Jankowski et al, 2004).

#### **3.2 Intensity**

208 Diagnostics and Rehabilitation of Parkinson's Disease

Perceptual analysis is subject to a large degree of subjectivity and inter or intra individual differences. However it can capture all functions involved in speech production system and is main foundation of parkinsonian dysarthria evaluation. On perceptual side, major disorder of parkinsonian speech is dysprosody. Prosody is defined as using of three vocal parameters (pitch, intensity and duration) which variations contribute to emotional and linguistic information. Parkinsonian voice is often described as low, monotonous, altered in timbre, too slow with hoarse character and difficult starting (Hartelius & Svensson, 1994). In addition articulations'problems were reported including a certain loss of identity of phonemes, the most suitable example being realization of plosives (/ t /, / d /) as fricatives (/ s /, / z /) due to insufficient closure of vocal tract (Robert & Spezza, 2005). These disorders can occur very early during disease's course, perhaps as early as the clinical onset of it even at the presymptomatic stage (Harel et al., 2004). Dysphonia is first manifestation that appears early. It is secondarily complemented by articulatory disorders and airflow dysfunctions (Ho et al. 1998; Logeman et al., 1978). However articulatory disorders and airflow dysfunctions alter intelligibility more than dysphonia. Chronological order of disorders appearing suggests abnormalities progression down to up of the vocal tract during disease's course. Disorders begin at laryngeal level and end with bilabial constriction via lingual and palate constriction also. In all cases, perceptual marks of parkinsonian dysarthria were well reported by Selby (1968). Points of emphasis disappear, voice volume decreases, while consonants pronunciation is deteriorating and sentence ends in a whisper. At clinical onset of parkinsonian dysarthria, voice is low, monotonous (no variation in height). After, progressive worsening of dysarthria leads to inaudible and unintelligible diction. In some cases general slowness of movement is also reflected in speech rate. In others cases patients talk quickly, tangle words and sometimes carry words acceleration until sentence ending, imitating feast walking. Perceptual disturbances of Parkinsonian speech could also be summarized by identifying two clusters. On the one hand, a main cluster of prosodic insufficiency that combines monotony of pitch and intensity, accent reduction, quick acceleration, variable flow and consonants imprecision. On the other hand,

an accessory cluster of phonatory incompetence that is related to voice disturbances.

by more objective methods of assessment.

between patients and normal subjects.

been studied by measuring several parameters.

**3. Acoustic analysis** 

**3.1 Fundamental frequency**

Despite large amount of information it provides, perceptual analysis must be supplemented

Instrumental methods are generally limited in their analysis field. Despite this limitation, they allow, from quantified data, complex functions evaluation and objective comparisons

On acoustic side, perceptual impressions physical basis of Parkinsonian dysarthria have

Measurements of voice fundamental frequency (F0) reported mixed results. However, most studies concluded that a F0 average increase in PD patients during sustained vowel, text reading or spontaneous speech (Flint et al. 1992; Hertrich & Ackermann, 1993; Robert & Spezza, 2005). For example, Ludlow and Bassich (1984) found a F0 average of 165.8 Hz for

**2. Perceptual analysis** 

Regarding the vocal intensity, the results of perceptual analysis and acoustic measurements are not always consistent. For example, Fox and Ramig (1997) reported that the sonorous volume of PD patients was significantly lower than control subjects, around 2 to 4 decibels during speech production or other speech tasks such as sustained vowel. This result demonstrates clearly the hypophonic caracter of parkinsonian dysarthria. The results of other acoustic studies showed no significant differences between PD patients and normal subjects (Canter, 1963; Metter & Hanson, 1986). The alteration or no of the sonorous volume rather depend on the degree of severity of illness (Ludlow & Bassich, 1984). Despite these mixed results, however, there would be leaning towards a small reduction of mean intensity which falls within the phonatory incompetence associated with the subglottic pressure decreasing. The shimmer is for intensity what the jitter is for frequency, and it reflects intensity variability of sound vibration from cycle to cycle. A shimmer increasing in the task of sustained vowel has been reported in Parkinson's disease (PD) patients compared with control subjects, indicating an alteration of laryngeal stability control (Jimenez et al., 1997). These findings suggest that a reflex part of speech production control appears to be intact, contrary to the dysfunction of voluntary control directly induced by the disease.

#### **3.3 Abnormalities of vocal timbre**

Acoustic measurements during sustained vowel confirmed the perceptual abnormalities of timbre (blown, frayed or tremulous character) in addition to showing F0 and intensity increasing variability from cycle to cycle, changes longer term due mainly to the tremor, with a reduction of signal/noise ratio (Viallet & Teston, 2007).

#### **3.4 Speed of speech**

The speed of speech of PD patients is highly variable from one subject to another (Darley et al, 1975). Some studies showed no significant difference between parkinsonian and normal subjects (Ackermann & Ziegler, 1991; Ludlow et al. 1987). Other studies have reported a

Relevance of Aerodynamic Evaluation in Parkinsonian Dysarthria 211

The rigidity of the laryngeal musculature is a major determinant of hypophonia associated with parkinsonian dysarthria. It has been demonstrated by studies in laryngoscopy which provided direct light on the anomalies of the larynx. Larynx anomalies include glottal gap by chord adherence default, sometimes hypertonia of ventricular bands and tremor which can be localized at chordal level or above glottal part of vocal tract (Jiang et al., 1999; Yuceturk et al., 2002). Laryngeal rigidity induces a particularly curved form of vocal cords responsible for the unusually large and constant aperture of the vocal tract (Smith et al.,

On physiological side it is mainly explored by electromyographic and kinematic methods. In fact electromyographic and kinematic methods permit to analyze strenght and movement of

The mobility of articulatory organs of speech, like other movements, is disturbed by two

The rigidity incrimination has been strengthened on the basis of certain works. For example Hunker et al. (1982) were able to evaluate a coefficient of rigidity by applying known forces on labial muscles and observing the resulting displacement. The lower lip of PD patients showed a significantly higher rigidity than control subjects, whereas for the upper lip, there was no significant difference between the two groups. Moreover a correlation between the degree of rigidity and the movements' reduction was observed by recording the lips movement with a strain- gauge system in connection with the muscular activities of inferior orbicularis and mentalis, (Barlow et al., 1983). However, this rigidity is not expressed identically on all articulatory organs, affecting preferentially muscles which are poor in neuro-muscular spindles and without stretch reflex such as the tongue comparatively to other muscles which are richer in neuro-muscular spindles and with monosynaptic reflex

The hypokinetic character of some articulatory movements during parkinsonian speech is reported in particular by Ackermann et al. (1993). In this study recording the lips and tongue movements with an electromagnetic system during the repetition of syllables [pa] and [ta], there was an increased frequency and decreased amplitude of articulatory movements during the repetition of the syllable [ta] and no anomaly was found during the repetition of the syllable [pa]. This result suggests that there may be different mechanical properties between the tongue and lips. Kinematic studies also showed that hypokinesia of muscles, thus the nature of motor performance, may depend on factors such as familiarity of the task, the existence of visual guidance (Connor and Abbs, 1991) or even speed of speech (Caligiuri, 1989). Finally the kinematic studies have also confirmed, in PD patients, lack of coordination between different muscles involved in the complex activity that is speech production. Indeed the kinematic analysis of jaw, upper and lower lip showed a different motor behavior of these three structures. The lower lip was working normally when the upper lip and jaw had velocity peaks and/or reduced amplitude of movement (Connor et

articulatory organs in order to better understand the motor speech disorders

major symptoms of Parkinson's disease: rigidity and hypokinesia.

activity such as the jaw elevators (Abbs et al., 1987).

**4.2 Phonatory system**

**4.3 Articulatory system**

**4.3.1 Articulatory organs movement**

1995).

al., 1989).

faster speech rate in PD patients (Weismes, 1984). Finally, the speech speed can also be slower (Volkmann et al., 1992). These differences reflect not only the variability between subjects, but also the possible variation of results depending on the task (Ho et al., 1998). In all cases, Parkinsonian speech is marked by abnormalities which are described a long time and may impact on speed: festination, palilalia and pseudo-stuttering with dysfluences (Monfrais-Pfauwadel, 2005). What is more, the fine analysis of the acoustic signal from read speech extracts with attentive listening has led to a better study of the Parkinsonian speech temporal organization: the speed of speech tends to be slower. This slowness seems correlated with a longer pause time, duration of breaks was found significantly higher in PD patients compared with control subjects (Duez, 2005). In addition breaks inside of words have been observed in PD patients and not in controls subjects. Finally many dysfluences, such as omissions, repetitions and false starts, were found almost exclusively in PD patients. Numerous breaks and dysfluences not only slow the speed of speech, but also deconstruct the language units, disrupt perceptive waiting of listeners and finally degrade intelligibility.

#### **3.5 Imprecise consonants**

The most typical perceptual error articulatory in Parkinsonian dysarthria, namely the realization of consonants as fricatives was also confirmed by the acoustic analysis. In effect during these tests, it is found, instead of a silence due to normally carried out occlusion, a signal corresponding to a low intensity friction noise due to air passage and defined as the spirantisation phenomenon. Similarly, the lack of acoustic contrasts reflecting a lack of articulation is a common feature of parkinsonian speech spectrograms (Kent & Rosenbek, 1982).

#### **3.6 Other anomalies**

Finally, other deviations were reported always in acoustic studies: the reduced duration of formant transitions (Connor et al. 1989; Forrest et al., 1989), the voicing of voiceless consonants assigned to the rigidity of the larynx, a control loss of voice onset time (VOT), that is to say, the time between the release of the consonant and the beginning of voicing, resulting in a lack of coordination between the larynx and articulatory organs (Forrest et al., 1989; Lieberman et al., 1992).

## **4. Physiological analysis**

It essentially uses electromyographic methods, vidéocinematographic, kinematic and aerodynamic. It provides quantitative data on respiratory plans, phonatory and articulatory (Teston, 2007).

#### **4.1. Respiratory system**

Kinematic studies have measured the thoracic and abdominal movements. The spirometric measurements allowed to assess the volumes of mobilized air during inspiration and expiration. At rest PD patients respiration is characterized by a shortening of respiratory cycle at the expense of expiration and, moreover, a relative decline of thoracic participation in respiratory movement. During speech production, it was noted in PD patients an inspiratory volume reduction of the thoracic cage, and an increase in inspiratory abdominal volume, which suggests an alteration of expiratory airflow necessary to set the appropriate contribution of laryngeal vibrator (Solomon & Hixon, 1993).

### **4.2 Phonatory system**

210 Diagnostics and Rehabilitation of Parkinson's Disease

faster speech rate in PD patients (Weismes, 1984). Finally, the speech speed can also be slower (Volkmann et al., 1992). These differences reflect not only the variability between subjects, but also the possible variation of results depending on the task (Ho et al., 1998). In all cases, Parkinsonian speech is marked by abnormalities which are described a long time and may impact on speed: festination, palilalia and pseudo-stuttering with dysfluences (Monfrais-Pfauwadel, 2005). What is more, the fine analysis of the acoustic signal from read speech extracts with attentive listening has led to a better study of the Parkinsonian speech temporal organization: the speed of speech tends to be slower. This slowness seems correlated with a longer pause time, duration of breaks was found significantly higher in PD patients compared with control subjects (Duez, 2005). In addition breaks inside of words have been observed in PD patients and not in controls subjects. Finally many dysfluences, such as omissions, repetitions and false starts, were found almost exclusively in PD patients. Numerous breaks and dysfluences not only slow the speed of speech, but also deconstruct the language units, disrupt perceptive waiting of listeners and finally degrade intelligibility.

The most typical perceptual error articulatory in Parkinsonian dysarthria, namely the realization of consonants as fricatives was also confirmed by the acoustic analysis. In effect during these tests, it is found, instead of a silence due to normally carried out occlusion, a signal corresponding to a low intensity friction noise due to air passage and defined as the spirantisation phenomenon. Similarly, the lack of acoustic contrasts reflecting a lack of articulation is a common feature of parkinsonian speech spectrograms (Kent & Rosenbek,

Finally, other deviations were reported always in acoustic studies: the reduced duration of formant transitions (Connor et al. 1989; Forrest et al., 1989), the voicing of voiceless consonants assigned to the rigidity of the larynx, a control loss of voice onset time (VOT), that is to say, the time between the release of the consonant and the beginning of voicing, resulting in a lack of coordination between the larynx and articulatory organs (Forrest et al.,

It essentially uses electromyographic methods, vidéocinematographic, kinematic and aerodynamic. It provides quantitative data on respiratory plans, phonatory and articulatory

Kinematic studies have measured the thoracic and abdominal movements. The spirometric measurements allowed to assess the volumes of mobilized air during inspiration and expiration. At rest PD patients respiration is characterized by a shortening of respiratory cycle at the expense of expiration and, moreover, a relative decline of thoracic participation in respiratory movement. During speech production, it was noted in PD patients an inspiratory volume reduction of the thoracic cage, and an increase in inspiratory abdominal volume, which suggests an alteration of expiratory airflow necessary to set the appropriate

contribution of laryngeal vibrator (Solomon & Hixon, 1993).

**3.5 Imprecise consonants**

**3.6 Other anomalies**

1989; Lieberman et al., 1992).

**4. Physiological analysis**

**4.1. Respiratory system**

(Teston, 2007).

1982).

The rigidity of the laryngeal musculature is a major determinant of hypophonia associated with parkinsonian dysarthria. It has been demonstrated by studies in laryngoscopy which provided direct light on the anomalies of the larynx. Larynx anomalies include glottal gap by chord adherence default, sometimes hypertonia of ventricular bands and tremor which can be localized at chordal level or above glottal part of vocal tract (Jiang et al., 1999; Yuceturk et al., 2002). Laryngeal rigidity induces a particularly curved form of vocal cords responsible for the unusually large and constant aperture of the vocal tract (Smith et al., 1995).

## **4.3 Articulatory system**

On physiological side it is mainly explored by electromyographic and kinematic methods. In fact electromyographic and kinematic methods permit to analyze strenght and movement of articulatory organs in order to better understand the motor speech disorders

### **4.3.1 Articulatory organs movement**

The mobility of articulatory organs of speech, like other movements, is disturbed by two major symptoms of Parkinson's disease: rigidity and hypokinesia.

The rigidity incrimination has been strengthened on the basis of certain works. For example Hunker et al. (1982) were able to evaluate a coefficient of rigidity by applying known forces on labial muscles and observing the resulting displacement. The lower lip of PD patients showed a significantly higher rigidity than control subjects, whereas for the upper lip, there was no significant difference between the two groups. Moreover a correlation between the degree of rigidity and the movements' reduction was observed by recording the lips movement with a strain- gauge system in connection with the muscular activities of inferior orbicularis and mentalis, (Barlow et al., 1983). However, this rigidity is not expressed identically on all articulatory organs, affecting preferentially muscles which are poor in neuro-muscular spindles and without stretch reflex such as the tongue comparatively to other muscles which are richer in neuro-muscular spindles and with monosynaptic reflex activity such as the jaw elevators (Abbs et al., 1987).

The hypokinetic character of some articulatory movements during parkinsonian speech is reported in particular by Ackermann et al. (1993). In this study recording the lips and tongue movements with an electromagnetic system during the repetition of syllables [pa] and [ta], there was an increased frequency and decreased amplitude of articulatory movements during the repetition of the syllable [ta] and no anomaly was found during the repetition of the syllable [pa]. This result suggests that there may be different mechanical properties between the tongue and lips. Kinematic studies also showed that hypokinesia of muscles, thus the nature of motor performance, may depend on factors such as familiarity of the task, the existence of visual guidance (Connor and Abbs, 1991) or even speed of speech (Caligiuri, 1989). Finally the kinematic studies have also confirmed, in PD patients, lack of coordination between different muscles involved in the complex activity that is speech production. Indeed the kinematic analysis of jaw, upper and lower lip showed a different motor behavior of these three structures. The lower lip was working normally when the upper lip and jaw had velocity peaks and/or reduced amplitude of movement (Connor et al., 1989).

Relevance of Aerodynamic Evaluation in Parkinsonian Dysarthria 213

We used in this study the vocal evaluation system EVA 2 developed by the Laboratory of Speech and Language and sold by SQ-Lab society. EVA 2 operates as a workstation PC in the Windows environment (**See Figures 1 and 2**) with different software applications

The recording device includes an acoustic channel and two aerodynamic channels: one for measurement of mean oral airflow (MOAF), the other for the IOP measurement. It is thus possible to measure IOP during holding of a voiceless plosive. As a reminder, IOP is the

Architecture eva2

**Eva 2 architecture** 

Capteurs Sensors

dedicated to acoustic and aerodynamic analysis of speech production.

Fig. 1. General Feature of Eva 2 (workstation PC in the Windows environment)

Oral airflow sensors

Nasal airflow

Sensors

Fig. 2. EVA 2 hand piece with accessories (microphone, mouth, sensors etc.)

**5.2 Equipement and measurement technique**

**5.2.1 Equipement** 

IOP Sensors

Microphone

Mouth

Nose mouth

SGP Sensors

Connection cable to interface

estimated subglottic pressure.

PC avec carte d'acquisition 16 canaux

PC with acquisition card 16 channels

Interface de conditionnement des signaux

Interface signal conditioning

## **4.3.2 The articulatory organs forces**

It is usually assessed by using force transducers (Barlow et al., 1983). Muscle abnormalities are also detectable by using electromyographic explorations (Leanderson et al., 1971). The latter, despite their relative inaccessibility to non-medical researchers and the difficulties attached to their technical realization and interpretation, can provide a wealth of information on the chronology of muscular events and agonist-antagonists relation (Teston, 2007). It has been noted in parkinsonian dysarthria abnormal electromyographic signal during the study of orbicularis upper lip activity in repetition of the syllable [pa]. Indeed, in PD patients comparatively with control subjects and during repetition, the short bursts of muscle activity associated with each syllable had duration of shorter and shorter with an associated reduction in their amplitude (Netsel et al., 1975).

These physiological analysis concerning only one level of peripheral production of speech should be more and more replaced by the combined study of at least two levels; example of such a combined analysis is provided by the study of pneumophonic coordination.

## **4.4 The pneumophonic coordination**

It reflects the synergy of action that must exist during speech production between respiratory and laryngeal levels. The measurement of subglottic pressure (SGP) is a good indicator of this pneumophonic coordination. Indeed, the SGP is evaluable indirectly via the intraoral pressure (IOP) during the production of plosives and depends on both the expiratory airflow and laryngeal resistance. In other words, SGP results from a dynamic conflict between air thrust forces and laryngeal resistance, so the evaluation of its trend in a group of breath can give a powerful index of the speaker pneumophonic coordination (Teston, 2007). So such a parameter, related to the aerodynamic side of speech production with in addition its non-invasive character, can be relevant in the assessment of parkinsonian dysarthria.

## **5. Relevance of the evaluation of aerodynamic parameters**

Our research team has experience of using aerodynamic parameters in the assessment of parkinsonian dysarthria. The measurement of such parameters has been performed in PD patients and control subjects by using the voice evaluation system Eva 2 of SQ LAB society in Aix-En-Provence.

#### **5.1 Used parameters**

We worked primarily on three parameters: the intra-oral pressure (IOP), the mean oral airflow (MOAF) and laryngeal resistance (LR).

IOP is an indirect reflection of subglottic pressure which is itself nothing other than the pressure exerted by the expiratory air column on the vocal cords. Subglottic pressure is an important aerodynamic parameter and could allow a better understanding of some dysfunctions in speech production system (Baken & Orlikoff, 2000).

The MOAF is another important aerodynamic parameter associated with the laryngeal function and speech production. MOAF and subglottic pressure allow together a better description of the aerodynamic component of speech production.

Finally, the LR is the ratio of IOP on the MOAF and should be able to give an idea about the functioning level of laryngeal stage.

## **5.2 Equipement and measurement technique 5.2.1 Equipement**

212 Diagnostics and Rehabilitation of Parkinson's Disease

It is usually assessed by using force transducers (Barlow et al., 1983). Muscle abnormalities are also detectable by using electromyographic explorations (Leanderson et al., 1971). The latter, despite their relative inaccessibility to non-medical researchers and the difficulties attached to their technical realization and interpretation, can provide a wealth of information on the chronology of muscular events and agonist-antagonists relation (Teston, 2007). It has been noted in parkinsonian dysarthria abnormal electromyographic signal during the study of orbicularis upper lip activity in repetition of the syllable [pa]. Indeed, in PD patients comparatively with control subjects and during repetition, the short bursts of muscle activity associated with each syllable had duration of shorter and shorter with an

These physiological analysis concerning only one level of peripheral production of speech should be more and more replaced by the combined study of at least two levels; example of

It reflects the synergy of action that must exist during speech production between respiratory and laryngeal levels. The measurement of subglottic pressure (SGP) is a good indicator of this pneumophonic coordination. Indeed, the SGP is evaluable indirectly via the intraoral pressure (IOP) during the production of plosives and depends on both the expiratory airflow and laryngeal resistance. In other words, SGP results from a dynamic conflict between air thrust forces and laryngeal resistance, so the evaluation of its trend in a group of breath can give a powerful index of the speaker pneumophonic coordination (Teston, 2007). So such a parameter, related to the aerodynamic side of speech production with in addition its non-invasive character, can be relevant in the assessment of

Our research team has experience of using aerodynamic parameters in the assessment of parkinsonian dysarthria. The measurement of such parameters has been performed in PD patients and control subjects by using the voice evaluation system Eva 2 of SQ LAB society

We worked primarily on three parameters: the intra-oral pressure (IOP), the mean oral

IOP is an indirect reflection of subglottic pressure which is itself nothing other than the pressure exerted by the expiratory air column on the vocal cords. Subglottic pressure is an important aerodynamic parameter and could allow a better understanding of some

The MOAF is another important aerodynamic parameter associated with the laryngeal function and speech production. MOAF and subglottic pressure allow together a better

Finally, the LR is the ratio of IOP on the MOAF and should be able to give an idea about the

such a combined analysis is provided by the study of pneumophonic coordination.

**4.3.2 The articulatory organs forces**

**4.4 The pneumophonic coordination** 

parkinsonian dysarthria.

in Aix-En-Provence.

**5.1 Used parameters** 

airflow (MOAF) and laryngeal resistance (LR).

functioning level of laryngeal stage.

associated reduction in their amplitude (Netsel et al., 1975).

**5. Relevance of the evaluation of aerodynamic parameters**

dysfunctions in speech production system (Baken & Orlikoff, 2000).

description of the aerodynamic component of speech production.

We used in this study the vocal evaluation system EVA 2 developed by the Laboratory of Speech and Language and sold by SQ-Lab society. EVA 2 operates as a workstation PC in the Windows environment (**See Figures 1 and 2**) with different software applications dedicated to acoustic and aerodynamic analysis of speech production.

The recording device includes an acoustic channel and two aerodynamic channels: one for measurement of mean oral airflow (MOAF), the other for the IOP measurement. It is thus possible to measure IOP during holding of a voiceless plosive. As a reminder, IOP is the estimated subglottic pressure.

Fig. 1. General Feature of Eva 2 (workstation PC in the Windows environment)

Fig. 2. EVA 2 hand piece with accessories (microphone, mouth, sensors etc.)

Relevance of Aerodynamic Evaluation in Parkinsonian Dysarthria 215

Fig. 4. Oral mouth firmly against the underside of the face

 **Mask** 

 **Nasal tip** 

**1 and 2 : airflow sensors** 

Fig. 5. Note the suction probe for taking IOP (indicated by red arrow)

withdrawal of L-dopa for at least 12 h (condition called OFF DOPA).

The study included 24 subjects with PD who had an average age of 59 years (SD = 5.65) with a mean duration of disease about 9, 9 years (SD = 3.27). Patients were recorded after

**5.3 Patients and control subjects** 

## **5.2.2 Technique**

The measurement technique derives from the general theory of fluids dynamic applied to the airway. According to this theory it is possible by adjustments of valves to estimate pressure-flow upstream from the direct measurement of pressure-flow downstream of the target site. The adjustments of valves in question occur naturally during the pronunciation of certain sounds. For example, during production of the consonant / p / the lips are closed while the glottis is open. In contrary during pronunciation of the vowel / a / the lips are open while the glottis is closed. The different conformation of these examples of valves located on the airway (glottis and lips) has a physical impact on the pressure and flow dynamics prevailing inside airway. So during the realization of a voiceless plosive (/ p /), there is a momentary equilibration of intra-oral and subglottic pressures. This equilibration allows indirect assessment of SGP (upstream) via the direct measurement of IOP (downstream). The momentary equilibration of subglottic and intra-oral pressures occurs when holding the voiceless plosive because at this moment there is no phonation, the lips are closed and the glottis is open. Thus the peak pressure generated by holding a voiceless plosive may be considered as a "snapshot" of the subglottic pressure immediately preceding phonation. Similarly during the realization of the vowel (/ a /) following a voiceless plosive (lips are open and glottis is closed), it is possible to consider the oral airflow as a snapshot of translaryngeal airflow because of continuity of flow through the upper airway when the mouth is open. Once we got the two parameters, it suffices to calculate the ratio of intra-oral pressure on the oral airflow to determine the laryngeal resistance (Smitheran & Hixon, 1981; Demolin et al, 1997) (**See Figure 3**).

Fig. 3. Evaluation of the subglottic pressure.

Intraoral pressure (IOP) is equivalent to the subglottic pressure (PSG) during the labial occlusion of phoneme "p". Subglottic pressure is estimated indirectly by "Interrupted Airway Method" (Smitheran & Hixon, 1981), method validated notably by Demolin et al. (1997).

Measurements were performed while the subject produced at a constant rate the sentence "Papa ne m'a pas parlé de beau-papa" (Daddy did not speak to me about daddy-in-law). During this production, oral mouth was firmly against the underside of the face to minimize air leakage (**see Figure 4**). Taking IOP is performed using a disposable suction catheter approximately 4 mm (**See Figure 5**). The probe was placed between the incisors and should not be crushed between the teeth or be obstructed by saliva.

The measurement technique derives from the general theory of fluids dynamic applied to the airway. According to this theory it is possible by adjustments of valves to estimate pressure-flow upstream from the direct measurement of pressure-flow downstream of the target site. The adjustments of valves in question occur naturally during the pronunciation of certain sounds. For example, during production of the consonant / p / the lips are closed while the glottis is open. In contrary during pronunciation of the vowel / a / the lips are open while the glottis is closed. The different conformation of these examples of valves located on the airway (glottis and lips) has a physical impact on the pressure and flow dynamics prevailing inside airway. So during the realization of a voiceless plosive (/ p /), there is a momentary equilibration of intra-oral and subglottic pressures. This equilibration allows indirect assessment of SGP (upstream) via the direct measurement of IOP (downstream). The momentary equilibration of subglottic and intra-oral pressures occurs when holding the voiceless plosive because at this moment there is no phonation, the lips are closed and the glottis is open. Thus the peak pressure generated by holding a voiceless plosive may be considered as a "snapshot" of the subglottic pressure immediately preceding phonation. Similarly during the realization of the vowel (/ a /) following a voiceless plosive (lips are open and glottis is closed), it is possible to consider the oral airflow as a snapshot of translaryngeal airflow because of continuity of flow through the upper airway when the mouth is open. Once we got the two parameters, it suffices to calculate the ratio of intra-oral pressure on the oral airflow to determine the laryngeal resistance (Smitheran & Hixon, 1981;

*Airflow Intensité Intensi* 

*Intensity* 

*IOP*

*IOP*

*IOP* 

*SGP* 

**[p] [a]** 

*SGP*

*SGP*

**[a]**

2 1 2

**5.2.2 Technique** 

Demolin et al, 1997) (**See Figure 3**).

*Intra-oral pressure*

*Subglottic pressure*

(1997).

Fig. 3. Evaluation of the subglottic pressure.

Cords

not be crushed between the teeth or be obstructed by saliva.

Lips

*IOP*

*SGP*

Intraoral pressure (IOP) is equivalent to the subglottic pressure (PSG) during the labial occlusion of phoneme "p". Subglottic pressure is estimated indirectly by "Interrupted Airway Method" (Smitheran & Hixon, 1981), method validated notably by Demolin et al.

Measurements were performed while the subject produced at a constant rate the sentence "Papa ne m'a pas parlé de beau-papa" (Daddy did not speak to me about daddy-in-law). During this production, oral mouth was firmly against the underside of the face to minimize air leakage (**see Figure 4**). Taking IOP is performed using a disposable suction catheter approximately 4 mm (**See Figure 5**). The probe was placed between the incisors and should

1 **[p]**

Fig. 4. Oral mouth firmly against the underside of the face

Fig. 5. Note the suction probe for taking IOP (indicated by red arrow)

## **5.3 Patients and control subjects**

The study included 24 subjects with PD who had an average age of 59 years (SD = 5.65) with a mean duration of disease about 9, 9 years (SD = 3.27). Patients were recorded after withdrawal of L-dopa for at least 12 h (condition called OFF DOPA).

Relevance of Aerodynamic Evaluation in Parkinsonian Dysarthria 217

Fig. 6. Curve of the intra-oral pressure (IOP) in control subjects (CTRL) and OFF DOPA

0,17 (0,08)

0,21 (0,07)

Table 2. Average of mean oral airflow (MOAF) in control subjects (CTRL) and OFF DOPA

Fig. 7. Curve of mean oral airflow (MOAF) in control subjects (CTRL) and OFF DOPA

Finally for the LR, the graphical representation of mean values at six points of measurement in control subjects (CTRL) and OFF DOPA patients showed on one hand a more linear overall appearance of the control-subjects 'curve, on the other hand, a curve of OFF Dopa patients above that of control subjects from P1 to P4 and then, below, beyond P4. In addition standard deviations were significantly larger in OFF DOPA patients than in control subjects (**See Table 3 and Figure 8**). The comparison between the two groups was statistically significant (p <0.05).

**DAOM 1 DAOM 2 DAOM 3 DAOM 4 DAOM 5 DAOM 6** 

0,17 (0,08)

0,20 (0,08)

0,19 (0,07)

0,21 (0,06)

0,2 (0,08)

0,2 (0,06)

patients at six measurement points.

0,2 (0,09)

(0,08)

patients at six measurement points.

0,16 (0,08)

0,21 (0,07)

**NB:** DAOM is the french abbreviation of mean oral air flow (MOAF)

patients at six measurement points. Standard deviations are in parentheses.

**OFF DOPA** 

**CTRL** 0,2

50 healthy subjects served as controls. They had an average age of 61 years (SD = 10, 5).

## **5.4 Statistical analysis**

Statistical comparisons between groups (CTRL vs. OFF DOPA) were conducted on the basis of a linear mixed model (software "R" version 2.6.2, http://www.r-project.org). This model emerged as best suited to the analysis of grouped data. Indeed, the repeated measurements, longitudinal studies are data that are presenting a group structure. And in our case, a single individual is undergoing multiple measures, and structured data in this way no longer meet one of the fundamental prerequisites for the validity of a classical linear model, namely the independence of measures. We set our statistical comparisons as follows: measurements of aerodynamic parameters (IOP, MOAF and LR) accounted for the numerical factor of the model, the group (CTRL, OFF DOPA), the position of the consonant / p / in the sentence produced (P1, P2, P3, P4, P5, P6) and the subject (patients, controls) were the three factors model variability.

A p-value less than 0.05 was accepted as statistical significance.

## **5.5 Results**

In a study that involved 20 male patients registered in terms ON / OFF STIM and 11 control subjects, measurement of IOP showed a statistically significant fall of this parameter in OFF STIM patients compared to controls. The stimulation of Subthalamic nucleus (STN) improved partially IOP with a statistically significant difference at the first two measurement points whereas there was an effect of convergence on the third point (Sarr et al., 2009).

In another study that focused on 24 patients registered in OFF DOPA condition and compared with 50 control subjects, three parameters (IOP, MOAF and LR) were measured on six / P / (P1 to P6) of the sentence « Papa ne m'a pas parlé de beau papa » that subjects pronounced at a constant rate.

Here too, there was, as regards the IOP, a statistically significant decrease in patients compared to controls (p = 0.0001) (**See Table 1 and Figure 6**).


Table 1. Average of intraoral pressure (IOP) in control subjects (CTRL) and OFF DOPA patients at six measurement points. Standard deviations are in parentheses.

Concerning mean oral airflow (MOAF) the curve of mean values at six points of measurement in control subjects (CTRL) and OFF DOPA patients showed an convergent aspect at extremities so that P1 and P6 while at the other measurement points (P2 to P5), the two curves were clearly separated: that of control subjects remain above that of OFF DOPA patients (**see Table 2 and Figure 7**). The comparison between the two groups was statistically significant (p = 0.001).

Statistical comparisons between groups (CTRL vs. OFF DOPA) were conducted on the basis of a linear mixed model (software "R" version 2.6.2, http://www.r-project.org). This model emerged as best suited to the analysis of grouped data. Indeed, the repeated measurements, longitudinal studies are data that are presenting a group structure. And in our case, a single individual is undergoing multiple measures, and structured data in this way no longer meet one of the fundamental prerequisites for the validity of a classical linear model, namely the independence of measures. We set our statistical comparisons as follows: measurements of aerodynamic parameters (IOP, MOAF and LR) accounted for the numerical factor of the model, the group (CTRL, OFF DOPA), the position of the consonant / p / in the sentence produced (P1, P2, P3, P4, P5, P6) and the subject (patients, controls) were the three factors

In a study that involved 20 male patients registered in terms ON / OFF STIM and 11 control subjects, measurement of IOP showed a statistically significant fall of this parameter in OFF STIM patients compared to controls. The stimulation of Subthalamic nucleus (STN) improved partially IOP with a statistically significant difference at the first two measurement points whereas there was an effect of convergence on the third point (Sarr et

In another study that focused on 24 patients registered in OFF DOPA condition and compared with 50 control subjects, three parameters (IOP, MOAF and LR) were measured on six / P / (P1 to P6) of the sentence « Papa ne m'a pas parlé de beau papa » that subjects

Here too, there was, as regards the IOP, a statistically significant decrease in patients

4,46 (1,8)

5,73 (1,90)

Concerning mean oral airflow (MOAF) the curve of mean values at six points of measurement in control subjects (CTRL) and OFF DOPA patients showed an convergent aspect at extremities so that P1 and P6 while at the other measurement points (P2 to P5), the two curves were clearly separated: that of control subjects remain above that of OFF DOPA patients (**see Table 2 and Figure 7**). The comparison between the two groups was

Table 1. Average of intraoral pressure (IOP) in control subjects (CTRL) and OFF DOPA

**P1 P2 P3 P4 P5 P6** 

4,7 (1,9)

5,9 (1,93) 4,49 (1,9)

6,06 (1,98)

4,26 (1,7)

5,67 (2,00)

50 healthy subjects served as controls. They had an average age of 61 years (SD = 10, 5).

A p-value less than 0.05 was accepted as statistical significance.

compared to controls (p = 0.0001) (**See Table 1 and Figure 6**).

6,22 (2,2)

6,97 (2,15)

patients at six measurement points. Standard deviations are in parentheses.

**5.4 Statistical analysis** 

model variability.

**5.5 Results** 

al., 2009).

pronounced at a constant rate.

**OFF DOPA** 3,84

**CTRL** 5,23

(1,9)

(2,00)

statistically significant (p = 0.001).

Fig. 6. Curve of the intra-oral pressure (IOP) in control subjects (CTRL) and OFF DOPA patients at six measurement points.


Table 2. Average of mean oral airflow (MOAF) in control subjects (CTRL) and OFF DOPA patients at six measurement points. Standard deviations are in parentheses.

**NB:** DAOM is the french abbreviation of mean oral air flow (MOAF)

Fig. 7. Curve of mean oral airflow (MOAF) in control subjects (CTRL) and OFF DOPA patients at six measurement points.

Finally for the LR, the graphical representation of mean values at six points of measurement in control subjects (CTRL) and OFF DOPA patients showed on one hand a more linear overall appearance of the control-subjects 'curve, on the other hand, a curve of OFF Dopa patients above that of control subjects from P1 to P4 and then, below, beyond P4. In addition standard deviations were significantly larger in OFF DOPA patients than in control subjects (**See Table 3 and Figure 8**). The comparison between the two groups was statistically significant (p <0.05).

Relevance of Aerodynamic Evaluation in Parkinsonian Dysarthria 219

subjects. The decrease of IOP, found in a previous study (Sarr et al., 2009), seems to confirm definitively the alteration of this parameter in parkinsonian dysarthria. The decrease of IOP in patients is due to dopamine deficiency inherent in PD. Dopamine deficiency induces a dysfunction of the respiratory muscles that is partly responsible for the dysarthria (Murdoch et al., 1989). Indeed there are, within overall poor control of expiratory airflow, an alteration of the air quantity needed for the vibration of vocal cords (Jiang et al., 1999a ; Solomon & Hixon, 1993). However, the SGP is the result of a surge in air column by the pressure of lung with laryngeal resistance (Crevier-Buchman, 2007; Solomon, 2007). In the particular context of this study, when measuring IOP via the GSP, the glottis is open, at that time so it's a pressure gradient which is measured and not a static value. This gradient is the result of coordinated action between the respiratory muscles and laryngeal floor, so it indicates pneumophonic coordination quality. In PD, the fall in pulmonary pressure associated with hypokinetic movements of laryngeal muscles induced an alteration of the SGP. So we have shown in this study that it is possible to consider the GSP, or IOP, as a strong indicator of Parkinsonian dysarthria in general and its pneumophonic side particularly. We confirm in same time the results already published in a preliminary study (Sarr et al., 2009). Therefore, the measurement of IOP may allow together, comparing OFF DOPA patients and control subjects, assessment of the disease impact on speech disorders and contribution to evaluation of somes therapies such as L-dopa and subthalamic nucleus stimulation on parkinsonian dysarthria. As a reminder in our study (Sarr et al., 2009), STN stimulation

Regarding the mean oral airflow (MOAF), no difference was found between patients in OFF DOPA and control subjects at the first and last measurement point (P1 and P6). That means patients and control subjects would develop the same speed to start and finish the sentence « Papa ne m'a pas parlé de beau-papa ». Difference between the two groups was only noted during the course of sentence production. Indeed at other measurement points (P2 to P5), the curve of control subjects is well above that of patients in OFF DOPA, the difference between the two groups was significant (p = 0.001). It is also found that the curve of control subjects had a more stable pace with its roughly more linear shape **(See figure 7).** This could reflect a greater mastery of oral airflow by control subjects. In other words, the relatively greater irregularity of the curve of average values of MOAF in patients could reinforce the idea of a less good control of the MOAF. The reported decrease of MOAF could merely be a consequence of the fall in IOP. For example, assuming that laryngeal resistance is constant, the drop in IOP is necessarily associated with diminution in MOAF. However it seems exist in this study a large variability in laryngeal resistance in patients, as an overview was provided us in the morphological analysis of their value curves. This suggests a relatively fluctuating fall in MOAF which may also be related to tissue properties, configuration of the glottis and impedance of the vocal apparatus (Jiang & Tao, 2007). It is reported more generally in extrapyramidal syndromes glottic and supraglottic disorders such as movement disorders. These disorders can obstruct completely or partially the upper airway to induce sometimes severe airflow decrease (Vincken et al., 1984). The MOAF decline during speech production of PD patients could also be explained by similar mechanisms, among others. Finally for the laryngeal resistance (LR), Parkinson's disease could induce a greater variability of this parameter in patients compared to control subjects, as evidenced by the general morphology of control subjects and OFF DOPA patients' curves. In other words, control subjects would have more stable values of LR, which would mean that Parkinson's disease induces instability of laryngeal resistance. The values of standard deviations

improves IOP significantly in the initial part of the expiratory phase.


Table 3. Mean of laryngeal resistance in control subjects (CTRL) and OFF DOPA patients at six measurement points. Standard deviations are in parentheses.


Fig. 8. Curve of mean values of laryngeal resistance (LR) in control subjects (CTRL) and OFF DOPA patients at six measurement points.

Fig. 9. Histogram of mean values of laryngeal resistance (LR) in control subjects (CTRL) and patients OFF DOPA at six measurement points, with standard deviations. The histogram allows to better see the standard deviations significantly larger in patients.

#### **5.6 Discussions**

This new study that examined 24 patients and 50 control subjects confirms the decrease of IOP on all six measurement points of the sentence when comparing patients with control

33,77 (23,1)

30,81 (11,83)

Table 3. Mean of laryngeal resistance in control subjects (CTRL) and OFF DOPA patients at

Fig. 8. Curve of mean values of laryngeal resistance (LR) in control subjects (CTRL) and OFF

Fig. 9. Histogram of mean values of laryngeal resistance (LR) in control subjects (CTRL) and patients OFF DOPA at six measurement points, with standard deviations. The histogram

This new study that examined 24 patients and 50 control subjects confirms the decrease of IOP on all six measurement points of the sentence when comparing patients with control

allows to better see the standard deviations significantly larger in patients.

**OFF DOPA** 

28,05 (20,50)

(16,92)

DOPA patients at six measurement points.

**5.6 Discussions** 

**CTRL** 25,75

51,22 (40,14)

35,38 (13,54)

six measurement points. Standard deviations are in parentheses. **NB**: RL is the french abbreviation of laryngeal resistance (LR)

**RL 1 RL 2 RL 3 RL 4 RL 5 RL 6** 

33,99 (21,97)

33,40 (14,09)

27,09 (16,27)

30,84 (11,64)

29,21 (26,74)

33,64 (13,51) subjects. The decrease of IOP, found in a previous study (Sarr et al., 2009), seems to confirm definitively the alteration of this parameter in parkinsonian dysarthria. The decrease of IOP in patients is due to dopamine deficiency inherent in PD. Dopamine deficiency induces a dysfunction of the respiratory muscles that is partly responsible for the dysarthria (Murdoch et al., 1989). Indeed there are, within overall poor control of expiratory airflow, an alteration of the air quantity needed for the vibration of vocal cords (Jiang et al., 1999a ; Solomon & Hixon, 1993). However, the SGP is the result of a surge in air column by the pressure of lung with laryngeal resistance (Crevier-Buchman, 2007; Solomon, 2007). In the particular context of this study, when measuring IOP via the GSP, the glottis is open, at that time so it's a pressure gradient which is measured and not a static value. This gradient is the result of coordinated action between the respiratory muscles and laryngeal floor, so it indicates pneumophonic coordination quality. In PD, the fall in pulmonary pressure associated with hypokinetic movements of laryngeal muscles induced an alteration of the SGP. So we have shown in this study that it is possible to consider the GSP, or IOP, as a strong indicator of Parkinsonian dysarthria in general and its pneumophonic side particularly. We confirm in same time the results already published in a preliminary study (Sarr et al., 2009). Therefore, the measurement of IOP may allow together, comparing OFF DOPA patients and control subjects, assessment of the disease impact on speech disorders and contribution to evaluation of somes therapies such as L-dopa and subthalamic nucleus stimulation on parkinsonian dysarthria. As a reminder in our study (Sarr et al., 2009), STN stimulation improves IOP significantly in the initial part of the expiratory phase.

Regarding the mean oral airflow (MOAF), no difference was found between patients in OFF DOPA and control subjects at the first and last measurement point (P1 and P6). That means patients and control subjects would develop the same speed to start and finish the sentence « Papa ne m'a pas parlé de beau-papa ». Difference between the two groups was only noted during the course of sentence production. Indeed at other measurement points (P2 to P5), the curve of control subjects is well above that of patients in OFF DOPA, the difference between the two groups was significant (p = 0.001). It is also found that the curve of control subjects had a more stable pace with its roughly more linear shape **(See figure 7).** This could reflect a greater mastery of oral airflow by control subjects. In other words, the relatively greater irregularity of the curve of average values of MOAF in patients could reinforce the idea of a less good control of the MOAF. The reported decrease of MOAF could merely be a consequence of the fall in IOP. For example, assuming that laryngeal resistance is constant, the drop in IOP is necessarily associated with diminution in MOAF. However it seems exist in this study a large variability in laryngeal resistance in patients, as an overview was provided us in the morphological analysis of their value curves. This suggests a relatively fluctuating fall in MOAF which may also be related to tissue properties, configuration of the glottis and impedance of the vocal apparatus (Jiang & Tao, 2007). It is reported more generally in extrapyramidal syndromes glottic and supraglottic disorders such as movement disorders. These disorders can obstruct completely or partially the upper airway to induce sometimes severe airflow decrease (Vincken et al., 1984). The MOAF decline during speech production of PD patients could also be explained by similar mechanisms, among others.

Finally for the laryngeal resistance (LR), Parkinson's disease could induce a greater variability of this parameter in patients compared to control subjects, as evidenced by the general morphology of control subjects and OFF DOPA patients' curves. In other words, control subjects would have more stable values of LR, which would mean that Parkinson's disease induces instability of laryngeal resistance. The values of standard deviations

Relevance of Aerodynamic Evaluation in Parkinsonian Dysarthria 221

points. So the important rise of patients' laryngeal resistance in the first half of the sentence, beyond the intrinsic behavior of larynx, may result from a larger drop of their airflow as we had also seen. Therefore the decline in patients' IOP in the second half of the sentence would induce the consequent decline of their laryngeal resistance. That's why the global evolutionary pace of patients' curve shows increased laryngeal resistance in the first half of the sentence and significant drop in the second half. These high laryngeal resistances in the beginning of the sentence could be related to a lack of pneumophonic coordination, that is to say a kind of phase shift between the air expiratory thrust and resistance state of the larynx. Everything would go as if, when the expiratory air exerts its thrust, the larynx is still at resistance level higher than normal. The larynx would amount only to a resistance normal level later, which would explain the decrease of laryngeal resistance in the second half of the sentence. In short, this phenomenon simply imitate, but this time at the pneumophonic floor, the mechanism of control loss of voice onset time (VOT) which reflects a lack of coordination between the larynx and articulatory organs (Forrest et al. 1989; Lieberman et

It thus appears that there is in Parkinson's disease pneumophonic coordination impairments which are evidenced by the fall in IOP and that of MOAF in patients compared with control subjects. And it follows from the alteration of these two parameters a greater instability of laryngeal resistance which is none other than ratio of two above mentioned parameters. For didactic sake, we attempted to separately discuss the different parameters (IOP, MOAF and LR). However it should be borne in mind that these parameters are closely related functionally, and that any change in one inevitably has repercussions on the other two. Indeed, the SGP (reflected here by the IOP) depends on the air expiratory column thrust and laryngeal resistance (LR) while translaryngeal airflow (reflected here by the MOAF) is merely the result of the conflict between expiratory thrust forces (SGP) and laryngeal resistance (LR) forces (Crevier-Buchman, 2007; Solomon, 2007). Reported disturbances in the three parameters pose the problem of events' real chronology because of parameters' correlation. Is it the increase in LR at the beginning of the sentence which induces a fall in MOAF or, conversely, would it fall in MOAF resulting of expiratory thrust poor dynamic that would cause the increase in LR? It could probably be a simultaneous mechanism combining both alteration of expiratory dynamic (leading to fall in IOP and MOAF) and elevated laryngeal resistance notably at sentence beginning (reinforcing the fall in MOAF). Such a mechanism would both explain decrease in IOP and initial elevation of laryngeal resistance which both lead to a decline in MOAF that patients would be tempted to correct by vocal abuse. Finally, such a mechanism would fit perfectly to a lack of pneumophonic coordination imitating, as we noted above, the lack of coordination in phono-articulatory

Parkinson's disease, given the study of these three parameters, likely induces an alteration of pneumophonic coordination involving a decrease in IOP, a decrease in MOAF and instability of the LR. So the measurements of these three aerodynamics parameters, by reflecting the dysfunction induced by disease, may well be relevant factors in parkinsonian dysarthria evaluation. These parameters can also be valuable in evaluation of several therapies used in Parkinson's disease treatment in general and dysarthria in particular. A limit of the present work is the lack of acoustic parameters assessment. In fact we thought

al., 1992).

stage which induces the voice onset time (VOT).

**6. Conclusion** 

significantly larger in OFF DOPA patients than control subjects, again reflecting greater variability in the values of LR at all measurement points, seem to confirm this trend (**See Figure 9**). The study of LR values distribution histogram in the two groups seems to be in the same direction. Indeed, the histogram shows a fairly symmetrical distribution for control subjects where OFF DOPA patients have more skewed distributions, with thus a tendency to give most often higher LR values compared to control subjects (**See Figure 10**).

Fig. 10. Histogram of the distribution of values of laryngeal resistance (Zlaryng). There is a fairly symmetrical distribution for control subjects, while values distributions are more skewed in OFF DOPA patients.

Laryngeal resistance is equal to the ratio of IOP on MOAF; its greater constancy among control subjects may indicate a more perfect mastery of these two parameters. Besides this relative constancy of laryngeal resistance in control subjects was found in the measures performed by Smitheran and Hixon (1981). Smitheran and Hixon measurements were performed to compare laryngeal resistance values in non-invasive technique of measurement with those of invasive procedures. The mean laryngeal resistance in their patients was 35.7 + / - 3.3 cm H20/LPS (all measurements are between 30 and 43, 1). Blosser et al. (1992) reported similar values with a mean of 38.4 + / - 7.43 cm H20/LPS. In addition laryngeal resistance may reflect the larynx subject behavior. This has been demonstrated in a canine animal model which is able of maintaining, like humans, a constant subglottic pressure during phonation. In this model it was found a significant rise in laryngeal resistance when increasing the recurrent laryngeal nerve stimulation while the same nerve paralysis induced a significant drop of laryngeal resistance (Nasri et al., 1994). This significant rise in LR was also found in other disease involving larynx impairment with patients 'average to 65 + / - 8.15 cm H20/LPS (Blosser et al., 1992). We can therefore assume that the instability of laryngeal resistance in OFF DOPA patients reflects a more variable behavior of their larynx, but also a greater fluctuation in IOP and MOAF. We know, as seen previously, that patients have IOP lower than those of control subjects at all measurement

significantly larger in OFF DOPA patients than control subjects, again reflecting greater variability in the values of LR at all measurement points, seem to confirm this trend (**See Figure 9**). The study of LR values distribution histogram in the two groups seems to be in the same direction. Indeed, the histogram shows a fairly symmetrical distribution for control subjects where OFF DOPA patients have more skewed distributions, with thus a tendency

ctr dopa\_off

to give most often higher LR values compared to control subjects (**See Figure 10**).

Fig. 10. Histogram of the distribution of values of laryngeal resistance (Zlaryng).

0 50 100 150 200 250 300

more skewed in OFF DOPA patients.

0

10

20

30

Percent of Total

There is a fairly symmetrical distribution for control subjects, while values distributions are

Zlaryng

Laryngeal resistance is equal to the ratio of IOP on MOAF; its greater constancy among control subjects may indicate a more perfect mastery of these two parameters. Besides this relative constancy of laryngeal resistance in control subjects was found in the measures performed by Smitheran and Hixon (1981). Smitheran and Hixon measurements were performed to compare laryngeal resistance values in non-invasive technique of measurement with those of invasive procedures. The mean laryngeal resistance in their patients was 35.7 + / - 3.3 cm H20/LPS (all measurements are between 30 and 43, 1). Blosser et al. (1992) reported similar values with a mean of 38.4 + / - 7.43 cm H20/LPS. In addition laryngeal resistance may reflect the larynx subject behavior. This has been demonstrated in a canine animal model which is able of maintaining, like humans, a constant subglottic pressure during phonation. In this model it was found a significant rise in laryngeal resistance when increasing the recurrent laryngeal nerve stimulation while the same nerve paralysis induced a significant drop of laryngeal resistance (Nasri et al., 1994). This significant rise in LR was also found in other disease involving larynx impairment with patients 'average to 65 + / - 8.15 cm H20/LPS (Blosser et al., 1992). We can therefore assume that the instability of laryngeal resistance in OFF DOPA patients reflects a more variable behavior of their larynx, but also a greater fluctuation in IOP and MOAF. We know, as seen previously, that patients have IOP lower than those of control subjects at all measurement points. So the important rise of patients' laryngeal resistance in the first half of the sentence, beyond the intrinsic behavior of larynx, may result from a larger drop of their airflow as we had also seen. Therefore the decline in patients' IOP in the second half of the sentence would induce the consequent decline of their laryngeal resistance. That's why the global evolutionary pace of patients' curve shows increased laryngeal resistance in the first half of the sentence and significant drop in the second half. These high laryngeal resistances in the beginning of the sentence could be related to a lack of pneumophonic coordination, that is to say a kind of phase shift between the air expiratory thrust and resistance state of the larynx. Everything would go as if, when the expiratory air exerts its thrust, the larynx is still at resistance level higher than normal. The larynx would amount only to a resistance normal level later, which would explain the decrease of laryngeal resistance in the second half of the sentence. In short, this phenomenon simply imitate, but this time at the pneumophonic floor, the mechanism of control loss of voice onset time (VOT) which reflects a lack of coordination between the larynx and articulatory organs (Forrest et al. 1989; Lieberman et al., 1992).

It thus appears that there is in Parkinson's disease pneumophonic coordination impairments which are evidenced by the fall in IOP and that of MOAF in patients compared with control subjects. And it follows from the alteration of these two parameters a greater instability of laryngeal resistance which is none other than ratio of two above mentioned parameters. For didactic sake, we attempted to separately discuss the different parameters (IOP, MOAF and LR). However it should be borne in mind that these parameters are closely related functionally, and that any change in one inevitably has repercussions on the other two. Indeed, the SGP (reflected here by the IOP) depends on the air expiratory column thrust and laryngeal resistance (LR) while translaryngeal airflow (reflected here by the MOAF) is merely the result of the conflict between expiratory thrust forces (SGP) and laryngeal resistance (LR) forces (Crevier-Buchman, 2007; Solomon, 2007). Reported disturbances in the three parameters pose the problem of events' real chronology because of parameters' correlation. Is it the increase in LR at the beginning of the sentence which induces a fall in MOAF or, conversely, would it fall in MOAF resulting of expiratory thrust poor dynamic that would cause the increase in LR? It could probably be a simultaneous mechanism combining both alteration of expiratory dynamic (leading to fall in IOP and MOAF) and elevated laryngeal resistance notably at sentence beginning (reinforcing the fall in MOAF). Such a mechanism would both explain decrease in IOP and initial elevation of laryngeal resistance which both lead to a decline in MOAF that patients would be tempted to correct by vocal abuse. Finally, such a mechanism would fit perfectly to a lack of pneumophonic coordination imitating, as we noted above, the lack of coordination in phono-articulatory stage which induces the voice onset time (VOT).

#### **6. Conclusion**

Parkinson's disease, given the study of these three parameters, likely induces an alteration of pneumophonic coordination involving a decrease in IOP, a decrease in MOAF and instability of the LR. So the measurements of these three aerodynamics parameters, by reflecting the dysfunction induced by disease, may well be relevant factors in parkinsonian dysarthria evaluation. These parameters can also be valuable in evaluation of several therapies used in Parkinson's disease treatment in general and dysarthria in particular. A limit of the present work is the lack of acoustic parameters assessment. In fact we thought

Relevance of Aerodynamic Evaluation in Parkinsonian Dysarthria 223

Fox C, Ramig LO. *Vocal sound pressure level and self-perception of speech and voice in men and* 

Gentil M, Pollak P, Perret J. *La dysarthrie parkinsonienne*. Rev. Neurol. (Paris) 1995, 151, 2, 105-112. Harel B, Cannizzaro M, Snyder PJ. *Variability in fundamental frequency during speech in* 

Hartelius L, Svensson P. *Speech and swallowing symptoms associated with Parkinson's disease and* 

Hertrich I, Ackermann H. *Acoustic analysis of speech prosody in Huntington's and Parkinson's* 

Ho AK, Iansek R, Bradshaw JL. *Regulation of parkinsonian speech volume: the effect of interlocutor distance*. J Neurol Neurosurg Psychiatry 1999, 67 :199-202. Ho AK, Iansek R, Marigliani C, Bradshaw JL, Gates S. *Speech impairment in a large sample of* 

Hunker CJ, Abbs JH, Barlow SM. *The relationship between parkinsonian rigidity and hypokinesia in the oro-facial system: a quantitative analysis*. Neurology 1982, 32: 749-754. Jankowski L, Purson A, Teston B, Viallet F. *Effets de la L-Dopa sur la dysprosodie et le* 

Jiang J, Tao C. *The minimum glottal airflow to initiate vocal fold oscillation*. J Acoust Soc Am

Jiang J, Lin E, Wang J, Hanson DG. *Glottographic measures before and after Levodopa treament in* 

Jiang J, O'Mara T, Chen HJ, Stern JI, Vlagos D, Hanson D. *Aerodynamic measurements of* 

Jimenez-Jimenez FJ, Gamboa J, Nieto A, Guerrero J, Orti-Pareja M, Molina JA, Garcia-Albea

Kent RD, Rosenbek JC. *Prosodic disturbance and neurologic lesions*. Brain Lang 1982, 15: 259-291. Leanderson R, Persson A, Ohman S. *Electromyographic studies of facial muscle activity in speech*.

Lieberman P, Kako E, Friedman J, Tajchman G, Feldman LS, Jiminez EB. *Speech production,* 

Logemann JA, Fisher HB, Boshes B, Blonsky ER. *Frequency and cooccurrence of vocal tract* 

Ludlow CL, Connor NP, Bassich CJ. *Speech timing in Parkinson's and Huntington's disease*.

Ludlow CL, Bassich CJ. (1984). *Relationship between perceptual ratings and acoustic measures of* 

Metter EJ, Hanson WR. *Clinical and acoustical variability in hypokinetic dysarthria*. J Comm

E, Cobeta I. *Acoustic voice analysis in untreated patients with Parkinson's disease*.

*syntax comprehension and cognitive deficits in Parkinson's disease*. Brain Lang 1992, 43:

*dysfunctions in the speech of a large sample of Parkinson patients*. J Speech Hear Dis

*hypokinetic speech.* In: McNeil M, Rosenbek JC, Aronson AE (eds), The dysarthrias : Physiology, acoustic, perception, management, pp. 163-196. College Hill Press, San

*fonctionnement laryngien de patients parkinsoniens*. Actes des XXVèmes journées

*Multiple sclerosis : A survey*. Folia Phoniatr Logop 1994, 46: 9-17.

*disease: a preliminary report*. Clin Ling Phonetics 1993, 7:285-297.

*patients with Parkinson's disease*. Behav Neurol 1998, 11: 131-137.

d'études sur la parole, Fès, Maroc, 19-22 avril 2004.

*Parkinson's disease*. Laryngoscope 1999, 109: 1287-1294.

Parkinsonism & Related Disorders 1997, 3: 111-116.

Acta Oto-laryngologica 1971, 72: 361-369.

Brain and Language 1987, 32: 195-214.

Disorders 1986, 19: 347-366.

*patients with Parkinson's desease*. J Voice 1999a, 13:583-591.

Pathology 1997, 6: 85-94.

2004, 56: 24-29.

2007, 121(5):2873-81.

169-189.

Diego.

1978, 43: 47-57.

*women with idiopathic Parkinson disease*. American Journal of Speech-Language

*prodromal and incipient Parkinson's disease: a longitudinal case study*. Brain Cognition

that the sentence "Papa ne m'a pas parlé de beau-papa" is less appropriate than other tasks such as sustained vowel for evaluation of acoustic parameters. In any case, increasingly, methods for assessing parkinsonian dysarthria should be larger, including both central and peripheral levels of speech production. Future research to better understand and assess parkinsonian dysarthria would benefit from taking more account of a more global study of dysarthria contours.

#### **7. References**


that the sentence "Papa ne m'a pas parlé de beau-papa" is less appropriate than other tasks such as sustained vowel for evaluation of acoustic parameters. In any case, increasingly, methods for assessing parkinsonian dysarthria should be larger, including both central and peripheral levels of speech production. Future research to better understand and assess parkinsonian dysarthria would benefit from taking more account of a more global study of

Abbs JH, Hartman DE, Vishwanat B. *Orofacial motor control impairment in Parkinson's disease*.

Ackermann H, Grone BF, Hoch G, Schonle PW. *Speech freezing in Parkinson's disease: a* 

Ackermann H, Ziegler W. *Articulatory deficits in parkinsonian dysarthria: an acoustic analysis*. Journal of Neurology, Neurosurgery, and Psychiatry 1991, 54: 1093-1098. Baken RJ and Orlikoff RF (2000). *Clinical Measurement of Speech and Voice*, 2nd ed. (Singular,

Barlow SM, Abbs JH. *Force transducers for the evaluation of labial, lingual, and mandibular motor impairments*. Journal of Speech and Hearing Research 1983, 26: 616-621. Blosser S, Wigley FM, Wise RA. *Increase in translaryngeal resistance during phonation in* 

Caligiuri M. *The influence of speaking rate on articulatory hypokinesia in parkinsonian dysarthria*.

Canter GJ. *Speech characteristics of patients with Parkinson's disease: I Intensity, pitch and* 

Connor NP, Abbs JH. *Task-dependent variations in parkinsonian motor impairment*. Brain 1991,

Connor NP, Abbs JH, Cole KJ, Gracco VL. *Parkinsonian deficits in serial multiarticulate* 

Crevier-Buchman L. *Modélisation du fonctionnement laryngé*. In Auzou P, Monnoury-Rolland V, Pinto S, Özsancak C (eds),Les dysarthries. Solal. Marseille. 2007 pp 91-100. Darley FL, Aronson AE, Brown JR. *Differential diagnostic patterns of dysarthria*. Journal of

Darley FL, Aronson AE, Brown JR. (1975). *Motor speech disorders*, pp. 171-197. Saunders WB,

Demolin D, Giovanni A, Hassid S, Heim C, Lecuit V, Socquet A (1997). "*Direct and indirect measurements of subglottic pressure*", Proc. Larynx 97, Marseille, p.69-72. Duez D. *Organisation temporelle de la parole et dysarthrie parkinsonienne*. In Özsancak C, Auzou

Flint AJ, Black SE, Campbell-Taylor I, Gailey GF, Levinton C. *Acoustic analysis in the* 

Forrest K, Weismer G, Turner GS. *Kinematic, acoustic and perceptual analyses of connected speech* 

P (ed). Les troubles de la parole et de la déglutition dans la maladie de Parkinson.

*differentiation of Parkinson's disease and major depression*. Journal of Psycholinguistic

*produced by parkinsonian and normal geriatric adults*. J Acoust Soc Am 1989, 85: 2608-

*kinematic analysis of orofacial movement by means of electromyographic articulography*.

dysarthria contours.

Neurology 1987, 37: 394-398.

Folia phoniat 1993, 45: 84-89.

Thomson Learning, San Diego).

Brain lang 1989, 36: 493-502.

114: 321-332.

Philadelphia.

2622.

*rheumatoid arthritis*. Chest 1992, 102(2):387-90.

*duration*. J Speech Hearing Dis 1963, 28: 221-229.

*movements for speech*. Brain 1989, 112: 997-1009.

Speech and Hearing Research 1969, 12: 246-269.

Solal. Marseille. 2005, pp 195-211.

Research 1992, 21: 383-389.

**7. References** 


**11** 

*1,2Japan, 3China* 

**Objective Evaluation of the Severity** 

**Temporal Auto-Correlation of Activity** 

Weidong Pan1,3, Yoshiharu Yamamoto2 and Shin Kwak1

*1Department of Neurology, Graduate School of Medicine, The University of Tokyo,* 

*2Educational Physiology Laboratory, Graduate School of Education, The University of Tokyo, 3Department of Neurology, Shuguang Hospital Affiliated to Shanghai University of TCM,* 

Parkinson disease (PD) is a neurodegenerative disorder not only with motor symptoms, including resting tremor, rigidity, bradykinesia and postural instability, but also with nonmotor symptoms, including autonomic disturbance, sleep disturbance and depression. Due to the lack of objective biomarkers like the blood glucose level for diabetes mellitus, severity of parkinsonism has been evaluated by using the symptom-based Unified Parkinson Disease Rating Scale (UPDRS) (Martinez-Martin et al., 1994) that covers the various aspects of symptoms in patients with PD. Although the UPDRS is the standard method for the assessment of parkinsonism and the evaluation of drug effects, the scoring is not free from

Wearable accelerometers enable long-term recording of patient's movement during activities of daily living, and hence might be a suitable device for quantitative assessment of the disease severity and progression. Alterations in locomotor-activity levels and disturbances in rest-activity rhythms have long been recognized as integral signs of major psychiatric and neurological disorders (Teicher, 1995; Witting et al., 1990). Improvement of ambulatory activity monitors (actigraph) has enabled precise calibration and storage of thousands of activity measurements acquired at predetermined times, hence enabled longterm recording of patient's movement during ordinary daily living (Katayama, 2001; Korte et al., 2004; Mormont et al., 2000; Okawa et al., 1995; Teicher, 1995; Tuisku et al., 2003; van Someren et al., 1996). It has been demonstrated that use of these devices is useful for the quantitative estimation of human behavior properties in normal subjects and patients with a variety of diseases, including depression, pain syndrome, and PD (Jean-Louis et al., 2000; Korszun et al., 2002; Nakamura et al., 2007; Ohashi et al., 2003; Pan et al., 2007; van Someren et al., 1993; 1998; 2006). However, because the pattern of daily activity greatly influences the recording with accelerometers, recorded activity levels may not adequately reflect the disease severity (Fig 1). Therefore, reliable analytical methods of the body acceleration signal free from the level of activity are required to describe the characteristics of body activity during daily living. Recently, fractal analysis was shown to be a robust tool to

inter-rater variance or the fluctuation of the symptoms.

**1. Introduction** 

**of Parkinsonism Using Power-Law** 


## **Objective Evaluation of the Severity of Parkinsonism Using Power-Law Temporal Auto-Correlation of Activity**

Weidong Pan1,3, Yoshiharu Yamamoto2 and Shin Kwak1

*1Department of Neurology, Graduate School of Medicine, The University of Tokyo, 2Educational Physiology Laboratory, Graduate School of Education, The University of Tokyo, 3Department of Neurology, Shuguang Hospital Affiliated to Shanghai University of TCM, 1,2Japan, 3China* 

#### **1. Introduction**

224 Diagnostics and Rehabilitation of Parkinson's Disease

Monfrais-Pfauwadel MC. *Palilalies et pseudobégaiements*. In Özsancak C, Auzou P (eds). Les

Murdoch BE, Chenery HJ, Bowler S, Ingram JC. *Respiratory function in parkinson's subjects* 

Nasri S, Namazie A, Kreiman J, Sercarz JA, Gerrat BR, Berke GS. *A pressure-regulated model of normal and pathologic phonation*. Otolaryngol Head Neck Surg. 1994, 111 (6):807-15. Netsell R, Daniel B, Celesia GG. *Acceleration and weakness in parkinsonian dysarthria*. J Speech

Robert D, Spezza C. *La dysphonie parkinsonienne et les troubles articulatoires dans la dysarthrie* 

déglutition dans la maladie de Parkinson. Solal. Marseille. 2005, pp 131-159. Sanabria J, Garcia Ruiz P, Guttierez R. *The effect of Levodopa on vocal function in Parkinson's* 

Sarr MM, Pinto S, Jankowski L, Teston B, Purson A, Ghio A, Régis J, Peragut JC, Viallet F.

Selby G. (1968). *Parkinson's disease*. In: PJ Winken & GW Bruyn, Handbook of clinical neurology, pp. 173-211. North Holland Publishing Company, Amsterdam. Smith ME, Ramig LO, Dromey C, Perez KS, Samandari R. *Intensive voice treatment in Parkinson disease: laryngostroboscopic finding*. Journal of Voice 1995, 9: 453-459. Smitheran J., Hixon T. *A Clinical method for estimating laryngeal airway resistance during vowel* 

Solomon NP. *La fonction respiratoire dans la production de parole*. In Auzou P, Monnoury-Rolland V, Pinto S, Özsancak C (eds),Les dysarthries. Solal. Marseille. 2007 pp 44-55. Solomon NP, Hixton TJ. *Speech breathing in Parkinson's disease*. Journal of Speech and Hearing

Teston B. *L'étude instrumentale des gestes dans la production de la parole : Importance de* 

Viallet F, Teston B. *La dysarthrie dans la maladie de Parkinson*. In Auzou P, Monnoury-Rolland V, Pinto S, Özsancak C (eds),Les dysarthries. Solal. Marseille. 2007 pp 375-382. Viallet F, Teston B, Jankowski L, Purson A, Peragut J, Régis J, Witgas T. *Effects of* 

Vincken WG, Gauthier SG, Dollfuss RE, Hanson RE, Darauay CM, Cosio MG. *Involvement of* 

Volkmann J, Hefter H, Lange HW, Freund HJ. *Impairment of temporal organization of speech in* 

Weismer G. (1984) *Articulatory characteristics of parkinsonian dysarthria*. In: The dysarthrias:

Yuceturk AV, Yilmaz H, Egrilmez M, Karaca S. *Voice analysis and videolaryngostroboscopy in patients with Parkinson's disease*. Eur Arch Otorhinolaryngol 2002, 259: 290-293.

*l'aérophonométrie*. In Auzou P, Monnoury-Rolland V, Pinto S, Özsancak C (eds),Les

*pharmacological versus electrophysical treatements on parkinsonian dysprosody*. In :

*upper-airway muscles in extrapyramidal disorders. A cause of airflow limitation.* N Engl J

Physiology, acoustics, perception, management (McNeil M, Rosenbek J, Aronson

Marseille. 2005, pp 213-222.

Hearing Dis 1975, 40: 170-178.

*disease*. Clin Neuropharmacol 2001, 24: 99-102.

*production*. J Speech Hear Dis 1981, 46:138-146.

dysarthries. Solal. Marseille. 2007 pp 248-258.

Speech prosody, pp 679-682. Aix-en-Provence (2002)

*basal ganglion diseases*. Brain Lang 1992, 43: 386-399.

AE, eds), pp 101-130. San Diego: College Hill Press.

Dis 1989, 54:610-626.

165: 1055 – 1061.

Research 1993, 36: 294-310.

Med. 1984, 311(7):438-42.

troubles de la parole et de la déglutition dans la maladie de Parkinson. Solal.

*exhibiting a perceptible speech deficit: a kinematic and spirometric analysis*. J Speech Hear

*parkinsonienne*. In Özsancak C, Auzou P (eds). Les troubles de la parole et de la

*Contribution de la mesure de la pression intra-orale pour la compréhension des troubles de la coordination pneumophonique dans la dysarthrie parkinsonienne*. Rev Neurol 2009,

> Parkinson disease (PD) is a neurodegenerative disorder not only with motor symptoms, including resting tremor, rigidity, bradykinesia and postural instability, but also with nonmotor symptoms, including autonomic disturbance, sleep disturbance and depression. Due to the lack of objective biomarkers like the blood glucose level for diabetes mellitus, severity of parkinsonism has been evaluated by using the symptom-based Unified Parkinson Disease Rating Scale (UPDRS) (Martinez-Martin et al., 1994) that covers the various aspects of symptoms in patients with PD. Although the UPDRS is the standard method for the assessment of parkinsonism and the evaluation of drug effects, the scoring is not free from inter-rater variance or the fluctuation of the symptoms.

> Wearable accelerometers enable long-term recording of patient's movement during activities of daily living, and hence might be a suitable device for quantitative assessment of the disease severity and progression. Alterations in locomotor-activity levels and disturbances in rest-activity rhythms have long been recognized as integral signs of major psychiatric and neurological disorders (Teicher, 1995; Witting et al., 1990). Improvement of ambulatory activity monitors (actigraph) has enabled precise calibration and storage of thousands of activity measurements acquired at predetermined times, hence enabled longterm recording of patient's movement during ordinary daily living (Katayama, 2001; Korte et al., 2004; Mormont et al., 2000; Okawa et al., 1995; Teicher, 1995; Tuisku et al., 2003; van Someren et al., 1996). It has been demonstrated that use of these devices is useful for the quantitative estimation of human behavior properties in normal subjects and patients with a variety of diseases, including depression, pain syndrome, and PD (Jean-Louis et al., 2000; Korszun et al., 2002; Nakamura et al., 2007; Ohashi et al., 2003; Pan et al., 2007; van Someren et al., 1993; 1998; 2006). However, because the pattern of daily activity greatly influences the recording with accelerometers, recorded activity levels may not adequately reflect the disease severity (Fig 1). Therefore, reliable analytical methods of the body acceleration signal free from the level of activity are required to describe the characteristics of body activity during daily living. Recently, fractal analysis was shown to be a robust tool to

Objective Evaluation

of the Severity of Parkinsonism Using Power-Law Temporal Auto-Correlation of Activity 227

where *t* is time. This is equivalent to using the Gaussian second derivative (so-called "Mexican hat") wavelet to examine the raw signals (Fig. 2), though the integration approach automatically removes the local mean and the local linear trend, as in DFA. By changing the scale of the wavelet, this "hat-shaped" template dilates or contracts in time, probing transient increases or decreases in activity records in different time scales. The transient increases (low-high-low activity patterns) yield local maxima of the wavelet coefficients at their time points, while the decreases (high-low-high activity patterns) yield local minima of the wavelet coefficients (see Fig. 2). Next, the squared wavelet coefficients at the local maxima or minima were averaged for all the available days. As the coefficient gives the

squared *W(S)* was used, again as in DFA. Finally, the power-law exponent (α) was obtained separately for local maxima and minima as the slope of a straight line fit in the doublelogarithmic plot of *S* vs. *W(S)*2. This method yields the same α–values as does DFA (Ohashi et al., 2003), but separately for periods with higher and lower activity levels. The power-law (scaling) exponent, α, reflects the probability of a simultaneous increase or decrease in the variability at two distant points in time in the time series, applied to all distances up to *longrange* time scales, thereby probing the nature of "switching'" patterns between high and low values in a statistical sense. Larger power-law exponents indicate positive temporal autocorrelation or *persistency* in the increase or decrease, and lower values correspond to

0 200 400 600 800 1000 Time (min)

Fig. 2. Conceptual explanation of the method to obtain power-law exponents for local maxima and minima. (*top*) Various widths of hat-shaped wavelets are slid along the data to detect local minima (*middle*) and local maxima (*bottom*) of the wavelet coefficients. Note that the local minima and maxima appear at the transient decreases and increases of the activity, respectively. The power-law exponents are calculated from the slope of the log-log plot of squared wavelet coefficients vs. the scale for local minima and maxima. In the actual

the "hat-shaped" wavelet. This yields the same power-law exponents as those obtained by

the DFA method for the same local maxima and minima as obtained in this figure.

*(t)* with different time scales, the

Squared Wavelet

Coefficients

*(t)*, i.e., the derivative of

magnitude of local fluctuations matching the shape of

negative auto-correlation or *anti-persistency* (Ohashi et al., 2003).

Activity Counts

analysis, we used an integrated, rather than raw, time series and

Reprinted with permission from (Pan et al., 2007).

Wide

Narrow Wide

Width of Wavelets

Narrow

disclose hidden auto-correlation patterns in biological data, such as heartbeat and limb movement (Ohashi et al., 2003; Pan et al., 2007; Peng et al., 1995; Sekine et al., 2004; Struzik et al., 2006). Power-law auto-correlation exponents for local maxima and minima of fluctuations of locomotor activity would be the most useful for our purpose, as they represent the level of persistency of movement patterns (Ohashi et al., 2003; Pan et al., 2007).

Fig. 1. Examples of 24 h actigraph recording. (left) Each vertical bar indicates activity counts per min. Sleep time is indicated in blue. Patients with approximately the same severity show different activity patterns and the activity counts (right: mean ± S.D.). UPDRS total/Part III.

In this review, we show how we can extract hidden autocorrelation patterns reflecting the severity of parkinsonism from the actigraph recording of patients' activity, and demonstrate that the analysis using power-law exponents is useful for the evaluation of effects of therapy on motor and non-motor symptoms of parkinsonism.

## **2. Analytical method of the motionlogger recordings for power-law autocorrelation exponents**

We analyzed patients' physical activity records collected by an actigraph device using power-law exponents probing temporal auto-correlation of the activity counts. The powerlaw exponent for local maxima most sensitively and reliably reflects disability without being influenced by the presence of tremor or the patterns of daily living (Pan et al., 2007).

To examine temporal auto-correlation of the physical activity time series (i.e., *dynamic*  aspects of physical activity), we used an extended, random-walk analysis, the detrended fluctuation analysis (DFA) (Peng et al., 1995), with a modification (Ohashi et al., 2003) for various "real-world" signals including activity time series. Briefly, a daytime physical activity time series was integrated, as in DFA, and wavelets with different time scales (*S*) were slid along the time series and correlated with the data to obtain the wavelet coefficients (*W*(*S*)) at each point. The third derivative of the Gaussian function was used as the so-called "mother wavelet":

disclose hidden auto-correlation patterns in biological data, such as heartbeat and limb movement (Ohashi et al., 2003; Pan et al., 2007; Peng et al., 1995; Sekine et al., 2004; Struzik et al., 2006). Power-law auto-correlation exponents for local maxima and minima of fluctuations of locomotor activity would be the most useful for our purpose, as they represent the level of persistency of movement patterns (Ohashi et al., 2003; Pan et al., 2007).

65y F UPDRS 48/30 H & Y 1.5

71y M UPDRS 47/29 H & Y 1.5

53y M UPDRS 49/32 H & Y 1.5

on motor and non-motor symptoms of parkinsonism.

**correlation exponents** 

"mother wavelet":

52y F UPDRS 49/31 H & Y 1.5

<sup>0000</sup> <sup>0600</sup> <sup>1200</sup> <sup>1800</sup> <sup>1200</sup> awake-time sleep-time

Fig. 1. Examples of 24 h actigraph recording. (left) Each vertical bar indicates activity counts per min. Sleep time is indicated in blue. Patients with approximately the same severity show different activity patterns and the activity counts (right: mean ± S.D.). UPDRS total/Part III. In this review, we show how we can extract hidden autocorrelation patterns reflecting the severity of parkinsonism from the actigraph recording of patients' activity, and demonstrate that the analysis using power-law exponents is useful for the evaluation of effects of therapy

**2. Analytical method of the motionlogger recordings for power-law auto-**

influenced by the presence of tremor or the patterns of daily living (Pan et al., 2007).

Ψ

We analyzed patients' physical activity records collected by an actigraph device using power-law exponents probing temporal auto-correlation of the activity counts. The powerlaw exponent for local maxima most sensitively and reliably reflects disability without being

To examine temporal auto-correlation of the physical activity time series (i.e., *dynamic*  aspects of physical activity), we used an extended, random-walk analysis, the detrended fluctuation analysis (DFA) (Peng et al., 1995), with a modification (Ohashi et al., 2003) for various "real-world" signals including activity time series. Briefly, a daytime physical activity time series was integrated, as in DFA, and wavelets with different time scales (*S*) were slid along the time series and correlated with the data to obtain the wavelet coefficients (*W*(*S*)) at each point. The third derivative of the Gaussian function was used as the so-called

*(t) = t(3-t2)e-0.5t*

*2 ,2* where *t* is time. This is equivalent to using the Gaussian second derivative (so-called "Mexican hat") wavelet to examine the raw signals (Fig. 2), though the integration approach automatically removes the local mean and the local linear trend, as in DFA. By changing the scale of the wavelet, this "hat-shaped" template dilates or contracts in time, probing transient increases or decreases in activity records in different time scales. The transient increases (low-high-low activity patterns) yield local maxima of the wavelet coefficients at their time points, while the decreases (high-low-high activity patterns) yield local minima of the wavelet coefficients (see Fig. 2). Next, the squared wavelet coefficients at the local maxima or minima were averaged for all the available days. As the coefficient gives the magnitude of local fluctuations matching the shape of *(t)* with different time scales, the squared *W(S)* was used, again as in DFA. Finally, the power-law exponent (α) was obtained separately for local maxima and minima as the slope of a straight line fit in the doublelogarithmic plot of *S* vs. *W(S)*2. This method yields the same α–values as does DFA (Ohashi et al., 2003), but separately for periods with higher and lower activity levels. The power-law (scaling) exponent, α, reflects the probability of a simultaneous increase or decrease in the variability at two distant points in time in the time series, applied to all distances up to *longrange* time scales, thereby probing the nature of "switching'" patterns between high and low values in a statistical sense. Larger power-law exponents indicate positive temporal autocorrelation or *persistency* in the increase or decrease, and lower values correspond to negative auto-correlation or *anti-persistency* (Ohashi et al., 2003).

Fig. 2. Conceptual explanation of the method to obtain power-law exponents for local maxima and minima. (*top*) Various widths of hat-shaped wavelets are slid along the data to detect local minima (*middle*) and local maxima (*bottom*) of the wavelet coefficients. Note that the local minima and maxima appear at the transient decreases and increases of the activity, respectively. The power-law exponents are calculated from the slope of the log-log plot of squared wavelet coefficients vs. the scale for local minima and maxima. In the actual analysis, we used an integrated, rather than raw, time series and *(t)*, i.e., the derivative of the "hat-shaped" wavelet. This yields the same power-law exponents as those obtained by the DFA method for the same local maxima and minima as obtained in this figure. Reprinted with permission from (Pan et al., 2007).

Objective Evaluation

4C)

of the Severity of Parkinsonism Using Power-Law Temporal Auto-Correlation of Activity 229

When the effects of medication were examined, we found that all the patients who did not take any medication at the time of enrolement (n=6) showed lower α-values for local maxima on days more than three weeks after they received clinically effective antiparkinsonism medication than on those before (Fig. 3D). Although presence of tremor significantly increased the activity counts in the arms with tremor as compared with those without tremor (Fig. 4A), power-law scaling of the records from arms with tremor showed a linear correlation between log *S* and log *W*(*S*)2 in the range of 8 to 35 min (Fig. 4B) and αvalues for local maxima were the same between the arms with tremor and those without tremor (Fig. 4B) with significantly higher α-values in patient arms than in control arms (Fig.

Fig. 4. Effects of tremor on actigraph counts and the power-law exponents. (A) Daily profiles of physical activity for the arm affected with tremor and that without tremor of a patient with unilaterally predominant parkinsonism with continuous tremor on one side. (B) Average wavelet coefficients for local maxima among arms with tremor (tremor) and without tremor (no tremor) of 6 patients with tremor, 26 arms of 13 patients without tremor (without tremor) and 20 arms of 10 control subjects (control). (C) The power-law exponents for local maxima and minima. \*: *P* < 0. 05. Reprinted with permission from (Pan et al., 2007). Larger power-law exponents (α) indicate positive temporal auto-correlation, or *persistency*, in the increase or decrease in the variability of activity at two distant points in time in the time series, and lower values correspond to negative auto-correlation or *anti-persistency*  (Ohashi et al., 2003). In other words, a lower α for local maxima or minima of activity records reflects more frequent switching behavior from low to high or high to low physical activity, respectively, and the switching behavior from lower to higher activity levels is considered to be related to akinesia in patients with parkinsonism. We found lower α-values for local maxima during GC days than during BC days, in the mild group than in the severe group, and before medication than after medication. Thus, these results demonstrate that the power-law analyses accurately describe the well known phenomenon that under these conditions patients switch their physical activity from lower to higher levels more easily, in

This method enables to evaluate relationships between time scales and magnitudes of fluctuation within each time scale, eliminating *non-stationarity* in the input data (i.e., changes in the baseline and trends within the data windows at different scales) that could affect calculation of the magnitudes of fluctuation. Therefore, this approach is suitable for the analysis of the long-term data collected in ambulatory settings (Pan et al., 2007).

## **3. Quantitative analysis of parkinsonism using power-law auto-correlation exponents**

The data acquired during awake-time and sleep-time were separated with Action-W, Version 2 (Ambulatory Monitors Inc., Ardsley, NY) (Fig. 1) and the data during awake-time were used for analyses. Average wavelet coefficients for local maxima and minima of the severe and mild groups provided straight lines in the range of 8-35 min (Fig. 3A), indicative of very robust α-values. When the mean α-values for local maxima and minima were compared, they found a significantly lower α-value for local maxima in the mild group than in the severe group (Fig. 3B). All the patients (13 male and 9 female patients with Parkinson disease) in both the severe (Hoeh-Yahl score > 3.0; n=9)) and mild groups (H-Y score ≦ 3.0; n=10)) showed significantly lower α-values for local maxima on good-condition (GC) days than on bad-condition (BC) days that were classified according to diary scores, whereas there was no significant difference in the mean α-values for local minima (Fig. 3C).

Fig. 3. Local maxima and minima of fluctuation of physical activity. (A) Average wavelet coefficients, as a function of the wavelet scale, for local maxima and minima. The slopes are power-law exponents, . (B) Comparisons of the mean for the severe and the mild groups, (C) for BC and GC days and for individual patients, and (D) for days before and after antiparkinsonism medication and for each patient. \*: *P* < 0.05, \*\*: *P* < 0.01, and \*\*\*: *P* < 0.001. Reprinted with permission from (Pan et al., 2007).

This method enables to evaluate relationships between time scales and magnitudes of fluctuation within each time scale, eliminating *non-stationarity* in the input data (i.e., changes in the baseline and trends within the data windows at different scales) that could affect calculation of the magnitudes of fluctuation. Therefore, this approach is suitable for the

analysis of the long-term data collected in ambulatory settings (Pan et al., 2007).

there was no significant difference in the mean α-values for local minima (Fig. 3C).

log W(S)

2

7.4

BC GC

Fig. 3. Local maxima and minima of fluctuation of physical activity. (A) Average wavelet coefficients, as a function of the wavelet scale, for local maxima and minima. The slopes are power-law exponents, . (B) Comparisons of the mean for the severe and the mild groups, (C) for BC and GC days and for individual patients, and (D) for days before and after antiparkinsonism medication and for each patient. \*: *P* < 0.05, \*\*: *P* < 0.01, and \*\*\*: *P* < 0.001.

maxima minima maxima minima

8

A. B.

severe 8.6

severe mild slope=1.105 slope=0.977

maxima minima

maxima minima

8.6

8

7.4

BC GC

0.9 1.1 1.3 1.5 0.9 1.1 1.3 1.5 8 min log S 35 min log S 8 min 35 min

BC GC

Reprinted with permission from (Pan et al., 2007).

0 0.2 0.4 0.6 0.8 1 1.2 1.4

α

D.

maxima minima

before after

severe mild

BF AF BF AF

maxima minima

0 0.2 0.4 0.6 0.8 1 1.2 1.4

α

**exponents** 

C.

α

0.4 0.6 0.8 1 1.2 1.4 1.6 mild

slope=1.085 slope=0.847

**3. Quantitative analysis of parkinsonism using power-law auto-correlation** 

The data acquired during awake-time and sleep-time were separated with Action-W, Version 2 (Ambulatory Monitors Inc., Ardsley, NY) (Fig. 1) and the data during awake-time were used for analyses. Average wavelet coefficients for local maxima and minima of the severe and mild groups provided straight lines in the range of 8-35 min (Fig. 3A), indicative of very robust α-values. When the mean α-values for local maxima and minima were compared, they found a significantly lower α-value for local maxima in the mild group than in the severe group (Fig. 3B). All the patients (13 male and 9 female patients with Parkinson disease) in both the severe (Hoeh-Yahl score > 3.0; n=9)) and mild groups (H-Y score ≦ 3.0; n=10)) showed significantly lower α-values for local maxima on good-condition (GC) days than on bad-condition (BC) days that were classified according to diary scores, whereas When the effects of medication were examined, we found that all the patients who did not take any medication at the time of enrolement (n=6) showed lower α-values for local maxima on days more than three weeks after they received clinically effective antiparkinsonism medication than on those before (Fig. 3D). Although presence of tremor significantly increased the activity counts in the arms with tremor as compared with those without tremor (Fig. 4A), power-law scaling of the records from arms with tremor showed a linear correlation between log *S* and log *W*(*S*)2 in the range of 8 to 35 min (Fig. 4B) and αvalues for local maxima were the same between the arms with tremor and those without tremor (Fig. 4B) with significantly higher α-values in patient arms than in control arms (Fig. 4C)

Fig. 4. Effects of tremor on actigraph counts and the power-law exponents. (A) Daily profiles of physical activity for the arm affected with tremor and that without tremor of a patient with unilaterally predominant parkinsonism with continuous tremor on one side. (B) Average wavelet coefficients for local maxima among arms with tremor (tremor) and without tremor (no tremor) of 6 patients with tremor, 26 arms of 13 patients without tremor (without tremor) and 20 arms of 10 control subjects (control). (C) The power-law exponents for local maxima and minima. \*: *P* < 0. 05. Reprinted with permission from (Pan et al., 2007).

Larger power-law exponents (α) indicate positive temporal auto-correlation, or *persistency*, in the increase or decrease in the variability of activity at two distant points in time in the time series, and lower values correspond to negative auto-correlation or *anti-persistency*  (Ohashi et al., 2003). In other words, a lower α for local maxima or minima of activity records reflects more frequent switching behavior from low to high or high to low physical activity, respectively, and the switching behavior from lower to higher activity levels is considered to be related to akinesia in patients with parkinsonism. We found lower α-values for local maxima during GC days than during BC days, in the mild group than in the severe group, and before medication than after medication. Thus, these results demonstrate that the power-law analyses accurately describe the well known phenomenon that under these conditions patients switch their physical activity from lower to higher levels more easily, in

Objective Evaluation

of the Severity of Parkinsonism Using Power-Law Temporal Auto-Correlation of Activity 231

significant baseline (week 0) differences in UPDRS scores, Hoehn & Yahr stages, mean counts, power-law temporal exponent α values, or in the dosage of antiparkinonian drugs between the two groups. All the patients were evaluated at week 0, week 1, and week 13 for the actigraph recording, UPDRS and Secondary Symptom Score, which is conventionally used in China to evaluate the effects of antiparkinsonism drugs and consists of 8 parts, including the assessments of non-fluent speech, vertigo, insomnia/nightmares, headache, sweating or night sweats, tiredness, sense of cold, and dysuria (Long, 1992). The awake-time and sleep-time actigraph data were used separately for the power-law temporal analyses.

Fig. 5. Effects of TCM and placebo granules on actigraph recondings. (A) Counts of physical activity (mean ± S.D.). (B) Average wavelet coefficients, as a function of the wavelet scale for

The local power-law exponent α values during both awake-time and sleep-time were significantly decreased after taking ZAZ2 granule, but not after taking placebo granule (Table 1, Fig 5). The average wavelet coefficients exhibited linear relationships in the range of scales from 8 min to 35 min both for the ZAZ2 and placebo groups (Fig. 5B). The local power-law exponent α values during both awake-time and sleep-time were significantly decreased both week 1 and 13 in the ZAZ2 group, but not in the placebo group (Table 1 and Fig 5C, P<0.01; Bonferroni test). The beneficial effects of ZAZ2 were shown with UPDRS scores, as well; significant and persistent improvements were found in the scores of Part II, Part II + Part III, and Part IV (Table 1). These scores at week 13 were significantly different between the ZAZ2 group and the placebo group. As the exploratory outcome of this study, most of the secondary symptoms were improved after taking ZAZ2 granule, whereas only a

We evaluated the beneficial effects of TCM specifically on sleep disturbance of patients with parkinsonism. We used placebo-controlled, randomized study design, in which 48 patients

awake-time and sleep-time. The slopes are power-law exponents α. (C) Power-law

exponents α (mean ± S.D.). \*: *P* < 0.05, \*\*: *P* < 0.01. (Pan et al., 2011a)

few symptoms were transiently improved in the placebo group (Table 2).

other words they exhibit milder akinesia, when the parkinsonism is mild than when it is severe. It is worthy to note that lower α-values for local maxima were obtained for all the patients after medication than before, and when in good condition than in bad condition (Fig. 3C, D), thereby providing a temporal profile of parkinsonism in each individual patient.

These results thus suggest that analysis of power-law temporal auto-correlation of physical activity time series using the bi-directional extension (Ohashi et al., 2003) is applicable to patients with parkinsonism for the evaluation of motor dysfunction irrespective of the presence of tremor and may provide useful objective data necessary for the control of drug dosage in the out-patient clinic and also for the evaluation of new drugs for parkinsonism (Pan et al., 2007).

## **4. Evaluation of effects of traditional Chinese medicine on parkinsonian symptoms**

Conventional antiparkinsonism drugs effectively ameliorate the symptoms of patients with PD during the initial several years of onset, but become increasingly less effective and induce motor fluctuations including wearing-off, on-off, dopa-induced dyskinesia, and agonist-induced sleep attack (Arnulf et al., 2002; Comella, 2002; Hobson et al., 2002; Ondo et al., 2001; Pahwa et al., 2006). PD patients not infrequently suffer from non-motor symptoms, such as neuropsychiatric symptoms, autonomic symptoms, gastrointestinal symptoms, sensory symptoms, non-motor fluctuations (autonomic symptoms, cognitive or psychiatric symptoms, sensory symptoms including pain), fatigue, and sleep disturbance (Chaudhuri& Schapira, 2009; Miyasaki et al., 2006; Park& Stacy, 2009), and these non-motor symptoms may be intrinsic to the disease pathology or may be the result of treatment with dopaminergic agents. Several studies have established that the non-motor symptoms of PD are common, occur across all stages of PD, and are a key determinant of quality of life (Chaudhuri& Schapira, 2009).

Herbal remedies have a long history of use (particularly in East Asian countries) for alleviating various symptoms and have been increasingly used as alternative medicines worldwide, including the United States (De Smet, 2002). Traditional Chinese medicines (TCM) ameliorate various symptoms, particularly the ageing-related symptoms, and hence are likely to be beneficial for chronic diseases such as PD (Iwasaki et al., 2004; 2005a; 2005b). Good compliance for long-term use with few side effects may be another merit of TCM suitable for patients with PD (Lian& Luo, 2007; Zhao et al., 2007).

In order to evaluate the effects of TCM on symptoms of parkinsonism, we evaluated the effects of Zeng-xiao An-shen Zhi-chan 2 (ZAZ2) on patients with PD using this method together with conventional scales for parkinsonism (Pan et al., 2011a). ZAZ2 granule is made up of 14 kinds of herbs*; Uncaria rhynchophylla* 10 g, *Rehmanniae radix* 10 g, *Cornus officinalis* 8 g, *Asnaragus cochinchinensic* 10 g, *Paeonia lactiflora* 10 g, *Desertliving cistanche* 10 g, *Puerariae radix* 10 g, *Arisaema consanguineum Schott* 10 g, *Salviae Miltiorrhizae radix* 10 g, *Acorus tatarinowii* 10 g, *Curcuma longa Linn* 12 g, *Morindae officinalis radix* 10 g, *Rhizoma gastrodiae* 10 g, and*Rhizoma chuanxiong* 10 g. One hundred and fifteen patients with idiopathic PD took 8 g of either ZAZ2 granule or placebo granule that was not distinguished by appearance or taste for 13 weeks. Patients were randomly assigned to the ZAZ2 group (n=59) or placebo group (n=56). There was no difference in the mean age, gender ratio or disease duration between the ZAZ2 and placebo groups, and the post hoc test revealed no

other words they exhibit milder akinesia, when the parkinsonism is mild than when it is severe. It is worthy to note that lower α-values for local maxima were obtained for all the patients after medication than before, and when in good condition than in bad condition (Fig. 3C, D), thereby providing a temporal profile of parkinsonism in each individual

These results thus suggest that analysis of power-law temporal auto-correlation of physical activity time series using the bi-directional extension (Ohashi et al., 2003) is applicable to patients with parkinsonism for the evaluation of motor dysfunction irrespective of the presence of tremor and may provide useful objective data necessary for the control of drug dosage in the out-patient clinic and also for the evaluation of new drugs for parkinsonism

**4. Evaluation of effects of traditional Chinese medicine on parkinsonian** 

Conventional antiparkinsonism drugs effectively ameliorate the symptoms of patients with PD during the initial several years of onset, but become increasingly less effective and induce motor fluctuations including wearing-off, on-off, dopa-induced dyskinesia, and agonist-induced sleep attack (Arnulf et al., 2002; Comella, 2002; Hobson et al., 2002; Ondo et al., 2001; Pahwa et al., 2006). PD patients not infrequently suffer from non-motor symptoms, such as neuropsychiatric symptoms, autonomic symptoms, gastrointestinal symptoms, sensory symptoms, non-motor fluctuations (autonomic symptoms, cognitive or psychiatric symptoms, sensory symptoms including pain), fatigue, and sleep disturbance (Chaudhuri& Schapira, 2009; Miyasaki et al., 2006; Park& Stacy, 2009), and these non-motor symptoms may be intrinsic to the disease pathology or may be the result of treatment with dopaminergic agents. Several studies have established that the non-motor symptoms of PD are common, occur across all stages of PD, and are a key determinant of quality of life

Herbal remedies have a long history of use (particularly in East Asian countries) for alleviating various symptoms and have been increasingly used as alternative medicines worldwide, including the United States (De Smet, 2002). Traditional Chinese medicines (TCM) ameliorate various symptoms, particularly the ageing-related symptoms, and hence are likely to be beneficial for chronic diseases such as PD (Iwasaki et al., 2004; 2005a; 2005b). Good compliance for long-term use with few side effects may be another merit of TCM

In order to evaluate the effects of TCM on symptoms of parkinsonism, we evaluated the effects of Zeng-xiao An-shen Zhi-chan 2 (ZAZ2) on patients with PD using this method together with conventional scales for parkinsonism (Pan et al., 2011a). ZAZ2 granule is made up of 14 kinds of herbs*; Uncaria rhynchophylla* 10 g, *Rehmanniae radix* 10 g, *Cornus officinalis* 8 g, *Asnaragus cochinchinensic* 10 g, *Paeonia lactiflora* 10 g, *Desertliving cistanche* 10 g, *Puerariae radix* 10 g, *Arisaema consanguineum Schott* 10 g, *Salviae Miltiorrhizae radix* 10 g, *Acorus tatarinowii* 10 g, *Curcuma longa Linn* 12 g, *Morindae officinalis radix* 10 g, *Rhizoma gastrodiae* 10 g, and*Rhizoma chuanxiong* 10 g. One hundred and fifteen patients with idiopathic PD took 8 g of either ZAZ2 granule or placebo granule that was not distinguished by appearance or taste for 13 weeks. Patients were randomly assigned to the ZAZ2 group (n=59) or placebo group (n=56). There was no difference in the mean age, gender ratio or disease duration between the ZAZ2 and placebo groups, and the post hoc test revealed no

suitable for patients with PD (Lian& Luo, 2007; Zhao et al., 2007).

patient.

(Pan et al., 2007).

(Chaudhuri& Schapira, 2009).

**symptoms** 

significant baseline (week 0) differences in UPDRS scores, Hoehn & Yahr stages, mean counts, power-law temporal exponent α values, or in the dosage of antiparkinonian drugs between the two groups. All the patients were evaluated at week 0, week 1, and week 13 for the actigraph recording, UPDRS and Secondary Symptom Score, which is conventionally used in China to evaluate the effects of antiparkinsonism drugs and consists of 8 parts, including the assessments of non-fluent speech, vertigo, insomnia/nightmares, headache, sweating or night sweats, tiredness, sense of cold, and dysuria (Long, 1992). The awake-time and sleep-time actigraph data were used separately for the power-law temporal analyses.

Fig. 5. Effects of TCM and placebo granules on actigraph recondings. (A) Counts of physical activity (mean ± S.D.). (B) Average wavelet coefficients, as a function of the wavelet scale for awake-time and sleep-time. The slopes are power-law exponents α. (C) Power-law exponents α (mean ± S.D.). \*: *P* < 0.05, \*\*: *P* < 0.01. (Pan et al., 2011a)

The local power-law exponent α values during both awake-time and sleep-time were significantly decreased after taking ZAZ2 granule, but not after taking placebo granule (Table 1, Fig 5). The average wavelet coefficients exhibited linear relationships in the range of scales from 8 min to 35 min both for the ZAZ2 and placebo groups (Fig. 5B). The local power-law exponent α values during both awake-time and sleep-time were significantly decreased both week 1 and 13 in the ZAZ2 group, but not in the placebo group (Table 1 and Fig 5C, P<0.01; Bonferroni test). The beneficial effects of ZAZ2 were shown with UPDRS scores, as well; significant and persistent improvements were found in the scores of Part II, Part II + Part III, and Part IV (Table 1). These scores at week 13 were significantly different between the ZAZ2 group and the placebo group. As the exploratory outcome of this study, most of the secondary symptoms were improved after taking ZAZ2 granule, whereas only a few symptoms were transiently improved in the placebo group (Table 2).

We evaluated the beneficial effects of TCM specifically on sleep disturbance of patients with parkinsonism. We used placebo-controlled, randomized study design, in which 48 patients

Objective Evaluation

of the Severity of Parkinsonism Using Power-Law Temporal Auto-Correlation of Activity 233

Fig. 6. Effects of cerebral granule (TCM). (A) Daily profiles of actigraph count for three consecutive days before and after taking TCM. Dashed line: midnight. (B) Changes from baseline in actigraph counts. Columns and bars (mean ± S.D.) indicate sleep efficiency (%), sleep latency (min) and the 5 least active hours (counts/min). \*: p < 0.05; \*\*: p < 0.01.

Enhancing neuronal transmission is a possible non-pharmacological therapeutic strategy for neurological diseases. The cranial nerves send direct inputs to the brain, and their stimulation may lead to alterations in various central functions. Such stimulation may potentially be a therapeutic strategy for brain disorders due to the low invasiveness as compared to deep brain stimulation. Considering its central connections, the vestibular nerve can influence limbic-to-motor functions, and we applied non-invasive and nonnociceptive noisy galvanic vestibular stimulation (GVS) to the patients with parkinsonism. We successfully improved parkinsonian symptoms by using noisy GVS at a low-frequency range targeting the vestibular nerves of patients with levodopa responsive PD and levodopa unresponsive parkinsonism (Yamamoto et al., 2005). This effect is presumably through the demonstrated vestibule-cerebellar connections, and input noise played the beneficial role in sensitizing neural systems, possibly through a mechanism known as stochastic resonance, a basic physical mechanism underlying noise-enhanced responses of nonlinear systems to weak signals. It is hypothesized that a central circuit signaling the onset of movement of which the threshold is relatively increased due to the diseases may benefit from noisy emulation of the afferent firing rates. We analyzed whether the beneficial effects of GVS on

As previously described (Yamamoto et al., 2005), a portable GVS device was used to deliver currents using a bilateral unipolar configuration, in which electrodes were placed over the patient's bilateral mastoid processes with the reference electrodes placed on the forehead. The waveform, a zero-mean, linearly detrended noisy current with a 1/f-type power

**5. Assessment for effects of GVS for ameliorating parkinsonism** 

parkinsonism was reflected in a decrease of the -value for local maxima.

Reprinted with permission from (Pan et al., 2011b).


Data presented are mean ± SD.\*:P < 0.05; \*\*:P < 0.01 compared to week 0 (Repeated-measure ANOVAs).

#:P < 0.05; ##:P < 0.01 compared to placebo (Bonferroni test). UPDRS: Unified Parkinson's Disease Rating Scale. : power-law exponent.

Table 1. Measurements before and after taking test granules. (Pan et al., 2011a)


Data presented are mean ± SD.\*: P < 0.05, \*\*: P < 0.01 compared with 0 weeks (Repeated-measure ANOVAs).

#:P < 0.05, ##:P < 0.01 compared to placebo (Bonferroni test).

Table 2. Secondary symptom scores before and after taking test granules. (Pan et al., 2011a)

with idiopathic PD who had at least three awakenings per night occurring at least 3 nights per week participated. Patients wore the actigraph on the wrist of their non-dominant hand for seven consecutive days twice at week 0 (before) and week 6 of taking either one of the granule. For control, age-matched 25 patients with non-neurological diseases who had neither sleep disturbance nor parkinsonism wore the actigraph for seven consecutive days. Daily profiles of activity counts clearly demonstrated an improvement of the biological rhythm after the additional treatment in the TCM group but not in the placebo group (Fig. 6A). After treatment, sleep latency, median sleep efficiency and the median 5 least active hour, all of which were the parameters specifically reflected sleep disturbance (Pan et al., 2011b), shifted towards the values of the control group in the TCM group, but not in the placebo group (Fig 6B).

Scores in UPDRS Part II reflects the long-term outcome of the patients (Harrison et al., 2009). That both -values for local maxima and the scores in UPDRS Part II, Part II + Part III and Part IV improved after TCM suggested that -values for local maxima reflected patients' overall ADL, including motor symptoms and non-motor symptoms. Therefore, it is likely that analysis of the α-values is useful for the evaluation of drug effects on the long-term outcome of patient with PD (Pan et al., 2011a; 2011b).

UPDRS total score 46.6 ± 16.3 44.7 ± 15.3 45.9 ± 18.1 46.3 ± 17.1 37.1 ± 11.2\* ## 40.7 ± 15.1\* # UPDRS I 2.5 ± 0.7 2.3 ± 1.1 2.4 ± 1.2 2.6 ± 0.8 2.1 ± 0.7\* 2.3 ± 0.9 UPDRS II 15.7 ± 9.3 14.8 ± 11.2 15.3 ± 11.6 15.9 ± 11.3 12.5 ± 4.6\*# 13.4 ± 9.8\*# UPDRS III 25.5 ± 12.9 23.8 ± 10.6\* 24.9 ± 12.7 25.4 ± 10.1 19.3 ± 9.8\*# 21.6 ± 10.4\* UPDRS IV 3.1 ± 1.1 2.9 ± 1.6 3.0 ± 1.4 3.2 ± 1.4 2.6 ± 0.8\* # 2.7±1.3\*# Awake-time (counts/min) 98.5 ± 14.1 102.6 ± 18.9 100.7 ± 16.9 99.8 ± 17.8 126.7 ± 13.4\*## 118.4 ± 11.8\*## Sleep-time (counts/min) 42.9 ± 17.1 38.8 ± 15.6\* 40.1 ± 14.8 43.2 ± 11.6 35.6 ± 13.6\* # 32.8 ± 13.6\* # α(awake-time) 0.97 ± 0.21 0.95 ± 0.28 0.96 ± 0.18 0.97 ± 0.24 0.88 ± 0.21\* # 0.86 ± 0.19\* ## α(sleep-time) 1.19 ± 0.28 1.16 ± 0.27 1.15 ± 0.29 1.18 ± 0.26 1.04 ± 0.22\* # 1.02 ± 0.18\* ##

nightmare Headache Sweating or

week 0 1.08 ± 0.74 1.33 ± 0.83 2.77 ± 0.98 0.92 ± 0.56 2.11 ± 0.68 1.66 ± 0.57 1.90 ± 0.67 2.23 ± 0.69 TCM week 1 0.56 ± 0.28\* 0.84 ± 0.26\* # 2.03 ± 0.78\* 0.64 ± 0.28\* ## 1.38 ± 0.69\* # 1.21 ± 0.46\* 1.48 ± 0.57\* 1.43 ± 0.31\* # week 13 0.65 ± 0.33\* ## 0.95 ± 0.37\* 1.73 ± 0.38\* # 0.63 ±0.19\* # 1.48 ± 0.28\* ## 1.27 ± 0.51\* # 1.58 ± 0.61 1.46 ± 0.36\* ## week 0 1.12 ± 0.59 1.31 ± 0.97 2.67 ± 0.87 1.03 ± 0.75 2.13 ± 1.32 1.70 ± 0.97 1.78 ± 0.39 2.29 ± 1.02 Placebo week 1 0.69 ± 0.32\* 1.12 ± 0.69 2.40 ± 0.69\* 0.96 ± 0.36\* 1.87 ± 0.58 1.35 ± 0.69\* 1.39 ± 0.81 1.69 ± 0.92\* week 13 1.02 ± 0.36 1.28 ± 0.53 2.45 ± 0.38 0.99 ± 0.65 2.18 ± 0.56 1.58 ± 0.66 1.64 ± 0.58 2.18 ± 1.30

Table 2. Secondary symptom scores before and after taking test granules. (Pan et al., 2011a) with idiopathic PD who had at least three awakenings per night occurring at least 3 nights per week participated. Patients wore the actigraph on the wrist of their non-dominant hand for seven consecutive days twice at week 0 (before) and week 6 of taking either one of the granule. For control, age-matched 25 patients with non-neurological diseases who had neither sleep disturbance nor parkinsonism wore the actigraph for seven consecutive days. Daily profiles of activity counts clearly demonstrated an improvement of the biological rhythm after the additional treatment in the TCM group but not in the placebo group (Fig. 6A). After treatment, sleep latency, median sleep efficiency and the median 5 least active hour, all of which were the parameters specifically reflected sleep disturbance (Pan et al., 2011b), shifted towards the values of the control group in the TCM group, but not in the

Scores in UPDRS Part II reflects the long-term outcome of the patients (Harrison et al., 2009). That both -values for local maxima and the scores in UPDRS Part II, Part II + Part III and Part IV improved after TCM suggested that -values for local maxima reflected patients' overall ADL, including motor symptoms and non-motor symptoms. Therefore, it is likely that analysis of the α-values is useful for the evaluation of drug effects on the long-term

night sweats Tiredness Sense of

cold Dysuria

Data presented are mean ± SD.\*:P < 0.05; \*\*:P < 0.01 compared to week 0 (Repeated-measure ANOVAs). #:P < 0.05; ##:P < 0.01 compared to placebo (Bonferroni test). UPDRS: Unified Parkinson's Disease Rating Scale.

Table 1. Measurements before and after taking test granules. (Pan et al., 2011a)

Data presented are mean ± SD.\*: P < 0.05, \*\*: P < 0.01 compared with 0 weeks (Repeated-measure ANOVAs).

speech Vertigo Insomnia/

: power-law exponent.

Group Time Non-fluent

placebo group (Fig 6B).

#:P < 0.05, ##:P < 0.01 compared to placebo (Bonferroni test).

outcome of patient with PD (Pan et al., 2011a; 2011b).

Week 0 Week 1 Week 13 Week 0 Week 1 Week 13

Placebo(n = 54) TCM(n = 56)

Fig. 6. Effects of cerebral granule (TCM). (A) Daily profiles of actigraph count for three consecutive days before and after taking TCM. Dashed line: midnight. (B) Changes from baseline in actigraph counts. Columns and bars (mean ± S.D.) indicate sleep efficiency (%), sleep latency (min) and the 5 least active hours (counts/min). \*: p < 0.05; \*\*: p < 0.01. Reprinted with permission from (Pan et al., 2011b).

## **5. Assessment for effects of GVS for ameliorating parkinsonism**

Enhancing neuronal transmission is a possible non-pharmacological therapeutic strategy for neurological diseases. The cranial nerves send direct inputs to the brain, and their stimulation may lead to alterations in various central functions. Such stimulation may potentially be a therapeutic strategy for brain disorders due to the low invasiveness as compared to deep brain stimulation. Considering its central connections, the vestibular nerve can influence limbic-to-motor functions, and we applied non-invasive and nonnociceptive noisy galvanic vestibular stimulation (GVS) to the patients with parkinsonism. We successfully improved parkinsonian symptoms by using noisy GVS at a low-frequency range targeting the vestibular nerves of patients with levodopa responsive PD and levodopa unresponsive parkinsonism (Yamamoto et al., 2005). This effect is presumably through the demonstrated vestibule-cerebellar connections, and input noise played the beneficial role in sensitizing neural systems, possibly through a mechanism known as stochastic resonance, a basic physical mechanism underlying noise-enhanced responses of nonlinear systems to weak signals. It is hypothesized that a central circuit signaling the onset of movement of which the threshold is relatively increased due to the diseases may benefit from noisy emulation of the afferent firing rates. We analyzed whether the beneficial effects of GVS on parkinsonism was reflected in a decrease of the -value for local maxima.

As previously described (Yamamoto et al., 2005), a portable GVS device was used to deliver currents using a bilateral unipolar configuration, in which electrodes were placed over the patient's bilateral mastoid processes with the reference electrodes placed on the forehead. The waveform, a zero-mean, linearly detrended noisy current with a 1/f-type power

Objective Evaluation

permission from (Pan et al., 2008).

**6. Conclusion** 

**7. Acknowledgement** 

of Health, Labor, and Welfare of Japan to S.K.

application itself, not to an order effect.

of the Severity of Parkinsonism Using Power-Law Temporal Auto-Correlation of Activity 235

Fig. 8. The group average wavelet coefficients for local maxima (A) and minima (B) for GVS and control (CON) conditions. (C) Comparisons of the mean for GVS and CON (left) and the within-individual differences (right). The error bars represent SEM. Reprinted with

first and the second days were compared, significant differences were not observed either for local maxima or minima, suggesting that the above differences were due to the GVS

We confirmed that measurement of the mean for local maxima detected the improvement

Analysis of patients' physical activity records collected by an actigraph device using powerlaw exponents probing temporal autocorrelation of the activity counts provides methods for the evaluation of disability resulting from motor and non-motor parkinsonism without being influenced by the presence of tremor or different patterns of daily living (Pan et al., 2007). Sufficient sensitivity and reliability of this method warrants the objectivity in the evaluation of symptom severity (Pan et al., 2008; Pan et al., 2011a) , hence this method may

This study was supported in part by Grants-in-Aid for Scientific Research from the Ministry

of parkinsonism during GVS with sufficient sensitivity (Pan et al., 2008).

be useful for the evaluation of disease progression and efficacy of new drug.

spectrum (Struzik et al., 2006) within a range of 0.01-2.0 Hz or a constant zero current for control, with a duration of 300 sec was continuously repeated during the tests. The magnitude of noisy GVS was set to 60% of each subject's nociceptive threshold (0.29 ± 0.20 mA). Then either the noisy GVS or the control zero current was continuously applied for the first 24 hours, and then switched to the counter-part and applied for another 24 hours, while the patients' wrist activity was monitored continuously for 48 hours. The order of noisy GVS and the control zero current was determined for each patient by random selection.

Fig. 7. Illustrative examples of wrist activity data of a PD patient during the control (CON) period (A) and during GVS application (B). The wavelet coefficients (*W(S)*) of these data, as a function of the wavelet scale (*S*) are shown for local maxima (C) and minima (D). The slopes are power-law exponents . Reprinted with permission from (Pan et al., 2008).

The representative wrist activity data of a PD patient during the control period and during the application of GVS were shown in Fig. 7A, B. Compared to control, GVS was associated with more frequent switching between higher and lower levels of activity. This resulted in a higher wavelet power (*W(S)*2) with GVS (Fig. 7C, D), particularly at smaller scales (*S*), or higher frequencies, for local maxima (Fig.7C). The power-law exponent , given by the slope of the log *S* vs. Log *W(S)*2 relationship and characterizing the nature of switching patterns between high and low values in a statistical sense, was smaller with GVS than with control stimulation, especially for the local maxima (Fig. 7C,D).

The group average wavelet coefficients exhibited linear relationships in the range of scales (*S*) from 8 min to 35 min both for local maxima and minima and for GVS and control conditions (Fig. 8A, B). The slope for local maxima with noisy GVS was substantially less than that with control stimulation. For local maxima, the mean power-law exponent was significantly smaller for GVS than for the control (Fig. 8C). The difference in the mean for local minima was much less than that for the local maxima. When the mean -values for the

Fig. 8. The group average wavelet coefficients for local maxima (A) and minima (B) for GVS and control (CON) conditions. (C) Comparisons of the mean for GVS and CON (left) and the within-individual differences (right). The error bars represent SEM. Reprinted with permission from (Pan et al., 2008).

first and the second days were compared, significant differences were not observed either for local maxima or minima, suggesting that the above differences were due to the GVS application itself, not to an order effect.

We confirmed that measurement of the mean for local maxima detected the improvement of parkinsonism during GVS with sufficient sensitivity (Pan et al., 2008).

## **6. Conclusion**

234 Diagnostics and Rehabilitation of Parkinson's Disease

spectrum (Struzik et al., 2006) within a range of 0.01-2.0 Hz or a constant zero current for control, with a duration of 300 sec was continuously repeated during the tests. The magnitude of noisy GVS was set to 60% of each subject's nociceptive threshold (0.29 ± 0.20 mA). Then either the noisy GVS or the control zero current was continuously applied for the first 24 hours, and then switched to the counter-part and applied for another 24 hours, while the patients' wrist activity was monitored continuously for 48 hours. The order of noisy GVS

and the control zero current was determined for each patient by random selection.

Fig. 7. Illustrative examples of wrist activity data of a PD patient during the control (CON) period (A) and during GVS application (B). The wavelet coefficients (*W(S)*) of these data, as a function of the wavelet scale (*S*) are shown for local maxima (C) and minima (D). The slopes are power-law exponents . Reprinted with permission from (Pan et al., 2008).

The representative wrist activity data of a PD patient during the control period and during the application of GVS were shown in Fig. 7A, B. Compared to control, GVS was associated with more frequent switching between higher and lower levels of activity. This resulted in a higher wavelet power (*W(S)*2) with GVS (Fig. 7C, D), particularly at smaller scales (*S*), or higher frequencies, for local maxima (Fig.7C). The power-law exponent , given by the slope of the log *S* vs. Log *W(S)*2 relationship and characterizing the nature of switching patterns between high and low values in a statistical sense, was smaller with GVS than with control

The group average wavelet coefficients exhibited linear relationships in the range of scales (*S*) from 8 min to 35 min both for local maxima and minima and for GVS and control conditions (Fig. 8A, B). The slope for local maxima with noisy GVS was substantially less than that with control stimulation. For local maxima, the mean power-law exponent was significantly smaller for GVS than for the control (Fig. 8C). The difference in the mean for local minima was much less than that for the local maxima. When the mean -values for the

stimulation, especially for the local maxima (Fig. 7C,D).

Analysis of patients' physical activity records collected by an actigraph device using powerlaw exponents probing temporal autocorrelation of the activity counts provides methods for the evaluation of disability resulting from motor and non-motor parkinsonism without being influenced by the presence of tremor or different patterns of daily living (Pan et al., 2007). Sufficient sensitivity and reliability of this method warrants the objectivity in the evaluation of symptom severity (Pan et al., 2008; Pan et al., 2011a) , hence this method may be useful for the evaluation of disease progression and efficacy of new drug.

## **7. Acknowledgement**

This study was supported in part by Grants-in-Aid for Scientific Research from the Ministry of Health, Labor, and Welfare of Japan to S.K.

Objective Evaluation

83, ISSN 0885-3185

1002, ISSN 1526-632X

of the Severity of Parkinsonism Using Power-Law Temporal Auto-Correlation of Activity 237

Lian, X.F. & Luo, X.D. (2007). [Effect of TCM treatment according to syndrome differentiation

Miyasaki, J.M., Shannon, K., Voon, V., Ravina, B., Kleiner-Fisman, G., Anderson, K.,

Mormont, M.C., Waterhouse, J., Bleuzen, P., Giacchetti, S., Jami, A., Bogdan, A., Lellouch, J.,

Nakamura, T., Kiyono, K., Yoshiuchi, K., Nakahara, R., Struzik, Z.R. & Yamamoto, Y. (2007).

Ohashi, K., Nunes Amaral, L.A., Natelson, B.H. & Yamamoto, Y. (2003). Asymmetrical

Okawa, M., Mishima, K., Hishikawa, Y. & Hozumi, S. (1995). [Rest-activity and body-

Ondo, W.G., Dat Vuong, K., Khan, H., Atassi, F., Kwak, C. & Jankovic, J. (2001). Daytime

Pahwa, R., Factor, S.A., Lyons, K.E., Ondo, W.G., Gronseth, G., Bronte-Stewart, H., Hallett,

Pan, W., Liu, Y., Fang, Z., Zhu, X., Kwak, S. & Yamamoto, Y. (2011b). A compound belonging

Pan, W., Ohashi, K., Yamamoto, Y. & Kwak, S. (2007). Power-law temporal autocorrelation

*Res,* Vol. 6, No. 8, (Aug 2000), pp. 3038-3045, ISSN 1078-0432

13, (Sep 2007), pp. 138103, ISSN 0031-9007

(Jan 1995), pp. 18-23, ISSN 0009-918X

pp.1308-1313, ISSN 1531-8257

8, (Oct 2001), pp. 1392-1396, ISSN 0028-3878

*disease,* in press. doi:10.4061/2011/789506,ISSN 2042-0080

*Sleep Med,* Vol. 12, No. 3, (Mar 2011), pp. 307-308, ISSN 1878-5506

No. 6 Pt 2, (Dec 2003), pp. 065204, ISSN 1539-3755

*Yi Jie He Za Zhi,* Vol. 27, No. 9, (Sep 2007), pp. 796-799, ISSN 1003-5370 Long, C. (1992). Definition and evaluation of parkinsonism in TCM. *Journal of Beijing University of Traditional Chinese Medicine* Vol. 4, No. 15, (1992), pp. 39. Martinez-Martin, P., Gil-Nagel, A., Gracia, L.M., Gomez, J.B., Martinez-Sarries, J. & Bermejo,

in enhancing curative effect and reducing side-effect of madopa]. *Zhongguo Zhong Xi* 

F. (1994). Unified Parkinson's Disease Rating Scale characteristics and structure. The Cooperative Multicentric Group. *Mov Disord,* Vol. 9, No. 1, (Jan 1994), pp. 76-

Shulman, L.M., Gronseth, G. & Weiner, W.J. (2006). Practice Parameter: evaluation and treatment of depression, psychosis, and dementia in Parkinson disease (an evidence-based review): report of the Quality Standards Subcommittee of the American Academy of Neurology. *Neurology,* Vol. 66, No. 7, (Apr 2006), pp. 996-

Misset, J.L., Touitou, Y. & Levi, F. (2000). Marked 24-h rest/activity rhythms are associated with better quality of life, better response, and longer survival in patients with metastatic colorectal cancer and good performance status. *Clin Cancer* 

Universal scaling law in human behavioral organization. *Phys Rev Lett,* Vol. 99, No.

singularities in real-world signals. *Phys Rev E Stat Nonlin Soft Matter Phys,* Vol. 68,

temperature rhythm disorders in elderly patients with dementia--senile dementia of Alzheimer's type and multi-infarct dementia]. *Rinsho Shinkeigaku,* Vol. 35, No. 1,

sleepiness and other sleep disorders in Parkinson's disease. *Neurology,* Vol. 57, No.

M., Miyasaki, J., Stevens, J. & Weiner, W.J. (2006). Practice Parameter: treatment of Parkinson disease with motor fluctuations and dyskinesia (an evidence-based review): report of the Quality Standards Subcommittee of the American Academy of Neurology. *Neurology,* Vol. 66, No. 7, (Apr 2006), pp. 983-995, ISSN 1526-632X Pan, W., Kwak, S., Liu, Y., Sun, Y., Fang, Z., Qin, B. & Yamamoto, Y. (2011a). Traditional

Chinese medicine improves nocturnal activity in Parkinson's disease. *Parkinson's* 

to traditional Chinese medicine improves nocturnal activity in Parkinson's disease.

of activity reflects severity of parkinsonism. *Mov Disord,* Vol. 22, No. 9, (Jul 2007),

#### **8. References**


Arnulf, I., Konofal, E., Merino-Andreu, M., Houeto, J.L., Mesnage, V., Welter, M.L.,

Chaudhuri, K.R. & Schapira, A.H. (2009). Non-motor symptoms of Parkinson's disease:

Comella, C.L. (2002). Daytime sleepiness, agonist therapy, and driving in Parkinson disease.

Comella, C.L., Morrissey, M. & Janko, K. (2005). Nocturnal activity with nighttime pergolide

De Smet, P.A. (2002). Herbal remedies. *N Engl J Med,* Vol. 347, No. 25, (Dec 2002), pp. 2046-

Harrison, M.B., Wylie, S.A., Frysinger, R.C., Patrie, J.T., Huss, D.S., Currie, L.J. & Wooten,

Iwasaki, K., Kobayashi, S., Chimura, Y., Taguchi, M., Inoue, K., Cho, S., Akiba, T., Arai, H.,

*Psychiatry,* Vol. 66, No. 12, (Dec 2005), pp. 1612-1613, ISSN 0160-6689 Iwasaki, K., Satoh-Nakagawa, T., Maruyama, M., Monma, Y., Nemoto, M., Tomita, N., Tanji, H.,

*JAMA,* Vol. 287, No. 4, (Jan 2002), pp. 509-511, ISSN 0098-7484

Lacomblez, L., Golmard, J.L., Derenne, J.P. & Agid, Y. (2002). Parkinson's disease and sleepiness: an integral part of PD. *Neurology,* Vol. 58, No. 7, (Apr 2002), pp.

dopaminergic pathophysiology and treatment. *Lancet Neurol,* Vol. 8, No. 5, (May

in Parkinson disease: a controlled study using actigraphy. *Neurology,* Vol. 64, No. 8,

G.F. (2009). UPDRS activity of daily living score as a marker of Parkinson's disease progression. *Mov Disord,* Vol. 24, No. 2, (Jan 2009), pp. 224-230, ISSN 1531-8257 Hobson, D.E., Lang, A.E., Martin, W.R., Razmy, A., Rivest, J. & Fleming, J. (2002). Excessive

daytime sleepiness and sudden-onset sleep in Parkinson disease: a survey by the Canadian Movement Disorders Group. *JAMA,* Vol. 287, No. 4, (Jan 2002), pp. 455-

Cyong, J.C. & Sasaki, H. (2004). A randomized, double-blind, placebo-controlled clinical trial of the Chinese herbal medicine "ba wei di huang wan" in the treatment of dementia. *J Am Geriatr Soc,* Vol. 52, No. 9, (Sep 2004), pp. 1518-1521, ISSN 0002-8614 Iwasaki, K., Maruyama, M., Tomita, N., Furukawa, K., Nemoto, M., Fujiwara, H., Seki, T.,

Fujii, M., Kodama, M. & Arai, H. (2005a). Effects of the traditional Chinese herbal medicine Yi-Gan San for cholinesterase inhibitor-resistant visual hallucinations and neuropsychiatric symptoms in patients with dementia with Lewy bodies. *J Clin* 

Fujiwara, H., Seki, T., Fujii, M., Arai, H. & Sasaki, H. (2005b). A randomized, observerblind, controlled trial of the traditional Chinese medicine Yi-Gan San for improvement of behavioral and psychological symptoms and activities of daily living in dementia

& von Gizycki, H. (2000). Sleep estimation from wrist activity in patients with major depression. *Physiol Behav,* Vol. 70, No. 1-2, (Jul 2000), pp. 49-53, ISSN 0031-9384 Katayama, S. (2001). Actigraph analysis of diurnal motor fluctuations during dopamine

agonist therapy. *Eur Neurol,* Vol. 46, No. Suppl 1, (2001), pp. 11-17, ISSN 0014-3022

(2002). Use of actigraphy for monitoring sleep and activity levels in patients with fibromyalgia and depression. *J Psychosom Res,* Vol. 52, No. 6, (Jun 2002), pp. 439-

of neonates born by different delivery modes. *Chronobiol Int,* Vol. 21, No. 1, (Jan

patients. *J Clin Psychiatry,* Vol. 66, No. 2, (Feb 2005), pp. 248-252, ISSN 0160-6689 Jean-Louis, G., Mendlowicz, M.V., Gillin, J.C., Rapaport, M.H., Kelsoe, J.R., Zizi, F., Landolt, H.

Korszun, A., Young, E.A., Engleberg, N.C., Brucksch, C.B., Greden, J.F. & Crofford, L.A.

Korte, J., Hoehn, T. & Siegmund, R. (2004). Actigraphic recordings of activity-rest rhythms

**8. References** 

1019-1024, ISSN 0028-3878

2056, ISSN 1533-4406

463, ISSN 0098-7484

443, ISSN 0022-3999

2004), pp. 95-106, ISSN 0742-0528

2009), pp. 464-474, ISSN 1474-4422

(Apr 2005), pp. 1450-1451, ISSN 1526-632X


**12** 

*Japan* 

**Postural Control While Sitting** 

 Hiroaki Nagase4 and Shinji Ohara5 *1Department of Neurology, Okaya City Hospital,* 

*General Industrial Technology Center,* 

*General Industrial Technology Center* 

*Graduate School of Engineering Kobe University* 

**and Its Association with Risk of Falls** 

**in Patients with Parkinson's Disease** 

Ryoichi Hayashi1, Takeshi Hayashi2, Junpei Aizawa3,

*2Department of Computer Science and Systems Engineering,* 

*3Department of Material Technology, Nagano Prefecture* 

*5Department of Neurology, Matsumoto Medical Center* 

*4Department of Information Technology, Nagano Prefecture* 

Abnormal postures and falls in patients with Parkinson's disease (PD) have been well recognized from the time of its earliest clinical description (Parkinson, 1817). Abnormal postures in PD are observed in the entire body, including flexion to the anterior, lateral, or anterolateral direction of the trunk, neck flexion, flexion of the extremities, and abnormal postures of the hands, fingers, and toes. A lateral deviation of the spine and a corresponding tendency to lean to one side was reported (Duvoisin and Marsden 1975; Hayashi et al., 2010). These postural abnormalities cause postural instability and falls. In a recent study, approximately 50% of the PD patients had fallen, compared to about 15% of healthy elderly subjects, and approximately 75% of falls in PD patients occurred during activities associated with daily living, such as turning around, standing up, and bending forward (Bloem et al., 2001). Postural instability is caused by an inability to adjust the center of gravity quickly enough to account for perturbations in the environment (Shivitz et al., 2006). Maki and colleagues (1994) have suggested that increased lateral sway is associated with increased risk of falling in elderly persons. Several studies have quantified postural stability during quiet stance in patients with PD and revealed that PD patients have more difficulty in controlling lateral postural sway than anteroposterior sway (Mitchell et al., 1995; Rocchi et

In the standing position, postural adjustments can be accomplished with responses at the ankle, knee, hip, and trunk joints, independently or combined (Hodges et al., 2002; Krishnamoorthy et al., 2005) and these complex structures consisting of multiple linked segments must be maintained in a stable position on a relatively narrow support base

**1. Introduction** 

al., 2002).


## **Postural Control While Sitting and Its Association with Risk of Falls in Patients with Parkinson's Disease**

Ryoichi Hayashi1, Takeshi Hayashi2, Junpei Aizawa3, Hiroaki Nagase4 and Shinji Ohara5 *1Department of Neurology, Okaya City Hospital, 2Department of Computer Science and Systems Engineering, Graduate School of Engineering Kobe University 3Department of Material Technology, Nagano Prefecture General Industrial Technology Center, 4Department of Information Technology, Nagano Prefecture General Industrial Technology Center 5Department of Neurology, Matsumoto Medical Center Japan* 

## **1. Introduction**

238 Diagnostics and Rehabilitation of Parkinson's Disease

Pan, W., Soma, R., Kwak, S. & Yamamoto, Y. (2008). Improvement of motor functions by

Park, A. & Stacy, M. (2009). Non-motor symptoms in Parkinson's disease. *J Neurol,* Vol. 256,

Peng, C.K., Havlin, S., Stanley, H.E. & Goldberger, A.L. (1995). Quantification of scaling

Sekine, M., Akay, M., Tamura, T., Higashi, Y. & Fujimoto, T. (2004). Fractal dynamics of

Struzik, Z.R., Hayano, J., Sakata, S., Kwak, S. & Yamamoto, Y. (2004). 1/f scaling in heart

Struzik, Z.R., Hayano, J., Soma, R., Kwak, S. & Yamamoto, Y. (2006). Aging of complex heart rate dynamics. *IEEE Trans Biomed Eng,* Vol. 53, No. 1, (Jan 2006), pp. 89-94, ISSN 0018-9294 Teicher, M.H. (1995). Actigraphy and motion analysis: new tools for psychiatry. *Harv Rev Psychiatry,* Vol. 3, No. 1, (May-Jun 1995), pp. 18-35, ISSN 1067-3229) Tuisku, K., Holi, M.M., Wahlbeck, K., Ahlgren, A.J. & Lauerma, H. (2003). Quantitative rest

van Someren, E.J., Hagebeuk, E.E., Lijzenga, C., Scheltens, P., de Rooij, S.E., Jonker, C., Pot,

van Someren, E.J., Pticek, M.D., Speelman, J.D., Schuurman, P.R., Esselink, R. & Swaab, D.F.

van Someren, E.J., van Gool, W.A., Vonk, B.F., Mirmiran, M., Speelman, J.D., Bosch, D.A. &

van Someren, E.J., Vonk, B.F., Thijssen, W.A., Speelman, J.D., Schuurman, P.R., Mirmiran,

Witting, W., Kwa, I.H., Eikelenboom, P., Mirmiran, M. & Swaab, D.F. (1990). Alterations in

Yamamoto, Y., Struzik, Z.R., Soma, R., Ohashi, K. & Kwak, S. (2005). Noisy vestibular

Zhao, H., Li, W.W. & Gao, J.P. (2007). [Clinical trial on treatment of Parkinson's disease of

*Xi Yi Jie He Za Zhi,* Vol. 27, No. 9, (Sep 2007), pp. 780-784, ISSN 1003-5370

Vol. 27, No. 6, (Mar 1990), pp. 563-572, ISSN 0006-3223

255, No. 11, (Nov 2008), pp. 1657-1661, ISSN 0340-5354

No. Suppl 3, (Aug 2009), pp. 293-298, ISSN 1432-1459

*Phys,* Vol. 70, No. 5, (Nov 2004), pp. 050901, ISSN 1539-3755

Vol. 5, No. 1, (1995), pp. 82-87, ISSN 1054-1500

2004), pp. 8-15, ISSN 1741-2560

ISSN 0885-3185

ISSN 0364-5134

259-270, ISSN 0006-3223

(Aug 2006), pp. 1136-1143, ISSN 0885-3185

(Jul 1993), pp. 16-23, ISSN 0022-510X

1998), pp. 386-395, ISSN 0018-9294

noisy vestibular stimulation in central neurodegenerative disorders. *J Neurol,* Vol.

exponents and crossover phenomena in nonstationary heartbeat time series. *Chaos,* 

body motion in patients with Parkinson's disease. *J Neural Eng,* Vol. 1, No. 1, (Mar

rate requires antagonistic autonomic control. *Phys Rev E Stat Nonlin Soft Matter* 

activity in ambulatory monitoring as a physiological marker of restless legs syndrome: a controlled study. *Mov Disord,* Vol. 18, No. 4, (Apr 2003), pp. 442-448,

A.M., Mirmiran, M. & Swaab, D.F. (1996). Circadian rest-activity rhythm disturbances in Alzheimer's disease. *Biol Psychiatry,* Vol. 40, No. 4, (Aug 1996), pp.

(2006). New actigraph for long-term tremor recording. *Mov Disord,* Vol. 21, No. 8,

Swaab, D.F. (1993). Ambulatory monitoring of tremor and other movements before and after thalamotomy: a new quantitative technique. *J Neurol Sci,* Vol. 117, No. 1-2,

M. & Swaab, D.F. (1998). A new actigraph for long-term registration of the duration and intensity of tremor and movement. *IEEE Trans Biomed Eng,* Vol. 45, No. 3, (Mar

the circadian rest-activity rhythm in aging and Alzheimer's disease. *Biol Psychiatry,* 

stimulation improves autonomic and motor responsiveness in central neurodegenerative disorders. *Ann Neurol,* Vol. 58, No. 2, (Aug 2005), pp. 175-181,

Gan-Shen yin deficiency type by recipe for nourishing Gan-Shen]. *Zhongguo Zhong* 

Abnormal postures and falls in patients with Parkinson's disease (PD) have been well recognized from the time of its earliest clinical description (Parkinson, 1817). Abnormal postures in PD are observed in the entire body, including flexion to the anterior, lateral, or anterolateral direction of the trunk, neck flexion, flexion of the extremities, and abnormal postures of the hands, fingers, and toes. A lateral deviation of the spine and a corresponding tendency to lean to one side was reported (Duvoisin and Marsden 1975; Hayashi et al., 2010). These postural abnormalities cause postural instability and falls. In a recent study, approximately 50% of the PD patients had fallen, compared to about 15% of healthy elderly subjects, and approximately 75% of falls in PD patients occurred during activities associated with daily living, such as turning around, standing up, and bending forward (Bloem et al., 2001). Postural instability is caused by an inability to adjust the center of gravity quickly enough to account for perturbations in the environment (Shivitz et al., 2006). Maki and colleagues (1994) have suggested that increased lateral sway is associated with increased risk of falling in elderly persons. Several studies have quantified postural stability during quiet stance in patients with PD and revealed that PD patients have more difficulty in controlling lateral postural sway than anteroposterior sway (Mitchell et al., 1995; Rocchi et al., 2002).

In the standing position, postural adjustments can be accomplished with responses at the ankle, knee, hip, and trunk joints, independently or combined (Hodges et al., 2002; Krishnamoorthy et al., 2005) and these complex structures consisting of multiple linked segments must be maintained in a stable position on a relatively narrow support base

Postural Control While Sitting

point in front of them.

during the raising of each arm.

were averaged.

markers were recorded with a sampling rate of 30 Hz.

**2.4 Evaluation of truncal inclination while raising arm** 

**2.3 Evaluation of truncal inclination at rest** 

**2.2 Experimental setup and procedure** 

and Its Association with Risk of Falls in Patients with Parkinson's Disease 241

A stable stool was placed on a force platform (Kistler platform type 9281CA, Winterthur, Switzerland). The stool was high enough so that each subject's legs could hang down without touching the platform (Fig. 1). Subjects sat on the stool at ease. At that time, if subjects presented a tilt of the trunk away from the vertical position, they were not asked to correct their trunk to the vertical position. Subjects were asked to maintain a sitting posture on the stool for 1) 2 minutes at rest, and 2) 30 seconds with a lateral arm raised alternatively up to 90 degrees with their legs hanging down and their hand placed on each thigh except for a raising arm. Subjects were also asked to keep their eyes open and to focus on the target

Raising right arm At rest Raising left arm

Using the forced plate with a sampling frequency of 500Hz, the excursion of the centre of pressure (COP) was measured. The position of the body segments was also measured using a video image processing system, with the images being developed in our laboratory. Nine reflective markers (1 cm in diameter) were placed on the forehead, the upper part of the sternum, shoulders, wrists, knees, and the level of the umbilicus, and reflections from the

Both COP excursions and the body-marker displacement were recorded simultaneously for 2 minutes. Values for the initial 10 seconds and the last 10 seconds were averaged, and the difference was calculated for each trial. The values obtained by the two trials for each subject

After evaluation of truncal inclination at rest, the patient was studied when each of their arms were raised laterally up to 90 degrees two times in the following sequence: at rest for

Fig. 1. Example of postural changes both during sitting on a stool at resting posture and

formed by the feet. Lee and coworkers (1995) studied preparatory postural adjustments associated with a lateral leg-raising task in parkinsonian patients with postural instability. In the sitting position, the influence of hip joints and lower extremities can be minimized to study the postural control of the trunk. Van der Burg and coworkers (2006) studied the postural control of the trunk during unstable sitting in PD patients and revealed that PD patients showed difficulty in truncal control. They suggested that these changes may be related to postural instability and fall risk.

In this study, the body movement in PD patients during sitting was investigated by measuring center of pressure (COP) excursions and trunk deviations under two conditions: 1) sitting at rest for two minutes, 2) raising his/her arm laterally to 90 degrees. An additional aim was to study differences in trunk control between patients who had a history of falling (fallers) and patients who did not have a history of falling (non-fallers). The aim of this study is also to test the hypothesis that postural abnormality in a lateral direction may, or would be a high risk factor for falling during the daily activities of PD patients and further attempts to formulate the pattern of muscle tone abnormalities that may underline this disturbance.

## **2. Patients and methods**

## **2.1 Patients**

17 consecutive idiopathic PD patients and 8 age-matched normal controls were studied. These 17 patients received regular outpatient treatment every month over the course of a one-year follow-up period. All patients satisfied the following inclusion criteria: Hoehn and Yahr stage II or higher while off medication (II=1, III=11, IV=5), a clear history of significant responsiveness to levodopa and an absence of other neurologic diseases including significant dementia or autonomic dysfunction. The patients' clinical data are given in Table 1. All patients (mean age ± sd: 72±6 years) did not have any neurological or other diseases that might affect their postural stability or ability to perform the experimental tasks.


Table 1. Clinical characteristics of PD patient

This study was approved by the Okaya City Hospital Committee for Research on Human Subjects, and informed consent was obtained from all test subjects.

## **2.2 Experimental setup and procedure**

240 Diagnostics and Rehabilitation of Parkinson's Disease

formed by the feet. Lee and coworkers (1995) studied preparatory postural adjustments associated with a lateral leg-raising task in parkinsonian patients with postural instability. In the sitting position, the influence of hip joints and lower extremities can be minimized to study the postural control of the trunk. Van der Burg and coworkers (2006) studied the postural control of the trunk during unstable sitting in PD patients and revealed that PD patients showed difficulty in truncal control. They suggested that these changes may be

In this study, the body movement in PD patients during sitting was investigated by measuring center of pressure (COP) excursions and trunk deviations under two conditions: 1) sitting at rest for two minutes, 2) raising his/her arm laterally to 90 degrees. An additional aim was to study differences in trunk control between patients who had a history of falling (fallers) and patients who did not have a history of falling (non-fallers). The aim of this study is also to test the hypothesis that postural abnormality in a lateral direction may, or would be a high risk factor for falling during the daily activities of PD patients and further attempts to formulate the pattern of muscle tone abnormalities that may underline

17 consecutive idiopathic PD patients and 8 age-matched normal controls were studied. These 17 patients received regular outpatient treatment every month over the course of a one-year follow-up period. All patients satisfied the following inclusion criteria: Hoehn and Yahr stage II or higher while off medication (II=1, III=11, IV=5), a clear history of significant responsiveness to levodopa and an absence of other neurologic diseases including significant dementia or autonomic dysfunction. The patients' clinical data are given in Table 1. All patients (mean age ± sd: 72±6 years) did not have any neurological or other diseases

This study was approved by the Okaya City Hospital Committee for Research on Human

that might affect their postural stability or ability to perform the experimental tasks.

related to postural instability and fall risk.

this disturbance.

**2.1 Patients** 

**2. Patients and methods** 

Table 1. Clinical characteristics of PD patient

Subjects, and informed consent was obtained from all test subjects.

A stable stool was placed on a force platform (Kistler platform type 9281CA, Winterthur, Switzerland). The stool was high enough so that each subject's legs could hang down without touching the platform (Fig. 1). Subjects sat on the stool at ease. At that time, if subjects presented a tilt of the trunk away from the vertical position, they were not asked to correct their trunk to the vertical position. Subjects were asked to maintain a sitting posture on the stool for 1) 2 minutes at rest, and 2) 30 seconds with a lateral arm raised alternatively up to 90 degrees with their legs hanging down and their hand placed on each thigh except for a raising arm. Subjects were also asked to keep their eyes open and to focus on the target point in front of them.

Raising right arm At rest Raising left arm

Fig. 1. Example of postural changes both during sitting on a stool at resting posture and during the raising of each arm.

Using the forced plate with a sampling frequency of 500Hz, the excursion of the centre of pressure (COP) was measured. The position of the body segments was also measured using a video image processing system, with the images being developed in our laboratory. Nine reflective markers (1 cm in diameter) were placed on the forehead, the upper part of the sternum, shoulders, wrists, knees, and the level of the umbilicus, and reflections from the markers were recorded with a sampling rate of 30 Hz.

## **2.3 Evaluation of truncal inclination at rest**

Both COP excursions and the body-marker displacement were recorded simultaneously for 2 minutes. Values for the initial 10 seconds and the last 10 seconds were averaged, and the difference was calculated for each trial. The values obtained by the two trials for each subject were averaged.

## **2.4 Evaluation of truncal inclination while raising arm**

After evaluation of truncal inclination at rest, the patient was studied when each of their arms were raised laterally up to 90 degrees two times in the following sequence: at rest for

Postural Control While Sitting

patients at rest.

and Its Association with Risk of Falls in Patients with Parkinson's Disease 243

Fig. 2. The relationship between the absolute value of lateral COP displacement (Abs-COP) and the absolute value of trunk displacement (Abs- trunk displacement) obtained from 17

Fig. 3. The relationship between the value of lateral COP displacement (COP) and the value

of trunk displacement obtained from 17 patients during arm-raising examination.

30 seconds, raising the right arm for 30 seconds, at rest for 30 seconds, then raising the left arm for 30 seconds.

## **2.5 Statistical analysis**

For each posturographic parameter and clinical measurement, Student's t-test or an analysis of variance (ANOVA) was used to compare group means between normal controls and PD patients. Correlation between the degree of COP displacement and the degree of body displacement was evaluated with Spearman's rank correlation coefficient.

## **3. Results**

## **3.1 Clinical features**

On examination, in all 17 patients, the side of initial symptoms was also the side of dominant clinical signs. Over the one-year follow-up period, 6 of 17 patients (35%) experienced no falls. The remaining 11 patients (65%) fell at least two times. Five of patients experienced more than 5 falls during the one-year follow-up, and these patients were described as "frequent fallers" in this paper. Patients who experienced less than 5 falls were described as a "less frequent fallers." This study found a strong correlation between disease severity and frequency of falls. The mean value with standard deviation of the Hoehn and Yahr stage in non-fallers and "less frequent fallers" was 2.8+0.4, and that of the "frequent fallers" was 3.5+0.5 (p<0.02). There was no significant difference between "frequent fallers" and "non-fallers" and "less frequent fallers" in age or duration of illness (73.3+5.2 years versus 70.0+7.0 years, p=0.3 in age; 9.5+2.6 years versus 10.5+7.7 years, p=0.7 in duration).

## **3.2 Body inclination during sitting**

A consistency in lateral displacement was observed in all 17 patients. Table 2 shows the values of two parameters of body inclination in each group. There was a tendency toward increased values of both lateral COP displacement and trunk displacement in a group of PD patients compared with controls. However, there was no significant difference in these parameters between that of the control group and the PD-patient group. When each parameter obtained by "frequent fallers" is compared with controls, there was a significant difference in each parameter (lateral COP displacement p=0.01, trunk displacement p=0.01).


Table 2. This data represents the mean value with one standard deviation of each parameter of body inclination in control subjects and in PD patients at rest for 2 minutes. Each mean value was calculated using an absolute value of each parameter obtained from each subject.

Fig. 2 shows the relationship between changes of lateral COP displacement and trunk displacement obtained from all 17 patients. The amount of lateral COP displacement was correlated significantly with that of trunk displacement (r=0.94, p<0.0001).

30 seconds, raising the right arm for 30 seconds, at rest for 30 seconds, then raising the left

For each posturographic parameter and clinical measurement, Student's t-test or an analysis of variance (ANOVA) was used to compare group means between normal controls and PD patients. Correlation between the degree of COP displacement and the degree of body

On examination, in all 17 patients, the side of initial symptoms was also the side of dominant clinical signs. Over the one-year follow-up period, 6 of 17 patients (35%) experienced no falls. The remaining 11 patients (65%) fell at least two times. Five of patients experienced more than 5 falls during the one-year follow-up, and these patients were described as "frequent fallers" in this paper. Patients who experienced less than 5 falls were described as a "less frequent fallers." This study found a strong correlation between disease severity and frequency of falls. The mean value with standard deviation of the Hoehn and Yahr stage in non-fallers and "less frequent fallers" was 2.8+0.4, and that of the "frequent fallers" was 3.5+0.5 (p<0.02). There was no significant difference between "frequent fallers" and "non-fallers" and "less frequent fallers" in age or duration of illness (73.3+5.2 years versus 70.0+7.0 years, p=0.3 in age; 9.5+2.6 years versus 10.5+7.7 years, p=0.7 in duration).

A consistency in lateral displacement was observed in all 17 patients. Table 2 shows the values of two parameters of body inclination in each group. There was a tendency toward increased values of both lateral COP displacement and trunk displacement in a group of PD patients compared with controls. However, there was no significant difference in these parameters between that of the control group and the PD-patient group. When each parameter obtained by "frequent fallers" is compared with controls, there was a significant difference in each parameter (lateral COP displacement p=0.01, trunk displacement p=0.01).

Table 2. This data represents the mean value with one standard deviation of each parameter of body inclination in control subjects and in PD patients at rest for 2 minutes. Each mean value was calculated using an absolute value of each parameter obtained from each subject. Fig. 2 shows the relationship between changes of lateral COP displacement and trunk displacement obtained from all 17 patients. The amount of lateral COP displacement was

correlated significantly with that of trunk displacement (r=0.94, p<0.0001).

displacement was evaluated with Spearman's rank correlation coefficient.

arm for 30 seconds.

**3. Results** 

**3.1 Clinical features** 

**3.2 Body inclination during sitting** 

**2.5 Statistical analysis** 

Fig. 2. The relationship between the absolute value of lateral COP displacement (Abs-COP) and the absolute value of trunk displacement (Abs- trunk displacement) obtained from 17 patients at rest.

Fig. 3. The relationship between the value of lateral COP displacement (COP) and the value of trunk displacement obtained from 17 patients during arm-raising examination.

## **3.3 Postural change during arm raising**

Two patients, who were "frequent fallers" and showed a large lateral inclination, had difficulty in keeping their sitting posture for more than 10 seconds and had to be supported by experimenters to prevent them from falling during arm-raising test. Therefore, the following was analyzed: 1) the relationship between the lateral COP displacement and the trunk displacement using the value observed at 1 second after raising the arm to 90 degrees for all 17 patients, and 2) changes of the body axis during the arm-raising for 15 patients.

A postural change observed from one patient during the arm-raising test is shown in Figure 1. When the patient raised her arm laterally up to 90 degrees, the trunk marker shifted to the opposite side. In Figure 3, the relationship between the lateral COP displacement and the trunk displacement obtained from all 17 patients is shown.


Postural Control While Sitting

and Its Association with Risk of Falls in Patients with Parkinson's Disease 245

Fig. 4. A: A sequential change of lateral COP displacement (COP) and the value of trunk displacement (Trunk-marker) when one patient raised her arm alternatively. B: The R value

Table 4. Each body segment size or body weight, which were used to estimate the torque,

and the estimated value of torque both at rest and arm-raising.

change, which was calculated by our proposed equation (see Text).

Table 3. The absolute mean value with one standard deviation of each change of body inclination in control subjects and in PD patients when each subject raised each arm 90 degrees.

A positive relationship with a high correlation coefficient was observed, which was the same as the relationship observed during maintaining sitting posture at rest, only the shift of the initial position of the trunk marker. Table 3 shows the change of body axis associated with the arm-raising in each group. There was no significant difference between both the control group and the PD patient group, or between the control group and the "frequent faller" PD patient group.

In this test, R was defined as the following equation under a hypothesis that the relationship between the lateral COP displacement (ΔG) and the trunk displacement (ΔL) obtained during the sitting condition also applied to the lateral arm raising condition. In the sitting test, the relationship between ΔL and ΔG is expressed as following equation; ΔL=2.46\*ΔG (cf. Figure 2).

## R=ΔL -2.46\*ΔG

Figure 4 shows each parameter obtained from one subject during the arm raising test and the calculated results applied using the equation above. The calculated data showed a square change during the arm-raising phase (Figure 4B). The square change associated with the arm raising was observed in all patients except two patients who had fallen.

## **3.4 Estimation of trunk muscle tone**

In this paper, we proposed a simulation model (cf. Figure 7) and an estimation of trunk muscle tone was made using patient's data under following conditions: 1) to mimic a sitting posture at rest and 2) to mimic a sitting posture with an arm raising. A detail of the model and a calculation procedure are described in the Appendix of this paper. Each body segment

Two patients, who were "frequent fallers" and showed a large lateral inclination, had difficulty in keeping their sitting posture for more than 10 seconds and had to be supported by experimenters to prevent them from falling during arm-raising test. Therefore, the following was analyzed: 1) the relationship between the lateral COP displacement and the trunk displacement using the value observed at 1 second after raising the arm to 90 degrees for all 17 patients, and 2) changes of the body axis during the arm-raising for 15 patients. A postural change observed from one patient during the arm-raising test is shown in Figure 1. When the patient raised her arm laterally up to 90 degrees, the trunk marker shifted to the opposite side. In Figure 3, the relationship between the lateral COP displacement and the

Table 3. The absolute mean value with one standard deviation of each change of body inclination in control subjects and in PD patients when each subject raised each arm 90

A positive relationship with a high correlation coefficient was observed, which was the same as the relationship observed during maintaining sitting posture at rest, only the shift of the initial position of the trunk marker. Table 3 shows the change of body axis associated with the arm-raising in each group. There was no significant difference between both the control group and the PD patient group, or between the control group and the "frequent

In this test, R was defined as the following equation under a hypothesis that the relationship between the lateral COP displacement (ΔG) and the trunk displacement (ΔL) obtained during the sitting condition also applied to the lateral arm raising condition. In the sitting test, the relationship between ΔL and ΔG is expressed as following equation; ΔL=2.46\*ΔG (cf.

R=ΔL -2.46\*ΔG Figure 4 shows each parameter obtained from one subject during the arm raising test and the calculated results applied using the equation above. The calculated data showed a square change during the arm-raising phase (Figure 4B). The square change associated with

In this paper, we proposed a simulation model (cf. Figure 7) and an estimation of trunk muscle tone was made using patient's data under following conditions: 1) to mimic a sitting posture at rest and 2) to mimic a sitting posture with an arm raising. A detail of the model and a calculation procedure are described in the Appendix of this paper. Each body segment

the arm raising was observed in all patients except two patients who had fallen.

**3.3 Postural change during arm raising** 

degrees.

Figure 2).

faller" PD patient group.

**3.4 Estimation of trunk muscle tone** 

trunk displacement obtained from all 17 patients is shown.

Fig. 4. A: A sequential change of lateral COP displacement (COP) and the value of trunk displacement (Trunk-marker) when one patient raised her arm alternatively. B: The R value change, which was calculated by our proposed equation (see Text).


Table 4. Each body segment size or body weight, which were used to estimate the torque, and the estimated value of torque both at rest and arm-raising.

Postural Control While Sitting

at the rest position.

breadth 26.3 cm; arm length 60.4 cm.

**4. Discussion** 

and Its Association with Risk of Falls in Patients with Parkinson's Disease 247

In Figure 6, both the trunk displacement and the COP displacement were shown when the arm segment of the model was raised up to 90 degrees. Although we observed a transient response when the arm segment moved from a resting posture to a 90-degree arm raising posture or when the arm segment moved from the 90-degree position to the initial position, a constant value was observed when the arm segment was held at the 90-degree position or

Fig. 6. Simulation representing an arm-raising test. A sequential change of lateral COP displacement and the value of trunk displacement when the model raised an arm

alternately. A simulation was conducted using each of the following parameters calculated by averaging the data obtained from each of our 17 patients; body weight (except legs): 49.5 kg; arm weight 3.2 kg; sitting height 70.5 cm; shoulder biacromial breadth 32.3 cm; bicristal

In this study, we demonstrated that PD patients who fell frequently tended to have 1) a value of lateral COP displacement greater than the value of control subjects, 2) a geometrical relationship between the arm and the trunk was preserved in PD patients, the same as in control subjects, and 3) a significant difference in postural muscle tone between frequent fallers and non-fallers or less frequent fallers. These results suggest that the measurement of both lateral COP displacement during sitting and arm-raising would be useful in predicting

Postural instability is one of the major symptoms of Parkinson's disease, and the instability in the control of upright stance and posture in PD often results in falling (Bloem et al., 2001; Wood et al., 2002; Grimbergen et al., 2004; Bloem et al., 2006). Several studies reported that postural sway patterns during quiet stance in PD patients are different compared to those of healthy elderly subjects, with PD patients displaying larger lateral excursion compared with anteroposterior excursion (Mitchell et al., 1995; Morris et al., 2000; Van Wegen et al., 2001; Van der Burg et al., 2006). In our study, the same tendency of postural sway pattern during sitting was observed and the excursions in the lateral direction in PD patients were larger than those of control subjects. Regarding postural control during standing, several authors

the risk of falling and predicting the trunk rigidity in PD patients.

**4.1 Body inclination during sitting and in relation to falling frequency** 

size or body weight, which was used to estimate the torque, is shown in Table 4. There was a tendency toward increased value of the torque in PD patients who experienced falling frequently. The mean value with standard deviation of the torque at rest in non-fallers and less frequent fallers was 2.4 Nm1.9 Nm, and that of the frequent fallers was 8.6 Nm8.4 Nm (p<0.03). In Figure 5, the relationship between the value of trunk displacement and the estimated torque for each patient is shown. An estimated torque value was calculated using an averaged data value of all 17 patients when our model leaned. The data suggested that the trunk muscle tone was larger in the patients with high falling risk than the patients with less falling risk. In a simulation of the arm-raising, there was no significant difference between the non-fallers (42.92.3 Nm) and the fallers (42.12.0 Nm) (p>0.47).

Fig. 5. Relationship between the value of trunk displacement and the estimated torque for each patient. An estimated torque value to the trunk displacement is shown as a straight line: a simulation was conducted using each of the following parameters calculated by averaging the data obtained from each of our 17 patients; body weight (except legs): 49.5 kg; arm weight 3.2 kg; sitting height 70.5 cm; shoulder biacromial breadth 32.3 cm; bicristal breadth 26.3 cm; arm length 60.4 cm.

In Figure 6, both the trunk displacement and the COP displacement were shown when the arm segment of the model was raised up to 90 degrees. Although we observed a transient response when the arm segment moved from a resting posture to a 90-degree arm raising posture or when the arm segment moved from the 90-degree position to the initial position, a constant value was observed when the arm segment was held at the 90-degree position or at the rest position.

Fig. 6. Simulation representing an arm-raising test. A sequential change of lateral COP displacement and the value of trunk displacement when the model raised an arm alternately. A simulation was conducted using each of the following parameters calculated by averaging the data obtained from each of our 17 patients; body weight (except legs): 49.5 kg; arm weight 3.2 kg; sitting height 70.5 cm; shoulder biacromial breadth 32.3 cm; bicristal breadth 26.3 cm; arm length 60.4 cm.

## **4. Discussion**

246 Diagnostics and Rehabilitation of Parkinson's Disease

size or body weight, which was used to estimate the torque, is shown in Table 4. There was a tendency toward increased value of the torque in PD patients who experienced falling frequently. The mean value with standard deviation of the torque at rest in non-fallers and less frequent fallers was 2.4 Nm1.9 Nm, and that of the frequent fallers was 8.6 Nm8.4 Nm (p<0.03). In Figure 5, the relationship between the value of trunk displacement and the estimated torque for each patient is shown. An estimated torque value was calculated using an averaged data value of all 17 patients when our model leaned. The data suggested that the trunk muscle tone was larger in the patients with high falling risk than the patients with less falling risk. In a simulation of the arm-raising, there was no significant difference

between the non-fallers (42.92.3 Nm) and the fallers (42.12.0 Nm) (p>0.47).

Fig. 5. Relationship between the value of trunk displacement and the estimated torque for each patient. An estimated torque value to the trunk displacement is shown as a straight line: a simulation was conducted using each of the following parameters calculated by averaging the data obtained from each of our 17 patients; body weight (except legs): 49.5 kg; arm weight 3.2 kg; sitting height 70.5 cm; shoulder biacromial breadth 32.3 cm; bicristal

breadth 26.3 cm; arm length 60.4 cm.

In this study, we demonstrated that PD patients who fell frequently tended to have 1) a value of lateral COP displacement greater than the value of control subjects, 2) a geometrical relationship between the arm and the trunk was preserved in PD patients, the same as in control subjects, and 3) a significant difference in postural muscle tone between frequent fallers and non-fallers or less frequent fallers. These results suggest that the measurement of both lateral COP displacement during sitting and arm-raising would be useful in predicting the risk of falling and predicting the trunk rigidity in PD patients.

## **4.1 Body inclination during sitting and in relation to falling frequency**

Postural instability is one of the major symptoms of Parkinson's disease, and the instability in the control of upright stance and posture in PD often results in falling (Bloem et al., 2001; Wood et al., 2002; Grimbergen et al., 2004; Bloem et al., 2006). Several studies reported that postural sway patterns during quiet stance in PD patients are different compared to those of healthy elderly subjects, with PD patients displaying larger lateral excursion compared with anteroposterior excursion (Mitchell et al., 1995; Morris et al., 2000; Van Wegen et al., 2001; Van der Burg et al., 2006). In our study, the same tendency of postural sway pattern during sitting was observed and the excursions in the lateral direction in PD patients were larger than those of control subjects. Regarding postural control during standing, several authors

Postural Control While Sitting

・deg in PD patients).

**5. Conclusion** 

**6. Appendix** 

the waist.

raising phase even though the trunk inclined.

**4.3 Estimation of postural muscle tone in the body axis** 

and Its Association with Risk of Falls in Patients with Parkinson's Disease 249

geometric relationship between the raised arm and the trunk was preserved during the arm-

Rigidity is a continuous and uniform increase in muscle tone, felt as a constant resistance throughout the range of passive movement of a limb or a neck, and is a cardinal symptom of Parkinson's disease. Clinically, rigidity is usually assessed by passively flexing and extending a patient's limb. Objectively, most previous investigators examined rigidity in the muscles in PD patients, using either torque motor or isokinetic dynamometer (Nuyens et al., 2000; Hayashi et al., 2001). A few studies have done quantitatively to measure of postural muscle tone in the body axis of healthy test subjects (Kumar 2004; Gurfinkel et al., 2006) or to measure of trunk rigidity in parkinsonian patients (Mak et al., 2007). In these studies, the measurement procedures were different; each estimated value of muscle tone was close. Gurfinkel et al. (2006) reported that the value was 2 Nm to 9 Nm in standing, 40 Nm to 80 Nm in sitting (Kumar, 2004). Mak and coworkers (2007) reported PD patients had a significantly higher trunk muscle tone when compared with normal controls in the standing position, in both passive trunk flexion (17-22 Nm・deg in the control group, 27-40 Nm・deg in PD patients) and passive trunk extension (21-26 Nm・deg in the control group, 28-45 Nm

In this paper, we estimated the postural muscle tone in the body axis based on our model and showed that the positive relationship between the trunk displacement and the estimated torque value at rest condition. The estimated torque value in the high risk PD patients for falls was larger than that of lower risk PD patients. These results were consistent with that reported by Mak and coworkers (2007). In their study, it was reported that

Based on these results, we conclude that a measurement of body inclination for an extended period of 2 minutes as in this study is a valuable predictor for the risk of falling and is a simple and easy method to estimate the trunk muscle tone. This study also demonstrates that even PD patients with high falling risk are capable of controlling their postural

The proposed model, which simplified a sitting posture on a chair, was developed using an ODE (Open Dynamic Engine) simulator. This model is composed of the following 7 parts, including the upper part of the trunk, arm part, shoulder hinge joint, impedance joint set under the upper part, waist hinge joint, lower part of the trunk and the hip hinge joint. Body segment parameters used at the simulator are given by general physical data obtained from our patients, and an estimation value of each body segment was made based on the

The unique point of our model is using an "impedance joint", which consist of two parts: "elastic property" and "viscous properties" to artificially reproduce the muscles encompassing

unstable PD patients had a tendency to have high trunk muscle tone.

geometry between the raised arm and the trunk.

previous reports (Jensen et al., 1994, Okada et al., 1996).

suggested that instability in the anterior-posterior direction was compensated for by increasing excursion in the lateral direction in PD patients (Schieppati et al., 1994; Mitchell et al., 1995; Van Wegen et al., 2001). These mechanisms might be working during sitting and would have eventually increased the lateral inclination. Van Emmerik et al. (1999) also suggested that lateral impairment in postural control in patients with PD might be a reflection of axial rigidity.

Several studies suggested that increased lateral sway is associated with increased risk of falling in both elderly subjects (Maki et al., 1994) and patients with PD (Mitchell et al., 1995; Rocchi et al., 2002). Our study also showed that the degree of lateral inclination during sitting was significantly larger in PD patients with history of falls than PD patients without history.

Based on these results, we suggest the risk of falls would increase when compensation in the anterior-posterior body sway with the lateral body sway is difficult. Measurement of postural change in the lateral direction during sitting for a relatively long time, 2 minutes in this study, is a simple and effective method that can be used in daily clinical examinations to evaluate the risk of falling.

#### **4.2 Postural change associated with a lateral arm-raising task**

A large number of human motor actions cause potential displacement of the body center of gravity (COP) and it is well known that voluntary movement is preceded and accompanied by postural muscular activities. Several studies which fall into this category include activation of posterior trunk and leg muscles when the arms are raised in front of the body (Traub et al., 1980; Zattare & Bouisset, 1988) or when a leg is raised while standing (Lee et al., 1995).

There are several reports that postural adjustments of the upper extremities may also be disrupted in Parkinson's disease. Abnormalities in the timing or amplitude of anticipatory postural adjustments, which occur when rapid voluntary arm movements are made while in the standing position, have been reported in parkinsonian subjects (Dick et al., 1986). Lee and coworkers (1995) studied the preparatory postural adjustments associate with a lateral leg raising task in parkinsonian patients and described the amplitude of the initial displacement of COP was markedly reduced and the interval between the earliest force changes and the onset of leg elevation was prolonged. These authors focused on the initial phase of the preparatory postural adjustments in parkinsonian patients with postural instability. On the other hand, in our study, the postural change after elevating and holding the arm was examined. We found no significant difference in the trunk inclination, associated with arm-raising between normal controls and PD patients, or between raising the right arm and the left arm test in PD patients. These test results indicated that the postural controls during the arm raising were preserved in PD patients we studied.

Adopting an appropriate body orientation, and maintaining this posture to the displacing effects of gravity or external forces is essential for postural control. Patients with Parkinson's disease had difficulty in postural control, especially the control of body vertically (Vaugoyeau et al., 2007; Hayashi et al., 2010). Steiger et al. (1996) reported that PD patients had difficulty in coordinating the orientation of the axial segments along the spinal axis. Several investigators also reported that the proprioceptive feedback information to the static position and movement perception processing decreased in PD patients (Zia et al., 2002; Keijsers et al., 2005; Vaugoyeau et al., 2007). In this study, the body inclination was observed during the arm-raising test, and the R-value was constant. These results suggested that a geometric relationship between the raised arm and the trunk was preserved during the armraising phase even though the trunk inclined.

## **4.3 Estimation of postural muscle tone in the body axis**

Rigidity is a continuous and uniform increase in muscle tone, felt as a constant resistance throughout the range of passive movement of a limb or a neck, and is a cardinal symptom of Parkinson's disease. Clinically, rigidity is usually assessed by passively flexing and extending a patient's limb. Objectively, most previous investigators examined rigidity in the muscles in PD patients, using either torque motor or isokinetic dynamometer (Nuyens et al., 2000; Hayashi et al., 2001). A few studies have done quantitatively to measure of postural muscle tone in the body axis of healthy test subjects (Kumar 2004; Gurfinkel et al., 2006) or to measure of trunk rigidity in parkinsonian patients (Mak et al., 2007). In these studies, the measurement procedures were different; each estimated value of muscle tone was close. Gurfinkel et al. (2006) reported that the value was 2 Nm to 9 Nm in standing, 40 Nm to 80 Nm in sitting (Kumar, 2004). Mak and coworkers (2007) reported PD patients had a significantly higher trunk muscle tone when compared with normal controls in the standing position, in both passive trunk flexion (17-22 Nm・deg in the control group, 27-40 Nm・deg in PD patients) and passive trunk extension (21-26 Nm・deg in the control group, 28-45 Nm ・deg in PD patients).

In this paper, we estimated the postural muscle tone in the body axis based on our model and showed that the positive relationship between the trunk displacement and the estimated torque value at rest condition. The estimated torque value in the high risk PD patients for falls was larger than that of lower risk PD patients. These results were consistent with that reported by Mak and coworkers (2007). In their study, it was reported that unstable PD patients had a tendency to have high trunk muscle tone.

## **5. Conclusion**

248 Diagnostics and Rehabilitation of Parkinson's Disease

suggested that instability in the anterior-posterior direction was compensated for by increasing excursion in the lateral direction in PD patients (Schieppati et al., 1994; Mitchell et al., 1995; Van Wegen et al., 2001). These mechanisms might be working during sitting and would have eventually increased the lateral inclination. Van Emmerik et al. (1999) also suggested that lateral impairment in postural control in patients with PD might be a

Several studies suggested that increased lateral sway is associated with increased risk of falling in both elderly subjects (Maki et al., 1994) and patients with PD (Mitchell et al., 1995; Rocchi et al., 2002). Our study also showed that the degree of lateral inclination during sitting was significantly larger in PD patients with history of falls than PD patients without history. Based on these results, we suggest the risk of falls would increase when compensation in the anterior-posterior body sway with the lateral body sway is difficult. Measurement of postural change in the lateral direction during sitting for a relatively long time, 2 minutes in this study, is a simple and effective method that can be used in daily clinical examinations to

A large number of human motor actions cause potential displacement of the body center of gravity (COP) and it is well known that voluntary movement is preceded and accompanied by postural muscular activities. Several studies which fall into this category include activation of posterior trunk and leg muscles when the arms are raised in front of the body (Traub et al., 1980; Zattare & Bouisset, 1988) or when a leg is raised while standing (Lee et

There are several reports that postural adjustments of the upper extremities may also be disrupted in Parkinson's disease. Abnormalities in the timing or amplitude of anticipatory postural adjustments, which occur when rapid voluntary arm movements are made while in the standing position, have been reported in parkinsonian subjects (Dick et al., 1986). Lee and coworkers (1995) studied the preparatory postural adjustments associate with a lateral leg raising task in parkinsonian patients and described the amplitude of the initial displacement of COP was markedly reduced and the interval between the earliest force changes and the onset of leg elevation was prolonged. These authors focused on the initial phase of the preparatory postural adjustments in parkinsonian patients with postural instability. On the other hand, in our study, the postural change after elevating and holding the arm was examined. We found no significant difference in the trunk inclination, associated with arm-raising between normal controls and PD patients, or between raising the right arm and the left arm test in PD patients. These test results indicated that the

postural controls during the arm raising were preserved in PD patients we studied.

Adopting an appropriate body orientation, and maintaining this posture to the displacing effects of gravity or external forces is essential for postural control. Patients with Parkinson's disease had difficulty in postural control, especially the control of body vertically (Vaugoyeau et al., 2007; Hayashi et al., 2010). Steiger et al. (1996) reported that PD patients had difficulty in coordinating the orientation of the axial segments along the spinal axis. Several investigators also reported that the proprioceptive feedback information to the static position and movement perception processing decreased in PD patients (Zia et al., 2002; Keijsers et al., 2005; Vaugoyeau et al., 2007). In this study, the body inclination was observed during the arm-raising test, and the R-value was constant. These results suggested that a

**4.2 Postural change associated with a lateral arm-raising task** 

reflection of axial rigidity.

evaluate the risk of falling.

al., 1995).

Based on these results, we conclude that a measurement of body inclination for an extended period of 2 minutes as in this study is a valuable predictor for the risk of falling and is a simple and easy method to estimate the trunk muscle tone. This study also demonstrates that even PD patients with high falling risk are capable of controlling their postural geometry between the raised arm and the trunk.

## **6. Appendix**

The proposed model, which simplified a sitting posture on a chair, was developed using an ODE (Open Dynamic Engine) simulator. This model is composed of the following 7 parts, including the upper part of the trunk, arm part, shoulder hinge joint, impedance joint set under the upper part, waist hinge joint, lower part of the trunk and the hip hinge joint. Body segment parameters used at the simulator are given by general physical data obtained from our patients, and an estimation value of each body segment was made based on the previous reports (Jensen et al., 1994, Okada et al., 1996).

The unique point of our model is using an "impedance joint", which consist of two parts: "elastic property" and "viscous properties" to artificially reproduce the muscles encompassing the waist.

Postural Control While Sitting

quiet standing position.

will be stationary; thereby we can consider 0 *x* .

Where *Ph* is the length from the home position.

The torque applied to the hip hinge joint will be express as:

Right arm hinge torque can be represented excellently by:

1. to mimic a sitting posture at rest

torque and the hip hinge torque.

trajectories of the arm.

and Its Association with Risk of Falls in Patients with Parkinson's Disease 251

Winter et al. (1998) reported that a relationship between COP and COM (center of mass or barycenter) was expressed in the following equation in a stiffness control of balance in the

> *<sup>I</sup> COP COM x mgh*

Where *I* is the inertial around the mass, *m* is the mass of the object, *g* is the acceleration due to gravity, *h* is the length from the origin coordinates and *x* is the second derivative of *COM*. At this simulation, thereafter the arm is raised up to the 90-degree position, the body

( )/( ) ( ) /( ) *tx un un x up up x la la x ra ra x un up la ra tz un un z up up z la la z ra ra z un up la ra*

*P M P M P MP MP M M M M P M P M P MP MP M M M M* .

Where , ,, *Mun u M MM <sup>p</sup> la ra* is the mathamatical representation of "lower part of the trunk (21.5% of body weight)", "upper part of the trunk (28.5% of body weight)", "left arm (6.5% of body weight)", "right arm (6.5% of body weight)", and \_ \_\_ \_ \_ \_\_\_ , ,, , , ,, *P P PP P P PP un x up x la x ra x un z up z la z ra z* is the barycentric position of each part where *x* and *z* represent the horizontal component and the vertical component, respectively. The angle between "barycentric position" and "hip hinge joint" will be expression as:

angle arctan( /( )) *P PP tx tz h* .

torque (0 ) (0 ) *h Dh Vh Ph Ah G RG R* ,

2. to mimic a sitting posture with the right arm raising, we used the right arm hinge

torque ( )( ) *ra Da Va Va Pa Aa Aa GB R GB R*

And , *R R Va Aa* are a displacement and an angular rate and , *B B Va Aa* are the target

where *G G Da Pa* , are a derivative gain and a proportional gain of the torque.

where *G G Dh Ph* , are a derivative gain and a proportional gain of the torque.

And , *R R Vh Ah* are a displacement and an angular rate of the body.

The barycentric coordinate of the full body ( *P P tx tz* , ) will be expression as:

\_ \_\_ \_ \_ \_\_ \_

The mathematical expression of the each impedance joint is shown as:

$$I\ddot{\theta} = \mathfrak{r} - d\dot{\theta} - k\theta^\*$$

where *I* is the fictitious force, *θ* is the angle of the joint, *τ* is the muscle torque around the waist, *d* is the elastic property and *k* is the viscous property. By using this impedance joint, it becomes very easy to simulate the behavior of the body. The hinge joints work as an actuator, rotation sprig or a torque motor; which are used in the shoulder, waist and hip joint. It is also possible to obtain the data of each part and the whole body's barycentric coordinates using this simulator.

The procedure of the simulation is thus illustrated as follows: first, each parameter of the model such as weight or height et al. was estimated based on the previous Japanese reports (Okada et al., 1996) and our patients (cf. Table 4). Then a condition was imposed on the model to remain stable while a small disturbance to the body was applied. During this procedure the torque exerted on each of the joints was calculated, and using this data we were able to calculate the optimal solution of the stiffness for each joint.

At the next step, a stiffness condition was imposed on the model, which was evaluated from the previous simulation and then, set the same condition for the model to stay stable. This time, instead of applying a small disturbance, we requested the model to raise their arm up to the 90-degree position in 1 second. Through simulation, we gave careful consideration of the body sway by using the impedance joint and the hinge joints. Tuning the two impedance joints and the stiffness joints through the numerical simulations, we estimated an identified torque, which is needed to maintain posture maintenance.

Fig. 7. A proposed model

coordinates using this simulator.

The mathematical expression of the each impedance joint is shown as:

were able to calculate the optimal solution of the stiffness for each joint.

torque, which is needed to maintain posture maintenance.

*I dk*

The procedure of the simulation is thus illustrated as follows: first, each parameter of the model such as weight or height et al. was estimated based on the previous Japanese reports (Okada et al., 1996) and our patients (cf. Table 4). Then a condition was imposed on the model to remain stable while a small disturbance to the body was applied. During this procedure the torque exerted on each of the joints was calculated, and using this data we

At the next step, a stiffness condition was imposed on the model, which was evaluated from the previous simulation and then, set the same condition for the model to stay stable. This time, instead of applying a small disturbance, we requested the model to raise their arm up to the 90-degree position in 1 second. Through simulation, we gave careful consideration of the body sway by using the impedance joint and the hinge joints. Tuning the two impedance joints and the stiffness joints through the numerical simulations, we estimated an identified

 where *I* is the fictitious force, *θ* is the angle of the joint, *τ* is the muscle torque around the waist, *d* is the elastic property and *k* is the viscous property. By using this impedance joint, it becomes very easy to simulate the behavior of the body. The hinge joints work as an actuator, rotation sprig or a torque motor; which are used in the shoulder, waist and hip joint. It is also possible to obtain the data of each part and the whole body's barycentric

 Winter et al. (1998) reported that a relationship between COP and COM (center of mass or barycenter) was expressed in the following equation in a stiffness control of balance in the quiet standing position.

$$COP-COM = -\frac{I}{mgh}\ddot{x}$$

Where *I* is the inertial around the mass, *m* is the mass of the object, *g* is the acceleration due to gravity, *h* is the length from the origin coordinates and *x* is the second derivative of *COM*. At this simulation, thereafter the arm is raised up to the 90-degree position, the body will be stationary; thereby we can consider 0 *x* .

The barycentric coordinate of the full body ( *P P tx tz* , ) will be expression as:

$$P\_{tx} = \left(M\_{\text{un}} \cdot P\_{\text{un\\_x}} + M\_{\text{up}} \cdot P\_{\text{up\\_x}} + M\_{\text{la}} \cdot P\_{\text{la\\_x}} + M\_{\text{ra\\_}} \cdot P\_{\text{ra\\_x}}\right) / \left(M\_{\text{un}} + M\_{\text{up}} + M\_{\text{la}} + M\_{\text{ra}}\right)$$

$$P\_{tz} = \left(M\_{\text{un}} \cdot P\_{\text{un\\_}} + M\_{\text{up}} \cdot P\_{\text{up\\_}\text{-}z} + M\_{\text{la}} \cdot P\_{\text{la\\_}} + M\_{\text{ra}} \cdot P\_{\text{ra\\_}}\right) / \left(M\_{\text{un}} + M\_{\text{up}} + M\_{\text{la}} + M\_{\text{ra}}\right)$$

Where , ,, *Mun u M MM <sup>p</sup> la ra* is the mathamatical representation of "lower part of the trunk (21.5% of body weight)", "upper part of the trunk (28.5% of body weight)", "left arm (6.5% of body weight)", "right arm (6.5% of body weight)", and \_ \_\_ \_ \_ \_\_\_ , ,, , , ,, *P P PP P P PP un x up x la x ra x un z up z la z ra z* is the barycentric position of each part where *x* and *z* represent the horizontal component and the vertical component, respectively. The angle between "barycentric position" and "hip hinge joint" will be expression as:

$$\text{angle} = \arctan(P\_{tx} \;/\; (P\_{tz} - P\_h)) \;. \; .$$

Where *Ph* is the length from the home position.

1. to mimic a sitting posture at rest

The torque applied to the hip hinge joint will be express as:

$$\text{torque}\_{\text{\tiny \text{\tiny \text{\tiny \text{\tiny \text{\tiny \text{\text}}}}}} = G\_{D\text{\text{\textgreater}}}(0 - R\_{V\text{\textquotesingle}}) + G\_{P\text{\textquotesingle}}(0 - R\_{A\text{\textquotesingle}}) \text{ \textquotesingle}$$

where *G G Dh Ph* , are a derivative gain and a proportional gain of the torque.

And , *R R Vh Ah* are a displacement and an angular rate of the body. 2. to mimic a sitting posture with the right arm raising, we used the right arm hinge torque and the hip hinge torque.

Right arm hinge torque can be represented excellently by:

$$\text{torque}\_{ra} = -G\_{Da}(B\_{Va} - R\_{Va}) - G\_{Pa}(B\_{Aa} - R\_{Aa})$$

where *G G Da Pa* , are a derivative gain and a proportional gain of the torque. And , *R R Va Aa* are a displacement and an angular rate and , *B B Va Aa* are the target trajectories of the arm.

Postural Control While Sitting

pp 205-16.

2000, pp87-93.

267-74.

97.

1980, pp 393-412.

146, 2007, pp 852-63.

*Psychiatry*. 72, 2002, pp 721-5.

pp186-91.

ISBN 4-13-060134-3, Japan

*Neurophysiol*. 93, 1994, pp 286-98.

*Relat Disord*. 12, 2006, pp 492-8.

and Its Association with Risk of Falls in Patients with Parkinson's Disease 253

Morris, M., Iansek, R., Smithson, F., Huxham, F. Postural insatiability in Parkinson's

Nuyens, G., De Weerdt, W., Dom, R., Nieuwboer, A., Spaepen, A. Torque variations

Okada, H., Ae, M., Fujii, N., Morioka, Y. Body segument inertia properties of Japanese

Schieppati, M., Hugon, M., Grasso, M., Nardone, A., Galante, M. The limits of equilibrium in

Shivitz, N., Koop, MM., Fahimi, J., Heit, G, Bronte-Stewart HM. Bilateral subthalamic

Steiger, MJ., Thompson, PD., Marsden, CD. Disordered axial movement in Parkinson's

Traub, MM., Rothwell, JC., Marsden, CD. Anticipatory postural reflexes in Parkinson's

Van der Burg, JCE., van Wegen, EEH., Rietberg, MB., Kwakkel, G., van Dieën, JH. Postural

Van Emmerik, REA., Wagenaar, RC., Winogrodzka, A., Wolters, EC. Identification of axial

Van Wegen, EEH., van Emmerik, REA., Wagenaar, RC., Ellis, T. Stability boundaries and lateral postural control in Parknson's disease. *Motor Control.* 3, 2001, pp 254-69. Vaugoyeau, M., Viel, S., Assaiante, C., Amblard, B., Azulay, JP. Impaired vertical postural

Winter, DA., Patla, AE., Prince, F., Ishac, M., Gielo-Perczak K. Stiffness control of balance in

Wood, BH., Bilclough, JA., Bowron, A., Walker, RW. Incidence and prediction of falls in

disease. *J Neurol Neurosurg Psychiatry.* 61, 1996, pp 645-8.

quiet standing. *J Neurophysiol.* 80, 1998, pp-1211-21.

Parkinson, J. (1817). *An essay on the shaking palsy.* Sherwood, Neely, and Jones, London. Rocchi, L., Chiari, L., Horak, FB. Effects of deep brain stimulation and levodopa on

disease: a comparison with and without a concurrent task. *Gait Posture.* 12, 2000,

during repeated passive isokinetic movements of the knee in subjects with Parkinson's disease and healthy control subjects. *Parkinsonism Relat Disord.* 6,

elderly. *Biomechanisms* Vol.13, 1996, pp 125-139. The Society of Biomechanisms.

postural sway in Parkinson's disease. *J Neurol Neurosurg Psychiatry.* 73, 2002, pp

young and elderly normal subjects and in parkinsonians. *Electroencephalogr Clin* 

nucleus deep brain stimulation improves certain aspects of postural control in Parkinson's disease, whereas medication does not. *Mov Disord*. 21, 2006, pp 1088-

disease and other akinetic-rigid syndromes and in cerebellar ataxia. *Brain.* 103,

control of the trunk during unstable sitting in Parkinson's disease. *Parkinsonism* 

rigidity during locomotion in Parkinson disease. *Arch Phys Med Rehabil.* 80, 1999,

control and proprioceptive integration deficits in Parkinson's disease. *Neuroscience.*

Parkinson's disease: a prospective multidisciplinary study. *J Neurol Neurosurg* 

### **7. References**


Bloem, BR., Grimbergen, YAM., Cramer, M., Willemsen, M., Zwinderman, AH. Prospective assessment of falls in Parkinson's disease. *J Neurol.* 248, 2001, pp 950-8. Bloem, BR., Grimbergen, YAM., van Dijk, JG., Munneke, M. The "posture second"

Dick, JP., Rothwell, JC., Berardelli, A., Thompson, PD., Gioux, M., Benecke, R., Day, BL.,

Duvoisin, RC., Marsden, CD. Note on the scoliosis of Parkinsonism. *J Neurol Neurosurg* 

Grimbergen, YAM., Munneke, M., Bloem, BR. Falls in Parkinson's disease. *Curr Opin Neurol*.

Gurfinkel, V., Cacciatore, TW., Cordo, P., Horak, F., Nutt, J., Skoss, R. Postural muscle tone in the body axis of healthy humans. *J Neurophysiol.* 96, 2006, pp 2678-87. Hayashi, R., Hashimoto, T., Tada, T., Ikeda, S. Relation between changes in long-latency

Hodges, PW., Gurfinkel, VS., Brumagne, S., Smith, TC., Cordo, PC. Coexistence of stability

Jensen, RK., Fletcher, P. Distribution of mass to the segments of elderly males and females. J

Keijsers, NL., Admiraal, MA., Cools, AR., Bloem, BR., Gielen, CC. Differential progression of

Lee, RG., Tonolli, I., Viallet, F., Aurenty, R., Massion, J. Preparatory postural adjustments

Mak, MK., Wong, EC., Hui-Chan, CW. Quantitative measurement of trunk rigidity in

Maki, BE., Holliday, PJ., Topper, AK. A prospective study of postural balance and risk of

Mitchell. SL., Collins, JJ., De Luca, CJ., Burrows, A., Lipsitz, LA. Open-loop and closed-loop

Kumar, S. Ergonomics and biology of spinal rotation. *Ergonomics.* 47, 2004, pp 370-415. Krishnamoorthy, V., Yang, JF., Scholz, JP. Joint coordination during quiet stance: effects of

after unilateral pallidotomy. *Clin Neurophysiol.* 112, 2001, pp 1814-21. Hayashi, R., Aizawa, J., Nagase, H., Ohara, S. Lateral inclination of the trunk and falling

strategy: a review of wrong priorities in Parkinson's disease. *J Neurol Sci.* 248,

Marsden, CD. Associated postural adjustments in Parkinson's disease. *J Neurol* 

stretch reflexes and muscle stiffness in Parkinson's disease comparison before and

frequency in Parkinson's disease patients. *Electromyogr Clin Neurophysiol.* 50, 2010,

and mobility in postural control: evidence from postural compensation for

proprioceptive and visual information processing deficits in Parkinson's disease.

in parkinsonian patients with postural instability. *Can J Neurol Sci.* 22, 1995, pp

falling in an ambulatory and independent elderly population. *J Gerontol*. 49, 1994,

postural control mechanisms in Parkinson's disease: increased mediolateral activity

**7. References** 

2006, pp 196-204.

17, 2004, pp 405-15.

pp 195-202.

126-35.

pp M72-84.

*Neurosurg Psychiatry.* 49, 1986, pp 1378-85.

respiration. *Exp Brain Res.* 144, 2002, pp 293-302.

*Psychiatry.* 38, 1975, pp 787-93.

Biomech. 27,1994, pp 89-96.

*Eur J Neurosci.* 21, 2005, pp 239-48.

vision. *Exp Brain Res.* 164, 2005, pp 1-17.

parkinsonian patients. *J Neurol.* 254, 2007, pp 202-9.

during quiet standing. *Neurosci Lett*. 197, 1995, pp 133-6.


**Part 3** 

**Multidisciplinary Cognitive Rehabilitation** 

 **in Parkinson's Disease** 


## **Part 3**

## **Multidisciplinary Cognitive Rehabilitation in Parkinson's Disease**

254 Diagnostics and Rehabilitation of Parkinson's Disease

Zattara, M., Bouisset, S. Posturo-kinetic organisation during the early phase of voluntary

Zia, S., Cody, FW., O'Boyle, DJ. Identification of unilateral elbow-joint position is impaired

by Parkinson's disease. *Clin Anat.* 15, 2002, pp 23-3.

pp 956-65.

upper limb movement. 1. Normal subjects. *J Neurol Neurosurg Psychiatry*. 51, 1988,

**13** 

*1,3Germany 2Swiss* 

**Cognitive Rehabilitation in Parkinson's Disease** 

Idiopathic Parkinson`s disease (PD) is a neurodegenerative disorder characterized by basal ganglia dysfunction frequently being associated with frontostriatal dysfunction and cognitive impairment. The prevalence of PD increases with age and is estimated at 100- 200/100000 people (Chen et al., 2001; Schrag et al., 2000) worldwide. The clinical hallmarks of PD are akinesia, rigidity and tremor (Douglas et al., 1999; Hughes et al., 1992). In the past PD has been considered as a pure movement disorder, but in recent years the presence of non-motor symptoms in PD has been recognized. Non-motor symptoms include a variety of autonomic dysfunctions such as orthostatic hypotension, postural tachycardia, bladder dysfunction, sleep disturbances, psychiatric symptoms, i.e. depression, hallucinations or psychosis and cognitive impairment. Non-motor symptoms such as pain, depression or sleep disturbances might precede the onset of motor symptoms in PD and are sometimes even more disabling than motor deficits. For many years cognitive impairment and the occurrence of dementia have been considered as not typical for IPD. James Parkinson (Parkinson, 1817) wrote in his essay on the shaking palsy " the senses are not disturbed". However, there is now enough evidence in the literature that dementia might occur in up to 40% of PD-patients (Emre et al., 2004). PD dementia is the third most common reason for dementia. Dementia in PD has been associated with reduced quality of life, greater sensitivity to medication, higher risk of developing psychosis, shortened survival (Levy, 2002), increased caregivers stress and frequent transfer to nursing homes (Aarsland et al., 2000) compared to PD-patients without dementia. In contrast to dementia mild cognitive impairment might occur early in the course of the disease. Approximately, a quarter of PDpatients without dementia have mild cognitive impairment (PD-MCI) and 20% might have MCI at the time of diagnosis (Aarsland et al., 2011). The cognitive deficits in PD are specific and include executive dysfunction, attentional and visuospatial deficits. Executive functions include control, manipulation, and cognitive flexibility (Funahashi et al., 2001; Lezak, 1995) and is part of working memory (Carpenter et al., 2000). The executive system is thought to

**1. Introduction** 

**1.1 Cognitive impairment in Parkinson's disease** 

**Using Neuropsychological Training, Transfer** 

**Training and Sports Therapy** 

*3Dept. of Orthopedic Surgery, Klinikum Osnabrück,* 

I. Reuter1, S. Mehnert1, M. Oechsner2 and M. Engelhardt3 *1Dept. of Neurology, Justus-Liebig University, Giessen, Germany* 

*2Neurologisches Rehabilitationszentrum, HELIOS Klinik Zihlschlacht AG,* 

## **Cognitive Rehabilitation in Parkinson's Disease Using Neuropsychological Training, Transfer Training and Sports Therapy**

I. Reuter1, S. Mehnert1, M. Oechsner2 and M. Engelhardt3

*1Dept. of Neurology, Justus-Liebig University, Giessen, Germany 2Neurologisches Rehabilitationszentrum, HELIOS Klinik Zihlschlacht AG, 3Dept. of Orthopedic Surgery, Klinikum Osnabrück, 1,3Germany 2Swiss* 

## **1. Introduction**

#### **1.1 Cognitive impairment in Parkinson's disease**

Idiopathic Parkinson`s disease (PD) is a neurodegenerative disorder characterized by basal ganglia dysfunction frequently being associated with frontostriatal dysfunction and cognitive impairment. The prevalence of PD increases with age and is estimated at 100- 200/100000 people (Chen et al., 2001; Schrag et al., 2000) worldwide. The clinical hallmarks of PD are akinesia, rigidity and tremor (Douglas et al., 1999; Hughes et al., 1992). In the past PD has been considered as a pure movement disorder, but in recent years the presence of non-motor symptoms in PD has been recognized. Non-motor symptoms include a variety of autonomic dysfunctions such as orthostatic hypotension, postural tachycardia, bladder dysfunction, sleep disturbances, psychiatric symptoms, i.e. depression, hallucinations or psychosis and cognitive impairment. Non-motor symptoms such as pain, depression or sleep disturbances might precede the onset of motor symptoms in PD and are sometimes even more disabling than motor deficits. For many years cognitive impairment and the occurrence of dementia have been considered as not typical for IPD. James Parkinson (Parkinson, 1817) wrote in his essay on the shaking palsy " the senses are not disturbed". However, there is now enough evidence in the literature that dementia might occur in up to 40% of PD-patients (Emre et al., 2004). PD dementia is the third most common reason for dementia. Dementia in PD has been associated with reduced quality of life, greater sensitivity to medication, higher risk of developing psychosis, shortened survival (Levy, 2002), increased caregivers stress and frequent transfer to nursing homes (Aarsland et al., 2000) compared to PD-patients without dementia. In contrast to dementia mild cognitive impairment might occur early in the course of the disease. Approximately, a quarter of PDpatients without dementia have mild cognitive impairment (PD-MCI) and 20% might have MCI at the time of diagnosis (Aarsland et al., 2011). The cognitive deficits in PD are specific and include executive dysfunction, attentional and visuospatial deficits. Executive functions include control, manipulation, and cognitive flexibility (Funahashi et al., 2001; Lezak, 1995) and is part of working memory (Carpenter et al., 2000). The executive system is thought to

Cognitive Rehabilitation in Parkinson's Disease

reduced BDNF levels in the SNC (Howells et al., 2000)

**1.3 Treatment options for cognitive decline in PD** 

executive functions and tasks of the lower extremities.

multimodal cognitive training to improve cognitive functions.

patients.

Using Neuropsychological Training, Transfer Training and Sports Therapy 259

frontal-executive dysfunction. Dopaminergic frontal systems play a major role in working memory and executive function (Goldman-Rakic et al., 1992), especially the dorsolateral prefrontal lobe. However, dopaminergic medication has not shown to have a substantial effect on cognitive problems in PD (Fournet et al., 2000; Lewis et al. , 2005,). So far, medical treatment has not been effective enough to prevent PD dementia and restore executive dysfunction.Acetylcholine esterase inhibitors improve cognitive functioning only in some

Furthermore, there is a large body of studies on animals and humans in the literature showing a positive effect of exercise and sports on cognition (Abbott et al., 2004; Colombe et al., 2003a, 2003b, 2006; Laurin et al., 2001; Rolland et al., 2010). Several studies suggest an enhancement of cortical plasticity by exercise. It is assumed that physical exercise mediates increased expression of neurotrophic factors as glial-derived neurotrophic factor (GDNF), basic fibroblast growth factor (FGF-2) , or brain-derived neurotrophic factor (BDNF) (Kleim et al., 2003). BDNF is a member of the neurotrophin family of growth factors vital for trophic support of neurons within both the peripheral and central nervous system. BDNF signals through tyrosine kinase receptor B and through the p75 receptor. Both are expressed by dopamine neurons. Postmortem studies in PD have shown that PD is associated with

Executive functioning was found to be improved by aerobic endurance exercise (Colcombre & Kramer, 2003; Kramer et al., 1999). Motor training was reported to improve cortical plasticity and cortical reorganisation (Nelles 2004; Shepherd 2001). Physical exercise also was found to improve the quality of daily living (Baatile et al., 2000, Reuter et al., 1999) in PD-patients. Furthermore, Hausdorff et al (2005) have shown that higher cognitive functions correlate with gait variability while Ble et al. (2005) reported a close correlation between

Since patients with mild cognitive impairment have a higher risk to develop dementia, intervention at an early stage of cognitive decline is desirable. Patients who complain of cognitive problems suffer more often from cognitive deficits than patients without complaints (Dujardin et al., 2010). Therefore, these patients should be offered neuropsychological testing and treatment. However, according to our experience, it is difficult to convince patients to participate in cognitive training programmes. PD-patients noting declining cognitive performance are often anxious and ashamed of having cognitive problems. They rather deny their problems and try to avoid situations which make their problems obvious to other people. On the other hand the majority of PD-patients is very interested in exercise- and sport-programmes focusing on improvement of motor skills and mobility. Considering the correlation between cognitive function and motor tasks, it might be possible to improve cognitive function by physical training. Furthermore, achievements in cognitive training performed at a writing desk are often difficult to transfer into daily life. Therefore, we have chosen a comprehensive approach and designed a study using a

The aim of the present study was to compare the effect of a multimodal cognitive training regime including paper and pencil tasks combined with transfer tasks and a psychomotor training with a cognitive training performed at a writing desk and a cognitive training consisting of various tasks requiring executive functions combined with transfer tasks.

be involved in handling new situations outside the domain of automatic psychological processes (no reproduction of learned schedules or set behaviours). The theoretic model of the executive system has been modified several times over the years. Crucial contributions to the concept of executive functions came from Norman (1980, 2000), Shallice (1982), Baddeley (1986) and Miller & Cohen (2001). In summary, executive functions involve planning and decision making, influence our handling and the processing of information. Furthermore, they are involved in error corrections or troubleshooting, in situations which require new sequences of actions. Components of the executive systems are attention (focusing on relevant information), selective visual attention, inhibition (inhibition of irrelevant information)(Smith & Jonides, 1999), overcoming of strong habitual responses or resisting temptation (Burgess & Shallice, 1996), task and time management, monitoring and coding of information for processing in the working memory, flexibility, set maintenance and set shifting. The executive system can be viewed as a manager enabling the adaptation of the perceptive, cognitive and motor system to new tasks. Some authors have claimed that cognitive control is the primary function of the prefrontal cortex (Miller & Cohens, 2001). Cognitive control is implemented by increasing gain of sensory or motor neurons that are involved in task or goal relevant actions (Miller & Cohen, 2001).

Patients with impaired executive functions face many difficulties in everyday life. They have a low attention span, difficulties in problem solving and decision making, in dual tasking, in set shifting, in visuoconstructive tasks, in adaptation to new tasks and even in verbal learning and delayed recall. Thus, PD-patients with impairment of executive functions have difficulties in simultaneously driving a car and searching for a street or in preparing a meal for several people. They also have difficulties in keeping appointments. Relatives report that patients avoid difficult tasks and retreat from social life. Executive dysfunctions also affect the social components and the interaction with other people (Smith & Jonides, 1999). Patients are reported of being more irritable and having difficulties in suppressing inadequate behaviour.

It has been proposed that executive dysfunction underlies all manifestations of cognitive impairment in PD (Lewis et al., 2005) as part of the 'frontal-executive brain syndrome' (Godefroy, 2003). In accordance Colman et al. (2009) found that executive dysfunction also underlies the performance of PD-patients on verb production.

Pathophysiologically (Leverenz et al., 2009) cognitive impairment in PD might be either associated with catecholaminergic or indolaminergic neurotransmission or with Alzheimer´s disease (AD) related pathology. While the first form manifests mainly with non amnestic features like impaired EF, and might be correlated with Lewy related pathology in limbic and neocortical regions. The second type of CI manifests in amnestic CI and might derive from processes of AD intersecting with PD. 40% of patients develop dementia (Emre et al., 2004).

### **1.2 Pathophysiology of cognitive impairment in PD**

Decline of cognitive performance in PD might result from rupture of nigro-striatumthalamus cortical circuit interconnecting the striatum to the prefrontal cortex, cholinergic deficits through the differentiation of neurons in the nucleus basalis of Meynert and the pedunculopontine-lateral dorsal tegmental neurons (Calabresi et al., 2006).

In PD the production of dopamine (DA) in the substantia nigra (SN) is decreased. DA is a major neurotransmitter of the basal ganglia, contributing seriously to the development of

be involved in handling new situations outside the domain of automatic psychological processes (no reproduction of learned schedules or set behaviours). The theoretic model of the executive system has been modified several times over the years. Crucial contributions to the concept of executive functions came from Norman (1980, 2000), Shallice (1982), Baddeley (1986) and Miller & Cohen (2001). In summary, executive functions involve planning and decision making, influence our handling and the processing of information. Furthermore, they are involved in error corrections or troubleshooting, in situations which require new sequences of actions. Components of the executive systems are attention (focusing on relevant information), selective visual attention, inhibition (inhibition of irrelevant information)(Smith & Jonides, 1999), overcoming of strong habitual responses or resisting temptation (Burgess & Shallice, 1996), task and time management, monitoring and coding of information for processing in the working memory, flexibility, set maintenance and set shifting. The executive system can be viewed as a manager enabling the adaptation of the perceptive, cognitive and motor system to new tasks. Some authors have claimed that cognitive control is the primary function of the prefrontal cortex (Miller & Cohens, 2001). Cognitive control is implemented by increasing gain of sensory or motor neurons that are

Patients with impaired executive functions face many difficulties in everyday life. They have a low attention span, difficulties in problem solving and decision making, in dual tasking, in set shifting, in visuoconstructive tasks, in adaptation to new tasks and even in verbal learning and delayed recall. Thus, PD-patients with impairment of executive functions have difficulties in simultaneously driving a car and searching for a street or in preparing a meal for several people. They also have difficulties in keeping appointments. Relatives report that patients avoid difficult tasks and retreat from social life. Executive dysfunctions also affect the social components and the interaction with other people (Smith & Jonides, 1999). Patients are reported of being more irritable and having difficulties in suppressing

It has been proposed that executive dysfunction underlies all manifestations of cognitive impairment in PD (Lewis et al., 2005) as part of the 'frontal-executive brain syndrome' (Godefroy, 2003). In accordance Colman et al. (2009) found that executive dysfunction also

Pathophysiologically (Leverenz et al., 2009) cognitive impairment in PD might be either associated with catecholaminergic or indolaminergic neurotransmission or with Alzheimer´s disease (AD) related pathology. While the first form manifests mainly with non amnestic features like impaired EF, and might be correlated with Lewy related pathology in limbic and neocortical regions. The second type of CI manifests in amnestic CI and might derive from processes of AD intersecting with PD. 40% of patients develop dementia (Emre

Decline of cognitive performance in PD might result from rupture of nigro-striatumthalamus cortical circuit interconnecting the striatum to the prefrontal cortex, cholinergic deficits through the differentiation of neurons in the nucleus basalis of Meynert and the

In PD the production of dopamine (DA) in the substantia nigra (SN) is decreased. DA is a major neurotransmitter of the basal ganglia, contributing seriously to the development of

pedunculopontine-lateral dorsal tegmental neurons (Calabresi et al., 2006).

involved in task or goal relevant actions (Miller & Cohen, 2001).

underlies the performance of PD-patients on verb production.

**1.2 Pathophysiology of cognitive impairment in PD** 

inadequate behaviour.

et al., 2004).

frontal-executive dysfunction. Dopaminergic frontal systems play a major role in working memory and executive function (Goldman-Rakic et al., 1992), especially the dorsolateral prefrontal lobe. However, dopaminergic medication has not shown to have a substantial effect on cognitive problems in PD (Fournet et al., 2000; Lewis et al. , 2005,). So far, medical treatment has not been effective enough to prevent PD dementia and restore executive dysfunction.Acetylcholine esterase inhibitors improve cognitive functioning only in some patients.

Furthermore, there is a large body of studies on animals and humans in the literature showing a positive effect of exercise and sports on cognition (Abbott et al., 2004; Colombe et al., 2003a, 2003b, 2006; Laurin et al., 2001; Rolland et al., 2010). Several studies suggest an enhancement of cortical plasticity by exercise. It is assumed that physical exercise mediates increased expression of neurotrophic factors as glial-derived neurotrophic factor (GDNF), basic fibroblast growth factor (FGF-2) , or brain-derived neurotrophic factor (BDNF) (Kleim et al., 2003). BDNF is a member of the neurotrophin family of growth factors vital for trophic support of neurons within both the peripheral and central nervous system. BDNF signals through tyrosine kinase receptor B and through the p75 receptor. Both are expressed by dopamine neurons. Postmortem studies in PD have shown that PD is associated with reduced BDNF levels in the SNC (Howells et al., 2000)

### **1.3 Treatment options for cognitive decline in PD**

Executive functioning was found to be improved by aerobic endurance exercise (Colcombre & Kramer, 2003; Kramer et al., 1999). Motor training was reported to improve cortical plasticity and cortical reorganisation (Nelles 2004; Shepherd 2001). Physical exercise also was found to improve the quality of daily living (Baatile et al., 2000, Reuter et al., 1999) in PD-patients. Furthermore, Hausdorff et al (2005) have shown that higher cognitive functions correlate with gait variability while Ble et al. (2005) reported a close correlation between executive functions and tasks of the lower extremities.

Since patients with mild cognitive impairment have a higher risk to develop dementia, intervention at an early stage of cognitive decline is desirable. Patients who complain of cognitive problems suffer more often from cognitive deficits than patients without complaints (Dujardin et al., 2010). Therefore, these patients should be offered neuropsychological testing and treatment. However, according to our experience, it is difficult to convince patients to participate in cognitive training programmes. PD-patients noting declining cognitive performance are often anxious and ashamed of having cognitive problems. They rather deny their problems and try to avoid situations which make their problems obvious to other people. On the other hand the majority of PD-patients is very interested in exercise- and sport-programmes focusing on improvement of motor skills and mobility. Considering the correlation between cognitive function and motor tasks, it might be possible to improve cognitive function by physical training. Furthermore, achievements in cognitive training performed at a writing desk are often difficult to transfer into daily life. Therefore, we have chosen a comprehensive approach and designed a study using a multimodal cognitive training to improve cognitive functions.

The aim of the present study was to compare the effect of a multimodal cognitive training regime including paper and pencil tasks combined with transfer tasks and a psychomotor training with a cognitive training performed at a writing desk and a cognitive training consisting of various tasks requiring executive functions combined with transfer tasks.

Cognitive Rehabilitation in Parkinson's Disease

Using Neuropsychological Training, Transfer Training and Sports Therapy 261

Fig. 1. Study design: First phase of the study: randomisation into three treatment arms, in-

Posture, postural stability, alternating movements and leg agility were assessed by using the single items of the UPDRS motor scale. The score of each item ranges between 0 to 4 points.

The Goal Attainment Scaling (GAS) allows individualisation of realistic and feasible goals according to patient needs and expectations. All patients identified a task they want to improve by the training programme. In this study, GAS was measured using a 6-point scale, where 3 represented function that is worse than at the start of treatment, 2 was no change, 1 represented some improvement but did not meet the expected goal, 0 represented goal achievement and +1 or +2 represented over-achievement or exceeding the defined

patient treatment; second phase of the study: training at home

therapeutic goal (Royal College of Physicians, 2008).

**2.2.1.2 Goal attainment scale** 

## **2. Methods**

## **2.1 Subjects**

240 patients with idiopathic Parkinson`s disease according to the UK brain bank criteria (Hughes et al., 1991) and complaints about cognitive problems were recruited for the study at the Parkinson clinic Bad Nauheim. Exclusion criteria were severe concomitant diseases, which limit physical performances, and a second neurodegenerative disease. All patients were assessed by a movement disorder specialist. Medical treatment was optimised prior to the study. It was aimed at keeping medication stable during the study. Demographic data included age, body mass index (BMI), duration of disease, weekly sports activity, smoking habits, medication and concomitant diseases (hypertension, chronic obstructive pulmonary disease, thyroid disease, diabetes mellitus, hypercholesterinaemia, osteoarthritis).

## **2.2 Design**

The study was divided into two phases, the first part consisted of a 4-week in-patient stay on a rehabilitation unit with a supervised cognitive training conducted by physiotherapists, occupational therapists and two neuropsychologists.

Patients were randomly allocated to one of the three training groups. Randomisation was conducted by using a computer-generated sequence. All groups received a cognitive training regime using paper and pencil material and a multimedial PC-training. Group A received cognitive training only, while group B took part in a transfer training and a cognitive training. Group C conducted a cognitive training, transfer- and psychomotor training. Patients of group A and B had relaxation training in addition to compensate for the additional training times and occupational therapy without translation training. (Fig. 1) The ethical committee of the Justus-Liebig University has approved the study and all patients gave informed consent. At the baseline visit a medical history was taken and all patients underwent a neurological assessment. Severity of disease was assessed by using the Unified Parkinson`s Disease rating scale (UPDRS).

Demographic data included information about education, profession, family, onset and severity of disease, medication, history of psychosis and impairments in daily living. Patients kept an activity log one week prior to the training programme and one week prior to the third assessment. Sports activities and time spent sitting, doing light, moderate, heavy work were recorded.

### **2.2.1 Scales used for neurological and neuropsychological assessment of PD**

## **2.2.1.1 UPDRS**

For the assessment of the longitudinal course of the disease the Unified Parkinson`s disease rating scale (UPDRS) was applied. The UPDRS is the most frequently used outcome measure in clinical trials in Parkinson`s disease (Fahn et al., 1987). The UPDRS has four subscales: part 1, which has 4 questions on mentation, behaviour and mood (range 0-16 points), part 2, which has 13 questions on activities of daily living (ADL) (range 0-52 points); part 3, which has 14 questions on motor functions (range 0-108 points); and part 4, which has 11 questions on motor and other complications of advanced disease (0-23 points). The UPDRS-Sum score ranges from 0 to 199 points, with a higher score indicating greater problems.

240 patients with idiopathic Parkinson`s disease according to the UK brain bank criteria (Hughes et al., 1991) and complaints about cognitive problems were recruited for the study at the Parkinson clinic Bad Nauheim. Exclusion criteria were severe concomitant diseases, which limit physical performances, and a second neurodegenerative disease. All patients were assessed by a movement disorder specialist. Medical treatment was optimised prior to the study. It was aimed at keeping medication stable during the study. Demographic data included age, body mass index (BMI), duration of disease, weekly sports activity, smoking habits, medication and concomitant diseases (hypertension, chronic obstructive pulmonary disease, thyroid disease, diabetes mellitus,

The study was divided into two phases, the first part consisted of a 4-week in-patient stay on a rehabilitation unit with a supervised cognitive training conducted by physiotherapists,

Patients were randomly allocated to one of the three training groups. Randomisation was conducted by using a computer-generated sequence. All groups received a cognitive training regime using paper and pencil material and a multimedial PC-training. Group A received cognitive training only, while group B took part in a transfer training and a cognitive training. Group C conducted a cognitive training, transfer- and psychomotor training. Patients of group A and B had relaxation training in addition to compensate for the additional training times and occupational therapy without translation training. (Fig. 1) The ethical committee of the Justus-Liebig University has approved the study and all patients gave informed consent. At the baseline visit a medical history was taken and all patients underwent a neurological assessment. Severity of disease was assessed by using the Unified

Demographic data included information about education, profession, family, onset and severity of disease, medication, history of psychosis and impairments in daily living. Patients kept an activity log one week prior to the training programme and one week prior to the third assessment. Sports activities and time spent sitting, doing light, moderate, heavy

For the assessment of the longitudinal course of the disease the Unified Parkinson`s disease rating scale (UPDRS) was applied. The UPDRS is the most frequently used outcome measure in clinical trials in Parkinson`s disease (Fahn et al., 1987). The UPDRS has four subscales: part 1, which has 4 questions on mentation, behaviour and mood (range 0-16 points), part 2, which has 13 questions on activities of daily living (ADL) (range 0-52 points); part 3, which has 14 questions on motor functions (range 0-108 points); and part 4, which has 11 questions on motor and other complications of advanced disease (0-23 points). The UPDRS-Sum score ranges from 0 to 199 points, with a higher score indicating greater

**2.2.1 Scales used for neurological and neuropsychological assessment of PD** 

**2. Methods 2.1 Subjects** 

**2.2 Design** 

hypercholesterinaemia, osteoarthritis).

Parkinson`s Disease rating scale (UPDRS).

work were recorded.

**2.2.1.1 UPDRS** 

problems.

occupational therapists and two neuropsychologists.

Fig. 1. Study design: First phase of the study: randomisation into three treatment arms, inpatient treatment; second phase of the study: training at home

Posture, postural stability, alternating movements and leg agility were assessed by using the single items of the UPDRS motor scale. The score of each item ranges between 0 to 4 points.

#### **2.2.1.2 Goal attainment scale**

The Goal Attainment Scaling (GAS) allows individualisation of realistic and feasible goals according to patient needs and expectations. All patients identified a task they want to improve by the training programme. In this study, GAS was measured using a 6-point scale, where 3 represented function that is worse than at the start of treatment, 2 was no change, 1 represented some improvement but did not meet the expected goal, 0 represented goal achievement and +1 or +2 represented over-achievement or exceeding the defined therapeutic goal (Royal College of Physicians, 2008).

Cognitive Rehabilitation in Parkinson's Disease

*C: Mini Mental test (MMSE)* 

*D: Alters-Konzentrationstest* 

*F: Trail making test* 

*G: MEMO-Test* 

IQ of the patient.

was assumed for less than 20 points.

*E: Paced auditory serial addition test (PASAT)* 

current study the slower speed was used.

performance monitoring, were applied.

ascending order. Time and errors are recorded.

Using Neuropsychological Training, Transfer Training and Sports Therapy 263

Further tests requiring executive and memory functions for assessment of cognitive performance of the PD-patients at baseline, second and final assessment were conducted.

The Mini mental state examination (Folstein et al., 1975) was used as screening tool for dementia. The test assesses orientation, registration, attention, calculation, recall, language, writing and copying. The maximum score is 30 points; high scores indicate good performance. The cut off criteria for an abnormal result are 24 points and below. Dementia

For assessment of attention the Alters-Konzentrationstest (Gatterer et al., 1989) was applied. Patients are asked to mark specific figures out of other figures alike the target. Time to complete

The Paced Auditory Serial Addition Test (Gronwall et al., 1977) assesses auditory information processing speed and flexibility and ability to calculate. Single digits are presented either every 3 seconds (trial 1) or every 2 seconds (trial 2). The patient has to add each new digit to the one immediately prior to it. The maximal possible score adds up to 60, the individual test score is equal to the total number of correct sums in each trial. In the

The trail making test (Reitan, 1958) assesses visual attention and task switching. Numbers from 1 to 30 are spread over a sheet, the patient is asked to connect the numbers in

The MEMO-Test (Schaaf et al., 1994) assesses short-term verbal memory. Ten words are read to the patient. Five trials are performed. Patients are asked to repeat the words immediately, after each trial the words left out are read again. The following assessments are performed: UR: all words produced by short-term memory, ALZS: all words recalled from long term memory, UR + ALZS: all words recalled; KALZS: all words permanently recalled from long term memory; NKALZS: all words inconsistently recalled from long term memory; LZS: all

The BADS (Wilson et al., 1998) is a battery of tests assessing executive function and comprises several subtests. In this study the Rule Shift Card test was applied to identify perseverative tendencies and mental flexibility, the Zoo Map test assessing was used the ability to plan and the Modified Six Element test, a test of planning, task scheduling and

The MWT-B (Lehrl, 1989) serves as a control factor. A list consisting of 37 rows with 5 words is shown to the patient. Only one of the five words has a real meaning the others are fantasy words. The patients should mark the word with the meaning. The correct answers are added up to the sum-score. Each score is related to a standard score (z) which estimates the

words recalled from long term memory; delayed recall after 15 min..

*H: Behavioural assessment of the dysexecutive syndrome (BADS)* 

*I : Mehrfach-Wortschatz-Test (MWT-B ) Multiple choice word test* 

the test, number of correctly marked figures, number and type of mistakes are recorded.

## **2.2.1.3 Neuropsychological tests:**

For neuropsychological assessment all patients underwent a detailed cognitive test battery at the beginning of the study including the ADAS-Cog subscale and the SCOPA-COG as outcome measures.

#### *A: ADAS-Cog (Alzheimer Disease Assessment Scale-Cognition)*

Although the ADAS-Cog is not a specific test for cognitive impairment in Parkinson`s Disease the scale was chosen as primary outcome measure in the current study, because it was the primary outcome measure in earlier trials assessing effects of medication on cognitive function in PD (Tab.1).


Table 1. Structure of ADAS-Cog scale.

The ADAS-Cog scale was the primary outcome measure in many clinical trials (Rosen WG et al., 1984). The conceptual framework underlying the ADAS-Cog identifies three reproducible factors: memory, language, praxis (Talwalker et al., 1996). The ADAS-Cog score ranges in total from 0 to 70 points with higher scores indicating greater impairment. Language ability is tested by naming objects and fingers, observer rated comprehension of spoken language, expressive language and word finding (range 0-25 points; memoryis tested by recall of instructions, word list recall and recognition (range 0-27 points), test of praxis (range 0-10 points) consists of constructional praxis (copying geometric figures) and ideational praxis (preparing envelope to send to oneself), orientation is assessed for time and space orientation (range 0-8points).

*B: SCOPA-COG(Scales for Outcome of Parkinson`s disease-Cognition)* 

The SCOPA-COG is an instrument which was designed to assess the specific cognitive deficits found in Parkinson`s disease (Marinus et al., 2003). The scale consisting of 10 items covers the domains: memory and recall (verbal recall, digit span backward, indicate cubes), attention (counting backward, months backward), executive function (fist-edge-palm, semantic fluency, dice), visual-spatial functions (assembly pattern) and memory (delayed recall). The score ranges from 0 to 43 points with higher scores reflecting better performance. Further tests requiring executive and memory functions for assessment of cognitive performance of the PD-patients at baseline, second and final assessment were conducted.

#### *C: Mini Mental test (MMSE)*

262 Diagnostics and Rehabilitation of Parkinson's Disease

For neuropsychological assessment all patients underwent a detailed cognitive test battery at the beginning of the study including the ADAS-Cog subscale and the SCOPA-COG as

Although the ADAS-Cog is not a specific test for cognitive impairment in Parkinson`s Disease the scale was chosen as primary outcome measure in the current study, because it was the primary outcome measure in earlier trials assessing effects of medication on

No . Task Characteristics Score 1 Word recall The recall task of frequent, easily to imagine words 0-10p 2 Naming Naming of 12 presented objects and fingers on a hand 0-5p 3 Commands Task of understanding and fulfilling 0-5p 4 Constructional Drawing 4 geometric forms using praxis a pattern 0-5p 5 Ideational The task of ability to perform praxis a familiar but complex

6 Orientation Assessment of time and space orientation 0-8 p

9 Spoken language Assessment of the quality of patient.s age ability speech 0-5p

11 Comprehension The patients ability to understand the spoken speech 0-5p

The ADAS-Cog scale was the primary outcome measure in many clinical trials (Rosen WG et al., 1984). The conceptual framework underlying the ADAS-Cog identifies three reproducible factors: memory, language, praxis (Talwalker et al., 1996). The ADAS-Cog score ranges in total from 0 to 70 points with higher scores indicating greater impairment. Language ability is tested by naming objects and fingers, observer rated comprehension of spoken language, expressive language and word finding (range 0-25 points; memoryis tested by recall of instructions, word list recall and recognition (range 0-27 points), test of praxis (range 0-10 points) consists of constructional praxis (copying geometric figures) and ideational praxis (preparing envelope to send to oneself), orientation is assessed for time

The SCOPA-COG is an instrument which was designed to assess the specific cognitive deficits found in Parkinson`s disease (Marinus et al., 2003). The scale consisting of 10 items covers the domains: memory and recall (verbal recall, digit span backward, indicate cubes), attention (counting backward, months backward), executive function (fist-edge-palm, semantic fluency, dice), visual-spatial functions (assembly pattern) and memory (delayed recall). The score ranges from 0 to 43 points with higher scores reflecting better performance.

7 Word recognition The task of discriminating new words from the already

8 Instructions Ability to remember instructions from remembering the previous recognition task

10 Word finding Assessment of patients ability to difficulty communicate

**2.2.1.3 Neuropsychological tests:** 

cognitive function in PD (Tab.1).

*A: ADAS-Cog (Alzheimer Disease Assessment Scale-Cognition)* 

sequence of actions

presented ones

verbally

*B: SCOPA-COG(Scales for Outcome of Parkinson`s disease-Cognition)* 

Table 1. Structure of ADAS-Cog scale.

and space orientation (range 0-8points).

outcome measures.

The Mini mental state examination (Folstein et al., 1975) was used as screening tool for dementia. The test assesses orientation, registration, attention, calculation, recall, language, writing and copying. The maximum score is 30 points; high scores indicate good performance. The cut off criteria for an abnormal result are 24 points and below. Dementia was assumed for less than 20 points.

### *D: Alters-Konzentrationstest*

For assessment of attention the Alters-Konzentrationstest (Gatterer et al., 1989) was applied. Patients are asked to mark specific figures out of other figures alike the target. Time to complete the test, number of correctly marked figures, number and type of mistakes are recorded.

#### *E: Paced auditory serial addition test (PASAT)*

The Paced Auditory Serial Addition Test (Gronwall et al., 1977) assesses auditory information processing speed and flexibility and ability to calculate. Single digits are presented either every 3 seconds (trial 1) or every 2 seconds (trial 2). The patient has to add each new digit to the one immediately prior to it. The maximal possible score adds up to 60, the individual test score is equal to the total number of correct sums in each trial. In the current study the slower speed was used.

#### *F: Trail making test*

0-5p

0-12p

0-5p

0-5p

The trail making test (Reitan, 1958) assesses visual attention and task switching. Numbers from 1 to 30 are spread over a sheet, the patient is asked to connect the numbers in ascending order. Time and errors are recorded.

#### *G: MEMO-Test*

The MEMO-Test (Schaaf et al., 1994) assesses short-term verbal memory. Ten words are read to the patient. Five trials are performed. Patients are asked to repeat the words immediately, after each trial the words left out are read again. The following assessments are performed: UR: all words produced by short-term memory, ALZS: all words recalled from long term memory, UR + ALZS: all words recalled; KALZS: all words permanently recalled from long term memory; NKALZS: all words inconsistently recalled from long term memory; LZS: all words recalled from long term memory; delayed recall after 15 min..

#### *H: Behavioural assessment of the dysexecutive syndrome (BADS)*

The BADS (Wilson et al., 1998) is a battery of tests assessing executive function and comprises several subtests. In this study the Rule Shift Card test was applied to identify perseverative tendencies and mental flexibility, the Zoo Map test assessing was used the ability to plan and the Modified Six Element test, a test of planning, task scheduling and performance monitoring, were applied.

#### *I : Mehrfach-Wortschatz-Test (MWT-B ) Multiple choice word test*

The MWT-B (Lehrl, 1989) serves as a control factor. A list consisting of 37 rows with 5 words is shown to the patient. Only one of the five words has a real meaning the others are fantasy words. The patients should mark the word with the meaning. The correct answers are added up to the sum-score. Each score is related to a standard score (z) which estimates the IQ of the patient.

Cognitive Rehabilitation in Parkinson's Disease

min. Patients received at least 10 sessions of transfer training.

*B: Transfer tasks* 

*C: Motor training* 

**2.2.3 Education of caregivers** 

physiotherapist and a psychologist. **2.2.3.1 Phase II Continuing training** 

**2.2.3.2 Evaluation of the training** 

long-term effect.

Using Neuropsychological Training, Transfer Training and Sports Therapy 265

The aim of the training was to support patients to manage better their daily life and to become more self-confident. Therefore, patients were asked to practise competence in tasks of daily routines. The transfer training programme was composed according to the baseline test results. Special preferences of the patients were considered. The transfer training included a training of concentration, use of mnemonics, strategy (planning), navigational skills, impulse control, decision processes, listening training and memory, behaviour, calculating, handling of money, summarising of articles read or heard and decision making. Typical tasks were to find the way to the supermarket or to prepare a meal, to go to the bank, pay a bill and to use mnemonics. For better evaluation of the training tasks were allocated to different categories: concentration, strategy, improvement of orientation, planning, use of mnemonic devices. The training took place 3 times a week each lasted 90

Group C performed a motor training resembling psychomotor training lessons applied in children. Psychomotor training (Golubović et al., 2011; Oswald et al., 1996) reflects a relationship between cognitive functions and physical movements. It includes training of coordination, strength, speed, perception and orientation. Patients should discover their body and their feelings. The therapeutic approach is multidimensional and based on individual capabilities and needs. The aim of the training was to practise motor sequences, dual tasking (walking and bouncing or throwing a ball, orientation in a space, walking through a parcours to improve anticipation. In summer the training was conducted partly outdoors with inclusion of Nordic walking. Thus, the training combines aerobic and psychomotor components. The training included at least 10, maximal 12 sessions each lasting 60 minutes.

A long lasting training effect depends on continuing training. Thus, cognitive training and exercises need to be adapted to the home environment. Consequently, the caregivers most often the patients` family were included in the programme. The education for the caregivers consisted of 5 modules (information about Parkinson`s disease, psychological aspects and the role of a caregiver, information about help aids, information on care instructions, assessment of individual problems, support in cognitive (all groups) and transfer training (group A and B), NW and psychomotor training). Course instructors were a specialist nurse,

Corresponding to the allocation to the training groups patients got lessons for the cognitive training, transfer training and physical exercises for the training at home. Caregivers were advised how to organise the training but the hospital staff did not organise the training at home.

All patients were tested using a neuropsychological test battery: prior to the training and prior to discharge to assess the short term effect and 3 months after the training to assess the

Caregivers were asked regarding their own well being and regarding the cognitive competence of the patients in daily living. Patients and caregivers kept a diary to record

training lesions. The diaries were collected and analysed at the 3rd. assessment.

## *K: Hospital anxiety and depression scale*

The Hospital anxiety and depression scale (Zigmond & Snaith, 1983) was applied for exclusion of significant depression and anxiety. The scale consists of two subscales, an anxiety scale and a depression scale ranging from 0 to 21 points respectively. Patients are asked to choose one response from the four given for each question. Patients were strongly encouraged to respond promptly. Questions related to anxiety are marked with A and to depression with D. Depression and anxiety are scored separately. On each scale 0 to 7 points indicate a normal, 8 to 10 points a borderline abnormal and 11 or more points an abnormal result.

## *L: State Trait anxiety inventary (STAI)*

The STAI scales (Spielberger et al., 1970) assess the trait anxiety (X2) and the anxiety in a specific situation (X1). Each scale consists of 20 items. Both scales present the answers on a 4 point Likert scale. Both scales range from 20 to 80 points with high scores indicating a high anxiety level.

## *M: Parkinson`s disease Questionnaire 39 (PDQ 39)*

For assessment of health related quality of life patients filled in the PDQ 39 (Jenkinson et al., 1997, Peto et al., 1995). It consists of 8 subscales: subscale 1 mobility (max. 40 points); subscale 2 activities of daily living (max. 24 points), subscale 3 emotional well being (max. 24 points), subscale 4 stigma (max 16 points), subscale 5 social support (max 12 points), subscale 6 cognition (max.16 points), subscale 7 communication (max.12 points), subscale 8 bodily discomfort (max. 12 points) . The sum score of raw data ranges from 0 to 156 points, with high scores indicating lower health related quality of life. For better comparison of the results raw data were transformed and expressed in percentages of maximal possible sum score.

## **2.2.2 Training programmes**

## *A: Cognitive training*

The cognitive training content was individually tailored to patients` requirements based on the results of the baseline tests. Four individual (one to one) lessons took place each week each lasting 60 min. All patients received at least 14 cognitive training sessions.

The training included training of attention, concentration, biographical work, reasoning, memory, working memory, social rules, anticipation, cognitive information speed, prospective memory, cognitive estimation, problem solving, sequencing and planning, associations and coping with disease.

For the training programme a set of tasks requiring executive and memory functions were chosen from a variety of specific tests. Executive tasks of the BADS, which were not used for baseline tests were included in the training. Simple patterns of the "Raven`s Progressive Matrices" were used to establish problem solving strategies in the patients. Picture arrangement tasks, picture completion tasks, block design, and object assembly were adapted from the "Wechsler Intelligence test for children". For improvement of verbal fluency patients were encouraged to tell short stories or discuss short text-passages. Photos were used for training of working memory. Tasks including visual search, rule finding were practised by using a PC-based programme. The training methods were designed to improve the various cognitive deficits, diagnosed at baseline and focused on the executive functions. Task difficulty was adapted to the individual performance level of the patients.

#### *B: Transfer tasks*

264 Diagnostics and Rehabilitation of Parkinson's Disease

The Hospital anxiety and depression scale (Zigmond & Snaith, 1983) was applied for exclusion of significant depression and anxiety. The scale consists of two subscales, an anxiety scale and a depression scale ranging from 0 to 21 points respectively. Patients are asked to choose one response from the four given for each question. Patients were strongly encouraged to respond promptly. Questions related to anxiety are marked with A and to depression with D. Depression and anxiety are scored separately. On each scale 0 to 7 points indicate a normal, 8

The STAI scales (Spielberger et al., 1970) assess the trait anxiety (X2) and the anxiety in a specific situation (X1). Each scale consists of 20 items. Both scales present the answers on a 4 point Likert scale. Both scales range from 20 to 80 points with high scores indicating a high

For assessment of health related quality of life patients filled in the PDQ 39 (Jenkinson et al., 1997, Peto et al., 1995). It consists of 8 subscales: subscale 1 mobility (max. 40 points); subscale 2 activities of daily living (max. 24 points), subscale 3 emotional well being (max. 24 points), subscale 4 stigma (max 16 points), subscale 5 social support (max 12 points), subscale 6 cognition (max.16 points), subscale 7 communication (max.12 points), subscale 8 bodily discomfort (max. 12 points) . The sum score of raw data ranges from 0 to 156 points, with high scores indicating lower health related quality of life. For better comparison of the results raw data were transformed and expressed in percentages of maximal possible sum

The cognitive training content was individually tailored to patients` requirements based on the results of the baseline tests. Four individual (one to one) lessons took place each week

The training included training of attention, concentration, biographical work, reasoning, memory, working memory, social rules, anticipation, cognitive information speed, prospective memory, cognitive estimation, problem solving, sequencing and planning,

For the training programme a set of tasks requiring executive and memory functions were chosen from a variety of specific tests. Executive tasks of the BADS, which were not used for baseline tests were included in the training. Simple patterns of the "Raven`s Progressive Matrices" were used to establish problem solving strategies in the patients. Picture arrangement tasks, picture completion tasks, block design, and object assembly were adapted from the "Wechsler Intelligence test for children". For improvement of verbal fluency patients were encouraged to tell short stories or discuss short text-passages. Photos were used for training of working memory. Tasks including visual search, rule finding were practised by using a PC-based programme. The training methods were designed to improve the various cognitive deficits, diagnosed at baseline and focused on the executive functions.

each lasting 60 min. All patients received at least 14 cognitive training sessions.

Task difficulty was adapted to the individual performance level of the patients.

to 10 points a borderline abnormal and 11 or more points an abnormal result.

*K: Hospital anxiety and depression scale* 

*L: State Trait anxiety inventary (STAI)* 

*M: Parkinson`s disease Questionnaire 39 (PDQ 39)* 

anxiety level.

score.

**2.2.2 Training programmes** 

associations and coping with disease.

*A: Cognitive training* 

The aim of the training was to support patients to manage better their daily life and to become more self-confident. Therefore, patients were asked to practise competence in tasks of daily routines. The transfer training programme was composed according to the baseline test results. Special preferences of the patients were considered. The transfer training included a training of concentration, use of mnemonics, strategy (planning), navigational skills, impulse control, decision processes, listening training and memory, behaviour, calculating, handling of money, summarising of articles read or heard and decision making. Typical tasks were to find the way to the supermarket or to prepare a meal, to go to the bank, pay a bill and to use mnemonics. For better evaluation of the training tasks were allocated to different categories: concentration, strategy, improvement of orientation, planning, use of mnemonic devices. The training took place 3 times a week each lasted 90 min. Patients received at least 10 sessions of transfer training.

#### *C: Motor training*

Group C performed a motor training resembling psychomotor training lessons applied in children. Psychomotor training (Golubović et al., 2011; Oswald et al., 1996) reflects a relationship between cognitive functions and physical movements. It includes training of coordination, strength, speed, perception and orientation. Patients should discover their body and their feelings. The therapeutic approach is multidimensional and based on individual capabilities and needs. The aim of the training was to practise motor sequences, dual tasking (walking and bouncing or throwing a ball, orientation in a space, walking through a parcours to improve anticipation. In summer the training was conducted partly outdoors with inclusion of Nordic walking. Thus, the training combines aerobic and psychomotor components. The training included at least 10, maximal 12 sessions each lasting 60 minutes.

## **2.2.3 Education of caregivers**

A long lasting training effect depends on continuing training. Thus, cognitive training and exercises need to be adapted to the home environment. Consequently, the caregivers most often the patients` family were included in the programme. The education for the caregivers consisted of 5 modules (information about Parkinson`s disease, psychological aspects and the role of a caregiver, information about help aids, information on care instructions, assessment of individual problems, support in cognitive (all groups) and transfer training (group A and B), NW and psychomotor training). Course instructors were a specialist nurse, physiotherapist and a psychologist.

#### **2.2.3.1 Phase II Continuing training**

Corresponding to the allocation to the training groups patients got lessons for the cognitive training, transfer training and physical exercises for the training at home. Caregivers were advised how to organise the training but the hospital staff did not organise the training at home.

#### **2.2.3.2 Evaluation of the training**

All patients were tested using a neuropsychological test battery: prior to the training and prior to discharge to assess the short term effect and 3 months after the training to assess the long-term effect.

Caregivers were asked regarding their own well being and regarding the cognitive competence of the patients in daily living. Patients and caregivers kept a diary to record training lesions. The diaries were collected and analysed at the 3rd. assessment.

Cognitive Rehabilitation in Parkinson's Disease

in group B and 76 patients in group C.

**3. Results** 

was successful.

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Percentage

of Training [%]

*A: Cognitive Training:* 

Using Neuropsychological Training, Transfer Training and Sports Therapy 267

In total 222 patients (97.1%) completed the programme, 71 patients in group A, 75 patients

The patients were on average 64 ± 4 years old and c. 8 years diagnosed with PD. The patients did not differ significantly in demographic data (Tab. 2). There was no difference in

Patients of group A reported to perform 8.5 ± 2.6 hours very hard work per week, while patients of group B and C reported of 9.2 ± 2.8 and 9.8 ± 2.1hours very hard work respectively.

The neuropsychological baseline assessment did not reveal any differences between the groups. The multiple choice word test (MWT-B) was conducted as a measure for premorbid intelligence, the groups did not differ significantly, either. Thus the randomisation process

The groups differed in time of practising concentration tasks and sequencing and planning tasks (F = 3.60; df = 2; p < 0.03). Group A and B spent 12% respectively 15% of the training with concentration training, group c only 8%. In contrast group C spent 22% of the training time with sequencing and planning tasks while group A 16% and group B 17%. The other

> soc. Ru cog. inf. speed anticipation assoc. pros.mem cog.est PS,Pl,Se reasoning w orking memory memory concentration attention

prospecive memory

**3.1 General results, demographic data and accomplishment of the training** 

Group A managed 15.2 ± 4.5, Group B 14.9 ± 5 and Group C 15.1 ± 5.5 hard work.

PD specific impairment and in the progress of PD between the groups. The physical activity of the patients did not differ significantly either.

training areas did not differ significantly between the groups (Fig. 2).

123

Soc.Ru= social rules; cog.inf.speed= cognitive information speed, assoc.=association; pros.mem= prospetive Fig. 2. Group C spent more time of the training with sequencing and planning tasks, while

Group A Group B Group C

group A spent more time with concentration tasks.

## **2.3 Statistical analysis**

Statistical analysis was conducted using IBM SPSS Statistics 18.0 (IBM, Somers, USA) statistical software. Formal power analysis was performed prior to the study. The power analysis was based on an improvement of the ADAS-Cog by 3 points. The results indicated that a sample size of 60 subjects per group was sufficient. Since comprehensive training programmes including several assessments imply drop outs, a drop out rate of 20% was taken into account. Demographic data on ordinal level were analysed by using a nonparametric test (Kruskall-Wallis). The Kruskall-Wallis test was also applied for the analysis of depression and the BADS subscales. Continuous data were analysed by using a One – way-ANOVA. The repeated measure analysis provides information about "between and within subjects" effects. Within subject effects give information about training effects over the assessment period. Linear trends were extracted by orthogonal polynomials and analysed for days and for trials (Memo test). Linear trends showed if there was a systematic change of training effects over time. The interaction between groups and the linear trend of days yielded information about difference in the rate of improvement between groups. The between subject factor compared the overall treatment effect between the groups. Post hoc analysis was done using Bonferroni tests. Parametric data were tested for normal distribution by using the Kolmogorov-Smirnov test. Significance level was set at 0.05.


Table 2. Demographic data

## **3. Results**

266 Diagnostics and Rehabilitation of Parkinson's Disease

Statistical analysis was conducted using IBM SPSS Statistics 18.0 (IBM, Somers, USA) statistical software. Formal power analysis was performed prior to the study. The power analysis was based on an improvement of the ADAS-Cog by 3 points. The results indicated that a sample size of 60 subjects per group was sufficient. Since comprehensive training programmes including several assessments imply drop outs, a drop out rate of 20% was taken into account. Demographic data on ordinal level were analysed by using a nonparametric test (Kruskall-Wallis). The Kruskall-Wallis test was also applied for the analysis of depression and the BADS subscales. Continuous data were analysed by using a One – way-ANOVA. The repeated measure analysis provides information about "between and within subjects" effects. Within subject effects give information about training effects over the assessment period. Linear trends were extracted by orthogonal polynomials and analysed for days and for trials (Memo test). Linear trends showed if there was a systematic change of training effects over time. The interaction between groups and the linear trend of days yielded information about difference in the rate of improvement between groups. The between subject factor compared the overall treatment effect between the groups. Post hoc analysis was done using Bonferroni tests. Parametric data were tested for normal distribution by using the Kolmogorov-Smirnov test. Significance level was set at 0.05.

gender F=35 M = 36 F = 36 M = 39 F = 36 M = 40

II 8 6 10 III 55 59 58 IV 9 10 8

L-Dopa Yes = 68 Yes = 64 Yes = 59 Dopamine agonist Yes = 53 Yes = 56 Yes = 59 MA0 inhibitor N = 43 N = 38 N = 43 COMT inhibitor N = 33 N = 31 N = 34 Antidepressants N = 7 N = 8 N = 8 Neuroleptic drugs N = 5 N = 8 N = 7

Duration of PD (months) 98± 8 95 ± 9 100 ± 6

Formal education (years) 10 ± 1.2 11 ± 0.6 11 ± 1.0

BMI 27.5 ± 4 26.8 ± 7 27.2 ± 3

Own = 40 Renting = 32

Smoking Yes = 7 No = 65 Yes = 10 No = 65 Yes = 9 No = 67 Sports activities (min) 155 ± 17 163 ± 25 147 ± 17

> Coronary Heart disease N = 7 N = 6 N = 8 Hypertension N = 32 N = 33 N = 36 Diabetes mellitus N = 7 N = 10 N = 8 COPD N = 5 N = 6 N = 9 Thyroid disease N = 12 N = 10 N = 11 Hypercholesteriaemia N = 36 N = 32 N = 27 Osteoarthritis N = 27 N = 31 N = 34

Group A (N = 71) Group B (N = 75) Group C (N = 76)

m = 58 s = 9 p = 5 m = 61 s = 11 p = 3 m = 63 s = 9 p = 4

Own = 43 Renting = 32

Own = 40 Renting= 36

**2.3 Statistical analysis** 

Stage

(Hoehn & Yahr)

Medication

Home

Marital status m = married, s = single, c = partner

Table 2. Demographic data

(own home, renting)

Comorbidity

## **3.1 General results, demographic data and accomplishment of the training**

In total 222 patients (97.1%) completed the programme, 71 patients in group A, 75 patients in group B and 76 patients in group C.

The patients were on average 64 ± 4 years old and c. 8 years diagnosed with PD. The patients did not differ significantly in demographic data (Tab. 2). There was no difference in PD specific impairment and in the progress of PD between the groups.

The physical activity of the patients did not differ significantly either.

Patients of group A reported to perform 8.5 ± 2.6 hours very hard work per week, while patients of group B and C reported of 9.2 ± 2.8 and 9.8 ± 2.1hours very hard work respectively. Group A managed 15.2 ± 4.5, Group B 14.9 ± 5 and Group C 15.1 ± 5.5 hard work.

The neuropsychological baseline assessment did not reveal any differences between the groups. The multiple choice word test (MWT-B) was conducted as a measure for premorbid intelligence, the groups did not differ significantly, either. Thus the randomisation process was successful.

#### *A: Cognitive Training:*

The groups differed in time of practising concentration tasks and sequencing and planning tasks (F = 3.60; df = 2; p < 0.03). Group A and B spent 12% respectively 15% of the training with concentration training, group c only 8%. In contrast group C spent 22% of the training time with sequencing and planning tasks while group A 16% and group B 17%. The other training areas did not differ significantly between the groups (Fig. 2).

Soc.Ru= social rules; cog.inf.speed= cognitive information speed, assoc.=association; pros.mem= prospetive prospecive memory

Fig. 2. Group C spent more time of the training with sequencing and planning tasks, while group A spent more time with concentration tasks.

Cognitive Rehabilitation in Parkinson's Disease

**3.2 Neuropsychological results** 

home.

MMST

ADAS-Cog

SCOPA-COG

TMT

AKT (time)

BADS Zoo (profile)

BADS instruction

*F: Assessment of the training programme by the caregivers* 

Using Neuropsychological Training, Transfer Training and Sports Therapy 269

Caregivers felt more relaxed and competent to handle difficult situation, while patients accepted the guidance of their care, felt more confident and thought that the caregivers were more understanding. Both, patients and caregivers felt competent to continue the training at

Test Baseline T1 T2 Significance

Group B 27.6 ± 1.89 n.d. 27.1 ± 1.7 n.s.

Group B 21.37 ± 4.11 18.33 ± 3.67 18.5 ± 4.2 p< 0.001

Group B 29.68 ± 2.87 31.32 ± 3.24 30.71 ± 2.9 p< 0.001

Group B 34.19 ± 15.6 32.0 ± 14.5 31.4 ± 13.5 p < 0.001

Group B 44.96 ± 16.5 41.67 ± 16.4 42.5 ± 15.3 n.s.

Group A 2.5 ± 0.95 3.0 ± 1.2 2.4 ± 1.2 T1:

Group A 2.8 ± 1.3 3.3 ± 1.1 2.9 ± 0.8 T1:

Group A 27.36 ± 1.76 n.d. 26.4 ± 1.8

Group C 28.14 ± 1.81 n.d. 28.5 ± 1.8

Group A 21.51 ± 2.27 20.81 ± 2.77 20.5 ± 3.6

Group C 22.92 ± 4.02 17.98 ± 2.76 17.4 ± 2.5

Group A 29.07 ± 3.8 27.21 ± 3.6 26.86 ± 3.32

Group C 31.83 ± 3.21 39.15 ± 2.9 39.29 ± 2.72

Group A 34.23 ± 16.87 31.7 ± 13.79 32.6 ± 14.5

Group C 33.98 ± 15.8 26.2 ± 13.4 23.12 ± 9.8

Group A 40.76 ± 15.2 42.85 ± 15.2 42.3 ± 14.2

Group C 41.36 ± 15.23 40.98 ± 16.3 41.3 ± 16.3

Group B 2.4 ± 0.9 2.8 ± 1.1 2.6 ± 1.1 Group C 2.6 ± 0.98 3.54 ± 0.82 3.43 ± 1.0

Group B 2.6 ± 1.3 2.9 ± 1.2 3.2 ± 1.1 Group C 2.7 ± 1.1 3.5 ± 1.1 3.8 ± 0.9 between groups

Chi-square: 49.31; p < 0.001

Chi-square: 14.421; p > 0.001

Chi-square: 7.1;

Chi-square: 9.1 p

p < 0.03 T2:

> 0.01

T2:

## *B: Transfer training*

Fig. 3. There was no difference in the quantity and quality of transfer training between group B and C.

#### *C: Motor training*

Patients had many difficulties to cope with the tasks. They struggled to find strategies to solve the tasks on their own. The type of tasks and exercises were new to the majority of patients. The character of the tasks challenged the patients since PD-patients have both, deficits in proprioception and in perception of stimuli. The lessons were conducted as individual lessons. It was not possible to conduct group lessons. About 40% of the training took place outdoors, 60% in the gym.

#### *D: Training at home*

60% of patients of group A continued practising cognitive tasks 3 times a week, while 40% conducted the training only once or twice per week. All patients of group B tried to continue the transfer tasks learnt during the rehabilitation but further assessment showed that only 60% performed transfer tasks following a regular schedule. 90% of the patients practised cognitive tasks 3 times a week. Patients of Group C spent more time practising cognitive and transfer tasks than the other both groups. Patients conducted the physical training programme most often together with their spouses and very regularly.

### *E: Assessment of the training by the patients*

Patients were asked to evaluate the training programme. Patients of group A felt that the cognitive training was arduous at times. Some patients perceived the training as stressful. Patients of group B and C were asked to compare the training programmes. Patients of group C preferred the motor training to transfer training and cognitive pencil and paper tasks. 80% of patients judged the training as strenuous and felt sometimes exhausted. 30% of patients reported of being frustrated at times but did not ask for help or further explanations.

#### *F: Assessment of the training programme by the caregivers*

Caregivers felt more relaxed and competent to handle difficult situation, while patients accepted the guidance of their care, felt more confident and thought that the caregivers were more understanding. Both, patients and caregivers felt competent to continue the training at home.


268 Diagnostics and Rehabilitation of Parkinson's Disease

set shifting summary calculation Decision making slow ness navigational skills planning, strategy mnemonics concentration

1 2

Group B Group C

Fig. 3. There was no difference in the quantity and quality of transfer training between

Patients had many difficulties to cope with the tasks. They struggled to find strategies to solve the tasks on their own. The type of tasks and exercises were new to the majority of patients. The character of the tasks challenged the patients since PD-patients have both, deficits in proprioception and in perception of stimuli. The lessons were conducted as individual lessons. It was not possible to conduct group lessons. About 40% of the training

60% of patients of group A continued practising cognitive tasks 3 times a week, while 40% conducted the training only once or twice per week. All patients of group B tried to continue the transfer tasks learnt during the rehabilitation but further assessment showed that only 60% performed transfer tasks following a regular schedule. 90% of the patients practised cognitive tasks 3 times a week. Patients of Group C spent more time practising cognitive and transfer tasks than the other both groups. Patients conducted the physical training

Patients were asked to evaluate the training programme. Patients of group A felt that the cognitive training was arduous at times. Some patients perceived the training as stressful. Patients of group B and C were asked to compare the training programmes. Patients of group C preferred the motor training to transfer training and cognitive pencil and paper tasks. 80% of patients judged the training as strenuous and felt sometimes exhausted. 30% of patients reported of being frustrated at times but did not ask for help or further

programme most often together with their spouses and very regularly.

*B: Transfer training* 

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

took place outdoors, 60% in the gym.

*E: Assessment of the training by the patients* 

Percentage

group B and C. *C: Motor training* 

*D: Training at home* 

explanations.

of training

 [%]


Cognitive Rehabilitation in Parkinson's Disease

shown in table 4 and 5.

*D: Concentration* 

*E: Information processing* 

points

*C: GAS* 

Using Neuropsychological Training, Transfer Training and Sports Therapy 271

0.001). Group C improved most resulting in a significant difference between the groups (F[2, 220] = 31.4, df = 2; p < 0.001). Since the slopes of the improvements differed between the groups, a significant interaction between days and groups occurred (F [2, 220] = 65.63; p< 0.001). Post hoc tests revealed a significant difference between all groups (p< 0.001). Patients of Group A reached 28.8 ± 3.7 points, Group B 30.3 ± 2.7 points and group C 37.6 ± 3.4 points. After completion of the in-patient training programme 31% of group A, 64% of group B and 88% of group C had shown a significant improvement on the SCOPA-COG, six months later at the final assessment 70% of patients of group A, 80% of patients of group B

**\*\*\* \*\*\***

Group A Group B Group C

**\*\*\* p < 0.001**

and 94% of patients of group C had been able to keep their level of performance.

123

The GAS was performed on the final assessment. Group C reached more often the main goal than the other groups (Chi-square: 57.1; p < 0.001). The detailed analysis of the results is

The main cognitive impairments reported by the patients could be attributed to the following domains: dual tasking, planning of complex and sequential tasks, decision making, rule recognition and rule shifting problems with delayed recall, difficulties in finding misplaced items. The patients based the selection of the goals on their individual main impairment. Table 4 shows the goals patients had chosen and if they were obtained. More patients of group A compared to group B and C did not obtain the chosen goal or deteriorated compared to baseline while 27.6% of patients of group C obtained the goal and

In the Alterskonzentrationstest (AKT) no difference between the groups was detected. Patients did not differ in attention span neither at baseline nor at the final assessment.

In the ZVT no difference between the groups was detected at baseline assessment. The time to complete the test decreased in all groups after the training (Flin [2,220] = 17.71; p< 0.001).

Baseline T1 T2

Fig. 4. Group C improved significantly more than group A and B

39.4% exceeded the expectations mildly and 7.6% substantially.


Table 3. Summary of neuropsychological test results

## **3.2.1 Primary outcome measure**

#### *A: ADAS-Cog*

All groups improved on the ADAS-Cog significantly shown by a significant linear trend (F lin [1, 220] = 150; p< 0.001. Group C improved most indicated by a significant interaction between groups and days (Fgroups x days [1, 220] = 27.26; p < 0.001) and a significant group difference (F [2,220] = 7.7, p < 0.001). Further analysis showed that 78% of the patients showed some improvement at the second assessment, 51% of patients of group A, 85% of patients of group B and 96% of patients of group C. 50% of the patients reached a reduction of the ADAS-Cog score of 3 or more points, 18% of group A, 54% of group B and 76% of group C. Six months after discharge of the rehabilitation unit 35% of patients (50% of patients of group A, 31% of patients of group B and 28% of group C) showed a deterioration compared to the assessment at the end of the in-patient training programme. Further improvement was observed in 21% patients of Group A, 37% patients of group B and 50% patients of group C.

#### *B: SCOPA-COG*

In accordance the SCOPA-COG test showed a significant difference between the groups (Fig. 4). All groups improved, indicated by the linear trend of days (Flin[1, 220) = 46.09; p < 0.001). Group C improved most resulting in a significant difference between the groups (F[2, 220] = 31.4, df = 2; p < 0.001). Since the slopes of the improvements differed between the groups, a significant interaction between days and groups occurred (F [2, 220] = 65.63; p< 0.001). Post hoc tests revealed a significant difference between all groups (p< 0.001). Patients of Group A reached 28.8 ± 3.7 points, Group B 30.3 ± 2.7 points and group C 37.6 ± 3.4 points. After completion of the in-patient training programme 31% of group A, 64% of group B and 88% of group C had shown a significant improvement on the SCOPA-COG, six months later at the final assessment 70% of patients of group A, 80% of patients of group B and 94% of patients of group C had been able to keep their level of performance.

## *C: GAS*

270 Diagnostics and Rehabilitation of Parkinson's Disease

Chi-square: 39.4;

Chi-square: 25.3

p < 0.001 T2:

p > 0.01

Group A 2.8 ± 1.2 3.14 ± 0.89 3.1 ± 0.9 T1:

Group B 31.00 ± 13.32 37.43 ± 12.72 39.57 ± 13.65 p < 0.001

Group B 10.7 ± 2.5 11.2 ± 2.1 11.24 ± 2 p < 0.01

Group B 2.9 ± 1.2 3.0 ± 1.2 2.9± 1.1 Group C 3.0. ± 0.7 3.55 ± 0.8 3.6 ± 0.9

Group A 29.94 ± 14.32 32.8 ± 14.83 32.5 ± 13.87

Group C 30.4 ± 12.98 46.5 ± 11.5 49.2 ± 13.4

Group A 10.1 ± 2.4 11.2 ± 2.2 11.2 ± 2.1

Group C 11.0 ± 2.5 12.5 ± 2.2 12.8 ± 1.9

Group A 44.8 ± 10.9 39.62 ± 10.56 38.98 ± 10.4

Group C 43.53 ± 9.8 38.76 ± 9.8 36.87 ± 10.0

Group A 42.68 ± 10.42 40.03 ± 10.2 41.2 ± 10.1

Group C 40.72 ± 9.98 38.8 ± 9.6 38.2 ± 10.2

patients of Group A, 37% patients of group B and 50% patients of group C.

Table 3. Summary of neuropsychological test results

**3.2.1 Primary outcome measure** 

Group B 44.42 ± 11.43 42.43 ± 11.2 38.65 ± 9.87 n.s.

Group B 41.84 ± 10.44 39.8 ± 9.8 40.2 ± 9.8 n.s.

All groups improved on the ADAS-Cog significantly shown by a significant linear trend (F lin [1, 220] = 150; p< 0.001. Group C improved most indicated by a significant interaction between groups and days (Fgroups x days [1, 220] = 27.26; p < 0.001) and a significant group difference (F [2,220] = 7.7, p < 0.001). Further analysis showed that 78% of the patients showed some improvement at the second assessment, 51% of patients of group A, 85% of patients of group B and 96% of patients of group C. 50% of the patients reached a reduction of the ADAS-Cog score of 3 or more points, 18% of group A, 54% of group B and 76% of group C. Six months after discharge of the rehabilitation unit 35% of patients (50% of patients of group A, 31% of patients of group B and 28% of group C) showed a deterioration compared to the assessment at the end of the in-patient training programme. Further improvement was observed in 21%

In accordance the SCOPA-COG test showed a significant difference between the groups (Fig. 4). All groups improved, indicated by the linear trend of days (Flin[1, 220) = 46.09; p <

BADS 6 elements

PASAT

TKS (points)

STAI X1

STAI X2

*A: ADAS-Cog* 

*B: SCOPA-COG* 

The GAS was performed on the final assessment. Group C reached more often the main goal than the other groups (Chi-square: 57.1; p < 0.001). The detailed analysis of the results is shown in table 4 and 5.

The main cognitive impairments reported by the patients could be attributed to the following domains: dual tasking, planning of complex and sequential tasks, decision making, rule recognition and rule shifting problems with delayed recall, difficulties in finding misplaced items. The patients based the selection of the goals on their individual main impairment. Table 4 shows the goals patients had chosen and if they were obtained.

More patients of group A compared to group B and C did not obtain the chosen goal or deteriorated compared to baseline while 27.6% of patients of group C obtained the goal and 39.4% exceeded the expectations mildly and 7.6% substantially.

### *D: Concentration*

In the Alterskonzentrationstest (AKT) no difference between the groups was detected. Patients did not differ in attention span neither at baseline nor at the final assessment.

#### *E: Information processing*

In the ZVT no difference between the groups was detected at baseline assessment. The time to complete the test decreased in all groups after the training (Flin [2,220] = 17.71; p< 0.001).

Cognitive Rehabilitation in Parkinson's Disease

N = 71

GAS Group A

Table 5. Results of the GAS

*F: Memory* 

Using Neuropsychological Training, Transfer Training and Sports Therapy 273

 Total Percent Total Percent Total Percent Total Percent -3 12 16.7 6 8 2 2.6 20 8.9 -2 10 13.8 5 6.7 3 3.9 18 23.7 -1 28 40.3 21 28 18 23.6 67 30 0 13 18.1 19 25.3 23 30.2 55 24.7 1 8 11.1 24 32 24 26.3 56 25,112 2 0 0 0 0 6 7.9 6 2.7 71 75 76 223

the improvement of the groups differed there was also a significant interaction between days and groups (F [2, 154] = 18.99; p < 0.001). Group C improved significantly more than

Only 157 patients (Group A: 50, Group B: 53, Group C: 54) managed the PASAT test on the first assessment and were included in the statistical model. The other patients did not succeed in finding a strategy to cope with the task. On the second and third assessment 56

In the MEMO-Test the recall of words improved in all groups over the trials as well as over the assessment days. The subscales of all words permanently stored in the long-term store (F[2,220] = 2.95; p< 0.05).and the total number of words in the long-term store (F[2,220] =

12345

Fig. 5. Group B and C kept more words permanently in memory than group A.

T1Group C T0 Group C T1 Group B

p < 0.05

T0 Group B T1 Group A T0 Group A

Trial

group B (p < 0.03) and A (p< 0.001) (F [2, 154]= 15.46; p< 0.001).

3.27; p< 0.05) differed between the groups (Fig 4).

0

1

2

3 4

Words permanently

 recalled

5

6

7

8 9

patients of group A, 64 of group B and 71 of group C scored on the test.

Group C N = 76

Total

Group B N = 75


Table 4. Goals chosen by the patients

Group C was superior to group A and B (p < 0.003) while group A and B did not differ resulting in a significant group difference (F[2,220] = 7.81; p< 0.001).

In the PASAT test the groups produced on average 50% correct answers at the baseline assessment, group A improved only marginally. Group B and C benefitted from the training programme shown in a significant linear trend for days (Flin [1, 154] = 63.71; p < 0.001). Since


Table 5. Results of the GAS

the improvement of the groups differed there was also a significant interaction between days and groups (F [2, 154] = 18.99; p < 0.001). Group C improved significantly more than group B (p < 0.03) and A (p< 0.001) (F [2, 154]= 15.46; p< 0.001).

Only 157 patients (Group A: 50, Group B: 53, Group C: 54) managed the PASAT test on the first assessment and were included in the statistical model. The other patients did not succeed in finding a strategy to cope with the task. On the second and third assessment 56 patients of group A, 64 of group B and 71 of group C scored on the test.

#### *F: Memory*

272 Diagnostics and Rehabilitation of Parkinson's Disease

Total number Percentage

[%] Total number Percentage

15 21.2 3 20

13 18.3 4 30.8

[%]

Groups Goal Goals chosen Goals obtained

N = 71 Dual tasking 15 21.1 3 20

N = 75 14 18.7 9 6.4

N = 76 15 20 10 67

N = 75 16 21.3 9 56.3

N = 76 17 22.4 10 58.9

N = 71 Decision making 10 14.1 4 40

N = 75 11 14.7 6 54.5

N = 76 14 18.4 10 71.4

N = 75 16 21.3 7 43.8

N = 76 14 18.4 10 71.4

N = 71 Delayed recall 12 16.9 3 25

N = 75 12 16 7 58.3

N = 76 11 14.5 9 82

N = 71 Search strategies 6 8.5 4 67

N = 75 6 8 5 83.3

N = 76 5 6.6 4 80

resulting in a significant group difference (F[2,220] = 7.81; p< 0.001).

Group C was superior to group A and B (p < 0.003) while group A and B did not differ

In the PASAT test the groups produced on average 50% correct answers at the baseline assessment, group A improved only marginally. Group B and C benefitted from the training programme shown in a significant linear trend for days (Flin [1, 154] = 63.71; p < 0.001). Since

A

B

C

A

B

C

A

B

C

A

B

C

A

B

C

A

B

C

N = 71 Planning of complex tasks

N = 71 Rule recognition and rule shifting

Table 4. Goals chosen by the patients

In the MEMO-Test the recall of words improved in all groups over the trials as well as over the assessment days. The subscales of all words permanently stored in the long-term store (F[2,220] = 2.95; p< 0.05).and the total number of words in the long-term store (F[2,220] = 3.27; p< 0.05) differed between the groups (Fig 4).

Fig. 5. Group B and C kept more words permanently in memory than group A.

Cognitive Rehabilitation in Parkinson's Disease

\*\*\* \*\*\*

patients of group C reported less impairment due to PD. (Fig. 7)

\*\*

Mobility ADL Emotional

perception

**3.2.3 PD specific impairment** 

0

Baseline

Final assessment

Table 6. UPDRS

\*\*\* p < 0.001; \*\* p < 0.01; \* p < 0.05

**3.2.4 Performance in daily living** 

partners in conjoint sports activities.

10

20

30

PDQ 39 Profile

40

50

60

Using Neuropsychological Training, Transfer Training and Sports Therapy 275

The PDQ39 shows that patients of group C rated their health related quality of life higher than the other groups. 13.8% of patients of group A, 38% of patients of group B and 52% of

12 3 4 5 6 7 8

Fig. 7. PD-patients of group C reported less PD-specific impairment at the final assessment.

The UPDRS score showed a mild improvement in all groups at the final assessment but there was no significant difference between the groups indicating that cognitive improvement was

not an unspecific effect resulting from general physical improvement. (Tab. 5)

UPDRS Motor scale 38.56 ± 12.44 37.53 ± 10.76 38.4 ± 11.78 UPDRS Sum-Score 59.20 ± 12.4 60.3 ± 12.4 61.5 ± 12.8

UPDRS Motor scale 34.1 ± 11.4 34.2 ± 11.2 35.2 ± 12.4 UPDRS Sum-Score 55.4 ± 12.4 56.3 ± 11.5 57.2 ± 11.4

The patients of Group C reported that they had adapted a more active life style, felt more confident in activities of daily living and had taken over some more chores. They perceived their partners and caregivers as being helpful. They enjoyed the participation of their

Patients of Group B also regarded the training programme as helpful but reported of having still problems with activities of daily living. Patients of group A had more difficulties with transfer of skills into daily life and the carry over effect was smaller than in the other groups.

Group A N=71

\*\*

Group A Group B Group C

Soc. support

\*

stigma cognition Communi-

Group B N = 75

\*\*

cation

n.s.

Bodily discomfort

> Group C N = 76

n.s.

Fig. 6. Group B and C recalled more words than group A.

### *G: Executive function*

The subtests of the BADS (rule shift cards, zoo map, modified 6 elements test) showed the following results:

The baseline scores of the rule shift cards did not differ between the groups. There was a mild but significant difference between the groups at the second assessment (Chi-square = 7.1; p < 0.03) and final assessment (Chi-square = 9.1; p< 0.01).

At baseline assessment Group C showed a tendency to better performance on the BADS Zoo Test. The mean profile scores of all groups were higher at the second assessment, but significant more patients of group C improved compared to group A and B. There was a clear group difference at the second (Chi-square = 49.31; p < 0.03) and third assessment ( Chi-square = 14.42; 0.001)

There was no difference in the performance in the 6 elements test or set shifting test. All groups showed an increase of the average profile scores leading to significant group differences at the second (Chi-square = 39.3; p< 0.001) and third assessments (Chi-square = 25.3; p< 0.001).

### *H: TKS*

The competence in cognitive estimation did not improve in Group A but in group B and C resulting in a significant difference between groups (T1:Chi square = 11.98; df = 2; p < 0.03; T2: Chi square = 22.153; df= 2; p < 0.002:

## **3.2.2 Assessment of mental state**

15% of the patients in group A, 20% of group B and 18% of patients of group C reported to suffer from depression and received medication. The results on the HADS depression scale indicated in 20% of patients of group A and group C respectively and in 25% of patients of Group B the presence of a mild to moderate depression. The anxiety level was assessed by using the Hamilton anxiety scale and did not differ between the groups. The additional assessments of the current anxiety level at the time of assessment (STAI X1) and of the personality trait anxiety (STAI X2) did not reveal differences between the groups. The anxiety at the time of the assessments decreased mildly from baseline to the final assessment.

## **3.2.3 PD specific impairment**

274 Diagnostics and Rehabilitation of Parkinson's Disease

12345

The subtests of the BADS (rule shift cards, zoo map, modified 6 elements test) showed the

The baseline scores of the rule shift cards did not differ between the groups. There was a mild but significant difference between the groups at the second assessment (Chi-square =

At baseline assessment Group C showed a tendency to better performance on the BADS Zoo Test. The mean profile scores of all groups were higher at the second assessment, but significant more patients of group C improved compared to group A and B. There was a clear group difference at the second (Chi-square = 49.31; p < 0.03) and third assessment (

There was no difference in the performance in the 6 elements test or set shifting test. All groups showed an increase of the average profile scores leading to significant group differences at the second (Chi-square = 39.3; p< 0.001) and third assessments (Chi-square =

The competence in cognitive estimation did not improve in Group A but in group B and C resulting in a significant difference between groups (T1:Chi square = 11.98; df = 2; p < 0.03;

15% of the patients in group A, 20% of group B and 18% of patients of group C reported to suffer from depression and received medication. The results on the HADS depression scale indicated in 20% of patients of group A and group C respectively and in 25% of patients of Group B the presence of a mild to moderate depression. The anxiety level was assessed by using the Hamilton anxiety scale and did not differ between the groups. The additional assessments of the current anxiety level at the time of assessment (STAI X1) and of the personality trait anxiety (STAI X2) did not reveal differences between the groups. The anxiety

at the time of the assessments decreased mildly from baseline to the final assessment.

T1Group C T0 Group C T1 Group B T0 Group B T1 Group A T0 Group A

p < 0.05

0

*G: Executive function* 

following results:

Chi-square = 14.42; 0.001)

T2: Chi square = 22.153; df= 2; p < 0.002:

**3.2.2 Assessment of mental state** 

25.3; p< 0.001).

*H: TKS* 

Fig. 6. Group B and C recalled more words than group A.

7.1; p < 0.03) and final assessment (Chi-square = 9.1; p< 0.01).

1

2

3

4

5

6

The PDQ39 shows that patients of group C rated their health related quality of life higher than the other groups. 13.8% of patients of group A, 38% of patients of group B and 52% of patients of group C reported less impairment due to PD. (Fig. 7)

Fig. 7. PD-patients of group C reported less PD-specific impairment at the final assessment. \*\*\* p < 0.001; \*\* p < 0.01; \* p < 0.05

The UPDRS score showed a mild improvement in all groups at the final assessment but there was no significant difference between the groups indicating that cognitive improvement was not an unspecific effect resulting from general physical improvement. (Tab. 5)


Table 6. UPDRS

#### **3.2.4 Performance in daily living**

The patients of Group C reported that they had adapted a more active life style, felt more confident in activities of daily living and had taken over some more chores. They perceived their partners and caregivers as being helpful. They enjoyed the participation of their partners in conjoint sports activities.

Patients of Group B also regarded the training programme as helpful but reported of having still problems with activities of daily living. Patients of group A had more difficulties with transfer of skills into daily life and the carry over effect was smaller than in the other groups.

Cognitive Rehabilitation in Parkinson's Disease

programme.

programme and at the second assessment six months later.

Using Neuropsychological Training, Transfer Training and Sports Therapy 277

The BADS subscales especially the zoo map is a very demanding task requiring excellent planning skills. Even patients of group A and B with previously shown improvement on the BADS subscales lost most of that. Only patients of group C managed to keep their level of performance. The performance of group B and A dropped nearly to baseline level. Thus, Group C has been superior to group B and A immediately after completing the training

The difficulties patients experienced while solving the tasks have been in accordance with the results of other studies (Lewis et al., 2003, 2005). Most improvement has been observed in the LURIA, dice, assembly pattern, MOSAIC test of the SCOPA-COG. Mild improvement has been observed in the ZOO map and the PASAT-test. The pattern of improvement did not differ between the groups but the percentage of subjects showing an improvement differed significantly as well as the speed of recovery. The UPDRS – score improved in all groups slightly. There were no significant differences between the groups and no significant change of medication. Accordingly, the improvement in the neuropsychological tests cannot be referred to a better physical condition of one group and can be attributed to the training

90% of patients of group C pursued the training at home with the same quantity and intensity while only 75% of group B and 50% of group A did so. Patients of group C conducted a motor training programme three times/week and practised cognitive tasks twice a week for 45 min. Patients of group B and C continued with some tasks resembling the transfer tasks they had performed during the training programme. The partners of the patients of group B and C managed to support the patients in practising transfer tasks, they asked them to prepare a meal or to do the shopping. The majority of the spouses of patients of group C joined their partners in the sports programme. The support of the spouses alleviated the home training significantly. As known from a questionnaire sent to the patients social aspects are very important for PD- patients. It is difficult to decide whether the further improvement of cognitive performance which occurred in some tests was due to the quantity of training or the content of the training. However, group C was already superior to the other groups at the second assessment. Since patients were compliant with the programme during the in-patient stay and received the same quantity of training, the different performance might rather be due to the content of the training than to the quantity. The performance of the patients differed between the tasks suggesting that the different training schedules between the groups affect the training outcome. For example the BADS zoo map a very challenging test as mentioned above requires various training approaches to achieve an improvement. As a result only patients of group C obtained an improvement on this test. Depression might also influence the performance in neuropsychological tests. Klepac et al. (2009) had found that depression preceding PD motor signs might favour poorer cognitive abilities. However, there was no significant difference between the groups regarding the percentage of patients being depressed and the onset of depression. Thus, an

influence of depression and anxiety on cognitive performance could be ruled out.

disorder specialists conducting the tests were blinded to the treatment arms.

Assessment bias in favour for one treatment can be excluded because the movement

Thus, the findings of the study suggest that PD patients benefit from a specific cognitive training and that a multimodal training might be most suitable for improving cognitive performance in PD. As already shown in a previous study (Hullmann et al., 2004) the cognitive training needs to be specific. Therefore, we had chosen an individual approach based on the Patients` results in the neuropsychological test battery. The specificity of the

Sports activities were with 300min/week higher in Group C than in group B (196min/week) and A (176 min/ week).

Patients of group A reported to perform 7.4 ± 3.1 hours very hard work per week, while patients of group B and C reported of 10.4 ± 2.2 and 11.5 ± 2.7 hours very hard work respectively. Group A managed 14.2 ± 3.9, Group B 16.1 ± 4.3 and Group C 17.9 ± 4.1 hard work

In accordance with the patients` reports 65% of the caregivers of patients in Group C found competence and cognition of the patients improved. In group B 54% of caregivers and in group A 49%of caregivers confirmed an improvement. A deterioration of the performance in daily living was reported in 11% of group C, 17% of group B and 25% of group A. In summary patients who conducted a multimodal cognitive rehabilitation programme improved most and continued coping with daily tasks. Patients of group C were more active in daily living and took more often part in sports activities.

## **4. Discussion**

In summary 90 % of patients of group A, 93.8% of patients of group B and 95% of patients of group C completed the training. Data of patients who did not continue with the programme were not included into the statistical analysis. Although patients complained of a lack of concentration, they performed well on the AKT. The second and third assessment did not reveal further improvement. The lack of improvement might be due to a ceiling effect since the performance on this test at baseline was good in all groups. The same might apply for the MMST which did not differ significantly between baseline and follow-up assessments. The training programme did not affect the mood of the patients.

At the baseline assessment patients of all three groups had shown deficits mainly in tests addressing executive functions. Consecutively, the performance of the patients was worse on the subtests of the SCOPA-COG semantic fluency, LURIA, dice and assembly pattern of the SCOPA-COG, Zoo test of the BADS, PASAT and cognitive estimation. The memory tasks such as immediate and delayed word recall were only mildly disturbed. All groups showed some improvements at the assessment immediately after completion of the training programme in the following tests: TMT, BADS Rule shift cards, zoo map, modified 6 elements test, PASAT and TKS. The mean scores of the ADAS-Cog and SCOPA-COG test in group A were not significantly better compared to the baseline assessment although 18% of the patients reached an improvement of 3 or more points on the ADAS-Cog. The findings were similar for the SCOPA-COG test. 31% of the patients showed an improvement on the SCOPA-COG test. Clear differences between the groups were found for the following tests: TMT, BADS zoo map, BADS rule shift cards, BADS 6 elements, PASAT, TKS, 2 subtests of the Memo Test.

At the second outcome assessment, 6 months after completion of the training programme, 21% of the patients of group A showed a further improvement on the ADAS-Cog, on the other hand 50% of patients of group A deteriorated, 31% of patients of group B and 28% of patients of group C within the 6 months after discharge from the rehabilitation unit. Most of the patients of group B and C were able to keep their performance level between the second and third assessment on the SCOPA-COG, while group A deteriorated. Further improvements between the second and the final assessment were obtained in group B and C on the TKS, PASAT and MEMO test.

Sports activities were with 300min/week higher in Group C than in group B (196min/week)

Patients of group A reported to perform 7.4 ± 3.1 hours very hard work per week, while patients of group B and C reported of 10.4 ± 2.2 and 11.5 ± 2.7 hours very hard work respectively. Group A managed 14.2 ± 3.9, Group B 16.1 ± 4.3 and Group C 17.9 ± 4.1 hard

In accordance with the patients` reports 65% of the caregivers of patients in Group C found competence and cognition of the patients improved. In group B 54% of caregivers and in group A 49%of caregivers confirmed an improvement. A deterioration of the performance in daily living was reported in 11% of group C, 17% of group B and 25% of group A. In summary patients who conducted a multimodal cognitive rehabilitation programme improved most and continued coping with daily tasks. Patients of group C were more active

In summary 90 % of patients of group A, 93.8% of patients of group B and 95% of patients of group C completed the training. Data of patients who did not continue with the programme were not included into the statistical analysis. Although patients complained of a lack of concentration, they performed well on the AKT. The second and third assessment did not reveal further improvement. The lack of improvement might be due to a ceiling effect since the performance on this test at baseline was good in all groups. The same might apply for the MMST which did not differ significantly between baseline and follow-up assessments.

At the baseline assessment patients of all three groups had shown deficits mainly in tests addressing executive functions. Consecutively, the performance of the patients was worse on the subtests of the SCOPA-COG semantic fluency, LURIA, dice and assembly pattern of the SCOPA-COG, Zoo test of the BADS, PASAT and cognitive estimation. The memory tasks such as immediate and delayed word recall were only mildly disturbed. All groups showed some improvements at the assessment immediately after completion of the training programme in the following tests: TMT, BADS Rule shift cards, zoo map, modified 6 elements test, PASAT and TKS. The mean scores of the ADAS-Cog and SCOPA-COG test in group A were not significantly better compared to the baseline assessment although 18% of the patients reached an improvement of 3 or more points on the ADAS-Cog. The findings were similar for the SCOPA-COG test. 31% of the patients showed an improvement on the SCOPA-COG test. Clear differences between the groups were found for the following tests: TMT, BADS zoo map, BADS rule shift cards, BADS 6 elements, PASAT, TKS, 2 subtests of

At the second outcome assessment, 6 months after completion of the training programme, 21% of the patients of group A showed a further improvement on the ADAS-Cog, on the other hand 50% of patients of group A deteriorated, 31% of patients of group B and 28% of patients of group C within the 6 months after discharge from the rehabilitation unit. Most of the patients of group B and C were able to keep their performance level between the second and third assessment on the SCOPA-COG, while group A deteriorated. Further improvements between the second and the final assessment were obtained in group B and C

in daily living and took more often part in sports activities.

The training programme did not affect the mood of the patients.

and A (176 min/ week).

work

**4. Discussion** 

the Memo Test.

on the TKS, PASAT and MEMO test.

The BADS subscales especially the zoo map is a very demanding task requiring excellent planning skills. Even patients of group A and B with previously shown improvement on the BADS subscales lost most of that. Only patients of group C managed to keep their level of performance. The performance of group B and A dropped nearly to baseline level. Thus, Group C has been superior to group B and A immediately after completing the training programme and at the second assessment six months later.

The difficulties patients experienced while solving the tasks have been in accordance with the results of other studies (Lewis et al., 2003, 2005). Most improvement has been observed in the LURIA, dice, assembly pattern, MOSAIC test of the SCOPA-COG. Mild improvement has been observed in the ZOO map and the PASAT-test. The pattern of improvement did not differ between the groups but the percentage of subjects showing an improvement differed significantly as well as the speed of recovery. The UPDRS – score improved in all groups slightly. There were no significant differences between the groups and no significant change of medication. Accordingly, the improvement in the neuropsychological tests cannot be referred to a better physical condition of one group and can be attributed to the training programme.

90% of patients of group C pursued the training at home with the same quantity and intensity while only 75% of group B and 50% of group A did so. Patients of group C conducted a motor training programme three times/week and practised cognitive tasks twice a week for 45 min. Patients of group B and C continued with some tasks resembling the transfer tasks they had performed during the training programme. The partners of the patients of group B and C managed to support the patients in practising transfer tasks, they asked them to prepare a meal or to do the shopping. The majority of the spouses of patients of group C joined their partners in the sports programme. The support of the spouses alleviated the home training significantly. As known from a questionnaire sent to the patients social aspects are very important for PD- patients. It is difficult to decide whether the further improvement of cognitive performance which occurred in some tests was due to the quantity of training or the content of the training. However, group C was already superior to the other groups at the second assessment. Since patients were compliant with the programme during the in-patient stay and received the same quantity of training, the different performance might rather be due to the content of the training than to the quantity. The performance of the patients differed between the tasks suggesting that the different training schedules between the groups affect the training outcome. For example the BADS zoo map a very challenging test as mentioned above requires various training approaches to achieve an improvement. As a result only patients of group C obtained an improvement on this test. Depression might also influence the performance in neuropsychological tests. Klepac et al. (2009) had found that depression preceding PD motor signs might favour poorer cognitive abilities. However, there was no significant difference between the groups regarding the percentage of patients being depressed and the onset of depression. Thus, an influence of depression and anxiety on cognitive performance could be ruled out.

Assessment bias in favour for one treatment can be excluded because the movement disorder specialists conducting the tests were blinded to the treatment arms.

Thus, the findings of the study suggest that PD patients benefit from a specific cognitive training and that a multimodal training might be most suitable for improving cognitive performance in PD. As already shown in a previous study (Hullmann et al., 2004) the cognitive training needs to be specific. Therefore, we had chosen an individual approach based on the Patients` results in the neuropsychological test battery. The specificity of the

Cognitive Rehabilitation in Parkinson's Disease

strategies with improvement of frontal lobe functions.

transfer tasks and a physical training is recommended.

tests of executive tasks.

Using Neuropsychological Training, Transfer Training and Sports Therapy 279

Our results are in accordance with the authors` conclusions (Goebel et al., 2008): "Adding training time and scheduling repetitive, cue-initiated learning trials may further improve training effects. Such a procedure may lead to more automated, implicit strategy application that demands less executive control (e.g., Baddeley,1998; Norman & Shallice, 1986; Sammer et al., 2006) whereas instruction alone bears the risk of increasing working memory load". This is in accord with a work of Sinforiani et al (2004) who showed a significant improvement at verbal fluency, logic memory and Raven's matrices tests after a 6- week cognitive rehabilitiation training including cognitive and physical training. After the completion of the training a carry-over effect has been observed and the authors referred the effects to the combination of a cognitive and physical training. The authors suggested that the cognitive rehabilitation training exerts its positive effects by reinforcing cognitive

Therefore, emphasis should be placed on the reduction of cognitive load in psychological training programmes. The combination of cognitive training at the writing table with

Research over the last decade has shown that cognitive deficits affect motor performance. Patients with cognitive deficits had more difficulties in motor tests than patients without cognitive deficits (Goldmann, 1998). Hausdorff et al (2005) have found a close correlation between walking and executive functions. Yogev et al. (2005) have shown that gait variability in dual tasking is closely associated with the performance in neuropsychological

Therefore, one might speculate that motor functions might affect cognitive performance as well. There is a huge body of literature suggesting a prevention of cognitive decline by life long exercise or even an improvement of cognitive deficits by physical activity. Executive functions may be selectively maintained or improved in people with better physical condition provided by physical training (Churchill et al., 2002). The importance of aerobic physical exercise on cognitive functions, especially on executive functions has been shown (Kramer 1999, Colcombe et al., 2003, 2004, 2006). The studies have been mainly conducted in healthy elderly or patients with dementia. Tanaka et al. (2008) have shown that older people with PD can benefit their executive functions in the same way, as do their peers without PD. The results of some studies have shown that brain areas undergoing biological aging benefit most from endurance sports. Even structural changes have been observed (Colcombe et al., 2006). Exercise is thought to enhance brain plasticity. Neuroplasticity might be supported by BDNF release, which is exercise regulated. Physical exercise increases the release of growth hormone (GH) which represents the main stimulus for the release of insulin growth factor (IGF-1). IGF-1 is involved in processes regulating learning,, memory, neurogenesis and amyloid degradation ( Holzenberger et al., 2003, Carter&Ramsey, 2002). The release of IGF-1 is closely related to the release of BDNF. Several responses of the brain to exercise have been described. In animal studies comparing young and old animals a difference was shown in the location of the BDnF mRNA upregulation in the hippocampus. Young animal showed an increase of BDNF mRNA in dendate gyrus, hilus and Ca3 region, old animals in the Ca1 and Ca2 region. Long term potentiation which is relevant for memory and learning was also found. LTP was correlated with increased expression of mRNA of the NR2B receptor unit of the NMDA (N-methyl-D-aspertate) receptor. Increase of cerebral blood flow and reduction of cardiovascular risk factors might also contribute to the positive effects of sport on cognition. The reduction of cardiovascular risk factors does not play a role in the present study because of the short observation time. The release of dopamine by exercise might also

training for executive functions is also shown in the fact that an other functional domain such as attention was not influenced by the training. Home based cognitive exercises were sufficient to keep the performance of the second assessment in patients of group C. However, patients of group B and C were able to keep some improvements as well. Home based cognitive training without transfer and physical training as performed by group A was less attractive for the patients. However, the poorer results of group A were not due to fewer training lessons since the performance of group A was already poorer on the second assessment. During the in-patient stay the quantity of training lessons were similar in all groups, only the percentage of specific training differed. Thus, the content of the training might be responsible for the different performance of the groups. The superiority of group B compared to group A suggests the efficacy of the transfer tasks. The psychomotor training helps the group C to improve further, especially in the challenging executive tasks regarding rule cognition, set shifting and decision making. However, it is not clear whether patients of group B and C could also cope better with completely new situations.

In contrast to a study by Paris et al. (2011), the present study suggested a translation of improved cognitive performance on the neuropsychological tests into daily living. Group C scored much higher on the PDQ 39 than the other groups. Thus, health related quality of life was improved markedly in these patients. The patients` caregivers also reported an improved competence in real life. In addition the goal attainment scale had been used in the present study. The patients picked the goals according to the cognitive problems they experienced in daily living. Most often the cognitive problem, they suffered most of, was chosen as goal. Half of the patients managed to obtain the goals agreed on prior to the training programme. Patients of group C reached significantly more often the goal than the patients of group B or A

The cognitive training performed in the study of Paris et al. (2011) resembled the training of group A in the current study. Only 29% of patients of group A obtained the goal compared to 69.8% of group C.

Some goals seemed to be more difficult to achieve (see Table 4). Patients faced more difficulties in attaining goals regarding rule generation and rule shifting while goals like dual tasking and memory improvement were easier to obtain. Group C was more successful to achieve an improvement in planning of complex tasks, rule generation and decision finding than group B and C.

Goebel et al. (2010) compared the ability of PD-patients to internally initiate a strategy with their ability to utilize an externally provided strategy in a simple Numerosity judgement task. The data of the study showed a general slowdown after strategy instruction. Furthermore, some patients reported difficulties in applying the strategies. The authors referred the findings to a failure in metacognition. Inferior utilization of metacognitive memory strategies seems to induce problems of PD-patients in real-life situations (Johnson et al., 2005, Shimamura, 2000). External instruction might activate metacognitive control processes and slow down the system. However, when PD-patients had sufficient time to solve the tasks there was no general deficit in the ability to internally generate a cognitive strategy in PD. Patients of group C had sufficient time during the psychomotor training to work out strategies to solve tasks and had time to initiate internal strategies. The combination of the psychomotor training with the transfer training provided the patients with some guidance and instructions to solve the tasks. However, the guidance was not too restrictive, there was enough time to find individual solutions. Additionally, the training was less standardised and strongly tailored to the patients` needs.

training for executive functions is also shown in the fact that an other functional domain such as attention was not influenced by the training. Home based cognitive exercises were sufficient to keep the performance of the second assessment in patients of group C. However, patients of group B and C were able to keep some improvements as well. Home based cognitive training without transfer and physical training as performed by group A was less attractive for the patients. However, the poorer results of group A were not due to fewer training lessons since the performance of group A was already poorer on the second assessment. During the in-patient stay the quantity of training lessons were similar in all groups, only the percentage of specific training differed. Thus, the content of the training might be responsible for the different performance of the groups. The superiority of group B compared to group A suggests the efficacy of the transfer tasks. The psychomotor training helps the group C to improve further, especially in the challenging executive tasks regarding rule cognition, set shifting and decision making. However, it is not clear whether

patients of group B and C could also cope better with completely new situations.

patients of group B or A

finding than group B and C.

to 69.8% of group C.

In contrast to a study by Paris et al. (2011), the present study suggested a translation of improved cognitive performance on the neuropsychological tests into daily living. Group C scored much higher on the PDQ 39 than the other groups. Thus, health related quality of life was improved markedly in these patients. The patients` caregivers also reported an improved competence in real life. In addition the goal attainment scale had been used in the present study. The patients picked the goals according to the cognitive problems they experienced in daily living. Most often the cognitive problem, they suffered most of, was chosen as goal. Half of the patients managed to obtain the goals agreed on prior to the training programme. Patients of group C reached significantly more often the goal than the

The cognitive training performed in the study of Paris et al. (2011) resembled the training of group A in the current study. Only 29% of patients of group A obtained the goal compared

Some goals seemed to be more difficult to achieve (see Table 4). Patients faced more difficulties in attaining goals regarding rule generation and rule shifting while goals like dual tasking and memory improvement were easier to obtain. Group C was more successful to achieve an improvement in planning of complex tasks, rule generation and decision

Goebel et al. (2010) compared the ability of PD-patients to internally initiate a strategy with their ability to utilize an externally provided strategy in a simple Numerosity judgement task. The data of the study showed a general slowdown after strategy instruction. Furthermore, some patients reported difficulties in applying the strategies. The authors referred the findings to a failure in metacognition. Inferior utilization of metacognitive memory strategies seems to induce problems of PD-patients in real-life situations (Johnson et al., 2005, Shimamura, 2000). External instruction might activate metacognitive control processes and slow down the system. However, when PD-patients had sufficient time to solve the tasks there was no general deficit in the ability to internally generate a cognitive strategy in PD. Patients of group C had sufficient time during the psychomotor training to work out strategies to solve tasks and had time to initiate internal strategies. The combination of the psychomotor training with the transfer training provided the patients with some guidance and instructions to solve the tasks. However, the guidance was not too restrictive, there was enough time to find individual solutions. Additionally, the training

was less standardised and strongly tailored to the patients` needs.

Our results are in accordance with the authors` conclusions (Goebel et al., 2008): "Adding training time and scheduling repetitive, cue-initiated learning trials may further improve training effects. Such a procedure may lead to more automated, implicit strategy application that demands less executive control (e.g., Baddeley,1998; Norman & Shallice, 1986; Sammer et al., 2006) whereas instruction alone bears the risk of increasing working memory load".

This is in accord with a work of Sinforiani et al (2004) who showed a significant improvement at verbal fluency, logic memory and Raven's matrices tests after a 6- week cognitive rehabilitiation training including cognitive and physical training. After the completion of the training a carry-over effect has been observed and the authors referred the effects to the combination of a cognitive and physical training. The authors suggested that the cognitive rehabilitation training exerts its positive effects by reinforcing cognitive strategies with improvement of frontal lobe functions.

Therefore, emphasis should be placed on the reduction of cognitive load in psychological training programmes. The combination of cognitive training at the writing table with transfer tasks and a physical training is recommended.

Research over the last decade has shown that cognitive deficits affect motor performance. Patients with cognitive deficits had more difficulties in motor tests than patients without cognitive deficits (Goldmann, 1998). Hausdorff et al (2005) have found a close correlation between walking and executive functions. Yogev et al. (2005) have shown that gait variability in dual tasking is closely associated with the performance in neuropsychological tests of executive tasks.

Therefore, one might speculate that motor functions might affect cognitive performance as well. There is a huge body of literature suggesting a prevention of cognitive decline by life long exercise or even an improvement of cognitive deficits by physical activity. Executive functions may be selectively maintained or improved in people with better physical condition provided by physical training (Churchill et al., 2002). The importance of aerobic physical exercise on cognitive functions, especially on executive functions has been shown (Kramer 1999, Colcombe et al., 2003, 2004, 2006). The studies have been mainly conducted in healthy elderly or patients with dementia. Tanaka et al. (2008) have shown that older people with PD can benefit their executive functions in the same way, as do their peers without PD. The results of some studies have shown that brain areas undergoing biological aging benefit most from endurance sports. Even structural changes have been observed (Colcombe et al., 2006). Exercise is thought to enhance brain plasticity. Neuroplasticity might be supported by BDNF release, which is exercise regulated. Physical exercise increases the release of growth hormone (GH) which represents the main stimulus for the release of insulin growth factor (IGF-1). IGF-1 is involved in processes regulating learning,, memory, neurogenesis and amyloid degradation ( Holzenberger et al., 2003, Carter&Ramsey, 2002). The release of IGF-1 is closely related to the release of BDNF. Several responses of the brain to exercise have been described. In animal studies comparing young and old animals a difference was shown in the location of the BDnF mRNA upregulation in the hippocampus. Young animal showed an increase of BDNF mRNA in dendate gyrus, hilus and Ca3 region, old animals in the Ca1 and Ca2 region. Long term potentiation which is relevant for memory and learning was also found. LTP was correlated with increased expression of mRNA of the NR2B receptor unit of the NMDA (N-methyl-D-aspertate) receptor. Increase of cerebral blood flow and reduction of cardiovascular risk factors might also contribute to the positive effects of sport on cognition. The reduction of cardiovascular risk factors does not play a role in the present study because of the short observation time. The release of dopamine by exercise might also

Cognitive Rehabilitation in Parkinson's Disease

**6. Strengths of the study** 

**7. Conclusion** 

compliance with the training at home.

*Geriatr Soc* , 48, pp. 938-942.

37, pp. 529-534

Parkinson`s disease. Curr Neurol Neurosci Rep.

Baddeley AD (1986). *Working memory*. Oxford: Clarendon Press

**8. Acknowledgement**

**9. References** 

these confounders a larger sample of patients will be needed.

stable avoiding confounding effects by the change of medication.

Using Neuropsychological Training, Transfer Training and Sports Therapy 281

or antidepressant medication might influence cognitive performance. In order to correct for

To our knowledge, the results of the current study show for the first time, that a multidisciplinary cognitive training in patients with Parkinson`s disease can lead to improvements of cognitive function which translate into everyday life and are not only shown by improvements on neuropsychological scales. We want to emphasize that a blinded randomised design and a standardised neuropsychological test battery were employed. Furthermore, the cohort of patients undergoing the study protocol was big and the dropout rate was low for this type of study. In addition the training of the caregivers guaranteed a supporting environment in all groups. We were able to keep the medication

In conclusion, we have shown that PD-patients with cognitive deficits benefit from a multidisciplinary cognitive training. A multimodal training is superior to a paper and pencil based cognitive treatment. We have shown a translation of improvements in cognitive tests into performance in real life. Although the multimodal training is time consuming and requires high motivation, it is worth to pursue the training considering the secondary diseases, loss of quality of life and the costs following the diagnosis of dementia. The role of the caregivers has also to be emphasized, the involvement of the family improves the

We thank the Dr. Werner Jackstädt-Stiftung, the support made the present study possible.

Aarsland D, Larsen JP, Tandberg E. & Laake K. (2000). Predictors of nursing home

Aarsland D, Brønnick K, & Fladby T. (2011 Epub ahead). Mild cognitive impairment in

Abbott RD, White LR, Webster R, Masaki KH, Curb JD. & Petrovitch H. (2004). Walking and dementia in physically capable elderly men. *JAMA,* 292 (12), pp. 1447-1453 Baatile J, Langbein WE, Weaver F, Maloney C & Jost MB. (2000). Effect of exercise on

Ble A, Volpato S, Zuliani G, Guralnik JM, Bandinelli S, Lauretani F, Bartali B, Maraldi C,

older persons: the InCHIANTI study.*J Am Geriatr Soc*, 53(3), pp. 410-5.

placement in Parkinson`s disease: A population based prospective study*. J Am* 

perceived quality of life in individuals with Parkinson`s disease. *J Rehabil Res Dev,* 

Fellin R, & Ferrucci L.(2005). Executive function correlates with walking speed in

play a role. An increase in the activity of antioxidant enzymes, and thus increases the capacity to defend against the stress of oxidation in the central nervous system (Rodák et al., 2001) might also support neuroplasticity.

It is not specified so far which type of exercise might be most promising for improving cognitive performance. The role of endurance training has been shown, whether a combination of aerobic training with a cognitive challenging physical training is of advantage needs further research. The physical training in the present study provided both, a training of strategies to solve tasks and an aerobic training. Since intermittent training schedules have been shown to be as effective as daily training, the frequency of training sessions should have been sufficient as well.

It is not clear so far how cognitive training at the writing table might improve cognitive performance. The destruction of the nigrostriatal dopaminergic pathways is often about 75% and involves the ventral tegmental area, which innervates the prefrontal cortex. Therefore, it is unlikely that cognitive training reconstructs the dopaminergic system.

The multimodal training of cognitive functions is time- consuming and put demands on resources. Due to the quantity and quality of the trainings sessions it will also be costly. On the other hand dementia is a risk factor for falls and transfer to nursing –homes, which increases the costs for the patients´ care substantially and jeopardizes the patients` quality of life. Considering the sequelae of dementia such as increased dependence on care givers, high morbidity and increased mortality it is justified to spend more time and effort into prevention of dementia. Therefore, provision of adequate financing is also required.

## **5. Limitations of the study**

One might criticise that we compared three different treatment arms and did not include a control group without cognitive training in this study. Patients were enrolled into the study during their stay in a rehabilitation unit and complained of a deterioration of their cognitive performance. For this reason it was not possible to withhold treatment. Further, we had shown in a previous study (Hullmann et al., 2004) the superiority of a cognitive training compared to standard treatment in control subjects. Therefore, we compared three different treatment arms with increasing stimulus modality.

Another limitation is that there are no evidence based data for the transfer training. Further research is necessary to evaluate and validate which transfer exercises are useful tools. The psychomotor training has been used for many years in children and has been used in patients with dementia (Oswald WP, 1996). However, it has not been validated in PDpatients so far. The selection of tasks had been based on the clinical experience of the therapists and medical staff and the published data based on the work with children.

One might also argue which improvement might be clinically relevant. However, the scales we used are all validated and had been often applied in clinical studies. The clinical relevance of the improvements is also shown by the observed translation into real life. One might criticize that the patients were not tested regarding their performance in completely new situations. Patients, caregivers and the neurologist supervising the treatment agreed on certain goals at the beginning of the study. Hence, situations resembling the agreed goals were trained during the study.

Furthermore, due to the short follow up period of 6 months we cannot report on long-term results. However, studies assessing long-term results are very difficult to conduct since it is very difficult to keep the medication stable. A change in dopaminergic (Fournet et al., 2000) or antidepressant medication might influence cognitive performance. In order to correct for these confounders a larger sample of patients will be needed.

## **6. Strengths of the study**

280 Diagnostics and Rehabilitation of Parkinson's Disease

play a role. An increase in the activity of antioxidant enzymes, and thus increases the capacity to defend against the stress of oxidation in the central nervous system (Rodák et al.,

It is not specified so far which type of exercise might be most promising for improving cognitive performance. The role of endurance training has been shown, whether a combination of aerobic training with a cognitive challenging physical training is of advantage needs further research. The physical training in the present study provided both, a training of strategies to solve tasks and an aerobic training. Since intermittent training schedules have been shown to be as effective as daily training, the frequency of training

It is not clear so far how cognitive training at the writing table might improve cognitive performance. The destruction of the nigrostriatal dopaminergic pathways is often about 75% and involves the ventral tegmental area, which innervates the prefrontal cortex. Therefore, it

The multimodal training of cognitive functions is time- consuming and put demands on resources. Due to the quantity and quality of the trainings sessions it will also be costly. On the other hand dementia is a risk factor for falls and transfer to nursing –homes, which increases the costs for the patients´ care substantially and jeopardizes the patients` quality of life. Considering the sequelae of dementia such as increased dependence on care givers, high morbidity and increased mortality it is justified to spend more time and effort into

One might criticise that we compared three different treatment arms and did not include a control group without cognitive training in this study. Patients were enrolled into the study during their stay in a rehabilitation unit and complained of a deterioration of their cognitive performance. For this reason it was not possible to withhold treatment. Further, we had shown in a previous study (Hullmann et al., 2004) the superiority of a cognitive training compared to standard treatment in control subjects. Therefore, we compared three different

Another limitation is that there are no evidence based data for the transfer training. Further research is necessary to evaluate and validate which transfer exercises are useful tools. The psychomotor training has been used for many years in children and has been used in patients with dementia (Oswald WP, 1996). However, it has not been validated in PDpatients so far. The selection of tasks had been based on the clinical experience of the

One might also argue which improvement might be clinically relevant. However, the scales we used are all validated and had been often applied in clinical studies. The clinical relevance of the improvements is also shown by the observed translation into real life. One might criticize that the patients were not tested regarding their performance in completely new situations. Patients, caregivers and the neurologist supervising the treatment agreed on certain goals at the beginning of the study. Hence, situations resembling the agreed goals

Furthermore, due to the short follow up period of 6 months we cannot report on long-term results. However, studies assessing long-term results are very difficult to conduct since it is very difficult to keep the medication stable. A change in dopaminergic (Fournet et al., 2000)

therapists and medical staff and the published data based on the work with children.

prevention of dementia. Therefore, provision of adequate financing is also required.

is unlikely that cognitive training reconstructs the dopaminergic system.

2001) might also support neuroplasticity.

sessions should have been sufficient as well.

**5. Limitations of the study** 

were trained during the study.

treatment arms with increasing stimulus modality.

To our knowledge, the results of the current study show for the first time, that a multidisciplinary cognitive training in patients with Parkinson`s disease can lead to improvements of cognitive function which translate into everyday life and are not only shown by improvements on neuropsychological scales. We want to emphasize that a blinded randomised design and a standardised neuropsychological test battery were employed. Furthermore, the cohort of patients undergoing the study protocol was big and the dropout rate was low for this type of study. In addition the training of the caregivers guaranteed a supporting environment in all groups. We were able to keep the medication stable avoiding confounding effects by the change of medication.

## **7. Conclusion**

In conclusion, we have shown that PD-patients with cognitive deficits benefit from a multidisciplinary cognitive training. A multimodal training is superior to a paper and pencil based cognitive treatment. We have shown a translation of improvements in cognitive tests into performance in real life. Although the multimodal training is time consuming and requires high motivation, it is worth to pursue the training considering the secondary diseases, loss of quality of life and the costs following the diagnosis of dementia. The role of the caregivers has also to be emphasized, the involvement of the family improves the compliance with the training at home.

## **8. Acknowledgement**

We thank the Dr. Werner Jackstädt-Stiftung, the support made the present study possible.

## **9. References**


Baddeley AD (1986). *Working memory*. Oxford: Clarendon Press

Ble A, Volpato S, Zuliani G, Guralnik JM, Bandinelli S, Lauretani F, Bartali B, Maraldi C, Fellin R, & Ferrucci L.(2005). Executive function correlates with walking speed in older persons: the InCHIANTI study.*J Am Geriatr Soc*, 53(3), pp. 410-5.

Cognitive Rehabilitation in Parkinson's Disease

Using Neuropsychological Training, Transfer Training and Sports Therapy 283

Fahn S, Elton RL and the members of the UPDRS Development Committee. (1987). Unified

Fitzpatrick R, Peto V, Greenhall R& Hyman N. (1997). The Parkinson`s diasease

Funahashi S. (2001). Neuronal mechanisms of executive control by prefrontal cortex.

Gatterer G, Fischer P, Simany M& Danielczyk W. (1989). The A-K-T (Alters-Konzentrations-Test a new psychometric test for geriatric patients. FunctNeurol; 4 (3), p. 273-276 Goldman-Rakic PS, Lidow MS, Smiley JF & Williams MS. (1992) The anatomy of dopamine in monkey and human prefrontal cortex. *J of Transmission*, suppl. 36, p. 163-177 Godefroy O. (2003).Frontal syndrome and disorders of executive functions*. J Neurol,* 250 (1),

Goebel S, Mehdorn HM& Leplow B. (2010). Strategy instruction in Parkinson's disease: Influence on cognitive performance.*Neuropsychologia,* 48(2), p. 574-80 Gronwall, D.M.A. (1977). Paced auditory serial-addition task: A measure of recovery from

Hausdorff JM, Yogev G, Springer S, Simon ES & Giladi N. (2005). Walking is more like

Heinz-Martin S., Oberauer K., Wittmann WW, Wilhelm O. & Schulze R. (2002). Working-

Hoehn MM, Yahr MR, Parkinsonism: onset, progression and mortality. (1967). *Neurology,*

Holzenberger M, Dupont J (2000). IGF-I receptor regulates lifespan and resistance to

Howells DW, Porrit MJ & Wong JY. (2000). Reduced BDNF mRNA expression in the

Hughes AJ, Daniel SE, Kilford L& Lees AJ. (1992). Accuracy of clinical diagnosis of

Hullmann K, Sammer G & Reuter I. (2004). Training of executive functions in Parkinson`s

Johnson AM, Pollard CC, Vernon PA, Tomes JL& Jog MS. (2005) Memory perception and strategy use in Parlinson`s disease. *Parkinsonism Relat. Disord*. 2005; 1, p. 111-115 Kleim JA, Jones TA& Schallert T. (2003). Motor enrichment and the induction of plasticity

idiopathic Parkinson`s disease. A clinicopathological study of 100 cases*. JNNP*, 55,

catching than tapping: gait in the elderly as a complex cognitive task. Exp Brain

memory capacity explains reasoning ability and a little bit more. *Intelligence*; 30, p.

dopaminergic medication. *Neuropsycholog*y, 14, p. 247-253

summary index score. *Age and Ageing*; 26, p. 1757-1769

concussion. *Perceptual and Motor Skills*, 44, p. 367-373.

oxidative exercise in mice. *Nature* 421 (6919), p. 182-187

disease. *Medimont International Proceedings*, p. 143-148.

before and after brain injury. *Neurochem Res,* 28, p. 1757-1769

Parkinson`s disease substantia nigra. *Exp. Neurol*; 166,p. 127-135

*Neuroscience Research*, 39,p. 147-165.

p. 1-6.

261-288.

17,p. 427-442.

p. 181-184

Res. 164 (4); p. 541-8.

Parkinson`s Disease rating scale. In: Fahn S, Marsden CD, Goldstein M et al. Eds. Recent developments in Parkinson`s disease II. New York. McMillan; p. 153-163 Folstein MF& Folstein SE. (1975). ""Mini Mental State". A practical method for grading the cognitive state of patients for the clinician*. J Psychiat Res* 12, p. 189-198. Fournet N, Moreaud O, Roulin J, Naegele B & Pellat J. (2000). Working memory functioning

in medicated Parkinson`s disease patients and the effect of withdrawal of

Questionnaire (PDQ 39): Development and validation of a Parkinson`s disease


Brand M, Kalbe E &Kessler J. (2002). Test zum kognitiven Schätzen (TKS). Göttingen:

Burgess PW& Shallice T. (1996) Response suppression, initiation and strategy use following

Calabresi P, Picconi B, Parnetti L &Di Filippo M. (2006). A convergent model for cognitive

Caradoc-Davies TH, Weatherall M & Dixon GS. (1992). Is the prevalence of Parkinson´s disease in New Zealand really changing? *Acta Neurol Scand* 86, p. 40-44 Carpenter PA, Just MA &Reichle ED. ( 2000). Working memory and executive function: Evidence from neuroimaging. *Curr opinion in Neurobiol,,* 10,p. 195-199. Carter S & Ramsey MM. (2002). A critical analysis of the role of growth hormone and IGF-I

Chen RC, Chang SF, Su CL, Chen TH, Yen MF, Wu HM, Chen ZY & Liou HH (2001).

Churchill JD, Calvez R, Colcombe S, Swain RA, Kramer AF & Greenough WT. (2002). .Exercise, experience and the aging brain. *Neurobiol of aging,* 23,p. 941-955. Colman KSF, Koerts J, Beilen M, Leenders KL, Post WJ& Bastiaanse R. (2009). The impact of

Colcombe S&Kramer AF. (2003). Fitness effects on cognitive function of older adults: a meta-

Colcombe S, Erickson KI&Raz N. (2003). Aerobic fitness reduces brain tissue loss in aging

Colcombe S, Erickson KI, Scalf PE, Kim JS, Prakash R, McAuley E, Elavsky S, Marquez DX,

Colcombe SJ, Kramer AF, Erickson KI, McAuley E, Cohen NJ, Webb A, Jerome GJ &

Colman KS, Koerts J, van Beilen M, Leenders KL, Post WJ & Bastiaanse R. (2009). The impact

Douglas J. Gelb, Eugene Oliver, Sid Gilman (1999) Diagnostic Criteria for Parkinson Disease.

Dujardin K, Duhamel A, Delliaux M, Thomas-Antérion C, Destée A & Defebvre L. (2010).

Elgh E, Domello¨ f M, Linder J, Edstrom M, Stenlund H & ForsgrenL. (2009). Cognitive

Emre M, Aarsland D, Albanese A, Byrne EJ, Deuschl G, De Deyn PP, Durif F, Kulisevsky J,

Prevalence, incidence and mortality of PD: A door-to-door survey in Ilna county,

executive functions on verb production in patients with Parkinson`s disease. *Cortex*

Hu L& Kramer AF. (2006). Aerobic exercise training increases brain volume in

Marquez DX. (2004). Cardiovascular fitness, cortical plasticity, and aging. *Proc Natl* 

of executive functions on verb production in patients with Parkinson's

Cognitive complaints in Parkinson`s disease. Its relationship with objective

function in early Parkinson's disease: a populationbased study. *Eur J Neurol* 16, p.

van Laar T, Lees A, Poewe W, Robillard A, Rosa MM, Wolters E, Quarg P, Tekin S& Lane R. (2004). Rivastigmine for Dementia associated with Pakinson`s disease.

dysfunctions in Parkinson's disease: the critical dopamine-acetylcholine synaptic

frontal lobe lesions. Neuropsychologia, 34(4), p.263-72

in aging and lifespan. *Trends genet,* 18 (6), p. 295-301

balance. *Lancet Neurol,* 5(11), p.974-83

Taiwan. *Neurol,* 57, p. 679-1686

*Acad Sci* 10(9), p. 3316-3321.

disease.*Cortex,* 45(8), p. 930-42.

*N Engl J Med,* 351, p. 2509-18.

cognitive decline. J Neurol. 257 (1), p. 79-84

*Arch* Neurol, 56, p. 33-39

1278–1284.

analytic study. *Psychol Sci* , 14, p. 125-130

humans. *J Gerontol A Biol Sci Med Sci* 58A, p. 176-180.

Aging Humans*. J Gerontology* 61A (11), p. 1166-1170.

45, p. 930-942.

Hogrefe.


Cognitive Rehabilitation in Parkinson's Disease

*Neurochenmistry international* 38, 17-23

activity. *Clin Geriatr Med* 26,p. 75-87

*Psychiatr* , 141, p. 1356-1364

Germany.

p. 387–391.

p. 1657-1661

39.46.

2805-2814

298 (1089), p. 199-209.

*and Cognition*, 9, 313–323.

Using Neuropsychological Training, Transfer Training and Sports Therapy 285

Rodák Z, Kaneko T, Tahara S, Nakamoto H, Pucsok J& Sasvari M. (2001). Regular exercise

Rolland Y, v Khan GA & Vellas B. (2010) Healthy brain aging: Role of exercise and physical

Rosen WG, Mohs RC& Davis KL (1984): A new rating scale for Alzheimer's disease. *Am J* 

Royal College of Physicians, the Intercollegiate Stroke Working Party. (2008). National

Sammer, G, Reuter, I., Hullmann, K., Kaps, M., & Vaitl, D. (2006). Training of executive.

Schaaf, A., Kessler, J., Grond, M.&Fink, G.R. (1994). Memo-Test. Hogrefe, Goettingen,

Schrag A, Ben-Shlomo Y& Quinn NP (2000). Cross sectional prevalence survey of idiopathic Parkinson´s disease and Parkinsonism in London*. BMJ,* 321, p. 21-22 Shepherd RB. (2001). Exercise and training to optimise functional motor performance in

Shallice T. (1982). Specific impairments of planning. Philos Trans R Soc Lond B Biol Sci 25,

Shimamura, A. P. (2000). Toward a cognitive neuroscience of metacognition. *Consciousness* 

Sinforiani, E., Banchieri, L., Zucchella, C., Pacchetti, C. & Sandrini, G. (2004). Cognitive

Smith EE, Jonides J. (1999). Storage and executive processes in the frontal lobes. *Science,* 283,

Spielberger CD, Gorsuch RL, Lushene RE. (1970): STAI; Manual for the State-Trait-Anxiety-

Talwalker S.: Assessment of AD with the ADAS-cog.(1996) . *J. Geriatr. Psychiat*. *Neurol.*, 9, p.

Teri L, Gibbons LE & McCurry SM (2003). Exercise plus behaviour management in patients

Trejo JL, Carro E, Torres-Aleman EL. (2001). Circulating insulin-like growth factor mediates

Willis SL, Tennstedt S, Marsiske M.Ball K, Ekias J, Koepke KM, Morris JN, Rebok GW,

Wilson BA, Evans JJ, Emslie H,Alderman N & P. Burgess. (1998). The development of an

Yogev G, Giladi N, Peretz C, Springer S, Simon ES & Hausdorff JM. (2005). Dual tasking,

with Alzheimer disease: A randomised controlled trial. *JAMA* 290 (15), p. 2015-2022

exercise-induced increases in the number of new neurons in the adult

Unverzagt FW, Stoddard AM Wright E, Active Study group. Long-term effects of cognitive training on everyday functional outcomes in elder adults. *JAMa,* ; 296, p.

ecologically valid test for assessing patients with a dysexecutive syndrome.

gait rhythmicity, and Parkinson`s Disease: which aspects of gait are attention

rehabilitation in Parkinson's disease. *Archives of Gerontology and Geriatrics:* Suppl 9,

clinical guideline for stroke. Third edition. London: RCP

functions in Parkinson's disease. *J Neurol Sci*, 248, p. 115–119.

stroke: driving neural reorganisation. *Neural Plast* 8, p. 121-129

Inventory. Consulting Psychologist Press, Palo Alto.

hippocampus, *J Neurosci* 25, p. 1628-1634

*Neuropsychol Rehabilitation 8 (3),* p: 213–228.

demanding? *Eur J Neurosci*, 22(5), p. 1248-56.

improves cognitive function and decreases oxidative damage in the rat brain.


Kramer AF, Hahn S, Cohen NJ, Banich MT, McAuley E, Harisson CR Chason J, Vakil E,

Kramer AF, Colcombe SJ, Erickson KI, Paige P. (2006). Fitness training and the brain: From

Larson EB, Wang L, Bowen JD, McCormick WC, Teri L, Crane P &Kukull W. (2006). Exercise

Laurin D., Verreault R., Lindsay J, MacPherson K& Rockword K. (2001). Physical Activity

*Lehrl, S. (1989).* Mehrfach-Wortschatz-Intelligenztest: MWT-B. Perimed Fachbuch-

Leverenz JB, Quinn JF, Zabetian C, Zhang Jing, Montine K S& Montine T J. (2009). Cognitive

Levy G, Tang MX, Louis ED, et al. (2002) The association of incident dementia with

Lewis SJG, SlaboszA, Robbins TW, Barker RA & Owen AM. (2005). Dopaminergic basis for

Logsdon RG, McCurry SM, &Teri L. (2005). A home health care approach to exercise for

Norris, G. & Tate, R.L. (2000). The behavioural assessment of the dysexecutive syndrome

Marinus J, Visser N, Verwey FRJ, Middelkoop HAM, Stiggelbout AM &van Hilten JJ. (2003). Assessment of cognition in Parkinson`s disease. *Neurology,* 61, p. 1222-1228 Mc Curry SM, Gibbons LE, Logsdon RG, Vitiello MV&Teri L. (2005). Nighttime Insomnia

Miller BT & D'Esposito M (2005). "Searching for "the top" in top-down control". *Neuron,* 48

Miller EK & Cohen JD (2001). "An integrative theory of prefrontal cortex function". *Annu Rev* 

Nelles G. (2004). Cortical reorganisation – effects of intensive therapy. *Restor Neurol Neurosci*;

Norman DA,&Shallice T (2000). "(1980) Attention to action: Willed and automatic control of behaviour". In Gazzaniga MS. *Cognitive neuroscience: a reader*. Oxford: Blackwell

Reitan, R. M. (1958) Validity of the Trail Making Test as an indicator of organic brain

Reuter I, Engelhardt M, Freiwald J& Baas H. (1999) Therapeutic value of exercise training in

Parkinson J. An essay on the shaking palsy. 1817London: Sherwood Neely and Jones

damage. *Perceptual and Motor Skills* 8. 271-276

Parkinson`s disease*. Med & Sci Sports Exerc.* 9: 1544-1549.

Lezak MF. (1995). Neuropsychological assessment. Oxford: Oxford University Press.

persons with Alzheimer disease. *Care Manage J*, 6 (2), p. 90-97

Georgia Atlanta GA: Georgia Institute of Technology.

age or older. *Ann Intern Med,* 144, p. 73-81.

mortality in PD*. Neurology,* 59, p. 1708–1713.

Verlagsgesellschaft mbH, Erlangen

*Neuropsychologica*, 43, p. 823-832

*Rehabilitation,* 10 (1),p. 33-45.

*Am Geriatr Soc,* 53, p. 793-802.

*Neurosci,* 24 (1), p. 167–202

(4),p. 535–8.

22, p. 239-244

function. *Nature*, 400, p. 418-419

58, p. 498-504

9,p. 903-912

Bardell L, Boileau RA & Colcombe A. (1999). Ageing, fitness and neurocognitive

molecules to Minds. *Proceedings of the 206 Cognitive Aging Conference*, Atlanta,

is associated with reduced risk for incident dementia among persons 65 years of

and Risk of Cognitive Impairment and Dementia in elderly persons. *Arch Neurol* ;

impairment and Dementia in Patients with Parkinson Disease. *Curr Top Med Chem,*

deficits in working memory but not attentional set-shifting in Parkinson`s disease.

(BADS): ecological, concurrent and construct validity. *Neuropsychological* 

Treatment and education for Alzheimner`s disease: A randomised controlled trial*. J* 


**Mobile Systems as a Challenge** 

Laura Pastor-Sanz, Mario Pansera, Jorge Cancela,

 *Spain* 

**– The Case of Parkinson's Disease** 

**for Neurological Diseases Management** 

Matteo Pastorino and María Teresa Arredondo Waldmeyer *Life Supporting Technologies, Universidad Politécnica de Madrid* 

Nowadays the importance of bio-medical engineering and mobile applications for healthcare is amazingly growing. During the last decades many devices and technological solutions have become available on the market and the interest in applying those technologies to the treatment of several kinds of pathologies has consequently increased. This chapter addresses the problem of continuous monitoring of patients affected by Parkinson's Disease (PD) and proposes a set of technologies to improve the following and

PD is a neurodegenerative disorder of the central nervous system that affects motor skills and speech (Tolosa, 1998). The primary biochemical abnormality in PD is a deficiency of dopamine due to degeneration of neurons in the substantia nigra pars compact (D. G. Standaert & Young, 2001). The characteristic motor features of the disease include bradykinesia (i.e. slowness of movement), tremor, rigidity (i.e. resistance to externally imposed movements), flexed posture, postural instability and freezing of gait. Furthermore, PD is usually characterised by the loss of normal prosody of the speech (Darkins et al.,

According to the World Health Organisation [WHO], 2002), there are more than six million people worldwide affected by PD. The syndrome typically appears around the age of 60. It affects Europeans and North Americans more often than Asians or Africans and it is more common in men than in women. PD affects about 2% of the population over the age of 65 years, figure that is expected to double by 2020 (de Lau & Breteler, 2006). For those reasons, PD poses a significant public health burden, which is likely to increase in the coming years. Annual medical care, including doctors' visits, physical therapies and treatment for cooccurring illnesses -such as depression- is estimated at \$2,000 to \$7,000 for people in early stages of the disease, and it is probably much higher for advanced stages. Surgical treatments for PD can cost \$25,000 or more. As the disease progresses, institutional care at an assisted-living facility or nursing home may be required, and the related costs can exceed

Technology in general and specifically ICT might be an affordable alternative for PD's patients' treatment and management. The development of platforms for remote health

**1. Introduction** 

1988).

management of such subjects.

\$100,000, per person annually.

Zigmond AS& Snaith RP. (1983). The Hospital Anxiety and Depression Scale. *Acta Psychiatr.Scand,* 67, p. 361-370 **14** 

## **Mobile Systems as a Challenge for Neurological Diseases Management – The Case of Parkinson's Disease**

Laura Pastor-Sanz, Mario Pansera, Jorge Cancela, Matteo Pastorino and María Teresa Arredondo Waldmeyer *Life Supporting Technologies, Universidad Politécnica de Madrid Spain* 

## **1. Introduction**

286 Diagnostics and Rehabilitation of Parkinson's Disease

Zigmond AS& Snaith RP. (1983). The Hospital Anxiety and Depression Scale. *Acta* 

Nowadays the importance of bio-medical engineering and mobile applications for healthcare is amazingly growing. During the last decades many devices and technological solutions have become available on the market and the interest in applying those technologies to the treatment of several kinds of pathologies has consequently increased. This chapter addresses the problem of continuous monitoring of patients affected by Parkinson's Disease (PD) and proposes a set of technologies to improve the following and management of such subjects.

PD is a neurodegenerative disorder of the central nervous system that affects motor skills and speech (Tolosa, 1998). The primary biochemical abnormality in PD is a deficiency of dopamine due to degeneration of neurons in the substantia nigra pars compact (D. G. Standaert & Young, 2001). The characteristic motor features of the disease include bradykinesia (i.e. slowness of movement), tremor, rigidity (i.e. resistance to externally imposed movements), flexed posture, postural instability and freezing of gait. Furthermore, PD is usually characterised by the loss of normal prosody of the speech (Darkins et al., 1988).

According to the World Health Organisation [WHO], 2002), there are more than six million people worldwide affected by PD. The syndrome typically appears around the age of 60. It affects Europeans and North Americans more often than Asians or Africans and it is more common in men than in women. PD affects about 2% of the population over the age of 65 years, figure that is expected to double by 2020 (de Lau & Breteler, 2006). For those reasons, PD poses a significant public health burden, which is likely to increase in the coming years. Annual medical care, including doctors' visits, physical therapies and treatment for cooccurring illnesses -such as depression- is estimated at \$2,000 to \$7,000 for people in early stages of the disease, and it is probably much higher for advanced stages. Surgical treatments for PD can cost \$25,000 or more. As the disease progresses, institutional care at an assisted-living facility or nursing home may be required, and the related costs can exceed \$100,000, per person annually.

Technology in general and specifically ICT might be an affordable alternative for PD's patients' treatment and management. The development of platforms for remote health

Mobile Systems as a Challenge for Neurological

**3. Assessment - state of the art** 

patients themselves make contact (R. Greenlaw et al., 2009).

features useful in clinical decision making (Group, 2001).

scheduled one hour appointment (R. Greenlaw et al., 2009).

types of solutions are reviewed.

**3.1 Clinical solution** 

dosing schedule.

Diseases Management – The Case of Parkinson's Disease 289

2005). Recent studies suggest that this procedure might be suited for the treatment of falls and freezing. In addition to motor symptoms, an improvement in rapid eye movement sleep in patient with PD treatment with PPN DBS has been reported (Lim et al., 2009). In the future, a tailored approach to patient's specific symptoms may be possible (Lozano, 2009).

The assessment of PD can be performed through clinical and technological methods. Both

In Europe, each neurologist or general practitioner (GP) normally cares for 50 to 800 patients with PD. The range in workload is a result of diversity both in national health systems and in the availability of clinical resources across Europe. Even at 50 patients per clinician, this represents a serious challenge to homecare monitoring for specialised conditions. PD's patients normally visit their specialised clinician or GP every 4-6 months. As a result, any changes in the patient's conditions may not be recognised for several months, unless the

In clinical practice, information about motor fluctuations is usually obtained by asking patients to recall the number of hours of ON (i.e. when medications effectively attenuate tremor) and OFF time (i.e. when medications are not effective). This kind of self–report is subject to perceptual bias (e.g. patients often have difficulty distinguishing dyskinesia from other symptoms) and recall bias. Another approach is the use of patient diaries, which can improve reliability by recording symptoms as they occur, but does not capture many of the

Certainly for PD there is the additional complication of symptoms which vary throughout the day (swinging between ON and OFF states). During the short office visit in this neurologist the patient may appear very well and he misses to report symptoms of wearing off. As a result, treatment modifications are not undertaken in time. Besides, it is disempowering for the patient to be asked to present a true picture of their disease in a pre-

The actual emergence of dyskinesias throughout the day mainly depends on the intermittent dopaminergic drug intake, even in influence by timing and quantity of each individual dose of levodopa. While other phenomena, such as delayed response or no-response depends also on stress, food intake and many other factors. In this case, patients will greatly benefit from quantitative objective assessment of their motor status in daily life in relation to the

In an attempt to solve these problems and to find more objective assessment, several rating scales have been designed and used. Among them, the Unified Parkinson's Disease Rating Scale (UPDRS) is the most widely used (Goetz et al., 2004). This rating tried to quantify selected symptoms and signs of Parkinsonism in a 5-points scoring system (with 0 for no signal and 4 for a marked severity of the sign). Unfortunately, the use of the UPDRS scale, like any other semi-objective rating scale presents some limitations like intra and inter observer inconsistencies. Besides, it can be time consuming and ca be biased by subjectivity issues related to historical information. Moreover, the pattern and severity of PD symptoms may vary considerably during the day, while clinical rating scales only provide moment-to moment assessment; and finally, measurements of motor fluctuations made in the clinic

status monitoring, the qualitative and quantitative assessment and treatment personalization for people suffering from neurodegenerative diseases is expecting to provide in the future a remarkable improvement in patients' management as well as a substantial cutting-off of the economic burden generated by the disease. New technologies allow monitoring the evolution of the disease through the employment of a wide range of wearable and user-friendly micro-sensors. Moreover, the last advances in data processing and data mining algorithms is bound to provide more accurate information about the diverse aspects of PD evolution. Finally, it is important to highlight the huge potential in costs reduction that such platforms could yield. Furthermore, it is worth mentioning that the reduction in costs of hospitalization and treatment represents an attractive asset for the market forces involved in the development of biomedical applications.

## **2. Treatment**

Current clinical treatment of Parkinson's disease is performed through ersatz dopamine administration or by using Deep Brain Stimulation (DBS) (Singh et al., 2007).

#### **2.1 Dopamine treatment**

Current therapy is based on augmentation or replacement of dopamine, using the biosynthetic precursor levodopa or drugs that activate dopamine receptors. These therapies are effective at the beginning of treatment. However, after a variable period of time, this initially excellent response is complicated by the appearance of MRCs. Complications include wearing-off, the abrupt loss of efficacy at the end of each dosing interval and diskinesias (de la Fuente-Fernández et al., 2004). Wearing off and dyskinesias produce substantial disability and frequently interfere with medical therapies (A.E. Lang & A.M. Lozano, 1998). Usually, motor fluctuations appear first, as a shortening of the initially smooth and long lasting dopaminergic response. In the typical case, few hours after drug intake the patients start to realize the re-emergence of signs and symptoms of the disease. This is known as end of dose deterioration or wearing off. This may happen several times a day; therefore the patient may actually several hours per day in the off state (Lees, 1989).

#### **2.2 Deep brain stimulation**

DBS is a surgical treatment involving the implantation of a medical device called a brain pacemaker, which sends electrical impulses to specific parts of the brain (Vaillancourt et al., 2003). The introduction of DBS as a therapeutic tool for advanced PD has revolutionised the clinical management of this condition. Due to its safety profile and efficacy, DBS has evolved from a last-resort therapeutic option to a modality that is now routinely offered to patients. Over the years, surgical candidates and the outcome expected with this procedure has become well established (Deuschl et al., 2006). Overall improvement that might be expected with surgery is similar to that provided by levodopa without the associated involuntary movements (Group, 2001). As the diseases progresses, however, nondopaminergic symptoms (gait, postural instability, sleep disorders and depression, among others) become more prominent, leading to a significant increase in morbidity. To overcome some of the problems, the use of different surgical targets has been advocated. Perhaps the most promising application of DBS on this regards involves the use of Pedunculopontine Nucleus (PPN) stimulation for the treatment of gait and postural instability (Mazzone et al., 2005). Recent studies suggest that this procedure might be suited for the treatment of falls and freezing. In addition to motor symptoms, an improvement in rapid eye movement sleep in patient with PD treatment with PPN DBS has been reported (Lim et al., 2009). In the future, a tailored approach to patient's specific symptoms may be possible (Lozano, 2009).

## **3. Assessment - state of the art**

The assessment of PD can be performed through clinical and technological methods. Both types of solutions are reviewed.

## **3.1 Clinical solution**

288 Diagnostics and Rehabilitation of Parkinson's Disease

status monitoring, the qualitative and quantitative assessment and treatment personalization for people suffering from neurodegenerative diseases is expecting to provide in the future a remarkable improvement in patients' management as well as a substantial cutting-off of the economic burden generated by the disease. New technologies allow monitoring the evolution of the disease through the employment of a wide range of wearable and user-friendly micro-sensors. Moreover, the last advances in data processing and data mining algorithms is bound to provide more accurate information about the diverse aspects of PD evolution. Finally, it is important to highlight the huge potential in costs reduction that such platforms could yield. Furthermore, it is worth mentioning that the reduction in costs of hospitalization and treatment represents an attractive asset for the

Current clinical treatment of Parkinson's disease is performed through ersatz dopamine

Current therapy is based on augmentation or replacement of dopamine, using the biosynthetic precursor levodopa or drugs that activate dopamine receptors. These therapies are effective at the beginning of treatment. However, after a variable period of time, this initially excellent response is complicated by the appearance of MRCs. Complications include wearing-off, the abrupt loss of efficacy at the end of each dosing interval and diskinesias (de la Fuente-Fernández et al., 2004). Wearing off and dyskinesias produce substantial disability and frequently interfere with medical therapies (A.E. Lang & A.M. Lozano, 1998). Usually, motor fluctuations appear first, as a shortening of the initially smooth and long lasting dopaminergic response. In the typical case, few hours after drug intake the patients start to realize the re-emergence of signs and symptoms of the disease. This is known as end of dose deterioration or wearing off. This may happen several times a day; therefore the patient may actually several hours per day in the off state (Lees, 1989).

DBS is a surgical treatment involving the implantation of a medical device called a brain pacemaker, which sends electrical impulses to specific parts of the brain (Vaillancourt et al., 2003). The introduction of DBS as a therapeutic tool for advanced PD has revolutionised the clinical management of this condition. Due to its safety profile and efficacy, DBS has evolved from a last-resort therapeutic option to a modality that is now routinely offered to patients. Over the years, surgical candidates and the outcome expected with this procedure has become well established (Deuschl et al., 2006). Overall improvement that might be expected with surgery is similar to that provided by levodopa without the associated involuntary movements (Group, 2001). As the diseases progresses, however, nondopaminergic symptoms (gait, postural instability, sleep disorders and depression, among others) become more prominent, leading to a significant increase in morbidity. To overcome some of the problems, the use of different surgical targets has been advocated. Perhaps the most promising application of DBS on this regards involves the use of Pedunculopontine Nucleus (PPN) stimulation for the treatment of gait and postural instability (Mazzone et al.,

market forces involved in the development of biomedical applications.

administration or by using Deep Brain Stimulation (DBS) (Singh et al., 2007).

**2. Treatment** 

**2.1 Dopamine treatment** 

**2.2 Deep brain stimulation** 

In Europe, each neurologist or general practitioner (GP) normally cares for 50 to 800 patients with PD. The range in workload is a result of diversity both in national health systems and in the availability of clinical resources across Europe. Even at 50 patients per clinician, this represents a serious challenge to homecare monitoring for specialised conditions. PD's patients normally visit their specialised clinician or GP every 4-6 months. As a result, any changes in the patient's conditions may not be recognised for several months, unless the patients themselves make contact (R. Greenlaw et al., 2009).

In clinical practice, information about motor fluctuations is usually obtained by asking patients to recall the number of hours of ON (i.e. when medications effectively attenuate tremor) and OFF time (i.e. when medications are not effective). This kind of self–report is subject to perceptual bias (e.g. patients often have difficulty distinguishing dyskinesia from other symptoms) and recall bias. Another approach is the use of patient diaries, which can improve reliability by recording symptoms as they occur, but does not capture many of the features useful in clinical decision making (Group, 2001).

Certainly for PD there is the additional complication of symptoms which vary throughout the day (swinging between ON and OFF states). During the short office visit in this neurologist the patient may appear very well and he misses to report symptoms of wearing off. As a result, treatment modifications are not undertaken in time. Besides, it is disempowering for the patient to be asked to present a true picture of their disease in a prescheduled one hour appointment (R. Greenlaw et al., 2009).

The actual emergence of dyskinesias throughout the day mainly depends on the intermittent dopaminergic drug intake, even in influence by timing and quantity of each individual dose of levodopa. While other phenomena, such as delayed response or no-response depends also on stress, food intake and many other factors. In this case, patients will greatly benefit from quantitative objective assessment of their motor status in daily life in relation to the dosing schedule.

In an attempt to solve these problems and to find more objective assessment, several rating scales have been designed and used. Among them, the Unified Parkinson's Disease Rating Scale (UPDRS) is the most widely used (Goetz et al., 2004). This rating tried to quantify selected symptoms and signs of Parkinsonism in a 5-points scoring system (with 0 for no signal and 4 for a marked severity of the sign). Unfortunately, the use of the UPDRS scale, like any other semi-objective rating scale presents some limitations like intra and inter observer inconsistencies. Besides, it can be time consuming and ca be biased by subjectivity issues related to historical information. Moreover, the pattern and severity of PD symptoms may vary considerably during the day, while clinical rating scales only provide moment-to moment assessment; and finally, measurements of motor fluctuations made in the clinic

Mobile Systems as a Challenge for Neurological

Diseases Management – The Case of Parkinson's Disease 291

The accuracy of measurements of the parameters above described depends on several technical features that are often in conflict with other needs such as usability, wearability,

In Table 2 a description of these desirable properties along with their conflicts is presented.

battery or storage capacity. **Smart sensor** Sensors possessing many characteristics are often bigger in

**Sensor storage capacity** Due to a limit in storage capacity, sensors have to upload

**Sensor processing capability** Because sensors do not often have large processing

communication channel. **Sensor communication range** Whilst sensors are only able to communicate over short

Most of the research work carried out in the field of PD monitoring focuses on the assessment of the motor status of PD's patients. During the last decade, many research groups have been trying to develop a system able to objectively quantify the severity of the motor disturbances using motion sensors (Patel et al., 2010; 2009; 2007). An important number of these studies is based on the study of various parameters of motor behaviour, in particular features related to the gait (R. Greenlaw et al., 2009; Salarian et al., 2007; 2004). Other studies focused on the identification of ON/OFF fluctuations through the assessment of tremor (Van Someren et al., 1993), dyskinesias (Keijsers et al., 2003; 2000) and bradykinesia (Papapetropoulos et al., 2010). Some groups are also committed to use

Additionally, in the literature there are examples of remote monitoring and patient management for PD (Tindall & Huebner, 2009), as well as the use of telematic services to facilitate the performance of motor tests remotely (Das, 2010; Dorsey et al., 2010; Giansanti et

Even though many advances have been done in the last years, it must be said that there is still a lack of an all-inclusive system able to provide reliable assessment of the status of PD patients being at the same time economically affordable. In particular it is crucial to provide:

size, expensive and consume more power

Therefore it is important to have an efficient

data frequently to the data personal server. So it is important to employ a good wireless communication technology that does not drain excessive power from the

capability, they may not be able to process all data before the upload to the personal server. This means that large amount of raw data should be stored and eventually sent.

range, it is crucial to define a specific radius of action.

especially for portability and mobility matters. However, small sensors may not have enough room for long-lasting

technical feasibility and the social acceptance of the devices used by the subjects.

**Small sensor** The size of sensor is definitely an important factor,

**Desirable properties Conflict** 

sensors.

Table 2. Wearable sensors desirable properties & conflicts

electromyogram (EMG) or voice analysis (Kimura et al., 2007).

al., 2008; Westin et al., 2010).

**3.2.1 Systems for Parkinson's disease monitoring** 

may not accurately reflect the actual functional disability experienced by the patient at home.

The accurate assessment of speech quality is a major research problem that has attracted attention in the field of speech communications for many years. Subjective quality measures given by professional personnel who have received special assessment training are necessarily time consuming and costly (Lingyun Gu et al., 2005).

## **3.2 Technological solution: sensors for motion analysis**

Over the past decades various technologies, methodologies and systems have been proposed for the monitoring and the assessment of the Parkinson's disease. A significant number of studies investigated various parameters of the gait of PD patients. Others focused on the evaluation and quantification of the patients' motor status and various disease symptoms by the use of computerized motion tests (e.g. handwriting, inserting pegs, and games). Table 1 describes some features of the human motion, as well as the characteristics which can be measured through the use of wearable sensors.


Table 1. Parkinson's disease – wearable sensors for human motion related measurements

may not accurately reflect the actual functional disability experienced by the patient at

The accurate assessment of speech quality is a major research problem that has attracted attention in the field of speech communications for many years. Subjective quality measures given by professional personnel who have received special assessment training are

Over the past decades various technologies, methodologies and systems have been proposed for the monitoring and the assessment of the Parkinson's disease. A significant number of studies investigated various parameters of the gait of PD patients. Others focused on the evaluation and quantification of the patients' motor status and various disease symptoms by the use of computerized motion tests (e.g. handwriting, inserting pegs, and games). Table 1 describes some features of the human motion, as well as the characteristics

Speed of Locomotion

Variability of the gait Motion sensor

Speed Motion sensor

Angle Amplitude Motion sensor

Motion sensor Frequency

Severity Motion sensor

Microphone EPE

Motion sensor Motion sensor Motion sensor

Motion sensor

**Features Characteristics Sensor** 

Rigidity of legs **Posture** Trunk inclination Motion sensor

> Length of Step Step Frequency

**Hand Movement** Speed Motion sensor

Stride

Amplitude

Duration Asymmetry **Fall** Fall detection Motion sensor **Freezing of Gait** Leg movement analysis Motion sensor

Duration

Severity

Duration

Asymmetry

Table 1. Parkinson's disease – wearable sensors for human motion related measurements

necessarily time consuming and costly (Lingyun Gu et al., 2005).

**3.2 Technological solution: sensors for motion analysis** 

which can be measured through the use of wearable sensors.

home.

**Gait** 

**Leg movement** 

**Tremor** 

**(LID)** 

**Bradykinesia** 

**Levodopa-Induced Dyskinesia** 

**Aphasia** Pitch

The accuracy of measurements of the parameters above described depends on several technical features that are often in conflict with other needs such as usability, wearability, technical feasibility and the social acceptance of the devices used by the subjects. In Table 2 a description of these desirable properties along with their conflicts is presented.

**Desirable properties Conflict Small sensor** The size of sensor is definitely an important factor, especially for portability and mobility matters. However, small sensors may not have enough room for long-lasting battery or storage capacity. **Smart sensor** Sensors possessing many characteristics are often bigger in size, expensive and consume more power **Sensor storage capacity** Due to a limit in storage capacity, sensors have to upload data frequently to the data personal server. So it is important to employ a good wireless communication technology that does not drain excessive power from the sensors. **Sensor processing capability** Because sensors do not often have large processing capability, they may not be able to process all data before the upload to the personal server. This means that large amount of raw data should be stored and eventually sent. Therefore it is important to have an efficient communication channel. **Sensor communication range** Whilst sensors are only able to communicate over short range, it is crucial to define a specific radius of action.

Table 2. Wearable sensors desirable properties & conflicts

## **3.2.1 Systems for Parkinson's disease monitoring**

Most of the research work carried out in the field of PD monitoring focuses on the assessment of the motor status of PD's patients. During the last decade, many research groups have been trying to develop a system able to objectively quantify the severity of the motor disturbances using motion sensors (Patel et al., 2010; 2009; 2007). An important number of these studies is based on the study of various parameters of motor behaviour, in particular features related to the gait (R. Greenlaw et al., 2009; Salarian et al., 2007; 2004).

Other studies focused on the identification of ON/OFF fluctuations through the assessment of tremor (Van Someren et al., 1993), dyskinesias (Keijsers et al., 2003; 2000) and bradykinesia (Papapetropoulos et al., 2010). Some groups are also committed to use electromyogram (EMG) or voice analysis (Kimura et al., 2007).

Additionally, in the literature there are examples of remote monitoring and patient management for PD (Tindall & Huebner, 2009), as well as the use of telematic services to facilitate the performance of motor tests remotely (Das, 2010; Dorsey et al., 2010; Giansanti et al., 2008; Westin et al., 2010).

Even though many advances have been done in the last years, it must be said that there is still a lack of an all-inclusive system able to provide reliable assessment of the status of PD patients being at the same time economically affordable. In particular it is crucial to provide:

Mobile Systems as a Challenge for Neurological

Diseases Management – The Case of Parkinson's Disease 293

Fig. 2. Kinetisense, Cleveland Medical Devices Inc. system for PD monitoring based on

Fig. 3. At Home Telemonitoring Device (AHTD), Intel Corporation system for PD

monitoring based on speech processing (Tsanas et al., 2010b)

Intel Corporation has developed a system called At home Telemonitoring Device (AHTD), based on nonlinear speech single processing methods around the use of discrete variational integrator. It could be used to perform speech analysis detect abnormalities and telemonitor

motor sensors and electromyography (Jovanov et al., 2001)

neurological disorders with voice singles.

*Intel Corporation* 


## **3.2.2 Available systems in the market/research**

There are several products produced by certain research groups or commercially available for the assessment of PD. Some examples follow:

### *Cleveland Medical Devices*

Cleveland Medical Devices Inc. commercialises Kinesia, a compact wireless system for monitoring the severity of PD motor symptoms. The system includes miniature motion sensors worn on the hand and it wirelessly transmits motor symptom information to a personal computer. Data is collected while patients follow computer based video instructions.

Fig. 1. Kinesia, Cleveland Medical Devices Inc. system for PD monitoring based on motor sensors (Jovanov et al., 2001)

Cleveland Medical Devices Inc. also commercialises Kinetisense, a compact wireless system for monitoring gait, posture or upper limb movement. The system integrates two channel Electromyography (EMG) and three orthogonal accelerometers and gyroscopes collecting data on three dimensional movements. The system is wearable and lightweight and is wirelessly connected to a computer for real-time data transmission. As alternative, data can be stored on a memory card for 12 hours recording and transmitted asynchronously. The system can be linked to different software tools for movement and posture analysis, detection of slips and falls and for the performance and monitoring of rehabilitation exercises.

Fig. 2. Kinetisense, Cleveland Medical Devices Inc. system for PD monitoring based on motor sensors and electromyography (Jovanov et al., 2001)

#### *Intel Corporation*

292 Diagnostics and Rehabilitation of Parkinson's Disease

An effective evaluation of PD symptoms through monitoring and testing routines while

 A personalised profile of the patient allowing the correlation between those factors affecting the severity of symptoms (i.e. medication schedule and meals) and the

 The clinician with a system able to manage more efficiently the patient by providing timely indications on the effectiveness of the therapy and suggestions on therapy

There are several products produced by certain research groups or commercially available

Cleveland Medical Devices Inc. commercialises Kinesia, a compact wireless system for monitoring the severity of PD motor symptoms. The system includes miniature motion sensors worn on the hand and it wirelessly transmits motor symptom information to a personal computer. Data is collected while patients follow computer based video

Fig. 1. Kinesia, Cleveland Medical Devices Inc. system for PD monitoring based on motor

falls and for the performance and monitoring of rehabilitation exercises.

Cleveland Medical Devices Inc. also commercialises Kinetisense, a compact wireless system for monitoring gait, posture or upper limb movement. The system integrates two channel Electromyography (EMG) and three orthogonal accelerometers and gyroscopes collecting data on three dimensional movements. The system is wearable and lightweight and is wirelessly connected to a computer for real-time data transmission. As alternative, data can be stored on a memory card for 12 hours recording and transmitted asynchronously. The system can be linked to different software tools for movement and posture analysis, detection of slips and

not interfering with the patient daily life.

**3.2.2 Available systems in the market/research** 

for the assessment of PD. Some examples follow:

evolution of the disease.

changes.

instructions.

*Cleveland Medical Devices* 

sensors (Jovanov et al., 2001)

Intel Corporation has developed a system called At home Telemonitoring Device (AHTD), based on nonlinear speech single processing methods around the use of discrete variational integrator. It could be used to perform speech analysis detect abnormalities and telemonitor neurological disorders with voice singles.

Fig. 3. At Home Telemonitoring Device (AHTD), Intel Corporation system for PD monitoring based on speech processing (Tsanas et al., 2010b)

Mobile Systems as a Challenge for Neurological

to improve the physical therapy. *Federal Polytechnic School of Lausanne* 

(Salarian et al., 2007).

(Salarian, et al., 2007b)

*Boston University* 

Diseases Management – The Case of Parkinson's Disease 295

exercises based on the personal rehabilitation and training program defined for the patient and stored in the PDA. The system can connect to the hospital or to the physician through the PDA and transmit daily motion data, which can be analysed by physicians and be used

The Federal Polytechnic School of Lausanne, Switzerland, has developed a system for motion monitoring based on a portable data-logger with three body-fixed inertial sensors.

Fig. 6. a) The trunk sensor used for physical activity monitoring b) The uni-axial gyroscopes

used for the gait analysis, The Federal Polytechnic School of Lausanne, Switzerland

Fig. 7. a) Trunk sensor used for physical activity monitoring b) The uni-axial gyroscopes used for the gait analysis, The Federal Polytechnic School of Lausanne, Switzerland

The Boston University National Institute of biomedical Imagining and Bioengineering has developed a wearable-sensor system for monitoring motor function. The system is composed by a device that can be worn unobtrusively by patients in their home to automatically detect the presence and severity of movement disorders associated with PD. The onset of the OFF status is based on the motor status of the patient that can be related to

#### *Karolinska Institute*

The Karolinska Institute, Sweden, has developed a prototype test battery for evaluating fluctuation motor symptoms in PD together with a decision support system as part of the Movistar TEVAL project (Westin et al., 2010). The system is based on a handheld device with built-in mobile communication, where combined patient diaries with on-screen motor tests are implemented. The data collected from the patient are transmitted to a central system where they are analysed through Artificial Intelligence methods. Besides, it originates alerts and advice via a web interface.

Fig. 4. TEVAL project prototype, Karolinska Institute, Sweden, based on patient diaries together with on-screen motor tests

#### *Twente University*

The University of Twente, Enschede, in the Netherlands, has developed a system called SensorShoe that is a mobile gait analysis tool. It is composed by a low-power sensor node equipped with movement sensors (3D accelerometers and 2D gyroscopes) connected to a PDA which provides immediate feedback to the patient while walking and suggest physical

Fig. 5. Sensor Shoe, University of Twente, Enschede, The Netherlands. Gait analysis based on movement sensors (Kauw-A-Tjoe et al., 2007)

exercises based on the personal rehabilitation and training program defined for the patient and stored in the PDA. The system can connect to the hospital or to the physician through the PDA and transmit daily motion data, which can be analysed by physicians and be used to improve the physical therapy.

## *Federal Polytechnic School of Lausanne*

294 Diagnostics and Rehabilitation of Parkinson's Disease

The Karolinska Institute, Sweden, has developed a prototype test battery for evaluating fluctuation motor symptoms in PD together with a decision support system as part of the Movistar TEVAL project (Westin et al., 2010). The system is based on a handheld device with built-in mobile communication, where combined patient diaries with on-screen motor tests are implemented. The data collected from the patient are transmitted to a central system where they are analysed through Artificial Intelligence methods. Besides, it

Fig. 4. TEVAL project prototype, Karolinska Institute, Sweden, based on patient diaries

The University of Twente, Enschede, in the Netherlands, has developed a system called SensorShoe that is a mobile gait analysis tool. It is composed by a low-power sensor node equipped with movement sensors (3D accelerometers and 2D gyroscopes) connected to a PDA which provides immediate feedback to the patient while walking and suggest physical

Fig. 5. Sensor Shoe, University of Twente, Enschede, The Netherlands. Gait analysis based

*Karolinska Institute* 

originates alerts and advice via a web interface.

together with on-screen motor tests

on movement sensors (Kauw-A-Tjoe et al., 2007)

*Twente University* 

The Federal Polytechnic School of Lausanne, Switzerland, has developed a system for motion monitoring based on a portable data-logger with three body-fixed inertial sensors.

Fig. 6. a) The trunk sensor used for physical activity monitoring b) The uni-axial gyroscopes used for the gait analysis, The Federal Polytechnic School of Lausanne, Switzerland (Salarian et al., 2007).

Fig. 7. a) Trunk sensor used for physical activity monitoring b) The uni-axial gyroscopes used for the gait analysis, The Federal Polytechnic School of Lausanne, Switzerland (Salarian, et al., 2007b)

#### *Boston University*

The Boston University National Institute of biomedical Imagining and Bioengineering has developed a wearable-sensor system for monitoring motor function. The system is composed by a device that can be worn unobtrusively by patients in their home to automatically detect the presence and severity of movement disorders associated with PD. The onset of the OFF status is based on the motor status of the patient that can be related to

Mobile Systems as a Challenge for Neurological

**4.2 The PERFORM medical and technological vision** 

medication change proposal is sent to the patient.

Other Info

Fig. 9. PERFORM medical and technological vision

Suggest Treatment Changes

Monitor Patient

New Treatment Regime

Test Devices

**4. PERFORM system** 

**4.1 Introduction** 

Diseases Management – The Case of Parkinson's Disease 297

PERFORM is a project partially funded by the European Commission under the Seventh Framework Program, aiming at providing an innovative and reliable tool that is able to monitor and evaluate motor neurodegenerative disease patients, such as PD patients. The PERFORM project is based on the development of an intelligent closed loop system that seamlessly integrates a wide range of wearable micro-sensors constantly monitoring several motor signals of the patients. Data acquired are pre-processed by advanced knowledge processing methods, integrated by fusion algorithms to allow health professionals to remotely monitor the overall status of the patients, adjust medication schedules and personalize treatment. Personalization of treatment occurs through PERFORM's capability to keep track of the timing and doses of the medication and meals that the patient is taking.

The information gathered by the inertial sensors (accelerometers and gyroscopes) is processed by several classifiers. As a result, it is possible to evaluate and quantify the PD symptoms that the patient presents as well as analyze the gait of the patient. Based on this information, together with information derived from tests performed with tests devices (e.g. virtual reality gloves) and information about the medication and food intake, a patient specific profile is built. Next steep is to compare the patient specific profile with his evaluation during the last week and last month, checking whether his status is stable, improving or worsening. Based on that, the system analyses whether a medication change is needed-always under medical supervision- and in this case, information about the

Detect & Quantify

Symptoms & Gait Build

Disease Progress

Patient Specific

disease profile Assess

the motor status of the patient with the medications assumptions. The system involves electromyographic (EMG) and accelerometric (ACC) body worn sensors, whose signals are analysed by a system using Artificial Intelligence methods.

Fig. 8. Boston University National Institute of Biomedical Imaging and Bioengineering, PD monitoring system through motor and EMG sensors (UB, 2011)

#### **3.3 Technological solution: non autonomous home based monitoring closed loop systems**

Close loop systems are those in which stimulation parameters are adjusted according to recorded signals. Talking about neurodegenerative diseases such as PD, close loop systems imply that medication doses and timing is adjusted based on the measurement of certain biomedical signals. Some examples follow.

## **3.3.1 PERFORM project**

A sophisticated multi-paRametric system FOR the continuous effective assessment and Monitoring of motor status in Parkinson's disease and other neurodegenerative diseases research project has developed an intelligent system that monitors several motor signals of the patients that are analyzed in a medical centre by a medical professional. The system is able to propose treatment changes, based on the clinical assessment. Further explanations are provided later in this chapter (Perform, 2008).

## **3.3.2 HELP project**

The HELP project (Home-based Empowered living for Parkinson's Disease Patients) aims at developing a comprehensive system able to administer drug therapy without patient intervention, in either continuous or on-demand basis in order to manage disease progression and to mitigate PD's symptoms (Help, 2011).

It is based on inertial sensors that capture inertial information about the patient's motion and compute spatiotemporal properties and Parkinson's related symptoms. At the point of care remote supervision of the patient is performed, together with Verification of the infusion algorithm and possible modification of its parameters. An intraoral device continuously administrates dopamine agonists to the mucosa from the mouth. Besides, a subcutaneous pump receives commands adapting the infusion rate of apomorphine, a nonselective dopamine agonist.

## **4. PERFORM system**

## **4.1 Introduction**

296 Diagnostics and Rehabilitation of Parkinson's Disease

the motor status of the patient with the medications assumptions. The system involves electromyographic (EMG) and accelerometric (ACC) body worn sensors, whose signals are

Fig. 8. Boston University National Institute of Biomedical Imaging and Bioengineering, PD

**3.3 Technological solution: non autonomous home based monitoring closed loop** 

Close loop systems are those in which stimulation parameters are adjusted according to recorded signals. Talking about neurodegenerative diseases such as PD, close loop systems imply that medication doses and timing is adjusted based on the measurement of certain

A sophisticated multi-paRametric system FOR the continuous effective assessment and Monitoring of motor status in Parkinson's disease and other neurodegenerative diseases research project has developed an intelligent system that monitors several motor signals of the patients that are analyzed in a medical centre by a medical professional. The system is able to propose treatment changes, based on the clinical assessment. Further explanations

The HELP project (Home-based Empowered living for Parkinson's Disease Patients) aims at developing a comprehensive system able to administer drug therapy without patient intervention, in either continuous or on-demand basis in order to manage disease

It is based on inertial sensors that capture inertial information about the patient's motion and compute spatiotemporal properties and Parkinson's related symptoms. At the point of care remote supervision of the patient is performed, together with Verification of the infusion algorithm and possible modification of its parameters. An intraoral device continuously administrates dopamine agonists to the mucosa from the mouth. Besides, a subcutaneous pump receives commands adapting the infusion rate of apomorphine, a non-

analysed by a system using Artificial Intelligence methods.

monitoring system through motor and EMG sensors (UB, 2011)

biomedical signals. Some examples follow.

are provided later in this chapter (Perform, 2008).

progression and to mitigate PD's symptoms (Help, 2011).

**3.3.1 PERFORM project** 

**3.3.2 HELP project** 

selective dopamine agonist.

**systems** 

PERFORM is a project partially funded by the European Commission under the Seventh Framework Program, aiming at providing an innovative and reliable tool that is able to monitor and evaluate motor neurodegenerative disease patients, such as PD patients.

The PERFORM project is based on the development of an intelligent closed loop system that seamlessly integrates a wide range of wearable micro-sensors constantly monitoring several motor signals of the patients. Data acquired are pre-processed by advanced knowledge processing methods, integrated by fusion algorithms to allow health professionals to remotely monitor the overall status of the patients, adjust medication schedules and personalize treatment. Personalization of treatment occurs through PERFORM's capability to keep track of the timing and doses of the medication and meals that the patient is taking.

## **4.2 The PERFORM medical and technological vision**

The information gathered by the inertial sensors (accelerometers and gyroscopes) is processed by several classifiers. As a result, it is possible to evaluate and quantify the PD symptoms that the patient presents as well as analyze the gait of the patient. Based on this information, together with information derived from tests performed with tests devices (e.g. virtual reality gloves) and information about the medication and food intake, a patient specific profile is built. Next steep is to compare the patient specific profile with his evaluation during the last week and last month, checking whether his status is stable, improving or worsening. Based on that, the system analyses whether a medication change is needed-always under medical supervision- and in this case, information about the medication change proposal is sent to the patient.

Fig. 9. PERFORM medical and technological vision

Mobile Systems as a Challenge for Neurological

(left) System placement on the body (right)

attachment/detachment of sensors.

assessment.

Diseases Management – The Case of Parkinson's Disease 299

Fig. 11. PERFORM System prototype including accelerometers, gyroscope and data logger

All sensors transmit using Zigbee protocol to the wearable device which is located on the patients' waist thus making up a body sensor network. Special attention is given to the sensors usage and the easy set up by the patient and the caregivers. The sensor size is no bigger than a small matchbox. Sensors on the arms and legs are attached on specially designed elastic Velcro bands, which allow fixation to any wrist or ankle size. The sensors are placed inside an elastic pocket on the band, which secures it firmly on the patient body avoiding motion artefact due to cloth movement. The sensor on the trunk in placed within a zipped elastic pocket on a vest. The vest is also equipped with Velcro straps to firmly adjust the sensor on the patient chest. The selected design allows the easy wearing and

**Test Module**. It consists of a set of devices (such as virtual reality gloves, microphones or video cameras) used to record patient information, while the patient is performing specific tests, as normally done at the clinician's office during an examination. The patient wears the test devices and performs the tests as instructed from the visual interface of the Base Module (Local Base Unit). The test module records the performed activities and identifies any abnormalities, such as wrong sensor or patient position. Finally, it processes the recorded data and extracts the information about the number of taps and hand movements per

**Patient interface**. Emphasis is given in designing an easy to use interface for the patient, considering the patient motor disabilities and limited computer familiarity. The designed interface inherits the feel and tough of the phone dialling pad, and all system choices are based on it. Patient use the interface to declare their subjective estimation of their own status, to gain access to relevant disease information, to receive instructions on life-style interventions, such as medication and good intake and on the execution of tests. Moreover, PD's patients declare medication intake information, which is useful for the patient status

second, the detection of hypophonia and neutral face expression.

## **4.3 The PERFORM architecture**

The system architecture proposed to meet the previously described medical and technological vision is presented in Fig. 10. It consists of two subsystems: the patient-side subsystem and the healthcare centre subsystem.

Fig. 10. PERFORM system architecture (R. Greenlaw et al., 2009)

The patient-side subsystem is responsible for the identification and quantification of the patient symptoms and the recording of other useful information for the evaluation of the patient status. The healthcare centre subsystem evaluates the disease progression and suggests appropriate treatment and changes, based on medical knowledge acquired from published medical guidelines.

The patient-side subsystem is composed by the following modules:

**Continuous monitoring Module**. It is used to monitor the patient motor status through the day. It consists of five accelerometers and a wearable device. The wearable device processes the recorded signals and detects patient falls in real time. The sensors position was hosen after careful examination and research on the targeted disease symptoms.

It is composed of four tri-axial accelerometers used to record the accelerations of the movements at each patient extremity, one accelerometer and gyroscope (on the trunk) used to record body/chest movement accelerations and angular body velocity during trunk and body turning, and a wearable device receiving all recorded signals. The sensors' position was chosen to allow all targeted symptoms detection and quantification with the minimum number of sensors.

The system architecture proposed to meet the previously described medical and technological vision is presented in Fig. 10. It consists of two subsystems: the patient-side

> Clinician Administrator

External Resources

2

Early Wearing Off Medication Change Stability-Worsening

Professional GUI

Patient Management

Scheduler

Freezing of Gait Gait Tremor LID Activity

Action Tremor

Central Unit Communicator

The patient-side subsystem is responsible for the identification and quantification of the patient symptoms and the recording of other useful information for the evaluation of the patient status. The healthcare centre subsystem evaluates the disease progression and suggests appropriate treatment and changes, based on medical knowledge acquired from

**Continuous monitoring Module**. It is used to monitor the patient motor status through the day. It consists of five accelerometers and a wearable device. The wearable device processes the recorded signals and detects patient falls in real time. The sensors position was hosen

It is composed of four tri-axial accelerometers used to record the accelerations of the movements at each patient extremity, one accelerometer and gyroscope (on the trunk) used to record body/chest movement accelerations and angular body velocity during trunk and body turning, and a wearable device receiving all recorded signals. The sensors' position was chosen to allow all targeted symptoms detection and quantification with the minimum

PERFORM

Central Unit Information Handler

Central Hospital Unit

Repository Interoperability Manager

Frequent Falls

User DB Login Manager Account Manager Permitted Access Level

Health Care Centre Subsystem

Patient List Index of Processed Info

Alert Manager

Clinical Decision Support Systems Patient Modelling Gait On – Off LID Tremor Bradykinesia

**4.3 The PERFORM architecture** 

Logger User-Hardware Interface Fall Detector Alert Manager

published medical guidelines.

number of sensors.

Wearable Sensors

Test Devices

subsystem and the healthcare centre subsystem.

Local Base Unit

Patient-Side Subsystem

Device Controller Test Processor

Information Handler

Patient GUI

Repository Bradykinesia

Communicator

Fig. 10. PERFORM system architecture (R. Greenlaw et al., 2009)

The patient-side subsystem is composed by the following modules:

after careful examination and research on the targeted disease symptoms.

Fig. 11. PERFORM System prototype including accelerometers, gyroscope and data logger (left) System placement on the body (right)

All sensors transmit using Zigbee protocol to the wearable device which is located on the patients' waist thus making up a body sensor network. Special attention is given to the sensors usage and the easy set up by the patient and the caregivers. The sensor size is no bigger than a small matchbox. Sensors on the arms and legs are attached on specially designed elastic Velcro bands, which allow fixation to any wrist or ankle size. The sensors are placed inside an elastic pocket on the band, which secures it firmly on the patient body avoiding motion artefact due to cloth movement. The sensor on the trunk in placed within a zipped elastic pocket on a vest. The vest is also equipped with Velcro straps to firmly adjust the sensor on the patient chest. The selected design allows the easy wearing and attachment/detachment of sensors.

**Test Module**. It consists of a set of devices (such as virtual reality gloves, microphones or video cameras) used to record patient information, while the patient is performing specific tests, as normally done at the clinician's office during an examination. The patient wears the test devices and performs the tests as instructed from the visual interface of the Base Module (Local Base Unit). The test module records the performed activities and identifies any abnormalities, such as wrong sensor or patient position. Finally, it processes the recorded data and extracts the information about the number of taps and hand movements per second, the detection of hypophonia and neutral face expression.

**Patient interface**. Emphasis is given in designing an easy to use interface for the patient, considering the patient motor disabilities and limited computer familiarity. The designed interface inherits the feel and tough of the phone dialling pad, and all system choices are based on it. Patient use the interface to declare their subjective estimation of their own status, to gain access to relevant disease information, to receive instructions on life-style interventions, such as medication and good intake and on the execution of tests. Moreover, PD's patients declare medication intake information, which is useful for the patient status assessment.

Mobile Systems as a Challenge for Neurological

Fig. 13. PERFORM medical interface

**Phase 1: Data collection with SHIMMER** 

support tool to validate the data used for this work.

**Phase 2: Data Collection with ANCO first release trainer classifier** 

**4.4 Evaluation methodology** 

Diseases Management – The Case of Parkinson's Disease 301

**Communication**. All patient-side data (signal features and patient inserted information) are

In order to test PERFORM system, several clinical trials have been arranged into 4 different

Eight subjects participated in this study, separated in two groups: four PD patients and four healthy subjects. The symptoms were rated by a professional neurologist with more than 20 years of experience with PD patients. Four accelerometers were placed on the right and left forearms and on the right and left calves, with a fifth accelerometer being placed on the trunk, at the base of the sternum. Motion data was collected using the SHIMMER platform (RealTime, 2011). SHIMMER is a small wireless sensor platform designed by Intel as a wearable device for health sensing applications. All sensors provide 3-axis accelerometer signals large storage, and low power-standards based communication capabilities. They also provide a Bluetooth protocol capability that allows SHIIMERs to stream the data to a computer. During the experiment, the accelerometry measurements were complemented by a reflective marker and a camera collection system. This complimentary analysis served as a

The data collection was performed with a network of wireless 3-axis ALA-6g sensors (Anco, 2011), located on the limbs, trunk and belt of the patient. During this phase, data were collected during tests with patients in a supervised environment, with the collaboration of the clinic's medical staff. Patients involved in this phase were required to be aged between 18 and 85 years old, suffering from PD, capable of complying with study requirements, receiving dopaminergic treatment and experiencing motor fluctuations. Dementia, Psychosis and significant systemic diseases (such as cancer) were the exclusion criteria

transmitted to the health-care subsystem once a day, through Internet.

phases, taking place between 2009 and 2011. Their description follows.

Fig. 12. PERFORM Patient Interface. Medication intake information (left). Food intake information (right)

**Communication:** the Base Module supports Bluetooth and Zigbee communication with the continuous monitoring system, and fixed line communication to the hospital centre over ISDN and xDSL.

**Symptom Detection**. This submodule processes received patient signals and detects the targeted patient symptoms (tremor, levodopa induced dyskinesia and off state). For each symptom dedicated submodule processes the relevant signals, detests the symptom episode and quantifies it into a severity scale from 1 to 4, according to the UPDRS scaling for PD patients (Cancela et al., 2010; Keijsers et al., 2000; Pansera et al., 2009). Other features such as duration, frequently and amplitude might also be provided for further clinician review and system evaluation.

The healthcare centre subsystem is composed of the following modules:

**Patient Modelling Module**. This module exploits the recorded patient information to build a patient symptom profile. For each main symptom (tremor, levodopa induced dyskinesia and on-off states), it produces a patient profile which describes the patient's common symptom features. When a new patient recording is processes, it is checked against the patient symptom profile. If significant differences are found, it might be due to two reasons: either a temporarily patient behaviour abnormality or a change to the patient profile. In the last case, the system checks whether a substantial number of similar situations are identified for the last time period for the specific patient and if that occurs, it creates an alert.

**Patient Management Module**. This module considers the detected symptoms and their characteristics, combines them with other recorded information and suggests appropriate treatment changes based on the accumulated specialists' knowledge on the management of PD.

**Medical Interface**: The system can be accessed either locally or remotely by the treating clinician and the general practitioner, using either a large screen access device (e.g. PC, laptop) or a small screen access device (e.g. PDA). Clinicians are directed to the home system screen, which presents the produced patient alerts to the patient specific screen, which provides the information needed to evaluate visually the patient condition. On request, the actual recorded signal and tests are downloaded from the patient-side to the healthcare centre for review. The focus is on the provision of an adequate visual description of the patient status within one screen, minimising the time spend by a clinician. Clinicians will access the system periodically to check patient status, but the option to be alerted when patient status changes are also available.

Fig. 12. PERFORM Patient Interface. Medication intake information (left). Food intake

The healthcare centre subsystem is composed of the following modules:

**Communication:** the Base Module supports Bluetooth and Zigbee communication with the continuous monitoring system, and fixed line communication to the hospital centre over

**Symptom Detection**. This submodule processes received patient signals and detects the targeted patient symptoms (tremor, levodopa induced dyskinesia and off state). For each symptom dedicated submodule processes the relevant signals, detests the symptom episode and quantifies it into a severity scale from 1 to 4, according to the UPDRS scaling for PD patients (Cancela et al., 2010; Keijsers et al., 2000; Pansera et al., 2009). Other features such as duration, frequently and amplitude might also be provided for further clinician review and

**Patient Modelling Module**. This module exploits the recorded patient information to build a patient symptom profile. For each main symptom (tremor, levodopa induced dyskinesia and on-off states), it produces a patient profile which describes the patient's common symptom features. When a new patient recording is processes, it is checked against the patient symptom profile. If significant differences are found, it might be due to two reasons: either a temporarily patient behaviour abnormality or a change to the patient profile. In the last case, the system checks whether a substantial number of similar situations are identified

**Patient Management Module**. This module considers the detected symptoms and their characteristics, combines them with other recorded information and suggests appropriate treatment changes based on the accumulated specialists' knowledge on the management of

**Medical Interface**: The system can be accessed either locally or remotely by the treating clinician and the general practitioner, using either a large screen access device (e.g. PC, laptop) or a small screen access device (e.g. PDA). Clinicians are directed to the home system screen, which presents the produced patient alerts to the patient specific screen, which provides the information needed to evaluate visually the patient condition. On request, the actual recorded signal and tests are downloaded from the patient-side to the healthcare centre for review. The focus is on the provision of an adequate visual description of the patient status within one screen, minimising the time spend by a clinician. Clinicians will access the system periodically to check patient status, but the option to be alerted when

for the last time period for the specific patient and if that occurs, it creates an alert.

information (right)

ISDN and xDSL.

system evaluation.

PD.

patient status changes are also available.


#### Fig. 13. PERFORM medical interface

**Communication**. All patient-side data (signal features and patient inserted information) are transmitted to the health-care subsystem once a day, through Internet.

## **4.4 Evaluation methodology**

In order to test PERFORM system, several clinical trials have been arranged into 4 different phases, taking place between 2009 and 2011. Their description follows.

#### **Phase 1: Data collection with SHIMMER**

Eight subjects participated in this study, separated in two groups: four PD patients and four healthy subjects. The symptoms were rated by a professional neurologist with more than 20 years of experience with PD patients. Four accelerometers were placed on the right and left forearms and on the right and left calves, with a fifth accelerometer being placed on the trunk, at the base of the sternum. Motion data was collected using the SHIMMER platform (RealTime, 2011). SHIMMER is a small wireless sensor platform designed by Intel as a wearable device for health sensing applications. All sensors provide 3-axis accelerometer signals large storage, and low power-standards based communication capabilities. They also provide a Bluetooth protocol capability that allows SHIIMERs to stream the data to a computer. During the experiment, the accelerometry measurements were complemented by a reflective marker and a camera collection system. This complimentary analysis served as a support tool to validate the data used for this work.

#### **Phase 2: Data Collection with ANCO first release trainer classifier**

The data collection was performed with a network of wireless 3-axis ALA-6g sensors (Anco, 2011), located on the limbs, trunk and belt of the patient. During this phase, data were collected during tests with patients in a supervised environment, with the collaboration of the clinic's medical staff. Patients involved in this phase were required to be aged between 18 and 85 years old, suffering from PD, capable of complying with study requirements, receiving dopaminergic treatment and experiencing motor fluctuations. Dementia, Psychosis and significant systemic diseases (such as cancer) were the exclusion criteria

Mobile Systems as a Challenge for Neurological

analysis of the entropy of the walking.

**4.6 Comparison with SoA systems** 

Federale de Lausanne is not detecting.

project has tried to do.

**5. Conclusions** 

recording data.

Diseases Management – The Case of Parkinson's Disease 303

body bradykinesia is quantified with a classification accuracy within the range of 70%-86% (depending on the number of sensors used in the measurements) compared to medical evaluation. Dyskinesia and tremor classification accuracy are respectively 93.6% and 97%. Those results are quite good in comparison with the subjective medical evaluation, with accuracy around 95%. The gait module has also proved to release useful information with acceptable level of accuracy. The gait analysis system provides a measure of stride length with a 7.3% of average error and a measurement of the complexity of movement through an

Phase IV of the pilots is expected to provide a more accurate validation with long-term

Compared to Kinesia system, a sophisticated multi-paRametric system FOR the continuous effective assessment and Monitoring of motor status in Parkinson's disease and other neurodegenerative diseases (PERFORM) presents the advantage of integrating other data from patient's monitoring, such as data from normal daily activity Kinesia monitors only upper extremity motor symptoms, while PERFORM is able to monitor the entire motor disorder, involving walking and moving in general, including freezing, falls risks, etc. The

In relation to Kinetisense commercial product, PERFORM links motion monitoring data to other information such as medication assumption, stress situations and historic data of the patient, in order to draw a complete picture and not only a snapshot such as Kinetisense does. The system is based on proprietary software and raw data may not be available. PERFORM integrates more data from continuous monitoring rather than on e-diary and

Compared with SensorShoe system, PERFORM provides link to medication and enables complex data interpretation and analysis. On the other hand, PERFORM system is also able to detect abnormalities in motion, which the prototype built by the École Polytechnique

In comparison with the system proposed by Boston university, PERFORM manages data

Finally, in comparison with the project HELP, PERFORM is able to provide an assessment for specific symptoms such as dyskinesia, bradykinesia, tremor and akynesia, delivering a quantification based on the UPDRS scale, which HELP is not taking into consideration. On the other hand, the system is not providing a feedback to adjust the medication directly on the patients. In other word, PERFORM is not closing the loop of monitoring/assessment/medication adjustment in an automatic way, that is what HELP

This chapter has presented a SoA review of the main methods used to monitor and assess PD's patients. On one hand, clinical assessment is usually performed through annotations in diaries and self-reports from the patient during the short office visit in this neurologist the patient may appear very well and he misses to report some symptoms, as a result, treatment modifications are not undertaken in time. The UPDRS is a scale that tries to quantify selected symptoms and signs of Parkinsonism in a 5-points scoring system (with 0 for no

system is based on proprietary software and raw data may not be available.

voluntary motion exercises as the TEVAL Movistar system does.

from more devices and is able to correlate different data input.

applied when selecting participants. The data set used in this study included trials with twenty PD patients, ten in Navarra (Spain) and ten in Ioannina (Greece). In order to comply with ethical requirements, all procedures were carried with the Clinic Institutional Review Board's permission. Data were collected following a standard clinical protocol in which patients carried out daily basic activities (i.e. walking, lying, sitting, etc.) during two cycles of on-off oscillations in response to levodopa during the same day and under the supervision of a clinician.

#### **Phase 3: Long time recording**

Data collection of phase 3 was performed with an updated version of the devices that includes a wearable and programmable logger that gave a better mobility to the patient and new ALA-6g accelerometers sensors equipped with an external battery allowing longer data collection session. Data were collected in a supervised environment and with the collaboration of the medical staff. Furthermore, patients involved in this phase, fulfilled with the age and medical specifications of the previous phase. The data were recorded with twenty-four PD patients, twelve in Navarra and twelve in Ioannina. Data were collected during a six-hour daily session in which patients carried out their normal daily activity. Moreover, four standard clinical protocol sessions were performed during two cycles of onoff oscillations in response to levodopa treatment and under the supervision of a clinician. At the end of the day, data were processed using the train set computed in the previous phase and the output were checked with the results provided by the clinicians

#### **Phase 4: Final system testing**

From March 2011, the integrated and final PERFORM system will be introduced to a new group of patients that will perform the final evaluation. The patient group will be constituted by 20 PD patients for regular tests and 4 PD patients for mid-term tests, all recruited from the Neurology Department from the Azienda Unita' Sanitaria Locale di Modena hospital in Modena, Italy.

## **4.5 Results**

PERFORM project has released promising results in patients monitoring and status assessment. Due to the short-term nature of the clinic trials that have been carried out it is difficult to determine the future impact on patient treatment; however it is possible to at least provide a quantification of the performances of the modules of the Patient –Side Subsystem. It is designed to assess the motor status of the patients and establish a direct connection with the physician. Its basic functions are:


The validation has proved that the first prototype of the Patient Side Subsystem is able to provide a very reasonable assessment of the daily activity of the patients using data classification techniques based on accelerometers. More specifically, the algorithms are able to discriminate activities such as walking, standing, laying or sitting with an accuracy of nearly 99%. The activity recognition is the base information needed to evaluate the symptoms related to movement. Besides, the clinical trials have proved that the algorithms are able to classify with an acceptable degree of accuracy the main symptoms of PD. General body bradykinesia is quantified with a classification accuracy within the range of 70%-86% (depending on the number of sensors used in the measurements) compared to medical evaluation. Dyskinesia and tremor classification accuracy are respectively 93.6% and 97%. Those results are quite good in comparison with the subjective medical evaluation, with accuracy around 95%. The gait module has also proved to release useful information with acceptable level of accuracy. The gait analysis system provides a measure of stride length with a 7.3% of average error and a measurement of the complexity of movement through an analysis of the entropy of the walking.

Phase IV of the pilots is expected to provide a more accurate validation with long-term recording data.

## **4.6 Comparison with SoA systems**

302 Diagnostics and Rehabilitation of Parkinson's Disease

applied when selecting participants. The data set used in this study included trials with twenty PD patients, ten in Navarra (Spain) and ten in Ioannina (Greece). In order to comply with ethical requirements, all procedures were carried with the Clinic Institutional Review Board's permission. Data were collected following a standard clinical protocol in which patients carried out daily basic activities (i.e. walking, lying, sitting, etc.) during two cycles of on-off oscillations in response to levodopa during the same day and under the

Data collection of phase 3 was performed with an updated version of the devices that includes a wearable and programmable logger that gave a better mobility to the patient and new ALA-6g accelerometers sensors equipped with an external battery allowing longer data collection session. Data were collected in a supervised environment and with the collaboration of the medical staff. Furthermore, patients involved in this phase, fulfilled with the age and medical specifications of the previous phase. The data were recorded with twenty-four PD patients, twelve in Navarra and twelve in Ioannina. Data were collected during a six-hour daily session in which patients carried out their normal daily activity. Moreover, four standard clinical protocol sessions were performed during two cycles of onoff oscillations in response to levodopa treatment and under the supervision of a clinician. At the end of the day, data were processed using the train set computed in the previous

From March 2011, the integrated and final PERFORM system will be introduced to a new group of patients that will perform the final evaluation. The patient group will be constituted by 20 PD patients for regular tests and 4 PD patients for mid-term tests, all recruited from the Neurology Department from the Azienda Unita' Sanitaria Locale di

PERFORM project has released promising results in patients monitoring and status assessment. Due to the short-term nature of the clinic trials that have been carried out it is difficult to determine the future impact on patient treatment; however it is possible to at least provide a quantification of the performances of the modules of the Patient –Side Subsystem. It is designed to assess the motor status of the patients and establish a direct

to provide a quantification of symptoms severity based on the UPRDS scale and present

The validation has proved that the first prototype of the Patient Side Subsystem is able to provide a very reasonable assessment of the daily activity of the patients using data classification techniques based on accelerometers. More specifically, the algorithms are able to discriminate activities such as walking, standing, laying or sitting with an accuracy of nearly 99%. The activity recognition is the base information needed to evaluate the symptoms related to movement. Besides, the clinical trials have proved that the algorithms are able to classify with an acceptable degree of accuracy the main symptoms of PD. General

such an information to the physician through remote communication

phase and the output were checked with the results provided by the clinicians

supervision of a clinician. **Phase 3: Long time recording** 

**Phase 4: Final system testing** 

Modena hospital in Modena, Italy.

connection with the physician. Its basic functions are:

to gather information about the daily life of the patient

to determine the activity of the subject

**4.5 Results** 

Compared to Kinesia system, a sophisticated multi-paRametric system FOR the continuous effective assessment and Monitoring of motor status in Parkinson's disease and other neurodegenerative diseases (PERFORM) presents the advantage of integrating other data from patient's monitoring, such as data from normal daily activity Kinesia monitors only upper extremity motor symptoms, while PERFORM is able to monitor the entire motor disorder, involving walking and moving in general, including freezing, falls risks, etc. The system is based on proprietary software and raw data may not be available.

In relation to Kinetisense commercial product, PERFORM links motion monitoring data to other information such as medication assumption, stress situations and historic data of the patient, in order to draw a complete picture and not only a snapshot such as Kinetisense does. The system is based on proprietary software and raw data may not be available. PERFORM integrates more data from continuous monitoring rather than on e-diary and voluntary motion exercises as the TEVAL Movistar system does.

Compared with SensorShoe system, PERFORM provides link to medication and enables complex data interpretation and analysis. On the other hand, PERFORM system is also able to detect abnormalities in motion, which the prototype built by the École Polytechnique Federale de Lausanne is not detecting.

In comparison with the system proposed by Boston university, PERFORM manages data from more devices and is able to correlate different data input.

Finally, in comparison with the project HELP, PERFORM is able to provide an assessment for specific symptoms such as dyskinesia, bradykinesia, tremor and akynesia, delivering a quantification based on the UPDRS scale, which HELP is not taking into consideration. On the other hand, the system is not providing a feedback to adjust the medication directly on the patients. In other word, PERFORM is not closing the loop of monitoring/assessment/medication adjustment in an automatic way, that is what HELP project has tried to do.

## **5. Conclusions**

This chapter has presented a SoA review of the main methods used to monitor and assess PD's patients. On one hand, clinical assessment is usually performed through annotations in diaries and self-reports from the patient during the short office visit in this neurologist the patient may appear very well and he misses to report some symptoms, as a result, treatment modifications are not undertaken in time. The UPDRS is a scale that tries to quantify selected symptoms and signs of Parkinsonism in a 5-points scoring system (with 0 for no

Mobile Systems as a Challenge for Neurological

pp. 2747-54. ISSN 1560-9545.

0885-3185.

ISSN: 1350-4533.

3185.

12.

1668. ISSN 0003-9942.

public document.

John Wiley & Sons. ISSN 0885-3185.

Diseases Management – The Case of Parkinson's Disease 305

Das, R. (2010). A comparison of multiple classification methods for diagnosis of Parkinson disease. *Expert Systems with Applications*, 37(2), 1568-1572. ISSN 0957-4174 . Deuschl, G., Schade-Brittinger, C., Krack, P., Volkmann, J., Schöfer, H. & Bötzel, K., (2006). A

Dorsey, E., Deuel, L. M., Voss, T. S., Finnigan, K., George, B. P. & Eason, S., (2010).

Fuente-Fernández, R. de la, Sossi, V., Huang, Z., Furtado, S., Lu, J.-Q. & Calne, D. B., (2004).

Giansanti, D., Maccioni, G., Cesinaro, S., Benvenuti, F., & Macellari, V. (2008). Assessment of

Giansanti, D., Maccioni, G., & Morelli, S. (2008). An Experience of Health Technology

Goetz, C. G., Poewe, W., Rascol, O., Sampaio, C., Stebbins, G. T., Counsell, (2004).

Greenlaw, R., Estrada, J., Pansera, M., Konitsiotis, S., Baga, D & Maziewski, P., (2009).

Group, P. S. (2001). Evaluation of dyskinesias in a pilot, randomized, placebo controlled trial

Help project (2011). Home Based Empowed Living for Parkinson Disease's Patients.

Jovanov, E., Raskovic, D., Price, J., Krishnamurthy, A., Chapman, J., & Moore, A. (2001).

Kauw-A-Tjoe, R. G., Thalen, J. P., Marin-Perianu, M., & Havinga, P. J. M. (2007). SensorShoe:

Keijsers, N L W, Horstink, M., Van Hilten, J. J., Hoff, J I, & Gielen, C. (2000). Detection and

Retrieved March 2011, from http://www.help-aal.com/HELP/

*Biomedical Sciences Instrumentation*, 37, 373-378. ISSN 0067-8856.

*Journal of Medicine*, 355(9), 896-908. ISSN 0028-4793.

randomized trial of deep-brain stimulation for Parkinson's disease. *New England* 

Increasing access to specialty care: A pilot, randomized controlled trial of telemedicine for Parkinson's disease. *Movement Disorders*, 25(11), 1652-1659. ISSN

Levodopa-induced changes in synaptic dopamine levels increase with progression of Parkinson's disease: implications for dyskinesias. *Journal of neurology*, 127(Pt 12),

fall-risk by means of a neural network based on parameters assessed by a wearable device during posturography. *Medical engineering & physics*, 30(3), pp. 367-372.

Assessment in New Models of Care for Subjects with Parkinson's Disease by Means of a New Wearable Device. *Telemedicine and e-health*, 14(5), 467-472. ISSN 1530-5627

Movement Disorder Society Task Force report on the Hoehn and Yahr staging scale: status and recommendations. *Movement Disorder*, 19(9), pp.1020-8. ISSN 0885-

PERFORM: Building and Mining Electronic Records of Neurological Patients Being Monitored in the Home. *Proceeding of World Congress on Medical Physics and Biomedical Engineering* (Vol. 25/9, pp. 533-535-535). Munich, Germany. September 7-

or remacemide in advanced Parkinson disease. *Archives of Neurology*, 58, pp. 1660-

Patient monitoring using personal area networks of wireless intelligent sensors.

Mobile Gait Analysis for Parkinson's Disease Patients. University of Innsbruck

assessment of the severity of levodopa-induced dyskinesia in patients with Parkinson's disease by neural networks. *Movement Disorders*, 15(6), pp. 1104-1111.

signal and 4 for a marked severity of the sign). Unfortunately, the UPDRS like any other semi-objective rating scale has limitations like intra and inter observer inconsistencies, can be time consuming and ca be biased by subjectivity issues related to historical information.

On the other hand, technological solutions are able to provide quantitative objective information, including the use of motor sensors, electromyography, position transducers, and speech recognition systems.

This chapter has presented PERFORM, a project partially funded by the European Commission under the 7th Framework Program, aiming at providing an innovative and reliable tool that is able to monitor and evaluate motor neurodegenerative disease patients, such as PD patients.

The PERFORM project is based on the development of an intelligent closed loop system that seamlessly integrates a wide range of wearable micro-sensors, constantly monitoring several motor and signals of the patients. Data acquired are pre-processed by advanced knowledge processing methods, integrated by fusion algorithms to allow health professionals to remotely monitor the overall status of the patients and adjust medication schedules and personalize treatment. Personalization of treatment occurs through PERFORM's capability to keep track of the timing and doses of the medication and meals that the patient is taking. The system architecture has been presented. A comparison with available related systems has been performed.

The system has already been tested in hospitals in Navarra (Spain) and Ioannina (Greece). The integrated tests of the system will be performed in Modena (Italy) from March 2011.

Obtained results so far suggest an overall valid closed loop system, able to detect PD symptoms based on motor signals and additional information, evaluate with a high accuracy level the overall status of the patient and propose medication changes accordingly. However, to achieve more improvements especially in the automation of close-loop mechanism, further improvements are needed in order to provide a complete reliable assessment system for symptoms severity.

## **6. Acknowledgment**

The authors thank PERFORM project Consortium for their valuable contribution to this work, and especially Prof. José Antonio Martín Pereda, Prof. Francisco del Pozo and Prof. Ana González Marcos from the Universidad Politécnica de Madrid for their collaboration and scientific support. PERFORM project (Contract Nr. 215952) is partially funded by the European Commission under the 7th Framework Programme.

## **7. References**

Anco. (2011). Retrieved March 2011, Available from http://www.anco.gr/


signal and 4 for a marked severity of the sign). Unfortunately, the UPDRS like any other semi-objective rating scale has limitations like intra and inter observer inconsistencies, can be time consuming and ca be biased by subjectivity issues related to historical information. On the other hand, technological solutions are able to provide quantitative objective information, including the use of motor sensors, electromyography, position transducers,

This chapter has presented PERFORM, a project partially funded by the European Commission under the 7th Framework Program, aiming at providing an innovative and reliable tool that is able to monitor and evaluate motor neurodegenerative disease patients,

The PERFORM project is based on the development of an intelligent closed loop system that seamlessly integrates a wide range of wearable micro-sensors, constantly monitoring several motor and signals of the patients. Data acquired are pre-processed by advanced knowledge processing methods, integrated by fusion algorithms to allow health professionals to remotely monitor the overall status of the patients and adjust medication schedules and personalize treatment. Personalization of treatment occurs through PERFORM's capability to keep track of the timing and doses of the medication and meals that the patient is taking. The system architecture has been presented. A comparison with available related systems

The system has already been tested in hospitals in Navarra (Spain) and Ioannina (Greece). The integrated tests of the system will be performed in Modena (Italy) from March 2011. Obtained results so far suggest an overall valid closed loop system, able to detect PD symptoms based on motor signals and additional information, evaluate with a high accuracy level the overall status of the patient and propose medication changes accordingly. However, to achieve more improvements especially in the automation of close-loop mechanism, further improvements are needed in order to provide a complete reliable

The authors thank PERFORM project Consortium for their valuable contribution to this work, and especially Prof. José Antonio Martín Pereda, Prof. Francisco del Pozo and Prof. Ana González Marcos from the Universidad Politécnica de Madrid for their collaboration and scientific support. PERFORM project (Contract Nr. 215952) is partially funded by the

Cancela, J, Pansera, M, Arredondo, M T, Estrada, J J, Pastorino, M., Pastor-Sanz, L., (2010). A

Darkins, A. W., Fromkin, V. A., & Benson, D. F. (1988). A characterization of the prosodic loss in Parkinson's disease. *Brain and Language*, 34(2), 315-327. ISSN 0093-934X.

comprehensive motor symptom monitoring and management system: the bradykinesia case. *Conference proceedings of Annual International Conference of the IEEE Engineering in Medicine and Biology Society*, pp. 1008-11. 2010, August 31 –

and speech recognition systems.

such as PD patients.

has been performed.

**6. Acknowledgment**

**7. References** 

September 4.

assessment system for symptoms severity.

European Commission under the 7th Framework Programme.

Anco. (2011). Retrieved March 2011, Available from http://www.anco.gr/


Mobile Systems as a Challenge for Neurological

13(6), 864-73. ISSN 0018-9294.

http://www.shimmer-research.com/

9294.

6290-6293). Lyon, France, 2007. August 23-26.

Perform. (2008). *Perform Project*. Grant Agreement Nr. 215952. RealTime. (2011). *Shimmer Research*. Retrieved March 2011, from

Diseases Management – The Case of Parkinson's Disease 307

Patel, Shyamal, Lorincz, Konrad, Hughes, Richard, Huggins, Nancy, Growdon, J., Standaert,

Salarian, A., Russmann, H., Vingerhoets, F. J. G., Burkhard, P. R., & Aminian, K. (2007).

Singh, N., Pillay, V., & Choonara, Y. E. (2007). Advances in the treatment of Parkinson's

Standaert, D. G., & Young, A. B. (2001). Treatment of CNS neurodegenerative diseases. In H.

Tindall, L. R., & Huebner, R. A. (2009). The Impact of an Application of Telerehabilitation Technology on Caregiver Burden. *International Journal of Telerehabilitation*, 1(1), 3-8. Tolosa. (1998). *Parkinson's disease and movement disorders*. ISBN 0-7817-7881-6. Baltimore:

Tsanas, A., Little, M. A., McSharry, P. E., & Ramig, L. O. (2010). Enhanced classical

UB. (2011). Boston University National Institute of Biomedical Imaging and Bioengineering.

Vaillancourt, D. E., Sturman, M. M., Verhagen Metman, L., Bakay, R. A. E., & Corcos, D. M.

features of essential tremor. *Neurology*, 61(7), pp.919-925. ISSN 0340-5354. Van Someren, E. J. W., Van Gool, W. A., Vonk, B. F. M., Mirmiran, M., Speelman, J. D.,

Westin, J., Dougherty, M., Nyholm, D., & Groth, T. (2010). A home environment test battery

disease. *Progress in neurobiology*, 81(1),pp.29-44. ISSN 0301-0082.

549-620). ISBN-10: 0071354697; New Your, Mc Graw-Hill.

Lippincott Williams & Wilkins Philadelphia.

Aires, Argentina, 2010. August 31 – September 4.

Retrieved March 2011, from http://www.bu.edu/bme/

*neurological sciences*, 117(1-2),pp.16-23. ISSN 0022-510X.

*methods and programs in biomedicine*, 98(1), 27-35. ISSN 0169-2607.

2007. *Proceeding of EMBS 2007. 29th Annual International Conference of the IEEE* (pp.

D., (2009). Monitoring motor fluctuations in patients with Parkinson's disease using wearable sensors. *IEEE transactions on information technology in biomedicine*,

Ambulatory monitoring of physical activities in patients with Parkinson's disease. *IEEE Transactions on Biomedical Engineering*, 54(12), pp. 2296-2299. ISSN 0018-9294. Salarian, A., Russmann, H., Vingerhoets, F. J. G., Dehollain, C., Blanc, Y., Burkhard, P. R.,

(2004). Gait assessment in Parkinson's disease: toward an ambulatory system for long-term monitoring. *IEEE Transactions on Biomedical Engineering*, 51(8). ISSN 0018-

and Limbird (Ed.), *Goodman and Gilman's Pharmacological Basis of Therapeutics* (pp.

dysphonia measures and sparse regression for telemonitoring of Parkinson's disease progression. *Proceeding of Acoustics Speech and Signal Processing (ICASSP), 2010 IEEE International Conference on Biomedical Engineering* (pp. 594-597). Buenos

(2003). Deep brain stimulation of the VIM thalamic nucleus modifies several

Bosch, D. A., (1993). Ambulatory monitoring of tremor and other movements before and after thalamotomy: a new quantitative technique. *Journal of the* 

for status assessment in patients with advanced Parkinson's disease. *Computer* 


Keijsers, N L W, Horstink, M.W.I.M., & Gielen, S. C. A. M. (2003). Automatic assessment of

Keijsers, N. L. W., Horstink, M. W. I. M., Hilten, J. J. van, Hoff, J. I., & Gielen, C. C. A. M.

Kimura, F., Horio, K., Hagane, Y., Yu, W., Katoh, R., Katane, T., (2007). Detecting

Lang, A.E., & Lozano, A.M. (1998). Parkinson's disease: First of two parts. *New England* 

Lau, L. M. L. de, & Breteler, M. M. B. (2006). Epidemiology of Parkinson's disease. *Lancet* 

Lees, A. J. (1989). The on-off phenomenon. Journal of Neurology, *Neurosurgery & Psychiatry*,

Lim, A. S., Moro, E., Lozano, A M, Hamani, C., Dostrovsky, J. O., Hutchison, W. D., (2009).

Mazzone, P., Lozano, A., Stanzione, P., Galati, S., Scarnati, E., Peppe, A., (2005).

Papapetropoulos, S., Heather, K., Scanlo, B. K., Guevara, A., Singer, C., & Levin, B. (2010).

Patel, S, Buckley, T., Rednic, R., McClure, D., Shih, L., Tarsy, D., et al. (2010). A Web-Based

Patel, S, Lorincz, K, Hughes, R, Huggins, N, Growdon, J. H., Welsh, M, (2007). Analysis of

target in Parkinson's disease. *Neuroreport*, 16(17), 1877. ISSN 0959-4965. Pansera, Mario, Estrada, Juan Jacobo, Pastor, L., Cancela, Jorge, Greenlaw, Reynold, &

the human pons. *Annals of neurology*, 66(1), pp.110-114. ISSN 0364-5134. Lingyun Gu, L., J.G, H., Shrivastav, R. M., & Sapienza, C. (2005). Disordered Speech

*Journal on Applied Signal Processing*, 9 (pp.1400–1409).ISSN 1110-8657. Lozano. (2009). Deep brain stimulation: current and future perspectives. *Neurosurgery Focus*,

Selective enhancement of rapid eye movement sleep by deep brain stimulation of

Assessment Using Automatic Methods Based on Quantitative Measures. *EURASIP* 

Implantation of human pedunculopontine nucleus: a safe and clinically relevant

Arredondo, Maria Teresa. (2009). Multi-parametric system for the continuous assessment and monitoring of motor status in Parkinson's disease: an entropybased gait comparison. *Conference proceedings of Annual International Conference of the IEEE Engineering in Medicine and Biology Society*, 2009, pp.1242-5. Minneapolis, USA.

Objective Quantification of Neuromotor Symptoms in Parkinson's Disease: Implementation of a Portable, Computerized Measurement Tool. *Parkinson's* 

System for Home Monitoring of Patients with Parkinson's Disease Using Wearable Sensors. *IEEE transactions on bio-medical engineering*, 58(3), 831-836. ISSN 0018-9294.

feature space for monitoring persons with Parkinson's Disease with application to a wireless wearable sensor system. Engineering in Medicine and Biology Society,

18(1), 70-80. ISSN 0885-3185.

1104-1111. ISSN 0885-3185.

*(ICPE).* Tokyo, Japan, 2007. August 16-18.

*neurology*, 5(6), 525-35. ISSN 1474-4422.

52(Suppl), 29. ISSN: 1359-5067.

27(1). ISSN 1092-0684

September 2-6.

*Disease*, 010 (Article ID 760196), 6.

*Jorunal of Medicine*, 339, 1044-1053. ISSN 0028-4793

levodopa-induced dyskinesias in daily life by neural networks. *Movement disorders*,

(2000). Detection and assessment of the severity of Levodopa-induced dyskinesia in patients with Parkinson's disease by neural networks. *Movement Disorders*, 15(6),

Perturbation Occurrence during Walking*. Conference proceeding of Towards sysnthesis of Micro-/Nano-systems: The 11th international conference on precision engineering*  2007. *Proceeding of EMBS 2007. 29th Annual International Conference of the IEEE* (pp. 6290-6293). Lyon, France, 2007. August 23-26.


**15** 

**An Investigation into the Impact of Parkinson's** 

*1School of Occupational Therapy and Social Work, Curtin Health Innovation Research* 

*3Department of Rehabilitation Medicine, IKE, Faculty of Health Sciences Linköping* 

PD is a severe neurodegenerative disease that can impair functional driving performance and increase the risk of accidents and fatalities on Australian roads (Austroads, 2000). In particular, cognitive symptoms of PD can have a substantial influence on driving performance due to the complicated and demanding nature of the task (Uitti, 2009). PD can affect the neural pathways that facilitate essential cognitive processes; such as attention, information processing speed, memory and risk assessment. These processes are all integral to the decision making process (Cools, et at., 2001). Previous research has highlighted that the ability to make accurate and timely decisions is essential for safe driving performance. However, this has not yet been researched in relation to people with PD (Devos, et al., 2007).

PD is the second most common neurological disease in Australia; causing impairments in motor control, cognitive functioning and sensation (Access Economics, 2010). PD usually affects people over the age of 50 years. However, the rate of disease progression and severity of symptoms can vary greatly between individuals (Australian Bureau of Statistics, 2004). Australia's aging population is expected to increase the prevalence rate of PD by 40%

Recent improvements in the medical and psychosocial treatment of PD has dramatically increased life expectancy, as people with PD now live approximately 12 to 20 years past diagnosis (Access Economics, 2010). PD is currently the sixth highest cause of diseaserelated driving cessation in Australia (Access Economics, 2010). People with PD generally stop driving at the age of 68; eight years earlier than the general population (Access Economics, 2010). Research into the impact of symptoms upon functional ability will enable

**1.1 Prevalence and aetiology of Parkinson's disease** 

by 2033 (refer to Figure 1) (Access Economics, 2010).

**1. Introduction** 

**Disease upon Decision Making Ability** 

*4School of Occupational Therapy, La Trobe University, Melbourne, VIC,* 

Jessica Davies1, Hoe Lee1 and Torbjorn Falkmer1,2,3,4

*2School of Health Sciences, Jönköping University, Jönköping,* 

**and Driving Performance** 

*Institute, Curtin University,* 

*University,* 

*2,3Sweden 1,4Australia* 


## **An Investigation into the Impact of Parkinson's Disease upon Decision Making Ability and Driving Performance**

Jessica Davies1, Hoe Lee1 and Torbjorn Falkmer1,2,3,4 *1School of Occupational Therapy and Social Work, Curtin Health Innovation Research Institute, Curtin University, 2School of Health Sciences, Jönköping University, Jönköping, 3Department of Rehabilitation Medicine, IKE, Faculty of Health Sciences Linköping University, 4School of Occupational Therapy, La Trobe University, Melbourne, VIC, 2,3Sweden 1,4Australia* 

## **1. Introduction**

308 Diagnostics and Rehabilitation of Parkinson's Disease

Westin, J., Ghiamati, S., Memedi, M., Nyholm, D., Johansson, A., Dougherty, M., (2010). A

*Journal of neuroscience methods*, 190(1), 143-148. ISSN 0165-0270. WHO. (2002). Integrating prevention into health care. Retrieved 2011, from http://www.who.int/mediacentre/factsheets/fs172/en/index.html

new computer method for assessing drawing impairment in Parkinson's disease.

PD is a severe neurodegenerative disease that can impair functional driving performance and increase the risk of accidents and fatalities on Australian roads (Austroads, 2000). In particular, cognitive symptoms of PD can have a substantial influence on driving performance due to the complicated and demanding nature of the task (Uitti, 2009). PD can affect the neural pathways that facilitate essential cognitive processes; such as attention, information processing speed, memory and risk assessment. These processes are all integral to the decision making process (Cools, et at., 2001). Previous research has highlighted that the ability to make accurate and timely decisions is essential for safe driving performance. However, this has not yet been researched in relation to people with PD (Devos, et al., 2007).

#### **1.1 Prevalence and aetiology of Parkinson's disease**

PD is the second most common neurological disease in Australia; causing impairments in motor control, cognitive functioning and sensation (Access Economics, 2010). PD usually affects people over the age of 50 years. However, the rate of disease progression and severity of symptoms can vary greatly between individuals (Australian Bureau of Statistics, 2004). Australia's aging population is expected to increase the prevalence rate of PD by 40% by 2033 (refer to Figure 1) (Access Economics, 2010).

Recent improvements in the medical and psychosocial treatment of PD has dramatically increased life expectancy, as people with PD now live approximately 12 to 20 years past diagnosis (Access Economics, 2010). PD is currently the sixth highest cause of diseaserelated driving cessation in Australia (Access Economics, 2010). People with PD generally stop driving at the age of 68; eight years earlier than the general population (Access Economics, 2010). Research into the impact of symptoms upon functional ability will enable

An Investigation into the Impact

required to confirm these claims.

over short and/or familiar distances (Devos, et al., 2007).

occurring in one third of people with only mild to moderate PD.

of Parkinson's Disease upon Decision Making Ability and Driving Performance 311

behaviours to be learnt, coordinated and continuously adapted in a constantly changing environment with time-based pressures (Elvik & Vaa, 2004). Driving therefore places extensive demands upon cognitive abilities, requiring high levels of vigilance, concentration, multitasking, complex reasoning and decision making even when driving

Physical symptoms of PD have been systematically researched in relation to driving performance. This has contributed to a comprehensive evidence base on the physical effects of PD symptoms upon driving performance (Cordell, et al., 2008; Jankovic, 2007). Drivers with PD have reduced strength and speed of movement, slower reaction times and a diminished ability to turn their head to check mirrors (Adler, et al., 2000; Heikkila, et al., 1997). Drivers with PD also have difficulty in negotiating roundabouts, turning across traffic, driving at high speeds and driving in urban environments (Cordell, et al., 2008; Radford, et al., 2004; Uc, et al., 2009). Drivers with PD are often aware of how their physical limitations influence their driving performance (Kulisevsky & Pagonabarraga, 2009). Consequently, many drivers with PD self-regulate their driving habits by avoiding potentially difficult or risky situations, such as not driving on the freeway, avoiding peak hour or having a co-pilot (Amick, et al., 2007). Factor and Weiner (2002) claimed that the main contributing factors to poor driving performance are PD-related deficits in cognition and visual processing as self-regulating behaviours are very effective in compensating for physical deficits. Uitti (2009) claimed that decline of visual sensitivity, motion perception and cognition are the largest contributing factors to unsafe driving. Further research is

Research into the impact of cognitive symptoms upon driving ability is limited and contradictory. It is difficult to detect the presence of cognitive impairment in PD and to determine the relationship and severity of cognitive impairment on driving performance. The exact prevalence of cognitive impairment amongst drivers with PD is unknown. People with mild to moderate PD have scored significantly lower upon psychomotor and cognitive assessments, showing that PD affects cognition and psychomotor ability at all stages of the disease (Heikkila et al., 1997). However, routine cognitive assessments, such as the Mini Mental Status Examination have low sensitivity, preventing the accurate detection of cognitive deficits in people with PD (Kulisevsky & Pagonabarraga, 2009). Adler and colleagues (2000) stated that 25 to 40% of people in the later stage of PD experience cognitive impairment whilst Factor and Weiner (2002) recorded a lower prevalence rate of 20% amongst another cohort in a similar stage of the disease. Tröster and Woods (2007), however, claimed that cognitive impairment is more common with an earlier onset,

It is known that the prevalence of cognitive impairment significantly increases with disease progression. However, the number of drivers with PD in Australia who have cognitive impairment is unknown (Amick, et al., 2007). Inability to accurately screen for cognitive impairment is of concern to road safety, since people who are affected may not be aware of it. If drivers with PD are not aware of the need to self regulate driving behaviour and/or compensate for performance alterations, the risk to road safety is increased (Amick, et al., 2007). Drivers may not seek medical advice and/or driving assessments may not be sought as needed, as the potential impacts upon driving performance are poorly understood (Betz & Fisher, 2009). Jones (2009) found that the most frequently self-identified cognitive areas affecting driving amongst people with PD were decision making, complex attention, visual search, impulse control, planning and divided attention. They also conducted a meta-

Adapted from: Access Economics, 2010

Fig. 1. Predicted prevalence of Parkinson's disease in Australia 2005-2033

the development of better screening tools and allow health professionals to differentiate between capable and unsafe drivers (Adler, et al., 2000). This may allow capable drivers with PD to retain their licences and current quality of life through active participation in occupations (Innes, et al., 2009). As the number of drivers with PD will rapidly increase due to the aging population, such an initiative will assist in improving road safety (Cordell, et al., 2008).

PD is caused by the progressive cellular death of dopaminergic neurons, predominantly in the basal ganglia in the brain (Arias-Carrión & Pöppel, 2007). Symptoms usually occur after the death of 70% of dopaminergic neurons; causing severe depletion of the neurotransmitter, dopamine (Jankovic, 2007). Dopamine has an extensive role in regulating movement, behaviour, mood and motivation; and may influence learning, time estimation, consequence prediction and awareness of the environment (Arias-Carrión & Pöppel, 2007). The cause of PD is unknown and as the disease cannot be detected prior to onset of symptoms, it is not currently possible to cure PD (Cools, et al., 2001). Severity of symptoms and rate of disease progression vary significantly between individuals. For example, some individuals may experience only minor symptoms 10 years after diagnosis, whilst other individuals may require full time high-support care within six months of being diagnosed with PD (Jankovic, 2007). It is not currently possible to predict how the disease will affect each individual's driving performance, and so assessment must be performed on a case-bycase basis (Jankovic, 2007).

#### **1.2 Physical and cognitive symptoms of Parkinson's disease that affect driving**

PD can cause a wide range of physical symptoms, which are known to affect driving ability. Common symptoms include motor tremors, bradykinesia, postural instability, rigidity, involuntary movements, generalised slowness and impaired balance (Adler, et al., 2000). People with PD can also experience alterations in sensation; including pain, burning, paresthesia and vestibular dysfunction (Jankovic, 2007). Driving is the most complicated activity of daily living, and even small mistakes can cause severe and potentially fatal crashes (Molnar, Marshall, & Man-Son-Hing, 2006). Driving requires numerous skills and

**138,000**

2005 2010 2033

**240,000**

Adapted from: Access Economics, 2010

**Year of Prediction**

0

50,000

100,000

150,000

Predicted Prevalance of PD

200,000

250,000

case basis (Jankovic, 2007).

al., 2008).

Fig. 1. Predicted prevalence of Parkinson's disease in Australia 2005-2033

**73,000**

the development of better screening tools and allow health professionals to differentiate between capable and unsafe drivers (Adler, et al., 2000). This may allow capable drivers with PD to retain their licences and current quality of life through active participation in occupations (Innes, et al., 2009). As the number of drivers with PD will rapidly increase due to the aging population, such an initiative will assist in improving road safety (Cordell, et

PD is caused by the progressive cellular death of dopaminergic neurons, predominantly in the basal ganglia in the brain (Arias-Carrión & Pöppel, 2007). Symptoms usually occur after the death of 70% of dopaminergic neurons; causing severe depletion of the neurotransmitter, dopamine (Jankovic, 2007). Dopamine has an extensive role in regulating movement, behaviour, mood and motivation; and may influence learning, time estimation, consequence prediction and awareness of the environment (Arias-Carrión & Pöppel, 2007). The cause of PD is unknown and as the disease cannot be detected prior to onset of symptoms, it is not currently possible to cure PD (Cools, et al., 2001). Severity of symptoms and rate of disease progression vary significantly between individuals. For example, some individuals may experience only minor symptoms 10 years after diagnosis, whilst other individuals may require full time high-support care within six months of being diagnosed with PD (Jankovic, 2007). It is not currently possible to predict how the disease will affect each individual's driving performance, and so assessment must be performed on a case-by-

**1.2 Physical and cognitive symptoms of Parkinson's disease that affect driving**  PD can cause a wide range of physical symptoms, which are known to affect driving ability. Common symptoms include motor tremors, bradykinesia, postural instability, rigidity, involuntary movements, generalised slowness and impaired balance (Adler, et al., 2000). People with PD can also experience alterations in sensation; including pain, burning, paresthesia and vestibular dysfunction (Jankovic, 2007). Driving is the most complicated activity of daily living, and even small mistakes can cause severe and potentially fatal crashes (Molnar, Marshall, & Man-Son-Hing, 2006). Driving requires numerous skills and behaviours to be learnt, coordinated and continuously adapted in a constantly changing environment with time-based pressures (Elvik & Vaa, 2004). Driving therefore places extensive demands upon cognitive abilities, requiring high levels of vigilance, concentration, multitasking, complex reasoning and decision making even when driving over short and/or familiar distances (Devos, et al., 2007).

Physical symptoms of PD have been systematically researched in relation to driving performance. This has contributed to a comprehensive evidence base on the physical effects of PD symptoms upon driving performance (Cordell, et al., 2008; Jankovic, 2007). Drivers with PD have reduced strength and speed of movement, slower reaction times and a diminished ability to turn their head to check mirrors (Adler, et al., 2000; Heikkila, et al., 1997). Drivers with PD also have difficulty in negotiating roundabouts, turning across traffic, driving at high speeds and driving in urban environments (Cordell, et al., 2008; Radford, et al., 2004; Uc, et al., 2009). Drivers with PD are often aware of how their physical limitations influence their driving performance (Kulisevsky & Pagonabarraga, 2009). Consequently, many drivers with PD self-regulate their driving habits by avoiding potentially difficult or risky situations, such as not driving on the freeway, avoiding peak hour or having a co-pilot (Amick, et al., 2007). Factor and Weiner (2002) claimed that the main contributing factors to poor driving performance are PD-related deficits in cognition and visual processing as self-regulating behaviours are very effective in compensating for physical deficits. Uitti (2009) claimed that decline of visual sensitivity, motion perception and cognition are the largest contributing factors to unsafe driving. Further research is required to confirm these claims.

Research into the impact of cognitive symptoms upon driving ability is limited and contradictory. It is difficult to detect the presence of cognitive impairment in PD and to determine the relationship and severity of cognitive impairment on driving performance. The exact prevalence of cognitive impairment amongst drivers with PD is unknown. People with mild to moderate PD have scored significantly lower upon psychomotor and cognitive assessments, showing that PD affects cognition and psychomotor ability at all stages of the disease (Heikkila et al., 1997). However, routine cognitive assessments, such as the Mini Mental Status Examination have low sensitivity, preventing the accurate detection of cognitive deficits in people with PD (Kulisevsky & Pagonabarraga, 2009). Adler and colleagues (2000) stated that 25 to 40% of people in the later stage of PD experience cognitive impairment whilst Factor and Weiner (2002) recorded a lower prevalence rate of 20% amongst another cohort in a similar stage of the disease. Tröster and Woods (2007), however, claimed that cognitive impairment is more common with an earlier onset, occurring in one third of people with only mild to moderate PD.

It is known that the prevalence of cognitive impairment significantly increases with disease progression. However, the number of drivers with PD in Australia who have cognitive impairment is unknown (Amick, et al., 2007). Inability to accurately screen for cognitive impairment is of concern to road safety, since people who are affected may not be aware of it. If drivers with PD are not aware of the need to self regulate driving behaviour and/or compensate for performance alterations, the risk to road safety is increased (Amick, et al., 2007). Drivers may not seek medical advice and/or driving assessments may not be sought as needed, as the potential impacts upon driving performance are poorly understood (Betz & Fisher, 2009). Jones (2009) found that the most frequently self-identified cognitive areas affecting driving amongst people with PD were decision making, complex attention, visual search, impulse control, planning and divided attention. They also conducted a meta-

An Investigation into the Impact

of Parkinson's Disease upon Decision Making Ability and Driving Performance 313

In Australia, like most of the developed countries, the guidelines regulating licence retainment and cancellation are based upon a system of subjective medical expert opinion (Adler, et al., 2000). There are no current national standards or requirements for how clinical driving assessments should be conducted (Innes, et al., 2009). Medical experts are often required to determine driving performance, even though the majority have not been trained in driving assessment, or actually observed their patient driving a car (Adler, et al., 2000). Specific clinical assessment batteries and criteria to renew or cancel driving licences have not been clearly defined in the Australian Assessing Fitness to Drive handbook; the combination of symptoms and/or the severity that could compromise driving ability are not defined (Cordell, et al., 2008). Therefore, the medical practitioner must make a subjective decision on the fitness to drive of their patients, even though they may not have been trained to do so (Cordell, et al., 2008). Most current methods of determining licence retainment or cancellation is through on-road driving tests and/or clinical psychometric assessments (National Road Transport Commission, 2003). On-road assessment is the gold standard. However, the process is costly and time consuming (Bedard, et al., 2010; Bryer, et al., 2006). A person who is unable to undergo a driving assessment as recommended by their medical professional is unlikely to be able to retain their licence (Anceaux, et al., 2008). The high assessment cost and need for drivers with PD to undergo annual driving reviews may

The cheapest, most accessible and commonly used method for determining driving ability is through clinical assessment. Tools, such as the Timed Up and Go (measures ability to stand up, walk for 3 metres and return to the chair), Unified Parkinson's Scale and Mini Mental Status Examination (MMSE) are commonly used (Cordell, et al., 2008). However, the predictive validity of using these tools in driving assessment is frequently questioned in the literature (Anceaux, et al., 2008; Betz & Fisher, 2009; Cordell, et al., 2008; Stolwyk, et al., 2006). Radford, Lincoln and Lennox (2004) stated that an objective and reliable assessment tool to measure driving ability do not currently exist. Based upon an extensive literature review, Molnar, Marshall and Man-Son-Hing (2006) concluded that no office-based test had validated cut-off scores that correlated to on-road driving performance amongst people with dementia. Ernst and Paulus (2005) noted that it is difficult to assess risk-taking behaviours in an indoor, clinical setting without actually watching the person drive. In a double blind study using 20 people with PD and 20 age-matched controls; it was found that there was a 35% inconsistency in clinical assessment results conducted by a neurologist, compared to on-road driving assessment results provided by a driving instructor and occupational therapist (Heikkila, et al., 1997). Although these results need to be interpreted with caution due to the small sample size; it does highlight that assessment processes need to be improved. Moreover, the Heikkila el al study (1997) did suggest that visual memory, choice reaction time and information processing speed tests could potentially be used to assess fitness to drive; once more research is conducted to establish validity and reliability. Betz and Fisher (2009) suggested that further research into the detection of cognitive impairment and its potential implications for road

safety is becoming more crucial in preventing fatal collisions as the population ages.

**1.4 Impact of poor decision making ability of PD drivers on driving performance**  PD-related cognitive deficits are believed to occur due the inefficient neurotransmission of dopamine-dependent neural connections between the basal ganglia and other areas of the brain (Tröster & Woods, 2007). The deprivation of dopamine, caused by the damage to the

contribute to the early cessation of driving (Access Economics, 2010).

analysis, and found that these six areas have been associated with previous incidents of unsafe driving and traffic errors (Amick, et al., 2007; Innes, et al., 2009).

In a study of 150 people with PD, it was found that cognitive impairment had a significant impact upon the crash rate per miles driven, irrespective of the actual disease severity (Devos, et al., 2007). Other studies have found that drivers with PD have increased indecision at T-junctions and when changing lanes, as well as a slower information processing speed, reaction time and decision making speed (Heikkila, et al., 1997; Stolwyk, et al., 2006). The current study focuses primarily upon decision making ability, which has been identified as one of the most important contributing factors to safe driving.

#### **1.3 Drivers with Parkinson's disease and road safety**

In 2008, traffic collisions caused 1,402 preventable deaths in Australia (Australian Bureau of Statistics, 2008). Deaths and disabilities caused by traffic collisions result in extensive, long term, social and emotional costs to families, friends and communities (Elvik & Vaa, 2004). Traffic collisions have vast financial implications; including healthcare services, insurance premiums, property damage and clean up services (Australian Bureau of Statistics, 2008). Therefore, improving road safety through research is of high importance to save lives and prevent disabilities. Although the majority of traffic collisions are preventable, the number of collisions is actually predicted to increase substantially in the future. Escalating population density in cities, increased usage of vehicles and number of cars per household are resulting in Australian road networks becoming more complicated and demanding (Australian Bureau of Statistics, 2009). The fastest growing population of Australian drivers are aged over 70 years, as improvements in healthcare have enabled drivers, including those with PD, to retain their licences for longer (Australian Bureau of Statistics, 2004). The ageing population demographics, in combination with the increased complexity of road systems, mean that the risk of collision for drivers over 65 years is predicted to triple by 2030 (Australian Bureau of Statistics, 2004). This older population are also more likely to sustain serious injuries or death during collisions due to age-related deterioration of musculoskeletal and cardiovascular systems (Adler, et al., 2000).

Longer licence retention can be very beneficial in improving the quality of life of older Australians, since they are able to maintain independence, access to the community and preserve their self-efficacy (Radford, et al., 2004). However, older drivers must be able to compensate for their age-related deficits, since the increasing complexity of road systems place additional demands on cognitive, physical and sensory systems (Elvik & Vaa, 2004). Drivers with PD face further challenges as the PD symptoms as well as side effects of medication can interfere with driving performance. Research, both on-road and using driving simulators, has shown that drivers with PD commit more risky faults and driving offences, and have a significantly increased number of collisions per kilometre driven when compared to the average population (Devos, et al., 2007; Radford, et al., 2004). Despite the challenges faced by drivers with PD in continuing to drive, it is unethical to cancel their licences based upon diagnosis of the disease alone (Tröster & Woods, 2007). Many drivers with PD are able to overcome barriers using their extensive driving experience and knowledge of road systems or they can compensate for the declining ability through selfmonitoring and self-regulation (Stolwyk, et al., 2006). For example, a person who becomes overwhelmed when driving at high speeds may change their route to avoid freeway driving (Tröster & Woods, 2007).

analysis, and found that these six areas have been associated with previous incidents of

In a study of 150 people with PD, it was found that cognitive impairment had a significant impact upon the crash rate per miles driven, irrespective of the actual disease severity (Devos, et al., 2007). Other studies have found that drivers with PD have increased indecision at T-junctions and when changing lanes, as well as a slower information processing speed, reaction time and decision making speed (Heikkila, et al., 1997; Stolwyk, et al., 2006). The current study focuses primarily upon decision making ability, which has

In 2008, traffic collisions caused 1,402 preventable deaths in Australia (Australian Bureau of Statistics, 2008). Deaths and disabilities caused by traffic collisions result in extensive, long term, social and emotional costs to families, friends and communities (Elvik & Vaa, 2004). Traffic collisions have vast financial implications; including healthcare services, insurance premiums, property damage and clean up services (Australian Bureau of Statistics, 2008). Therefore, improving road safety through research is of high importance to save lives and prevent disabilities. Although the majority of traffic collisions are preventable, the number of collisions is actually predicted to increase substantially in the future. Escalating population density in cities, increased usage of vehicles and number of cars per household are resulting in Australian road networks becoming more complicated and demanding (Australian Bureau of Statistics, 2009). The fastest growing population of Australian drivers are aged over 70 years, as improvements in healthcare have enabled drivers, including those with PD, to retain their licences for longer (Australian Bureau of Statistics, 2004). The ageing population demographics, in combination with the increased complexity of road systems, mean that the risk of collision for drivers over 65 years is predicted to triple by 2030 (Australian Bureau of Statistics, 2004). This older population are also more likely to sustain serious injuries or death during collisions due to age-related deterioration of

Longer licence retention can be very beneficial in improving the quality of life of older Australians, since they are able to maintain independence, access to the community and preserve their self-efficacy (Radford, et al., 2004). However, older drivers must be able to compensate for their age-related deficits, since the increasing complexity of road systems place additional demands on cognitive, physical and sensory systems (Elvik & Vaa, 2004). Drivers with PD face further challenges as the PD symptoms as well as side effects of medication can interfere with driving performance. Research, both on-road and using driving simulators, has shown that drivers with PD commit more risky faults and driving offences, and have a significantly increased number of collisions per kilometre driven when compared to the average population (Devos, et al., 2007; Radford, et al., 2004). Despite the challenges faced by drivers with PD in continuing to drive, it is unethical to cancel their licences based upon diagnosis of the disease alone (Tröster & Woods, 2007). Many drivers with PD are able to overcome barriers using their extensive driving experience and knowledge of road systems or they can compensate for the declining ability through selfmonitoring and self-regulation (Stolwyk, et al., 2006). For example, a person who becomes overwhelmed when driving at high speeds may change their route to avoid freeway driving

unsafe driving and traffic errors (Amick, et al., 2007; Innes, et al., 2009).

**1.3 Drivers with Parkinson's disease and road safety** 

musculoskeletal and cardiovascular systems (Adler, et al., 2000).

(Tröster & Woods, 2007).

been identified as one of the most important contributing factors to safe driving.

In Australia, like most of the developed countries, the guidelines regulating licence retainment and cancellation are based upon a system of subjective medical expert opinion (Adler, et al., 2000). There are no current national standards or requirements for how clinical driving assessments should be conducted (Innes, et al., 2009). Medical experts are often required to determine driving performance, even though the majority have not been trained in driving assessment, or actually observed their patient driving a car (Adler, et al., 2000). Specific clinical assessment batteries and criteria to renew or cancel driving licences have not been clearly defined in the Australian Assessing Fitness to Drive handbook; the combination of symptoms and/or the severity that could compromise driving ability are not defined (Cordell, et al., 2008). Therefore, the medical practitioner must make a subjective decision on the fitness to drive of their patients, even though they may not have been trained to do so (Cordell, et al., 2008). Most current methods of determining licence retainment or cancellation is through on-road driving tests and/or clinical psychometric assessments (National Road Transport Commission, 2003). On-road assessment is the gold standard. However, the process is costly and time consuming (Bedard, et al., 2010; Bryer, et al., 2006). A person who is unable to undergo a driving assessment as recommended by their medical professional is unlikely to be able to retain their licence (Anceaux, et al., 2008). The high assessment cost and need for drivers with PD to undergo annual driving reviews may contribute to the early cessation of driving (Access Economics, 2010).

The cheapest, most accessible and commonly used method for determining driving ability is through clinical assessment. Tools, such as the Timed Up and Go (measures ability to stand up, walk for 3 metres and return to the chair), Unified Parkinson's Scale and Mini Mental Status Examination (MMSE) are commonly used (Cordell, et al., 2008). However, the predictive validity of using these tools in driving assessment is frequently questioned in the literature (Anceaux, et al., 2008; Betz & Fisher, 2009; Cordell, et al., 2008; Stolwyk, et al., 2006). Radford, Lincoln and Lennox (2004) stated that an objective and reliable assessment tool to measure driving ability do not currently exist. Based upon an extensive literature review, Molnar, Marshall and Man-Son-Hing (2006) concluded that no office-based test had validated cut-off scores that correlated to on-road driving performance amongst people with dementia. Ernst and Paulus (2005) noted that it is difficult to assess risk-taking behaviours in an indoor, clinical setting without actually watching the person drive. In a double blind study using 20 people with PD and 20 age-matched controls; it was found that there was a 35% inconsistency in clinical assessment results conducted by a neurologist, compared to on-road driving assessment results provided by a driving instructor and occupational therapist (Heikkila, et al., 1997). Although these results need to be interpreted with caution due to the small sample size; it does highlight that assessment processes need to be improved. Moreover, the Heikkila el al study (1997) did suggest that visual memory, choice reaction time and information processing speed tests could potentially be used to assess fitness to drive; once more research is conducted to establish validity and reliability. Betz and Fisher (2009) suggested that further research into the detection of cognitive impairment and its potential implications for road safety is becoming more crucial in preventing fatal collisions as the population ages.

## **1.4 Impact of poor decision making ability of PD drivers on driving performance**

PD-related cognitive deficits are believed to occur due the inefficient neurotransmission of dopamine-dependent neural connections between the basal ganglia and other areas of the brain (Tröster & Woods, 2007). The deprivation of dopamine, caused by the damage to the

An Investigation into the Impact

(Adapted from: Kalis, et al., 2008; Lefy & Dubois, 2006)

indecisive driver's actions and react in time.

Fig. 2. Summary of the decision making process in driving

of Parkinson's Disease upon Decision Making Ability and Driving Performance 315

conditioning, passengers and the visual environment (Ernst & Paulus, 2005). Medication, fatigue, other PD symptoms, co-morbid conditions and environmental distractions can also

Decisions can be made either through conscious deliberation, for example, deciding if a parking space is large enough for the car, or through an unconscious process using previously learned behavioural patterns; for instance, automatically using the indicator when leaving a roundabout (Ernst & Paulus, 2005). PD can cause deficits in decision making ability at any of the decision making stages, and the resultant hesitancy, ambivalence or apathy may significantly impact upon road safety for the driver and other road users (Kalis, et al., 2008). As shown in Figure 2, if a driver is indecisive about whether to stop, slow down or to proceed through a roundabout, they could increase the risk of collision due to either incorrect use of signals, inappropriate speed or lane placement, sudden braking without checking review mirrors and/or impulsively increasing speed. All of these actions can directly result in a collision, especially as the other drivers may not be able to anticipate the

Numerous studies have identified that hesitancy and indecision contribute to a higher risk of crashing. However, the extent of the contribution is unknown (Bryer, Rapport, & Hanks, 2006; Stolwyk, et al., 2006). Drivers with PD frequently have a lack of cognitive flexibility and difficulty in shifting attention and multi-tasking, particularly when in stressful situations (Arias-Carrión & Pöppel, 2007). Drivers with PD often drive at slower speeds, have reduced reaction times and can fail to notice specific landmarks and traffic signs (Stolwyk, et al., 2006; Uc, et al., 2009). A study that surveyed 5,210 drivers with PD found that cognitive deficits are strongly associated with dangerous driving, with the most common causes of collision being indeciveness at T junctions and reduced usage of mirrors (Meindorfner, et al., 2005). A review of 42 driving studies concluded that the effect of a disease upon driving performance is difficult to determine due to numerous confounding factors. It is not currently possible to conduct an extensive randomised controlled trial into

intensify the deficits experienced by drivers with PD (Tröster & Woods, 2007).

basal ganglia, can directly affect the cognitive functions that are essential to decision making ability. These include; time estimation, working memory, executive function, compulsion, perseveration, attention, motivation and information processing speed (Cools, et al., 2001). Additionally, priority given to stimuli, error prediction, action planning, learning and interest in the environment are also affected (Ernst & Paulus, 2005). Furthermore, Nieoullon (2002) stated that the reduced amount of dopamine may interfere with a person's ability to perform an activity or behaviour, as well as alter a person's ability to adapt to environmental changes. Making decisions is a high-level cognitive function that involves the caudate nucleus and ventral striatum of the basal ganglia, as well as parts of the prefrontal cortex of the brain (Ernst & Paulus, 2005). The decision making process is reliant upon the neurotransmitter dopamine to transmit information via the mesocortical and mesolimbic pathways to the involved areas of the brain (Cools, et al., 2001). Due to the complexity of the decision making process, multiple high-level cerebral functions contribute to the ability to make a decision within a set period. These include attention, information processing speed and capacity, working memory, concentration, recall memory, planning, complex reasoning and risk assessment (Busemeyer & Stout, 2002; Kalis, et al., 2008). Fatigue, stress, emotions and medication can cause the speed and accuracy of decision making ability to fluctuate (Ernst & Paulus, 2005).

The Decision Making Process Model (see Figure 2) defines three important stages to making a decision: Option Generation, Option Selection and Action Initiation (Kalis, et al., 2008). PD can affect all of the components of decision making, although the severity of deficits vary from person to person (Stolwyk, et al., 2006; Tröster & Woods, 2007). This model has been employed in research to study PD in numerous activities other than driving (Levy & Dubois, 2006). Firstly, in Option Generation the person considers the requirements of the situation and thinks of possible courses of action. Then during the Option Selection stage, the person analyses each potential course of action for probable outcomes. Factors that can influence the selection of one course of action over the alternatives include: probability of the benefits and/or risks, the person's previous experiences, emotional state, values and preferences for one course of action (Ernst & Paulus, 2005). Finally, in Action Initiation, the decision is implemented through physical actions (Kalis, et al., 2008). The person then evaluates the results of the decision to promote learning for future situations. According to Busemeyer and Stout (2002), poor decisions can be due to a failure to anticipate consequences, poor perceptual sensitivity, problems in memory storage or retrieval, inability to determine possible courses of action, fatigue, poor concentration, difficulty in learning from mistakes, and/or impulsivity.

Decision making deficits have been recognised as a key area that could influence driving competence and safety amongst people with PD (Cools, et al., 2001). Dopamine has an important role in facilitating the cognitive processes that enable a person to make a decision. However, what this functionally entails for driving is poorly understood (Arias-Carrión & Pöppel, 2007; Cools, et al., 2001). Deficits in decision making are most apparent during activities, such as driving, that require spontaneous, complex information processing and reasoning within time constraints (Tröster & Woods, 2007). The driver may have to make multiple decisions in quick succession, which place extensive demands upon cognitive processes. The driver must quickly consider all components of the situation, generate and consider options, implement the choice, evaluate the result and then start the decision making process again (Busemeyer & Stout, 2002). The driver may also have to ignore multiple distracting auditory, visual and tactile stimuli from the car's radio, air

basal ganglia, can directly affect the cognitive functions that are essential to decision making ability. These include; time estimation, working memory, executive function, compulsion, perseveration, attention, motivation and information processing speed (Cools, et al., 2001). Additionally, priority given to stimuli, error prediction, action planning, learning and interest in the environment are also affected (Ernst & Paulus, 2005). Furthermore, Nieoullon (2002) stated that the reduced amount of dopamine may interfere with a person's ability to perform an activity or behaviour, as well as alter a person's ability to adapt to environmental changes. Making decisions is a high-level cognitive function that involves the caudate nucleus and ventral striatum of the basal ganglia, as well as parts of the prefrontal cortex of the brain (Ernst & Paulus, 2005). The decision making process is reliant upon the neurotransmitter dopamine to transmit information via the mesocortical and mesolimbic pathways to the involved areas of the brain (Cools, et al., 2001). Due to the complexity of the decision making process, multiple high-level cerebral functions contribute to the ability to make a decision within a set period. These include attention, information processing speed and capacity, working memory, concentration, recall memory, planning, complex reasoning and risk assessment (Busemeyer & Stout, 2002; Kalis, et al., 2008). Fatigue, stress, emotions and medication can cause the speed and accuracy of decision making ability to fluctuate

The Decision Making Process Model (see Figure 2) defines three important stages to making a decision: Option Generation, Option Selection and Action Initiation (Kalis, et al., 2008). PD can affect all of the components of decision making, although the severity of deficits vary from person to person (Stolwyk, et al., 2006; Tröster & Woods, 2007). This model has been employed in research to study PD in numerous activities other than driving (Levy & Dubois, 2006). Firstly, in Option Generation the person considers the requirements of the situation and thinks of possible courses of action. Then during the Option Selection stage, the person analyses each potential course of action for probable outcomes. Factors that can influence the selection of one course of action over the alternatives include: probability of the benefits and/or risks, the person's previous experiences, emotional state, values and preferences for one course of action (Ernst & Paulus, 2005). Finally, in Action Initiation, the decision is implemented through physical actions (Kalis, et al., 2008). The person then evaluates the results of the decision to promote learning for future situations. According to Busemeyer and Stout (2002), poor decisions can be due to a failure to anticipate consequences, poor perceptual sensitivity, problems in memory storage or retrieval, inability to determine possible courses of action, fatigue, poor concentration, difficulty in

Decision making deficits have been recognised as a key area that could influence driving competence and safety amongst people with PD (Cools, et al., 2001). Dopamine has an important role in facilitating the cognitive processes that enable a person to make a decision. However, what this functionally entails for driving is poorly understood (Arias-Carrión & Pöppel, 2007; Cools, et al., 2001). Deficits in decision making are most apparent during activities, such as driving, that require spontaneous, complex information processing and reasoning within time constraints (Tröster & Woods, 2007). The driver may have to make multiple decisions in quick succession, which place extensive demands upon cognitive processes. The driver must quickly consider all components of the situation, generate and consider options, implement the choice, evaluate the result and then start the decision making process again (Busemeyer & Stout, 2002). The driver may also have to ignore multiple distracting auditory, visual and tactile stimuli from the car's radio, air

(Ernst & Paulus, 2005).

learning from mistakes, and/or impulsivity.

conditioning, passengers and the visual environment (Ernst & Paulus, 2005). Medication, fatigue, other PD symptoms, co-morbid conditions and environmental distractions can also intensify the deficits experienced by drivers with PD (Tröster & Woods, 2007).

(Adapted from: Kalis, et al., 2008; Lefy & Dubois, 2006)

Fig. 2. Summary of the decision making process in driving

Decisions can be made either through conscious deliberation, for example, deciding if a parking space is large enough for the car, or through an unconscious process using previously learned behavioural patterns; for instance, automatically using the indicator when leaving a roundabout (Ernst & Paulus, 2005). PD can cause deficits in decision making ability at any of the decision making stages, and the resultant hesitancy, ambivalence or apathy may significantly impact upon road safety for the driver and other road users (Kalis, et al., 2008). As shown in Figure 2, if a driver is indecisive about whether to stop, slow down or to proceed through a roundabout, they could increase the risk of collision due to either incorrect use of signals, inappropriate speed or lane placement, sudden braking without checking review mirrors and/or impulsively increasing speed. All of these actions can directly result in a collision, especially as the other drivers may not be able to anticipate the indecisive driver's actions and react in time.

Numerous studies have identified that hesitancy and indecision contribute to a higher risk of crashing. However, the extent of the contribution is unknown (Bryer, Rapport, & Hanks, 2006; Stolwyk, et al., 2006). Drivers with PD frequently have a lack of cognitive flexibility and difficulty in shifting attention and multi-tasking, particularly when in stressful situations (Arias-Carrión & Pöppel, 2007). Drivers with PD often drive at slower speeds, have reduced reaction times and can fail to notice specific landmarks and traffic signs (Stolwyk, et al., 2006; Uc, et al., 2009). A study that surveyed 5,210 drivers with PD found that cognitive deficits are strongly associated with dangerous driving, with the most common causes of collision being indeciveness at T junctions and reduced usage of mirrors (Meindorfner, et al., 2005). A review of 42 driving studies concluded that the effect of a disease upon driving performance is difficult to determine due to numerous confounding factors. It is not currently possible to conduct an extensive randomised controlled trial into

An Investigation into the Impact

been taken to ensure that the data collected was valid.

**2.3.1 Initial screening of participant medical and driving history** 

**2.3 Equipment used in the study** 

**2.3.2 Psychometric assessment** 

driving.

of Parkinson's Disease upon Decision Making Ability and Driving Performance 317

cognitive or physical impairment, depression and/or psychiatric conditions. Participants were withdrawn from the study immediately if they requested to do so. A reason for withdrawal was not required. Fifteen drivers with PD and 17 control group participants were recruited were contacted by phone to establish suitability to participate in the study. To address Study Objectives 2 and 3, baseline-driving performance was established in Trial 1 and then a time constraint was imposed to create pressure upon the participants. In Trial 2, all participants were told to complete the same driving scenario 20% faster than in Trial 1. The percentage of reduction in time was based upon pilot study data. A 20% reduction represented a time that was perceived by the participants as being challenging, yet achievable within the driving assessment parameters. This time pressure forced the participants to make quicker decisions in response to the traffic conditions, without compromising on safety or breaking the road rules. Drivers with PD are more likely to experience decision making deficits whilst making complex decisions under pressure (Amick, et al., 2007). The study assumption is that drivers with PD are capable of making correct decisions; however, they require more time to do so. Important driving behaviours, such as appropriate signalling, use of mirrors and obeying the speed limit potentially could have been affected and/or forgotten as the participants concentrated upon negotiating the scenario faster. The driving performance of participants was measured using Driving Performance Score. A battery of psychometric assessment tools were administered to the participants to assess the cognitive processes that are essential to decision making ability. The cognitive processes included executive function, task switching, sustained, selected and divided attention, attention set shifting, memory, efficiency and accuracy of information processing systems, visual attention and decision making speed and accuracy. All psychometric assessment tools were time based, standardised instruments that measured speed and accuracy of response. The study assumption was that drivers with PD are capable of completing the assessments; however, they will require more time to do so. The confounding variables in the study are presented in Table 1 in next page. Measures have

The following section describes the tools used for initial screening of participants, and psychometric assessments for measuring the main components of decision making in

Standardised clinical assessment tools and a Medical History and Driving History Checklist were used to screen for potentially confounding factors (refer to Table 2 and Table 3 for details of assessments). All assessments were administered in a quiet, distraction free room as per the instruction manuals to ensure the reliability of data. The research assistant was trained in administering these assessments prior to commencement of the data collection.

Decision making ability cannot be directly measured. Instead, the main contributing components were all assessed using a battery of psychometric assessments. These components were attention set shifting, visual attention, memory, information processing speed and decision-making speed and accuracy (refer to Table 3). The psychometric

this area, since there is not yet enough information available to control all confounding variables (Elvik & Vaa, 2004). Therefore, the study reported in this chapter was valuable in trialling alternative assessment methodologies and making recommendations for future research projects. Information from the study may also contribute to the development of a successful assessment protocol for drivers with PD to improve road safety.

## **2. Methodology**

## **2.1 Purpose of study**

The aim of the research was to explore the impact of impaired decision making ability upon the driving performance of people with PD. To address the aim, a quantitative, pre-post case-control study design was employed to assess participants the decision making ability of drivers with PD and healthy controls, as well as their driving performance under time pressure, were examined. The objectives of the study are: **Objective 1:** To assess the decision making ability of drivers with PD using standardised psychometric assessment tools and the E-prime computer based assessment; **Objective 2:** To investigate the relationship between the decision making ability and driving performance of people with PD; and **Objective 3:** To compare the driving performance of people with PD to the healthy control group whilst driving under a time pressure in the driving simulator.

The first objective was addressed by administering an assessment battery of clinical psychometric tests to assess the main cognitive processes that contribute to decision making ability. The assumption was that drivers with PD would have lower scores on the psychometric assessments, due to PD-related cognitive impairments, when compared with the healthy control group. The second objective was addressed by assessing the driving performance of the groups on the driving simulator. The assumption was that drivers with PD would have poorer driving performance at baseline driving (Trial One) as well as driving under time pressure (Trial Two) when compared to the healthy control group. The third objective was addressed by analysing the results from stage one and two to determine if there is a correlation between driving performance and decision making ability. The assumption was that the ability of people with PD to make correct decisions whilst driving under time pressure would be significantly lower than the control group. Ethical approval was granted by the Curtin University Human Research Ethics. Data was collected from Sept 2009 until March 2010 at the Curtin University Driving Rehabilitation Clinic.

#### **2.2 Participants**

Convenience sampling was used to recruit participants by displaying advertisement posters at community centres, retirement villages, shopping centres and neurologists' offices. Advertisements were also placed in community newsletters, as well as the Western Australia Parkinson's Association newsletter. Study participants were required to be community living adults, aged 50 to 80 years old with a valid driving licence. They had to be current drivers, driving at least half an hour each week. To ensure adequate binocular acuity, a score of at least 6/12 corrected vision on the Snellen Acuity Chart was required. In the experimental group, each participant's diagnosis of PD had to have been confirmed by a general practitioner or neurologist. Participants were excluded from the study if they had severe hearing impairments or inadequate comprehension of written or verbal English as judged by the researcher, or any co-morbid diagnosis that may interfere with driving ability. Participants with the following conditions were excluded from the study: dementia, severe

this area, since there is not yet enough information available to control all confounding variables (Elvik & Vaa, 2004). Therefore, the study reported in this chapter was valuable in trialling alternative assessment methodologies and making recommendations for future research projects. Information from the study may also contribute to the development of a

The aim of the research was to explore the impact of impaired decision making ability upon the driving performance of people with PD. To address the aim, a quantitative, pre-post case-control study design was employed to assess participants the decision making ability of drivers with PD and healthy controls, as well as their driving performance under time pressure, were examined. The objectives of the study are: **Objective 1:** To assess the decision making ability of drivers with PD using standardised psychometric assessment tools and the E-prime computer based assessment; **Objective 2:** To investigate the relationship between the decision making ability and driving performance of people with PD; and **Objective 3:** To compare the driving performance of people with PD to the healthy control

The first objective was addressed by administering an assessment battery of clinical psychometric tests to assess the main cognitive processes that contribute to decision making ability. The assumption was that drivers with PD would have lower scores on the psychometric assessments, due to PD-related cognitive impairments, when compared with the healthy control group. The second objective was addressed by assessing the driving performance of the groups on the driving simulator. The assumption was that drivers with PD would have poorer driving performance at baseline driving (Trial One) as well as driving under time pressure (Trial Two) when compared to the healthy control group. The third objective was addressed by analysing the results from stage one and two to determine if there is a correlation between driving performance and decision making ability. The assumption was that the ability of people with PD to make correct decisions whilst driving under time pressure would be significantly lower than the control group. Ethical approval was granted by the Curtin University Human Research Ethics. Data was collected from Sept

Convenience sampling was used to recruit participants by displaying advertisement posters at community centres, retirement villages, shopping centres and neurologists' offices. Advertisements were also placed in community newsletters, as well as the Western Australia Parkinson's Association newsletter. Study participants were required to be community living adults, aged 50 to 80 years old with a valid driving licence. They had to be current drivers, driving at least half an hour each week. To ensure adequate binocular acuity, a score of at least 6/12 corrected vision on the Snellen Acuity Chart was required. In the experimental group, each participant's diagnosis of PD had to have been confirmed by a general practitioner or neurologist. Participants were excluded from the study if they had severe hearing impairments or inadequate comprehension of written or verbal English as judged by the researcher, or any co-morbid diagnosis that may interfere with driving ability. Participants with the following conditions were excluded from the study: dementia, severe

successful assessment protocol for drivers with PD to improve road safety.

group whilst driving under a time pressure in the driving simulator.

2009 until March 2010 at the Curtin University Driving Rehabilitation Clinic.

**2. Methodology 2.1 Purpose of study** 

**2.2 Participants** 

cognitive or physical impairment, depression and/or psychiatric conditions. Participants were withdrawn from the study immediately if they requested to do so. A reason for withdrawal was not required. Fifteen drivers with PD and 17 control group participants

were recruited were contacted by phone to establish suitability to participate in the study. To address Study Objectives 2 and 3, baseline-driving performance was established in Trial 1 and then a time constraint was imposed to create pressure upon the participants. In Trial 2, all participants were told to complete the same driving scenario 20% faster than in Trial 1. The percentage of reduction in time was based upon pilot study data. A 20% reduction represented a time that was perceived by the participants as being challenging, yet achievable within the driving assessment parameters. This time pressure forced the participants to make quicker decisions in response to the traffic conditions, without compromising on safety or breaking the road rules. Drivers with PD are more likely to experience decision making deficits whilst making complex decisions under pressure (Amick, et al., 2007). The study assumption is that drivers with PD are capable of making correct decisions; however, they require more time to do so. Important driving behaviours, such as appropriate signalling, use of mirrors and obeying the speed limit potentially could have been affected and/or forgotten as the participants concentrated upon negotiating the scenario faster. The driving performance of participants was measured using Driving Performance Score. A battery of psychometric assessment tools were administered to the participants to assess the cognitive processes that are essential to decision making ability. The cognitive processes included executive function, task switching, sustained, selected and divided attention, attention set shifting, memory, efficiency and accuracy of information processing systems, visual attention and decision making speed and accuracy. All psychometric assessment tools were time based, standardised instruments that measured speed and accuracy of response. The study assumption was that drivers with PD are capable of completing the assessments; however, they will require more time to do so. The confounding variables in the study are presented in Table 1 in next page. Measures have been taken to ensure that the data collected was valid.

## **2.3 Equipment used in the study**

The following section describes the tools used for initial screening of participants, and psychometric assessments for measuring the main components of decision making in driving.

#### **2.3.1 Initial screening of participant medical and driving history**

Standardised clinical assessment tools and a Medical History and Driving History Checklist were used to screen for potentially confounding factors (refer to Table 2 and Table 3 for details of assessments). All assessments were administered in a quiet, distraction free room as per the instruction manuals to ensure the reliability of data. The research assistant was trained in administering these assessments prior to commencement of the data collection.

#### **2.3.2 Psychometric assessment**

Decision making ability cannot be directly measured. Instead, the main contributing components were all assessed using a battery of psychometric assessments. These components were attention set shifting, visual attention, memory, information processing speed and decision-making speed and accuracy (refer to Table 3). The psychometric

An Investigation into the Impact

Medical Checklist

Driving History Checklist

Snellen Acuity

Chart

Cognistat (Kiernan, Mueller,

Langston, & Van Dyke, 1987)

Table 2. Outline of Screening Tools

**2.3.3 E-Prime computer based tool** 

of Parkinson's Disease upon Decision Making Ability and Driving Performance 319

Screening Tool Purpose of Assessment Tool Administration and justification for





al., 2008; Lee, et al., 2003). It took approximately 10 minutes to complete.

participants about how to answer the following question (refer to Figure 4).

cognitive impairment Subtests of attention, constructional ability, memory, calculations, reasoning and judgement were administered.

Use





(Adler, et al., 2000)

The E-Prime software has been used in 104 research studies since 2001; including research projects into simulated situations, older adults and neurological conditions (Psychology Software Tools, 2010). The E-Prime software is capable of millisecond precision and is frequently used in research to increase the accuracy and reliability of data (Ranzini, et al., 2009). In the present study, the E-Prime computer program was set up to measure the speed and accuracy of the participants' decision making ability by administering a series of multiple choice questions (refer to Figure 3). The questions were based upon traffic situations in which drivers with PD are known to experience difficulty; such as roundabouts, traffic lights, freeway driving, city driving, over taking and right hand turns (Allen, et al., 2003; Anceaux, et

The "red", "yellow" and "green" button system (refer to Figure 3) had buttons that were large, visually distinguishable, and highly sensitive to touch, to enable people, who experienced PDrelated physical symptoms to enter their decision as quickly as possible. The computer was placed in front of a blank, white wall and the researcher sat behind the participant, out of sight to prevent potential distractions. The questions were displayed in large, white writing on a black backdrop to improve readability. A black instruction screen was displayed to inform


Table 1. Confounding variables of the study

assessment battery comprised of the Symbol Digit Modalities Test (Smith, 2007), Digit Vigilance Test (Kelland & Lewis, 1996), Purdue Pegboard (Lafayette Instrument Company, 1985) and Trail Making Test – B (Corrigan & Hinkeldey, 1987).

The assessments were chosen based upon recommendations from literature to ensure high reliability, sensitivity, and/or validity of each test in assessing driving performance. For example, the Trail Making Test-B, Symbol Digit Modalities Test and Digit Vigilance are highly sensitive to detecting differences in cognitive performance (Smith, 2007). The Trail Making Test-B is one of the most frequently used tests in driving research and clinical settings, due to its high reliability and sensitivity to mild cognitive impairment (Arbuthnott & Frank, 2000; Ashendorf et al., 2008). The Symbol Digit Modalities Test was found in a study of 150 people to be the most reliable of 12 assessment tools in detecting mild cognitive impairment (Ashendorf, et al., 2008).


Table 2. Outline of Screening Tools

318 Diagnostics and Rehabilitation of Parkinson's Disease


(Radford, et al., 2004).

Screening procedure.

were offered.

performance.

performance.

assessment battery comprised of the Symbol Digit Modalities Test (Smith, 2007), Digit Vigilance Test (Kelland & Lewis, 1996), Purdue Pegboard (Lafayette Instrument Company,

The assessments were chosen based upon recommendations from literature to ensure high reliability, sensitivity, and/or validity of each test in assessing driving performance. For example, the Trail Making Test-B, Symbol Digit Modalities Test and Digit Vigilance are highly sensitive to detecting differences in cognitive performance (Smith, 2007). The Trail Making Test-B is one of the most frequently used tests in driving research and clinical settings, due to its high reliability and sensitivity to mild cognitive impairment (Arbuthnott & Frank, 2000; Ashendorf et al., 2008). The Symbol Digit Modalities Test was found in a study of 150 people to be the most reliable of 12 assessment tools in detecting mild cognitive






Variable Potential Impact Measures taken to improve validity






Driving performance usually decreases after the age of 60 years (Adler, et al., 2000)

1985) and Trail Making Test – B (Corrigan & Hinkeldey, 1987).

al., 2004).

2004).

2010).

et al., 2005).

Table 1. Confounding variables of the study

impairment (Ashendorf, et al., 2008).

Medication

Co Morbid Conditions

Fatigue

Driving Experience

Gender

Age

## **2.3.3 E-Prime computer based tool**

The E-Prime software has been used in 104 research studies since 2001; including research projects into simulated situations, older adults and neurological conditions (Psychology Software Tools, 2010). The E-Prime software is capable of millisecond precision and is frequently used in research to increase the accuracy and reliability of data (Ranzini, et al., 2009). In the present study, the E-Prime computer program was set up to measure the speed and accuracy of the participants' decision making ability by administering a series of multiple choice questions (refer to Figure 3). The questions were based upon traffic situations in which drivers with PD are known to experience difficulty; such as roundabouts, traffic lights, freeway driving, city driving, over taking and right hand turns (Allen, et al., 2003; Anceaux, et al., 2008; Lee, et al., 2003). It took approximately 10 minutes to complete.

The "red", "yellow" and "green" button system (refer to Figure 3) had buttons that were large, visually distinguishable, and highly sensitive to touch, to enable people, who experienced PDrelated physical symptoms to enter their decision as quickly as possible. The computer was placed in front of a blank, white wall and the researcher sat behind the participant, out of sight to prevent potential distractions. The questions were displayed in large, white writing on a black backdrop to improve readability. A black instruction screen was displayed to inform participants about how to answer the following question (refer to Figure 4).

An Investigation into the Impact

Fig. 3. The setup of the E-Prime Assessment tool

Practice Question 1

**Yellow** to slow down **Green** to proceed

Press any key to continue

Press **Red** to stop

You are driving the car with the yellow star about to turn left.

Fig. 4. E-Prime Instruction Screen

Fig. 5. The E-Prime Assessment Tool; displaying an example of a question

In each photograph there was a car labelled with a bright yellow star (refer to Figure 5). Participants were instructed to assume that he or she was the driver of the car with the yellow star and give the most appropriate response for each scenario. The participant had to decide whether they would 'stop', 'slow down' or 'proceed' based upon their interpretation of the hazards as shown in the photograph. Participants responded using one of the three

of Parkinson's Disease upon Decision Making Ability and Driving Performance 321


Table 3. Outline of the Psychometric Assessment Tools

Fig. 3. The setup of the E-Prime Assessment tool

320 Diagnostics and Rehabilitation of Parkinson's Disease


Hinkeldey, 1987). Participants join alternating dots of letters and numbers (1, A, 2, B etc.).




processing systems. Participants had to convert geometric shapes into numbers as quickly as

To assess decision making accuracy and response time - Multiple-choice questions based upon photographs of different driving scenarios.

Table 3. Outline of the Psychometric Assessment Tools

**Purpose and Administration Literature support for tool validity** 

2010).

1996).




Determines if participants were able to remember and attend to important information whilst disregarding excess



stimuli (Radford, et al., 2004).

(Wood, et al., 2005)

intra-rater reliability.

**Psychometric** 

Trail Making Test B (TMT-B) (Corrigan & Hinkeldey, 1987)

Purdue Peg Board (Lafayette Instrument Company, 1985)

Digit Vigilance

results.

sixes.

possible.

Symbol Digit Modalities Test (Smith, 2007)

E-Prime Computer Based Assessment (Psychology Software Tools,

2010)

Test (Kelland & Lewis, 1996)

**Tool** 


Fig. 4. E-Prime Instruction Screen

Fig. 5. The E-Prime Assessment Tool; displaying an example of a question

In each photograph there was a car labelled with a bright yellow star (refer to Figure 5). Participants were instructed to assume that he or she was the driver of the car with the yellow star and give the most appropriate response for each scenario. The participant had to decide whether they would 'stop', 'slow down' or 'proceed' based upon their interpretation of the hazards as shown in the photograph. Participants responded using one of the three

An Investigation into the Impact

instructions were consistent throughout data collection.

Assessment Guidelines was tabulated in Table 4.

statistical tests were checked to ensure that there were no violations.

education level were found to be not significantly different between groups.

**2.4 Data analysis** 

**3. Study results** 

**3.1 Participant demographics** 

of Parkinson's Disease upon Decision Making Ability and Driving Performance 323

driving simulator literature (Allen, et al., 2003; Factor & Weiner, 2002; Lee, et al., 2003; National Road Transport Commission, 2003). In the present study, the roadway geometry and intersections, position of traffic signals and markings, weather conditions, the responsiveness of vehicle controls, location of other vehicles and road users were all programmed to target decision making ability. The scenarios included small town, city and country driving, simple and complex intersections, curved roads, simulated emergency braking, varied speed control and visually obscured intersections. Auditory instructions were included in the simulator programming to ensure that all of the information and

To investigate the impact of PD-related decision making deficits upon driving performance, the scenario in this study was designed to specially assess hazard detection, risk assessment, impulsiveness and decision making ability (Bedard, et al., 2010; Elvik & Vaa, 2004). Traffic situations that are known to be affected by PD, such as driving at high speeds, turning corners, overtaking, merging and complex city intersections were included (Stolwyk, et al., 2006; Radford, et al., 2006). For example, during the scenario, a recorded verbal instruction told each participant to overtake three slow moving trucks whilst avoiding oncoming traffic. A similar process was used in a study by Amick et al., (2007) as they researched cognitive indicators of poor driving performance of drivers with PD. Driving Performance

Data was analysed using the Statistical Package for Social Sciences (SPSS) (SPSS Inc. 2009). Demographic information of participants was presented using descriptive statistics. The difference in total run time and driving performance score between groups was analysed using t-tests; whereas a Chi squared test and Fisher's Exact test was used to analyse ordinal variables, such as gender and number of collisions and infringements. A stepwise Multiple Linear Regression Model was used to analyse the driving performance and E-Prime scores; the driving performance score was the dependent variable and E-Prime (correct answers, time taken and participant group) were the independent variables. The psychometric assessments and the components of the Driving Performance Score were analysed using the non-parametric Wilcoxon 2-sample test. A repeated measure regression analysis was performed using the driving score as a dependent variable and the results of the psychometric assessments, simulator trial run number and group identifier (drivers with PD or control group) as independent variables. The least significant variables were then removed, one at a time, until the p-value associated with each of the remaining variables was less than 0.05. Prior to the analysis, normality of data and the assumptions of the

Seventeen people in the control group and 11 drivers with PD were assessed and their demographic data was tabulated in Table 5. In exploring the characteristics of the participants, it was identified that the number of years of driving experience was different between the comparison groups (p=0.042). The drivers with PD group had driven on average 7 years and 8 months longer than the control group. The participants' age, gender, employment status and

buttons. The accuracy of answers and response time was automatically recorded by the E-Prime software to determine decision making ability.

## **2.3.4 Driving simulator in curtin driving rehabilitation clinic**

A fixed-base, Systems Technology Incorporated (STI) driving simulator was used to assess driving performance in this study (Lee, et al., 2003). Driving simulators are frequently used in research and clinical practice to assess driving ability, since risk of injury and property damage is eliminated (Bedard, et al., 2010). The STISIM driving simulator enables the development of highly controlled and regulated traffic scenarios (Allen, et al., 2003). The STISIM simulator technology has been used in 61 different studies, whilst the STISIM simulator driving technology in particular has been used in at least 24 studies in the past eight years (Systems Techonology Inc, 2010). Low cost, fixed base driving simulators have been used in research on older drivers and on the effect of fatigue, drugs, cognitive impairment, Alzheimer's disease, PD, traumatic brain injury and numerous other conditions upon driving performance (Bedard, et al., 2010; Lee, et al., 2003). Driving simulators are becoming more affordable options; especially as on-road assessment costs are becoming more prohibitive due to the increasing fuel, car purchase and maintenance costs and higher insurance premiums (Bedard, et al., 2010).

Fig. 6. The Curtin University STISIM Driving Simulator

Simulators are capable of distinguishing between safe and unsafe drivers (De Winter, et al., 2008; Lee, et al., 2003). Numerous studies have found high transferability of simulator-based behaviours to on-road driving behaviours (De Winter, et al., 2008; Lee, et al., 2003). Factor and Weiner (2002) found that driving simulators have a greater accuracy in predicting driving ability than the clinical psychometric assessments currently used by medical practitioners. High inter-rater and intra-rater reliability (correlation coefficients were 0.87 and 0.83 respectively) were recently established by Bedard and colleagues (2010). They used the simulator-recorded data and data manually recorded by a laboratory assistant in a similar STISIM simulator. The validity of the driving simulator used in this study has been established for assessing older adults (Lee, et al., 2003). A photograph of a participant being assessed on the Curtin University STISIM Driving Simulator is shown in Fig. 6.

### **2.3.5 Development of the STISIM driving scenario**

Two driving scenarios were specially designed for the present study. They were based upon the Western Australian licensing standards, in combination with recommendations from driving simulator literature (Allen, et al., 2003; Factor & Weiner, 2002; Lee, et al., 2003; National Road Transport Commission, 2003). In the present study, the roadway geometry and intersections, position of traffic signals and markings, weather conditions, the responsiveness of vehicle controls, location of other vehicles and road users were all programmed to target decision making ability. The scenarios included small town, city and country driving, simple and complex intersections, curved roads, simulated emergency braking, varied speed control and visually obscured intersections. Auditory instructions were included in the simulator programming to ensure that all of the information and instructions were consistent throughout data collection.

To investigate the impact of PD-related decision making deficits upon driving performance, the scenario in this study was designed to specially assess hazard detection, risk assessment, impulsiveness and decision making ability (Bedard, et al., 2010; Elvik & Vaa, 2004). Traffic situations that are known to be affected by PD, such as driving at high speeds, turning corners, overtaking, merging and complex city intersections were included (Stolwyk, et al., 2006; Radford, et al., 2006). For example, during the scenario, a recorded verbal instruction told each participant to overtake three slow moving trucks whilst avoiding oncoming traffic. A similar process was used in a study by Amick et al., (2007) as they researched cognitive indicators of poor driving performance of drivers with PD. Driving Performance Assessment Guidelines was tabulated in Table 4.

## **2.4 Data analysis**

322 Diagnostics and Rehabilitation of Parkinson's Disease

buttons. The accuracy of answers and response time was automatically recorded by the E-

A fixed-base, Systems Technology Incorporated (STI) driving simulator was used to assess driving performance in this study (Lee, et al., 2003). Driving simulators are frequently used in research and clinical practice to assess driving ability, since risk of injury and property damage is eliminated (Bedard, et al., 2010). The STISIM driving simulator enables the development of highly controlled and regulated traffic scenarios (Allen, et al., 2003). The STISIM simulator technology has been used in 61 different studies, whilst the STISIM simulator driving technology in particular has been used in at least 24 studies in the past eight years (Systems Techonology Inc, 2010). Low cost, fixed base driving simulators have been used in research on older drivers and on the effect of fatigue, drugs, cognitive impairment, Alzheimer's disease, PD, traumatic brain injury and numerous other conditions upon driving performance (Bedard, et al., 2010; Lee, et al., 2003). Driving simulators are becoming more affordable options; especially as on-road assessment costs are becoming more prohibitive due to the increasing fuel, car purchase and maintenance costs and higher insurance premiums (Bedard, et al., 2010).

Simulators are capable of distinguishing between safe and unsafe drivers (De Winter, et al., 2008; Lee, et al., 2003). Numerous studies have found high transferability of simulator-based behaviours to on-road driving behaviours (De Winter, et al., 2008; Lee, et al., 2003). Factor and Weiner (2002) found that driving simulators have a greater accuracy in predicting driving ability than the clinical psychometric assessments currently used by medical practitioners. High inter-rater and intra-rater reliability (correlation coefficients were 0.87 and 0.83 respectively) were recently established by Bedard and colleagues (2010). They used the simulator-recorded data and data manually recorded by a laboratory assistant in a similar STISIM simulator. The validity of the driving simulator used in this study has been established for assessing older adults (Lee, et al., 2003). A photograph of a participant being

Two driving scenarios were specially designed for the present study. They were based upon the Western Australian licensing standards, in combination with recommendations from

assessed on the Curtin University STISIM Driving Simulator is shown in Fig. 6.

Prime software to determine decision making ability.

Fig. 6. The Curtin University STISIM Driving Simulator

**2.3.5 Development of the STISIM driving scenario** 

**2.3.4 Driving simulator in curtin driving rehabilitation clinic** 

Data was analysed using the Statistical Package for Social Sciences (SPSS) (SPSS Inc. 2009). Demographic information of participants was presented using descriptive statistics. The difference in total run time and driving performance score between groups was analysed using t-tests; whereas a Chi squared test and Fisher's Exact test was used to analyse ordinal variables, such as gender and number of collisions and infringements. A stepwise Multiple Linear Regression Model was used to analyse the driving performance and E-Prime scores; the driving performance score was the dependent variable and E-Prime (correct answers, time taken and participant group) were the independent variables. The psychometric assessments and the components of the Driving Performance Score were analysed using the non-parametric Wilcoxon 2-sample test. A repeated measure regression analysis was performed using the driving score as a dependent variable and the results of the psychometric assessments, simulator trial run number and group identifier (drivers with PD or control group) as independent variables. The least significant variables were then removed, one at a time, until the p-value associated with each of the remaining variables was less than 0.05. Prior to the analysis, normality of data and the assumptions of the statistical tests were checked to ensure that there were no violations.

## **3. Study results**

## **3.1 Participant demographics**

Seventeen people in the control group and 11 drivers with PD were assessed and their demographic data was tabulated in Table 5. In exploring the characteristics of the participants, it was identified that the number of years of driving experience was different between the comparison groups (p=0.042). The drivers with PD group had driven on average 7 years and 8 months longer than the control group. The participants' age, gender, employment status and education level were found to be not significantly different between groups.

An Investigation into the Impact

Weekly hours driving

Number of Collisions in last 2

Number of Infringements in last 2

Gender

years

years

Education Level – Tertiary Study – Year 12 High School

 – No answer Disease Symptoms – Tremors in legs – Tremors in arms – Mild Rigidity – Moderate Rigidity – Severe Rigidity – Mild Fatigue – Moderate Fatigue – Severe Fatigue

Categorical frequency (percentage)

any assistance.

Table 5. Results of Participant Demographic Data

**3.2 Psychometric assessment results** 

**Variable Drivers with** 

of Parkinson's Disease upon Decision Making Ability and Driving Performance 325

**PD n=11 Mean (SD)** 

0

4 (36%) 4 (36%) 3 (28%)

3 (28%) 7 (63%) 4 (36%) 4 (36%) 0 2 (18%) 7 (63%) 0

^ Chi squared test; # T-test; + Fisher's Exact test; \*Results were statistically significant (p<0.05) and

The drivers with PD had on average a diagnosis of PD for approximately 8 years and 4 months. Medications that were prescribed to the participants with PD included: Sinemet, Madopar, Cabaser, Sifrol and Selgene. Some participants with PD reported experiencing tremors in arms and legs as well as mild to moderate rigidity and fatigue (refer to Table 5). Six of the control participants reported experiencing mild to moderate fatigue, which was not related to PD. All participants with PD reported they required only minimal assistance to complete self-care activities, whilst none of the participants in the control group required

The results of four standardised, psychometric assessments and the E-Prime Assessment Tool are shown in Table 6. The only psychometric assessment tool that detected a difference between the groups was the Purdue Pegboard Both Hands subtest and Overall Score. These results indicate that there may be a difference in the speed and dexterity of upper limb

Age 68.2 (5.3) 65.6 (8.8) 0.427#

– Female 6 (55%) 7 (41%) 0.489^

 – Minimum 8.0 (8.5) 13.7 (9.9) 0.162# – Maximum 9 (8.6) 14.7 (11.3) 0.204# Years of Driving Experience 50.6 (5.5) 42.9 (9.2) 0.042\*

**Control n=17 Mean (SD)** 

0 3 (28%) 0.526+

2 (12%)

8 (47%) 6 (35%) 3 (18%)


**p-value** 

0.515+




(Bedard, et al., 2010; Bryer, et al., 2006; Elvik & Vaa, 2004; National Road Transport Commission, 2003). Table 4. Driving Performance Assessment Guidelines


**Assessment Frequency and Scoring Procedure** 

locations/events. One point deducted for each omission per

locations/events. Up to three points deducted depending on

Up to four points deducted depending on severity of error - Assessed at 23 locations. Points deducted for excess speed as per




severity of error - Assessed at 20 locations/events.

national guidelines.

by stimulator. One point deducted for each instance. - Number of deviations recorded

by simulator. One point deducted for each time.

for each omission

omission.

feedback


stimulator. One point deducted




mirror

**Definition of Required Behaviour/Skill** 











(Bedard, et al., 2010; Bryer, et al., 2006; Elvik & Vaa, 2004; National Road Transport Commission, 2003).

around obstacles and maintains a safety buffer

around vehicle.

speed limit.

roads

vehicle

lights.

recorded.

Table 4. Driving Performance Assessment Guidelines

movements.

diverging.

**Assessment Component** 

Smooth Manoeuvring around Obstacles

Appropriate Stopping

Maintains Appropriate

Maintains Correct Lane

Maintains Control of Vehicle on Turns

Appropriate Behaviours to Avoid Hazards

Appropriate Use of

Demonstrates Caution during Manoeuvres

Qualitative Feedback (Bedard, et al., 2010; Elvik & Vaa, 2004)

Indicators

Frequency of Appropriate Use of

Frequency of

Vehicle Speed

Distance

Position

Mirrors


^ Chi squared test; # T-test; + Fisher's Exact test; \*Results were statistically significant (p<0.05) and Categorical frequency (percentage)

Table 5. Results of Participant Demographic Data

The drivers with PD had on average a diagnosis of PD for approximately 8 years and 4 months. Medications that were prescribed to the participants with PD included: Sinemet, Madopar, Cabaser, Sifrol and Selgene. Some participants with PD reported experiencing tremors in arms and legs as well as mild to moderate rigidity and fatigue (refer to Table 5). Six of the control participants reported experiencing mild to moderate fatigue, which was not related to PD. All participants with PD reported they required only minimal assistance to complete self-care activities, whilst none of the participants in the control group required any assistance.

#### **3.2 Psychometric assessment results**

The results of four standardised, psychometric assessments and the E-Prime Assessment Tool are shown in Table 6. The only psychometric assessment tool that detected a difference between the groups was the Purdue Pegboard Both Hands subtest and Overall Score. These results indicate that there may be a difference in the speed and dexterity of upper limb

An Investigation into the Impact

Driving Performance Score Trial 1 Trial 2

the change in Scenario Completion Time for each trial.

**Variable Drivers with PD**

Between Group Comparison Comparison between Trials

Between Group Comparison Comparison between Trials



Scenario Completion Time (seconds)



was not unsafe or dangerous.

\* Results are statistically significant (p < 0.05) ^ Wilcoxon Two-sample Signed Rank Test

Trial 1 Trial 2
