**5. Obsessive-Compulsive Disorder**

Obsessive-compulsive disorder (OCD) has a lifetime prevalence of 2-3% (Weissman et al., 1994). OCD is characterized by persistent and recurrent thoughts that invade conscious awareness against a patient's will (obsessions) and is further usually accompanied by egodystonic, ritualistic behaviors that the patient is obliged to perform in order to prevent overwhelming anxiety (compulsions). Patients with OCD form a more homogeneous group than those with other psychiatric disorders, and this perhaps accounts for the fact that previous functional neuroimaging studies have provided relatively consistent findings on aberrant brain regions in this disorder, which include the OFC, ACC, caudate nuclei, thalamus and so on. That is, the etiology of OCD has been presumed to follow a cortico-subcortical model. Functional neuroimaging techniques have contributed substantially to the exploration of these areas relevant to the disorder. Furthermore, recent reports on treatment intervention for OCD have strongly suggested that selective serotonin reuptake inhibitors (SSRI) and cognitive behavior therapy (CBT), both established treatment approaches, raise some effects on patients' brain blood flow and metabolism, and further normalize aberrant regional perfusion and metabolism within these networks in treatment responders.

Resting State Blood Flow and Glucose Metabolism in Psychiatric Disorders 141

1994) and thalamus (Martinot et al., 1990; Lucey et al., 1995). These discrepancies were presumed to be due to varied treatment duration of serotonin reuptake inhibitors (SRIs) (Rubin et al., 1995), or to childhood- or adult-onset of the disease (Geller et al., 1995), presence or absence of comorbidity disorders such as MDD or tic disorder (Crespo-Faccoro et al., 1999; Hoehn-Saric et al., 2001) and the measurement of different parameters (brain blood flow or metabolism). Interestingly, whereas SPECT studies tended to indicate a decrease in rCBF, FDG-PET studies tended to show an increase in rGMR in the key regions in the disease, suggesting a possibility of uncoupling between brain blood flow and glucose utilization (Whiteside et al., 2004). At the very least, these regions are closely

Studies on the relation between the symptom severity and the degree of abnormality in these areas have presented very varied results and failed to provide consistent findings.

Previous studies have replicated well that aberrant findings of rCBF/rGMR relevant to OCD-related regions could be normalized by pharmacological intervention of SRIs. Treatment of clomipramine, a tricyclic antidepressant, over several months could normalize regional blood flow or metabolism in the OFC and/or caudate nucleus from significant increase level prior to intervention compared to normal controls (Benkelfat et al., 1990; Swedo et al., 1992; Rubin et al., 1995). Also, intervention by two SSRIs, paroxetine and fluoxetine, provided similar results to clomipramine; increased rCBF/rGMR in the OFC and/or caudate nucleus at baseline were reduced significantly following treatment with paroxetine (Saxena et al., 1999, 2002; Hansen et al., 2002; Diler et al., 2004) and increased rCBF/rGMR in the ACC/caudate nucleus/thalamus at baseline decreased significantly after fluoxetine treatment (Hoehn-Saric et al., 1991; Baxter et al., 1992). Furthermore, in most of these studies, responders in clinical symptoms to pharmacological intervention tended to show a significant decrease relative to baseline, whereas non-responders showed no change by the treatment (Benkelfat et al., 1990; Baxter et al., 1992; Swedo et al., 1992; Saxena et al., 1999; Hoehn-Saric et al., 2001; Diler et al., 2004; Ho Pian et al., 2005). With respect to response prediction, several studies have found that the lower the brain blood flow or metabolism in relevant regions prior to treatment was, the greater was the reduction in OCD symptoms (Benkelfat et al., 1990; Saxena et al., 1999). In addition, there were significant correlations between decrease of metabolism at baseline in the OFC or caudate nucleus and improvement of OCD symptoms (Benkelfat et al., 1990; Swedo et al., 1992; Baxter et al., 1992). However, studies on significant response predictors have been very restricted and

**5.2 Change following intervention by SRIs and cognitive-behavior therapy** 

reliable parameters on response prediction have never been explored to date.

state consequence occurring regardless of treatment approaches.

CBT, interestingly, also appears to normalize increased rCBF/rGMR in some relevant areas, including the caudate nucleus (Baxter et al., 1992; Schwartz et al., 1996; Nakatani et al., 2003) and thalamus (Saxena et al., 2009). Additionally, responders to CBT exhibited greater reduction in the caudate nucleus from baseline to CBT intervention than did non-responders (Schwartz et al., 1996). Although there have been few studies up to now on the alteration of brain function before and after CBT, growing notions on the effects of CBT on brain functions within subjects would address some important issues on whether the functional brain change induced by SRIs is a direct consequence of their pharmacological actions, or a

involved in the pathophysiology of OCD.

#### **5.1 Dysfunction of the orbitofrontal-subcortical circuit in OCD**

The basal ganglia is a candidate abnormal area in OCD to which great attention was initially paid. The reason for this is a high rate of patients with obsessive symptoms were found to have certain diseases, such as Von Economo encephalitis (Schilder, 1938), Sydenham's chorea (Swedo et al., 1989) and Tourette's syndrome (Nee et al., 1980), which have presumed to be impaired in the basal ganglia. Afterwards, functional neuroimaging studies on OCD have focused on the striatum, in particular caudate nucleus as aberrant region within patients' brains and concurrently have successively detected some abnormal brain areas such as the OFC, ACC and thalamus in patients with OCD, when compare them with normal healthy subjects. In this context, researchers have proposed a dysfunction of corticostriatum-thalamus-cortical network as an etiological model of OCD (Modell et al., 1989; Baxter et al., 1996; Saxena et al., 1998).

It has been classically recognized that the cortico-subcortical network consists of direct and indirect pathways. The thalamus in the network has a gating function which filters all stimuli from the outer world and receives two main inputs from the striatum. The one is the direct pathway where signals from the striatum input to the thalamus via the globus pallidus internal/substantial nigra and the other is the indirect pathway where signals from the striatum input to the globus pallidus internal/substantial nigra through the globus pallidus external or subthalamic nucleus, and are further sent to the thalamus. Afterwards, feedback signals from the thalamus are sent to the cortex. These pathways consist of neurotransmissions combined with excitatory signals by glutamate and inhibitory signals by GABA. The direct pathway inputting to the thalamus disinhibits the thalamus (reinforcement of positive feedback) and the indirect pathway inhibits the thalamus (negative feedback), thereby helping to maintain the balance of the system (Alexander and Crutcher, 1990). In patients with OCD, it is presumed that this circuit represents an imbalance of hyperactivity. In the dysfunctional network, impairment in the striatum leads to an insufficient gating function of the thalamus, resulting in cortical hyperactivities. In this context, the direct pathway in the patients with OCD predominates over the indirect pathway. In terms of symptom-relations, the striatum is essentially involved in unconscious acquisition of the initial process of action or behavior, and hypermobilization of the impaired striatum could lead to compulsive symptoms in the manner of ritual behaviors, in order to normalize the undesirable thoughts or anxieties occurring via the dysfunctional thalamus. On the other hand, these invasive thoughts and excess anxieties would relate with hyperactivity in the OFC and ACC, respectively.

Previous functional neuroimaging studies in subjects at rest or undergoing symptomprovocation have implicated an increase in rCBF/rGMR in the OFC (Baxter et al., 1987, 1988; Benkelfat et al., 1990; Horwitz et al., 1991; Rubin et al., 1992, 1995; McGuire et al., 1994; Alptekin et al., 2001), ACC (Swedo et al., 1989; Horwitz et al., 1991; Perani et al., 1995), caudate nucleus (Baxter et al., 1987, 1988; Diler et al., 2004; Saxena et al., 2004), putamen (Benkelfat et al., 1990; Perani et al., 1995) and thalamus (McGuire et al., 1994; Perani et al., 1995; Alptekin et al., 2001; Saxena et al., 2001, 2004), strongly suggesting hyperactivities in the cortico-subcortical loop in patients with OCD. However, other studies have demonstrated inverse results, i.e., decreases in the OFC (Crespo-Faccoro et al., 1999; Busatto et al., 2000), ACC (Busatto et al., 2000), caudate nucleus (Rubin et al., 1992, 1995; Edmonstone et al., 1994; Lucey et al., 1995, 1997), putamen (Edmonstone et al.,

The basal ganglia is a candidate abnormal area in OCD to which great attention was initially paid. The reason for this is a high rate of patients with obsessive symptoms were found to have certain diseases, such as Von Economo encephalitis (Schilder, 1938), Sydenham's chorea (Swedo et al., 1989) and Tourette's syndrome (Nee et al., 1980), which have presumed to be impaired in the basal ganglia. Afterwards, functional neuroimaging studies on OCD have focused on the striatum, in particular caudate nucleus as aberrant region within patients' brains and concurrently have successively detected some abnormal brain areas such as the OFC, ACC and thalamus in patients with OCD, when compare them with normal healthy subjects. In this context, researchers have proposed a dysfunction of corticostriatum-thalamus-cortical network as an etiological model of OCD (Modell et al., 1989;

It has been classically recognized that the cortico-subcortical network consists of direct and indirect pathways. The thalamus in the network has a gating function which filters all stimuli from the outer world and receives two main inputs from the striatum. The one is the direct pathway where signals from the striatum input to the thalamus via the globus pallidus internal/substantial nigra and the other is the indirect pathway where signals from the striatum input to the globus pallidus internal/substantial nigra through the globus pallidus external or subthalamic nucleus, and are further sent to the thalamus. Afterwards, feedback signals from the thalamus are sent to the cortex. These pathways consist of neurotransmissions combined with excitatory signals by glutamate and inhibitory signals by GABA. The direct pathway inputting to the thalamus disinhibits the thalamus (reinforcement of positive feedback) and the indirect pathway inhibits the thalamus (negative feedback), thereby helping to maintain the balance of the system (Alexander and Crutcher, 1990). In patients with OCD, it is presumed that this circuit represents an imbalance of hyperactivity. In the dysfunctional network, impairment in the striatum leads to an insufficient gating function of the thalamus, resulting in cortical hyperactivities. In this context, the direct pathway in the patients with OCD predominates over the indirect pathway. In terms of symptom-relations, the striatum is essentially involved in unconscious acquisition of the initial process of action or behavior, and hypermobilization of the impaired striatum could lead to compulsive symptoms in the manner of ritual behaviors, in order to normalize the undesirable thoughts or anxieties occurring via the dysfunctional thalamus. On the other hand, these invasive thoughts and excess anxieties would relate with

Previous functional neuroimaging studies in subjects at rest or undergoing symptomprovocation have implicated an increase in rCBF/rGMR in the OFC (Baxter et al., 1987, 1988; Benkelfat et al., 1990; Horwitz et al., 1991; Rubin et al., 1992, 1995; McGuire et al., 1994; Alptekin et al., 2001), ACC (Swedo et al., 1989; Horwitz et al., 1991; Perani et al., 1995), caudate nucleus (Baxter et al., 1987, 1988; Diler et al., 2004; Saxena et al., 2004), putamen (Benkelfat et al., 1990; Perani et al., 1995) and thalamus (McGuire et al., 1994; Perani et al., 1995; Alptekin et al., 2001; Saxena et al., 2001, 2004), strongly suggesting hyperactivities in the cortico-subcortical loop in patients with OCD. However, other studies have demonstrated inverse results, i.e., decreases in the OFC (Crespo-Faccoro et al., 1999; Busatto et al., 2000), ACC (Busatto et al., 2000), caudate nucleus (Rubin et al., 1992, 1995; Edmonstone et al., 1994; Lucey et al., 1995, 1997), putamen (Edmonstone et al.,

**5.1 Dysfunction of the orbitofrontal-subcortical circuit in OCD** 

Baxter et al., 1996; Saxena et al., 1998).

hyperactivity in the OFC and ACC, respectively.

1994) and thalamus (Martinot et al., 1990; Lucey et al., 1995). These discrepancies were presumed to be due to varied treatment duration of serotonin reuptake inhibitors (SRIs) (Rubin et al., 1995), or to childhood- or adult-onset of the disease (Geller et al., 1995), presence or absence of comorbidity disorders such as MDD or tic disorder (Crespo-Faccoro et al., 1999; Hoehn-Saric et al., 2001) and the measurement of different parameters (brain blood flow or metabolism). Interestingly, whereas SPECT studies tended to indicate a decrease in rCBF, FDG-PET studies tended to show an increase in rGMR in the key regions in the disease, suggesting a possibility of uncoupling between brain blood flow and glucose utilization (Whiteside et al., 2004). At the very least, these regions are closely involved in the pathophysiology of OCD.

Studies on the relation between the symptom severity and the degree of abnormality in these areas have presented very varied results and failed to provide consistent findings.

#### **5.2 Change following intervention by SRIs and cognitive-behavior therapy**

Previous studies have replicated well that aberrant findings of rCBF/rGMR relevant to OCD-related regions could be normalized by pharmacological intervention of SRIs. Treatment of clomipramine, a tricyclic antidepressant, over several months could normalize regional blood flow or metabolism in the OFC and/or caudate nucleus from significant increase level prior to intervention compared to normal controls (Benkelfat et al., 1990; Swedo et al., 1992; Rubin et al., 1995). Also, intervention by two SSRIs, paroxetine and fluoxetine, provided similar results to clomipramine; increased rCBF/rGMR in the OFC and/or caudate nucleus at baseline were reduced significantly following treatment with paroxetine (Saxena et al., 1999, 2002; Hansen et al., 2002; Diler et al., 2004) and increased rCBF/rGMR in the ACC/caudate nucleus/thalamus at baseline decreased significantly after fluoxetine treatment (Hoehn-Saric et al., 1991; Baxter et al., 1992). Furthermore, in most of these studies, responders in clinical symptoms to pharmacological intervention tended to show a significant decrease relative to baseline, whereas non-responders showed no change by the treatment (Benkelfat et al., 1990; Baxter et al., 1992; Swedo et al., 1992; Saxena et al., 1999; Hoehn-Saric et al., 2001; Diler et al., 2004; Ho Pian et al., 2005). With respect to response prediction, several studies have found that the lower the brain blood flow or metabolism in relevant regions prior to treatment was, the greater was the reduction in OCD symptoms (Benkelfat et al., 1990; Saxena et al., 1999). In addition, there were significant correlations between decrease of metabolism at baseline in the OFC or caudate nucleus and improvement of OCD symptoms (Benkelfat et al., 1990; Swedo et al., 1992; Baxter et al., 1992). However, studies on significant response predictors have been very restricted and reliable parameters on response prediction have never been explored to date.

CBT, interestingly, also appears to normalize increased rCBF/rGMR in some relevant areas, including the caudate nucleus (Baxter et al., 1992; Schwartz et al., 1996; Nakatani et al., 2003) and thalamus (Saxena et al., 2009). Additionally, responders to CBT exhibited greater reduction in the caudate nucleus from baseline to CBT intervention than did non-responders (Schwartz et al., 1996). Although there have been few studies up to now on the alteration of brain function before and after CBT, growing notions on the effects of CBT on brain functions within subjects would address some important issues on whether the functional brain change induced by SRIs is a direct consequence of their pharmacological actions, or a state consequence occurring regardless of treatment approaches.

Resting State Blood Flow and Glucose Metabolism in Psychiatric Disorders 143

Andreasen NC, O'Leary DS, Flaum M, Nopoulos P, Watkins GL, Boles Ponto LL, Hichwa

Ashton L, Barnes A, Livingston M, Wyper D; Scottish Schizophrenia Research Group.

Bartlett EJ, Wolkin A, Brodie JD, Laska EM, Wolf AP, Sanfilipo M. (1991). Importance of

Bauer M, London ED, Rasgon N, Berman SM, Frye MA, Altshuler LL, Mandelkern MA,

of levothyroxine alter regional cerebral metabolism and improve mood in bipolar

Baxter LR Jr, Phelps ME, Mazziotta JC, Schwartz JM, Gerner RH, Selin CE, Sumida RM.

Baxter LR Jr, Phelps ME, Mazziotta JC, Guze BH, Schwartz JM, Selin CE. (1987). Local

Baxter LR Jr, Schwartz JM, Mazziotta JC, Phelps ME, Pahl JJ, Guze BH, Fairbanks L. (1988).

Baxter LR Jr, Schwartz JM, Phelps ME, Mazziotta JC, Guze BH, Selin CE, Gerner RH,

Baxter LR Jr, Schwartz JM, Bergman KS, Szuba MP, Guze BH, Mazziotta JC, Alazraki A,

Baxter LR Jr, Saxena S, Brody AL, Ackermann RF, Colgan M, Schwartz JM, Allen-Martinez

Human and Nonhuman Primate. *Semin Clin Neuropsychiatry*, 1(1), 32-47. Benkelfat C, Nordahl TE, Semple WE, King AC, Murphy DL, Cohen RM. (1990). Local

Berman I, Merson A, Sison C, Allan E, Schaefer C, Loberboym M, Losonczy MF. (1996).

neuroleptic-naïve patients. *Lancet*, 14. 349(9067), 1730-1734.

episode schizophrenia. *Behav Neurol*, 12(1-2), 93-101.

schizophrenic patients. *Am J Psychiatry*, 155(3), 337-343.

compulsive disorder. *Am J Psychiatry*, 145(12), 1560-1563.

with clomipramine. *Arch Gen Psychiatry*, 47(9), 840-848.

three types of depression. *Arch Gen Psychiatry*, 46(3), 243-250.

Supraphysiological doses

42(5), 441-447.

211-218.

depression. *Mol Psychiatry*, 10(5), 456-469.

*Arch Gen Psychiatry*, 49(9), 681-689.

RD. (1997). Hypofrontality in schizophrenia: distributed dysfunctional circuits in

(2000). Cingulate abnormalities associated with PANSS negative scores in first

pharmacologic control in PET studies: effects of thiothixene and haloperidol on cerebral glucose utilization in chronic schizophrenia. *Psychiatry Res*, 40(2), 115-124. Bartlett EJ, Brodie JD, Simkowitz P, Schlösser R, Dewey SL, Lindenmayer JP, Rusinek H,

Wolkin A, Cancro R, Schiffer W. (1998). Effect of a haloperidol challenge on regional brain metabolism in neuroleptic-responsive and nonresponsive

Bramen J, Voytek B, Woods R, Mazziotta JC, Whybrow PC. (2005).

(1985). Cerebral metabolic rates for glucose in mood disorders. Studies with positron emission tomography and fluorodeoxyglucose F 18. *Arch Gen Psychiatry*,

cerebral glucose metabolic rates in obsessive-compulsive disorder. A comparison with rates in unipolar depression and in normal controls. *Arch Gen Psychiatry*, 44(3),

Cerebral glucose metabolic rates in nondepressed patients with obsessive-

Sumida RM. (1989). Reduction of prefrontal cortex glucose metabolism common to

Selin CE, Ferng HK, Munford P, et al. (1992). Caudate glucose metabolic rate changes with both drug and behavior therapy for obsessive-compulsive disorder.

Z, Fuster JM, Phelps ME. (1996). Brain Mediation of Obsessive-Compulsive Disorder Symptoms: Evidence From Functional Brain Imaging Studies in the

cerebral glucose metabolic rates in obsessive-compulsive disorder. Patients treated

Regional cerebral blood flow changes associated with risperidone treatment in elderly schizophrenia patients: a pilot study. *Psychopharmacol Bull*, 32(1), 95-100.

#### **5.3 Depression as a comorbidity with OCD**

Although most studies have been directed to the patients with OCD without MDD, in clinical practice OCD patients frequently have major depression as a comorbidity; approximately one-third of OCD patients also have MDD (Rasmussen and Eisen, 1992; Weismann et al., 1994), whereas 22-38% of patients with MDD have obsessive-compulsive symptoms (Kendell and DiScipio, 1970). Thus, notions acquired from studies performed on pure OCD patients without depression might deviate from the actual pathophysiology of OCD. Further, since SRIs and CBT are commonly effective for improvement of both OCD and MDD, exploration of the neuronal substrates shared by the two diseases might provide very valuable information for understanding the etiology.

Saxena et al. (1999) demonstrated that patients with concurrent OCD and MDD showed a significant reduction in metabolism in the hippocampus similar to that of patients with MDD alone. Furthermore, treatment with paroxetine for patients with concurrent OCD and MDD induced a reduction of rGMR in the ventral lateral prefrontal cortex, which was similar to the findings in patients with MDD alone, but did not show a decrease in the OFC and caudate nucleus like that seen in the patients with OCD alone (Saxena et al., 2002). These findings suggested that patients with concurrent OCD and MDD had the pathophysiology of MDD, and thus may constitute a distinctive subtype within OCD, such that both the etiologies of OCD and MDD should be considered carefully when devising a treatment strategy.

#### **5.4 Conclusion**

Functional neuroimaging studies on OCD have provided much more consistent findings than structural MRI studies. That is, in patients with OCD, some important regions in the cortical and subcortical areas present with hyperactivity and are normalized by pharmacotherapy. Since improvements by SRIs and CBT occur in only about half of patients (responders), further neuroimaging studies controlled by treatment intervention are strongly needed.
