**5. Postural control**

The mechanisms underlying PB have been attributed to a dysfunction of sensory (vertical) perception that leads to a postural reactive behavior (Karnath et al., 2000b; Karnath &Broetz, 2003; Saj et al., 2005; Johannsen et al., 2006c; Perennou, D. A. et al., 2008). These perceptions represent the subjective spatial perceptions, which include the haptic vertical (SHV), visual vertical (SVV), postural vertical (SPV) and the subjective straight ahead (SSA). Figure 2 shows the methodology and sensorial systems involved with these perceptions, and table 3 summarizes the available data about the subjective spatial perceptions of pusher patients published so far.

Karnath et al. found 5 patients with severe PB (SCP=6) who experience their body as being oriented "upright" when it is actually tilted about 18° towards the side of the brain lesion and with no SVV bias (Karnath et al., 2000b). According to the authors, the possible explanation for the PB is that when patients try to move their body to a subjectively 'upright' position, they became laterally unstable because their center of mass was shifted too far to the ipsilesional side and they react to this imbalance by pushing themselves to the contralesional side (Karnath et al., 2000b; Karnath, 2007).

In contrast, Pérrenou et al. recently found a contralesional bias of SPV in 6 pusher patients with an SCP score ranging from 3 to 6. Moreover, all these patients also presented with contralesional tilts in SHV and SVV (Perennou, D. A. et al., 2008). Their hypothesis was that pushing is an implicit active body postural alignment with the perceived vertical. Interestingly, Johannsen et al. demonstrated that patients with PB align their nonparetic leg upright when their trunks are actually tilted to the side opposite to the encephalic lesion (Johannsen et al., 2006a). The authors pointed out that observing the spontaneous posture of the body segments in a seated subject may be a reasonable approach to predict the subject's SPV (Johannsen et al., 2006a). However, future research is needed to verify the correlation between SPV and non-paretic leg orientation in the same sample of pusher patients.

The contradictory findings described above may reflect a difference in the methodology and inclusion/exclusion criteria. Karnath et al. (Karnath et al., 2000b) evaluated the SPV with the

A possible gender influence on the incidence of PB was initially suggested (Lafosse et al., 2005). Nevertheless, analysis of several studies performed in large samples of neurologic injured patients found no persistent gender predominance (Danells et al., 2004; Santos-

Paresis of the contralesional extremities seems to be more frequent and more severe in pusher patients than in control encephalic lesioned patients (Karnath et al., 2005). On the other hand, severe PB can occur despite mild degree of hemiparesis (Santos-Pontelli et al., 2007). This observation raises an interesting question: is hemiparesis necessary for the development of the pushing behavior? We reported a patient that the resolution of the contraversive pushing did not depend on the resolution of the hemiparesis (Santos-Pontelli et al., 2007). Therefore, it is possible that hemiparesis may be more properly considered a commonly associated symptom of PB rather than an essential component of the syndrome and its damaged graviceptive circuitry. Further studies involving patients with pusher syndrome controlled for the degree of hemiparesis may be necessary to clarify the impact of

The mechanisms underlying PB have been attributed to a dysfunction of sensory (vertical) perception that leads to a postural reactive behavior (Karnath et al., 2000b; Karnath &Broetz, 2003; Saj et al., 2005; Johannsen et al., 2006c; Perennou, D. A. et al., 2008). These perceptions represent the subjective spatial perceptions, which include the haptic vertical (SHV), visual vertical (SVV), postural vertical (SPV) and the subjective straight ahead (SSA). Figure 2 shows the methodology and sensorial systems involved with these perceptions, and table 3 summarizes the available data about the subjective spatial perceptions of pusher patients

Karnath et al. found 5 patients with severe PB (SCP=6) who experience their body as being oriented "upright" when it is actually tilted about 18° towards the side of the brain lesion and with no SVV bias (Karnath et al., 2000b). According to the authors, the possible explanation for the PB is that when patients try to move their body to a subjectively 'upright' position, they became laterally unstable because their center of mass was shifted too far to the ipsilesional side and they react to this imbalance by pushing themselves to the

In contrast, Pérrenou et al. recently found a contralesional bias of SPV in 6 pusher patients with an SCP score ranging from 3 to 6. Moreover, all these patients also presented with contralesional tilts in SHV and SVV (Perennou, D. A. et al., 2008). Their hypothesis was that pushing is an implicit active body postural alignment with the perceived vertical. Interestingly, Johannsen et al. demonstrated that patients with PB align their nonparetic leg upright when their trunks are actually tilted to the side opposite to the encephalic lesion (Johannsen et al., 2006a). The authors pointed out that observing the spontaneous posture of the body segments in a seated subject may be a reasonable approach to predict the subject's SPV (Johannsen et al., 2006a). However, future research is needed to verify the correlation between SPV and non-paretic leg orientation in the same sample of pusher

The contradictory findings described above may reflect a difference in the methodology and inclusion/exclusion criteria. Karnath et al. (Karnath et al., 2000b) evaluated the SPV with the

Pontelli et al., 2004; Lafosse et al., 2005).

**5. Postural control** 

published so far.

patients.

PB itself on long-term prognosis after neurologic conditions.

contralesional side (Karnath et al., 2000b; Karnath, 2007).

patients' legs hanging freely, while Pérennou et al. used a plantar support (Perennou, D. A. et al., 2008). Additionally, Pérrenou et al. did not screen for neglect. The influence of the presence of plantar support or neglect on the measurement of the SPV is unknown.

1. SHV: determined by manipulation of a wooden or metal rod to the earth-vertical position with the patients' eyes closed: this is essentially driven by proprioceptive afferences (Sharpe, 2003). 1a: with one hand.1b: with two hands.

2. SVV: assessed by the patients' verbal command to adjust a visible line in complete darkness. It depends only on vestibular information with the assistance of the visual cues, independent of the proprioceptors and truncal graviceptors when the subjects are positioned in alignment with Earth vertical (Anastasopoulos et al., 1997; Mittelstaedt, 1998; Trousselard et al., 2004; Lopez et al., 2011). 3. SPV: assessed with subjects seated on a tiltable chair that is capable of rotating in a particular plane and is immobilized by lateral stabilization to prevent postural reactions. The examiner asks the subjects to state, in absence of vision, when they feel their body as vertically oriented (Karnath et al., 2000b; Sharpe, 2003; Perennou, D. A. et al., 2008). The tilting velocity must be 1.5º/s to minimize semicircular canal stimulation (Sadeghi et al., 2007), and acoustic and vibration feedback should also be taken into account. This is determined essentially by interoceptive inputs (Mittelstaedt, 1998; Karnath et al., 2000b).

4. SSA: evaluated by asking the patient to point to the position they perceived as straight ahead and represents an egocentric reference framework (Richard et al., 2004; Saj et al., 2006).

Fig. 2. Methodology description and the sensorial systems involved with SHV, SVV, SPV and SSA.

The SVV (with a haptic component) and the SSA was found to be tilted to the side of the lesion in patients with neglect without PB and tilted to the contralesional side in patients with neglect and PB (Saj et al., 2005; Honore et al., 2009). Nevertheless, the SVV with no haptic influence conducted in a representative sample of pusher patients with and without neglect did not reveal a tilt of this perception46. Unfortunately, none of the above studies performed a systematic evaluation of the vestibular system for review see (Eggers & Zee, 2003). Although the dysfunction of the vestibular system is not assumed to be involved with PB (Perennou, D., 2005; Pontelli et al., 2005), its evaluation became imperative to dissociate

New Insights for a Better Understanding

unimodal misperception (Broetz et al., 2004).

support [for review see (Carr & Shepherd, 2006)].

unknown neurological disorder.

**7. Neuroimaging analysis** 

without evoking fear.

be trained to use conscious strategies to realign their body.

of the Pusher Behavior: From Clinical to Neuroimaging Features 249

because the orientation perception of visual cues in pusher patients is not impaired, they can

However, the contralesional tilts of SPV, SVV and SHV recently described in patients with PB raise the question about the utility of visual feedback treatment in all pusher patients (Pedersen et al., 1996). Some findings with healthy subjects have shown a difference in performance if the learner directs attention toward the effect of the movement (an external focus) instead of to the movement itself (an internal focus) (Wulf et al., 1998). It is possible that in pusher patients with multimodal misperception, we could induce the patient to perceive that their body position is tilted by showing the difference between the effect of the movement using their perceived (wrong) vertical reference and using the (somesthetic or verbal) reference given by the therapist. Broetz and Karnath recommended this demonstration of the ineffective result of the pathological pushing in patients with

Recently, Shepherd and Carr suggested that the behavioral development may be a natural adaptive response to rehabilitation methods that have the potential to increase the fear of falling and provoke defensive pushing (Shepherd &Carr, 2005). The fact that PB has been identified early after the encephalic lesion argues against this possibility. Additionally, we performed a systematic screening of PB in an acute neurological unit (Santos-Pontelli et al., 2004), and we often identified the PB while the patients were positioned sitting on the edge of the bed for the first time after the onset. Nevertheless, as pointed out by the authors, it is imperative to take the fear of falling into account and to be careful to perform the exercises

Other general evidence-based methods of intervention are naturally applied for pusher patients because other neurological deficits are present. So far, several studies suggest the following: task-oriented exercises, patients' focus on the actual activity, strength and skill training, specific strategies for spatial neglect (when present), patients' awareness of their deficits, attention to the intensity of skill practice and the extent of cardiovascular stress, proper rehabilitation environment, and the use of a treadmill with and without body weight

A consensus on neurological rehabilitation is that intervention requires specificity and that the postural balance is essential in regaining independence in the activities of daily living. Thus, exercises must be individualized, and the best therapeutic strategy for PB should be chosen based on the vertical misperception of each pusher patient as soon as possible. The absence of controlled trials that investigate the treatment of PB supports the need for further research. Moreover, we should be careful about making statements about the PB based on few samples. Multicenter researches could help PB investigative groups to perform more representative studies in order to clarify all the underlying aspects of this still largely

Several brain structures have been associated with PB. In this context, Pedersen et al.(Pedersen et al., 1996) and Santos-Pontelli et al. (Santos-Pontelli et al., 2011) have indicated a wide range of findings from no visible lesion to massive hemispheric lesions on neuroimaging scans in a large sample of PB patients. In these studies, radiologists and neurologists analyzed computed tomography or magnetic resonance imaging in order to determine the type and location of the encephalic lesions. Pedersen at al. determined the

size of the stroke by the largest diameter of the lesion (Pedersen et al., 1996).

vestibular dysfunction from the vertical misperceptions of pusher patients because SVV is essentially driven by this system (Anastasopoulos et al., 1997; Mittelstaedt, 1998; Trousselard et al., 2004). Other aspects to be considered for the evaluation of verticality perception are the learning effect and the number of trials performed. Therefore, in order to state which vertical perception is disturbed in pusher patients, the studies' designs require a meticulous methodology and a large sample of pusher patients. The underlying mechanisms of PB still remain unclear.


SD: Standard Deviation; RBD: Right Brain Damage; LBD: Left Brain Damage; \*(with haptic component) \*\* Mean and standard deviation calculated from the data available in the reference (Perennou, D. A. et al., 2008).

Table 3. Summarized available data about the subjective perceptions of pusher patients.
