Share this article on:

The ‘Pusher Syndrome’

Roller, Margaret L. DPT, MS

Journal of Neurologic Physical Therapy: March 2004 - Volume 28 - Issue 1 - p 29
doi: 10.1097/01.NPT.0000284775.32802.c0
Special Interest Papers

Pusher syndrome in patients post-stroke is characterized by leaning and active pushing toward the hemiplegic side with no compensation for the instability, and resistance to passive correction toward midline. Patients with pusher syndrome post-stroke have been found to require longer than average to reach independence in activities of daily living and ambulation. This behavior is found to be present primarily in patients with right hemispheric lesions centered in the area of the posterolateral thalamus. Patients with this disorder misperceive body orientation in space, believing that the upright position is approximately 18° tilted toward the ipsilesional side. This could be the result of ‘graviceptive neglect’ of signals originating in the contralesional trunk and pelvis. The ability to use visual information for the correction of body posture continues to exist in this population, allowing for the use of visual cues and cognitive strategies as potential rehabilitation tools.

Assistant Professor in the Department of Physical Therapy, California State University, Northridge, Northridge, CA (, and Physical Therapist, UCLA Medical Center Department of Rehabilitation Services, Los Angeles, CA

Back to Top | Article Outline


Pushing is one of the most perplexing perceptual-motor behaviors encountered in the rehabilitation of patients post-stroke. It is clinically displayed in 5% of all patients post-stroke, and in approximately 10% of patients with stroke referred for neurologic rehabilitation.1 These tend to be the most severely impaired patients with profound functional limitations in performing transfers, standing, and gait. Individuals with pusher syndrome also take a longer than average time to recover independence post-stroke.1,2 Evidence for the suspected nature and type of stroke responsible for this unusual behavioral disorder has been published, and may contribute to our knowledge of how to best treat these patients to facilitate more independent function.3–5 The purpose of this review is to describe the characteristics of the pusher syndrome in patients post-stroke, present current information related to the location of the stroke lesion and cause of this behavioral disorder, and to discuss physical therapy interventions for this patient population.

Patients with hemiparesis post-stroke are known to demonstrate some postural imbalance, and may lose balance in the lateral direction, typically falling toward the weaker side.2,5–8 These patients usually attempt to compensate for this imbalance by shifting body weight across midline, onto the ipsilesional side and away from the hemiplegic side. A small percentage of patients post-stroke demonstrate a behavioral phenomenon characterized by leaning and active pushing toward the hemiplegic side in all positions using the nonparetic arm and leg, and resistance to any attempt at passive correction of posture towards midline or across the body toward the nonaffected side. These patients subjectively report the impression of lateral instability and the fear of falling toward the non-paretic side.3 This behavior has been called the “pusher syndrome” by Davies,9 Bailey and Leivseth,10 and Premoselli et al.11 It also has been referred to as ‘ipsilateral pushing,’1,7 ‘contraversive pushing,’2,4 and ‘pusher behavior.’5

Back to Top | Article Outline


Observations of the clinical behaviors of patients with pusher syndrome were documented by Davies in her book entitled Steps to Follow.9 Davies categorized the typical body systems impairments of these patients into a syndrome that included left hemiplegia, spatial and sensory neglect of the hemiplegic side, postural asymmetries, and the characteristic shift of body weight toward the hemi-plegic side with no attempt to support or compensate for the resultant imbalance.9 These patients typically show no protective responses when active pushing leads to postural instability in sitting and standing, and are therefore at high risk for falls.3,9 Davies clinically observed that sensory neglect can be severe and include an impaired ability to perceive any sensory afferent information on the hemiplegic side including tactile, proprioceptive, visual, and auditory stimuli.9 Other investigators have found sensory impairments to be commonly associated with pushing syn-drome.12,13

The associated characteristics of patients who display this phenomenon have been a topic of debate from the time Davies defined it as ‘the pusher syndrome.’9 Davies observed that neglect of the physical and environmental space on the hemiplegic side, and the predominance of right brain lesion characterized this ‘syndrome.’9 Several investigators since that time have consistently found an association of neglect with impaired balance and particularly with pusher syndrome.3–5,14,15 In a study by Karnath and colleagues3 involving 23 patients with pusher behavior, 80% of 15 patients with right brain lesions and pushing displayed spatial neglect and somatosensory impairment. Neglect was determined to be severe in approximately 90% of these patients. Neglect was not present in 8 patients with left-sided lesions and pushing, however, aphasia was often present.3 Perennou et al5 found that 3 patients with pushing and left hemiplegia due to right hemispheric stroke also showed severe spatial neglect that included severe bodily and nonbodily neglect, tactile extinction, and hypoesthesia. These 3 patients also did not fear falling toward the hemiparetic side despite pronounced postural impairment.5 Pedersen et al1 compared characteristics of 34 patients with pusher syndrome post-stroke to patients post-stroke without the behavior. These investigators found no significant difference in the presence of neglect and aphasia in patients with pusher syndrome. Also, no significant difference was found between the two groups for symptoms of anosognosia or apraxia. These authors determined the results of this study to be contrary to the existence of a pusher syndrome that encompasses both physical and neuropsychological symptoms.

Pedersen et al1 also observed no significant association between the side of stroke lesion and the incidence of pushing. However other authors have shown a significantly greater incidence among patients with right brain lesions.3,11,12 In the study by Karnath et al3, 65% of the sample of 23 patients with pusher behavior experienced a right hemispheric lesion. Regardless of the side of the lesion, the directional bias of pushing appears to be predominantly toward the side contralateral to the lesion.1,4,5,11–13 A related but weaker tendency to fall posteriorly also can exist in this population.12,13,15

The severity of paresis of the hemiplegic extremities has not been shown to directly influence the presence of pushing or body orientation with respect to gravity while seated.8 However all 23 of the patients with contraversive pushing in the study by Karnath et al3 displayed severe paresis of both the upper and lower contralesional extremities.

Perennou et al5 noted that the presence of a mild pelvic tilt toward the contralesional side occurred in some non-pushers and concluded that a continuum in pushing behavior must exist, rather than a distinct occurrence of the behavior. If this assumption is true, then pusher syndrome is perhaps not simply an entity, but an extreme form of this disorder of postural misrepresentation that includes visual-spatial impairment and neglect.5

Back to Top | Article Outline


Active pushing with the nonparetic extremities helps to distinguish patients with pusher syndrome from patients with thalamic astasia13 and those with lateropulsion in Wallenberg's syndrome.15,16 The ability of patients with pusher syndrome to maintain correct head orientation toward vertical in the presence of a severe lateral lean helps to distinguish these patients from those with cerebral lesions in the ‘vestibular cortex.’17

Thalamic astasia, or the inability to stand unsupported, has been found to be a result of lesions of the posterolateral thalamus.13 In a study by Masdeu and Gorelick,13 8 patients with unilateral thalamic lesions could not sit independently, demonstrated obvious truncal instability, and fell backward or to the affected side from a sitting position when unsupported. When asked to sit up, patients with astasia typically grasp the side rail of the bed with the unaffected hand or with both hands and pull themselves up, rather than using the available power of the trunk.13 This behavior is different from pusher syndrome.5,13 When patients with pusher syndrome are asked to sit up they do the exact opposite by extending the unaffected arm and using it to actively push away from the nonparetic side. Another difference in these two types of patients is the severity of hemiparesis. Patients post-stroke who push have been found to have severe paresis of the contralateral arm and leg,3 whereas patients with thalamic astasia characteristically have very mild or nomotor weakness.13

Wallenberg's syndrome can be found in patients with acute unilateral brainstem infarctions in the area of the medulla. These patients display a tilt of the subjective visual vertical (visual perception of the upright position), lateropulsion (a tendency to fall sideways) without active pushing or resistance to passive correction, and a deviation of the center of gravity toward the ipsilesional, nonparetic side. The tendency of patients with Wallenberg's syndrome is to fall sideways toward the side of the lesion, or ipsiversively, compared to patients with pusher syndrome who fall to the side opposite the lesion, or contraversively.15,16

Patients with pusher syndrome post-stroke have been shown to display an intact perception of subjective visual vertical, indicating that their processing of visual and vestibular inputs for perception and orientation of the visual world is undisturbed.3,17 However, they display an impaired perception of subjective postural vertical by demonstrating a severe tilt of the pelvis with active pushing toward the hemiplegic side, leading to a contraversive tilt of the body toward the side opposite to the lesion and falling to that side.3,17 Patients with pusher syndrome are generally able to keep a correct head orientation toward vertical even in the presence of a severe lateral lean. This perceptual characteristic helps to differentiate this population from patients with cerebral lesions in the posterior insula, also called the ‘vestibular cortex,’ who commonly experience a perceived tilt of the subjective visual vertical.17 However, patients with lesions of this area of the cerebral cortex do not tend to produce impairments of postural control such as falling to one side, indicating intact perception and performance of the subjective postural vertical4,17–19 (Table 1).

Table 1

Table 1

Back to Top | Article Outline


An examination of MRI scans of the infarcted brain regions of 23 patients with severe contraversive pushing was conducted by Karnath et al.3 The overlapping area of these infarctions clearly centered on the posterolateral thalamus, an area that is anatomically distinct from the ‘vestibular cortex’ in the posterior insula,3,17 yet similar to that of patients with thalamic astasia.13 Karnath et al3 proposed that the posterolateral thalamus, including the ventral posterior nucleus, lateral posterior nucleus, and probably its cortical projections, seems to be fundamentally involved in the neural representation of the graviceptive system in humans that is critical for our control of upright body posture. This area extended from the posterolateral thalamus into the posterior crus of the internal capsule, which helps to explain the severe hemiparesis present in all 23 of the patients with contraversive pushing in the Karnath study.3 It is yet unclear whether slightly differing lesion locations in the posterolateral thalamus may result in different clinical syndromes due to the disturbance of different functional systems represented in this structure.3

It is thought that the occurrence of pusher syndrome in patients post-stroke may be associated with a disturbed neural representation of the truncal graviceptive system, which contributes to the control of upright posture of the body in space.4,5 A kinematic analysis performed by Perennou et al5 showed that pusher syndrome is not the result of disrupted processing of vestibular information, but rather results from a high-order disruption in the processing of somesthetic information originating in the contrale-sional hemibody. It was suspected that this could be gravi-ceptive neglect, defined as extinction of graviceptive signals originating in the contralesional part of the trunk and pelvis.5 This disruption then leads patients who push to actively adjust their body posture to a behavioral vertical biased to the side opposite the cerebral lesion.5

Researchers tend to agree that, whatever the cause of pusher syndrome, patients who demonstrate the behavior misperceive their body position in space.4,7,8,17 Evidence for this misperception can be observed clinically when a patient complains of a feeling of falling as a therapist moves the patient's body to a midline position over the base of support. When asked which direction they perceive to be falling, the patient will indicate toward the nonparetic side. If the therapist were to let go, the patient would instead fall to the hemiplegic side. The patient perceives and believes that the appropriate, upright position in space is located somewhere to the ipsilesional side of true center.4 It has been demonstrated in a laboratory environment without visual cues from the surroundings (with eyes closed, blindfolded, or wearing Frenzel lenses) that patients with pusher syndrome subjectively perceive their body as oriented ‘upright’ when it is tilted an average of 18° to the ipsilesional side.4 This experiment was strong evidence that the underlying cause of pusher behavior may be a severe misperception of body orientation in relation to gravity.4 Patients with pusher syndrome do not align their body with the visual vertical (which is intact and upright), with their perceived postural vertical (which is impaired at an average of 18? to the ipsilesional side), or with an intermediate posture. Instead, they move the body into the opposite direction, toward the hemiplegic side. Karnath et al4 expressed that they cannot definitively clarify this discrepancy. These investigators speculated that by pushing the longitudinal body axis toward the contralesional (hemiplegic) side these patients try to actively compensate for the mismatch between the intact visual vertical and the ‘tilted’ orientation of subjective body verticality. A therapist's attempt to correct a patient's body posture toward the objective upright position with eyes open seems to contradict the patient's effort to ‘compensate,’ which induces the feeling of lateral instability, fear of falling, and provokes active resistance to such attempts.4 Generally, a conflict between 2 sensory-perceptual reference systems is resolved either by suppressing one or both of them, or by a compromise (eg, by weighted summation).4 However, neither happens in the case of pusher syndrome.4

It is also possible that pushing is a secondary response to the patients' unexpected experience of loss of lateral balance when trying to get up and sit upright in a vertically oriented room.4 As a patient tries to get up and orient the body (subjectively) upright, they experience lateral instability because the center of mass is shifted too far to the ipsilesional side. Pushing the body to the contralesional side might be the reaction to this experience.4

Back to Top | Article Outline


Only recently has a clinical tool been developed to measure the degree of pushing behavior that is present in patients post-stroke. The Standardized Scale for Contraversive Pushing (SCP) was designed by Karnath et al4,20 to classify patients based on the severity of behavioral and clinical impairments (Table 2). This scale is based on criteria established by Davies4,9 and assesses the following: (1) posture: symmetry of spontaneous posture while sitting and standing, (2) extension: the use of the ipsilesional arm or leg to extend the area of physical contact to the ground while sitting and standing, and (3) resistance: resistance to passive correction of posture while sitting and standing. One point for sitting ability and one point for standing ability are attainable in each of the 3 sections for a maximum of 2 points for each section and 6 points total. Patients are scored as possessing contraversive pushing if all 3 criteria are present, reaching a total score of at least 1 for each cri-terion.4,20 This scale has not been studied to determine its psychometric properties, therefore clinicians are cautioned to use this tool as a guideline only until further investigation of its reliability and validity is performed with this population.

Table 2

Table 2

Back to Top | Article Outline


Bohannon7 found that some patients with pusher syndrome can ‘learn’ to balance in standing with their feet apart within a half-hour session. After the patients' positional misperception was made apparent to them by introducing them to the appropriate position over their feet, and providing them with practice and feedback, these patients typically achieved a short duration of independent static standing – first with their backs against a wall, and later without support and during movement.7 The type of practice and feedback supplied during these training sessions was not specified.

Both Karnath et al4 and Perennou and associates5 noted a preserved ability of patients with contraversive pushing to align the body axis to earth-vertical with the help of visual cues from the environment and speculated that this might be a useful rehabilitation tool. In the study by Karnath et al4, when patients with contraversive pushing and suspected impaired truncal graviception visually explored a structured laboratory environment, they were able to correctly align their longitudinal body axis to earth-vertical. This indicates that seeing upright orientation of surrounding objects and persons helped their impaired graviceptive system. This experiment was performed under controlled laboratory conditions in which the body was immobilized by lateral stabilization, the legs were hanging freely, and the patients were explicitly instructed to concentrate on visual information available in the laboratory environment. With the help of visual cues, these patients could only temporarily align their body axis to earth-vertical. It was concluded that visual input does not suffice to continuously control upright body posture in patients with pusher behavior.4 Although patients with pusher syndrome are typically unable to spontaneously use visual input to control upright body posture, the use of visual cues from the environment may be an effective training tool when combined with conscious strategies for the achievement of postural control.4

In a study of patients with hemispheric cerebral lesions, contralesional tilt of the pelvis, and impaired perception of postural vertical without pushing behavior by Perennou et al,8 transcutaneous electrical nerve stimulation (TENS) of short pulse duration, with intensity below motor threshold applied to the skin on the contralesional side of the neck reduced the bias of the perception of postural vertical. This effect was greater in patients who showed sensory loss or neglect and more severe contralesional bias of the postural vertical, which was defined as being greater than 3.0° off true vertical. It is believed that this method of applying TENS works by activating various afferent nerve fibers in the neck that convey stimuli to the contralateral cerebral hemisphere, thus unmasking the patients' latent postural capabilities. These authors believe that TENS probably acts as a powerful substitution stimulus that reactivates or reorganizes the damaged neural network devoted to body orientation with respect to gravity.8 This intervention may show promise for patients with pusher behavior who also demonstrate significant contralesional tilt of the pelvis and impairment in the perception of postural vertical.

Pedersen and colleagues1 speculated that pushing in different postures represented the degree of severity of the disorder. If this is accurate, then during remission post-stroke, pushing should first diminish in the supine position, then in the sitting position, and finally in the standing position. No studies have been done to support this con-tention.1

Pedersen et al1 determined that, upon admission to a stroke rehabilitation unit, subjects with pusher syndrome had more severe strokes as expressed by lower neurological scores on the Scandinavian Stroke Scale, and lower initial activities of daily living (ADL) function scores as measured by the Barthel Index. The presence of pusher syndrome was found to slow the recovery process post-stroke,1,2 requiring an average of 3.6 weeks (63%) longer for subjects with pushing behavior to reach their highest ADL function scores on the Barthel Index upon discharge from a rehabilitation facility.1 Ultimate functional outcome was not found to be influenced by pusher syndrome;1,2 however, since patients with this disorder have difficulty and take longer to achieve any level of independence in activities of daily living (ADL) and ambulation they have, in the past, been erroneously regarded as unsuitable for formal rehabilitation.1

Back to Top | Article Outline


Pusher syndrome in patients post-stroke is characterized by leaning and active pushing toward the hemiplegic side in all positions with no attempt to compensate for the imbalance, and resistance to any attempt at passive correction toward the ipsilesional side.3,9 Several authors agree that hemisensory impairments and severe neglect also characterize this syndrome.3,5,9 This perceptual-motor behavior is found to be present primarily in patients with right hemispheric lesions centered in the area of the posterolateral thalamus.3,11,12 Patients with pusher syndrome display a misperception of body orientation in space, believing that the upright position is approximately 18° tilted toward the ipsilesional side.4 It is suspected that this could be the result of ‘graviceptive neglect’ of signals originating in the contralesional trunk and pelvis.5 Pushing toward the contralesional side may be an attempt to compensate for a mismatch between an intact perception of the visual vertical and an impaired perception of postural vertical.4 Since the ability to use visual information for the correction of body posture continues to exist in this population, the use of visual cues, along with cognitive strategies, is recommended as a rehabilitation tool.4,5 The recovery process for patients with pusher syndrome post-stroke has been found to be slow, requiring longer than average to reach a level of independence in activities of daily living and ambulation.1

Back to Top | Article Outline


There are still many unanswered questions regarding the examination and treatment of patients post-stroke with pusher syndrome. Several suggestions for future research have been identified based on the findings of this review. First, the examination tool called the ‘Clinical Assessment Scale for Contraversive Pushing’ is yet to be investigated to determine its psychometric properties. This tool demonstrates the potential to identify the presence and severity of pushing in patients post-stroke and should be studied to establish its utility in the clinical setting. It is also unclear exactly what practice activities, parameters, and feedback are necessary to create long-term learning of vertical postures in patients with contraversive pushing. Visual feedback training has shown promise in the laboratory setting and should be studied in patients with pusher syndrome to determine whether it will decrease the amount of time it takes to reach a level of independent function in sitting ability, transfers, and ambulation. Lastly, the length of time to recover ADL and gross motor function was noted to be longer for patients with pusher syndrome, yet no studies were found that analyzed the cost of rehabilitation for these patients compared to other patients with post-stroke hemiplegia. Future research in these areas would greatly add to our ability to best provide physical therapy management for patients with this disorder.

Back to Top | Article Outline


1 Pedersen PM, Wandel A, Jorgensen HS, et al. Ipsilateral pushing in stroke: incidence, relation to neuropsycho-logical symptoms, and impact on rehabilitation. The Copenhagen Stroke Study. Arch Phys Med Rehabil. 1996;77:25–28.
2 Karnath HO, Johannsen L, Broetz D, et al. Prognosis of contraversive pushing. J Neurol. 2002;249:1250–1253.
3 Karnath HO, Ferber S, Dichgans J. The neural representation of postural control in humans. Proceedings of the National Academy of Sciences of the United States of America. 2000;97:13931–13936.
4 Karnath HO, Ferber S, Dichgans J. The origin of contra-versive pushing:evidence for a second graviceptive system in humans. Neurology. 2000;55:1298–1304.
5 Perennou DA, Amblard B, Laassel el M, et al. Understanding the pusher behavior of some stroke patients with spatial deficits: a pilot study. Arch Phys Med Rehabil. 2002;83:570–575.
6 Rode G, Tiliket C, Boisson D. Predominance of postural imbalance in left hemiparetic patients. Scand J Rehabil Med. 1997;29:11–16.
7 Bohannon RW. Ipsilateral pushing in stroke. Arch Phys Med Rehabil. 1996;77:524–525.
8 Perennou DA, Amblard B, Leblond C, et al. Biased postural vertical in humans with hemispheric cerebral lesions. Neurosci Lett. 1998;252:75–78.
9 Davies PM. Steps to Follow: A Guide to Treatment of Adult Hemiplegia. New York, NY: Springer-Verlag; 1985.
10 Bailey MJ, Leivseth L. The ‘pusher’ syndrome in elderly stroke patients… symposium on stroke. Br J Ther Rehabil. 2000;7:11–16.
11 Premoselli S, Cesana L, Cerri C. Pusher syndrome in stroke: clinical, neuropsychological and neurophysio-logical investigation. Europa Medicophysica. 2001; 37:143–151.
12 Bohannon RW, Cook AC, Larkin PA, et al. The listing phenomenon of hemiplegic patients. Neurol Report. 1986; 10:43–44.
13 Masdeu JC, Gorelick PB. Thalamic astasia: inability to stand after unilateral thalamic lesions. Ann Neurol. 1988;23:596–603.
14 Bohannon RW, Smith MB, Larkin PA. Relationship between independent sitting balance and side of hemi-paresis. Phys Ther. 1986;66:944–945.
15 Dieterich M, Brandt T. Wallenberg's syndrome: lat-eropulsion, cyclorotation, and subjective visual vertical in thirty-six patients. Ann Neurol. 1992;31:399–408.
16 Bjerver K, Silfverskiold BP. Lateropulsion and imbalance in Wallenberg's syndrome. Acta Neurol Scand. 1968; 44:91–100.
17 Brandt T, Dieterich M, Danek A. Vestibular cortex lesions affect the perception of verticality. Ann Neurol. 1994; 35:403–412.
18 Bisdorff AR, Wolsley CJ, Anastasopoulos D, et al. The perception of body verticality (subjective postural vertical) in peripheral and central vestibular disorders. Brain. 1996;119(Pt 5):1523–1534.
19 Anastasopoulos D, Haslwanter T, Bronstein A, et al. Dissociation between the perception of body vertical-ity and the visual vertical in acute peripheral vestibular disorder in humans. Neurosci Lett. 1997;233:151–153.
20 Karnath HO, Brotz D, Gotz A. [Clinical symptoms, origin, and therapy of the “pusher syndrome”]. Nervenarzt. 2001;72:86–92.

pusher syndrome; stroke; hemiplegia

© 2004 Neurology Section, APTA