Secondary Logo

Journal Logo

Systematic Reviews

Clinical Outcome Measures for Lateropulsion Poststroke: An Updated Systematic Review

Koter, Ryan PT, DPT; Regan, Sara PT, DPT; Clark, Caitlin PT, DPT; Huang, Vicki DPT; Mosley, Melissa PT, DPT; Wyant, Erin PT, DPT; Cook, Chad PT, PhD, MBA, FAAOMPT; Hoder, Jeffrey PT, DPT, NCS

Author Information
Journal of Neurologic Physical Therapy: July 2017 - Volume 41 - Issue 3 - p 145-155
doi: 10.1097/NPT.0000000000000194



Lateropulsion stems from the Latin ‘lateralis’ or side and ‘pulsus’ to push or drive. Clinically, the term refers to an inability to sustain midline of either the eyes (ocular lateropulsion) with an ocular drift, or, more frequently, to refer to misperception of postural verticality with an involuntary tendency to drift or fall to one side when positioned upright. With regards to postural lateropulsion, the site of the lesion determines whether a postural tilt or drift occurs towards the side of the lesion, ipsiversive lateropulsion, or away from the side of the lesion, contraversive lateropulsion. Ocular lateropulsion and ipsiversive postural lateropulsion are sequelae of the Lateral Medullary Syndrome or “Wallenberg's Syndrome” seen in posterior circulation brainstem strokes to the lateral medulla.1 They are related to damage to vestibular nuclei causing abnormal perceptions of ocular and postural vertical. Ipsiversive lateropulsion was not considered within the scope of this review.

Contraversive Lateropulsion, also referred to as contraversive pushing, pusher behavior, and pusher syndrome, is a postural control disorder that can manifest in individuals following stroke.2–5 Davies5 first described the tendency of certain persons poststroke to push “strongly towards his hemiplegic side in all positions and resist any attempt at passive correction of his posture; that is, correction which would bring his weight towards or over the midline of his body to the unaffected side.” The etiology of contraversive lateropulsion remains largely unknown, but Pérennou et al6 found that those who exhibit lateropulsion have an altered sense of postural verticality. These individuals define vertical from an egocentric reference system rather than that of the Earth. Reports on the prevalence of lateropulsion range from 10% to greater than 60%, with variable reports on prognosis.2,7–9 Some authors describe contraversive lateropulsion resolving within 6 months, whereas others report persistence for over 2 years.10,11 Babyar et al12 suggest that time to recovery from contraversive lateropulsion is related to the number and magnitude of stroke impairments present at admission including motor, proprioceptive, and visual-spatial orientation impairments.

Persistent contraversive lateropulsion towards the affected side interferes with motor recovery during rehabilitation and has been shown to protract recovery time when combined with additional postural control deficits.2,12 Early identification of contraversive lateropulsion may guide interventional treatment from physical therapists.10 The lack of a gold standard diagnosis as well as a reliable and valid outcome measure hinders clinicians and researchers from consistently quantifying the degree and presence of contraversive lateropulsion. Unclear quantification may coincide with a deficiency of studies that examine the efficacy of interventions used to ameliorate the disordered behavior. The identification of the most psychometrically sound outcome measures may increase the feasibility of future intervention trials. Use of a reliable measure with strong test-retest ability will render novel interventional research more relevant and clinically meaningful.

Babyar et al13 performed a systematic review to evaluate the clinimetric properties and clinical practicality of published outcome measures used to quantify the degree of contraversive lateropulsion. The review cited 4 articles and evaluated 3 clinical outcome measures: (1) the Scale for Contraversive Pushing (SCP), 2) the Modified Scale for Contraversive Pushing (M-SCP), and (3) the Burke Lateropulsion Scale (BLS).13–17 No recommendation was made for clinical use of a single scale. The review concluded that the SCP had been investigated most thoroughly, the BLS and M-SCP examine more functional positions than the SCP, and that more research is warranted with more heterogeneous clinical populations.13

Since the publication of Babyar and colleagues'18 review, the COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) checklist has been published. COSMIN is a tool with standardized language used to assess psychometric properties to improve reporting and to compare outcome tools.18 Subsequently, the purpose of this systematic review was to perform an updated, exhaustive search to identify any additional scales utilized for contraversive lateropulsion and to reevaluate literature that addresses clinimetric properties of scales previously identified using the COSMIN checklist.18 The authors strive to provide a useful resource for clinicians by making more specific recommendations for use of these tools in clinical practice and to promote interventional research that uses the most appropriate tool to measure responsiveness to treatment.


During the search and reporting phase, the authors prospectively applied the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Statement.19 The COSMIN checklist was used to evaluate the methodological quality of the measurement properties in each study.18 Percentage agreement is high for the COSMIN itself (68% of items had percentage agreement >80% in the Delphi study), indicating that raters often choose the same response option; however, low reported κ coefficients (61% below 0.40) may indicate a difficulty in distinguishing differences between articles on an item level.18,20 Content validity has been established in a large Delphi study performed by a panel of experts.21Table 1 displays accepted definitions of measurement property terminology utilized during COSMIN application.18,22

Table 1. - Accepted Definitions of Measurement Property Terminologya
Agreement A form of measurement reliability that indicates how close repeated measures are to one another
Construct validity The degree to which a measure matches the operational definition of the concept or constructs it is said to represent
Content validity The degree to which items in an instrument represent all of the facets of the variable being measured
Criterion validity The degree to which a measure of interest relates to a measure with established validity (a “gold” or “reference” standard)
Cross-cultural validity The degree to which the performance of the items on a translated or culturally adapted instrument are an adequate reflection of the performance of the items of the original version of the instrument
Face validity A subjective assessment of the degree to which an instrument appears to measure what it was designed to measure
Internal consistency The degree to which subsections of an instrument measure the same concept or construct; the degree of interrelatedness among the items
Reliability The extent to which repeated measurements agree with one another. This can include measurement over time (test-retest), by different persons on the same occasion (interrater), or by the same person(s) on different occasions (intrarater)
Responsiveness The ability of a measure to detect change over time in the phenomenon of interest
Validity The degree to which an instrument measures the construct that it is supposed to measure
aDefinitions adapted from Mokkink et al18 and Jewell.22

Search Strategy

PubMed, CINAHL, and Embase were searched from the inception of each database to March 2017. The search term was as follows: (“stroke”[MeSH Terms] OR “stroke”[tiab]) AND (lateropulsion[tiab] OR (lateral[tiab] AND propulsion[tiab]) OR “pusher syndrome”[All Fields] OR (pushing[tiab] AND syndrome[tiab]) OR “pusher behaviour”[All Fields] OR “pusher behavior”[tiab] OR “pushing behaviour”[tiab] OR “pushing behavior”[tiab] OR “contraversive pushing”[tiab] OR “postural vertical”[tiab] OR “visual vertical”[tiab] OR (haptic[tiab] AND vertical[tiab])).

Eligibility Criteria

Included studies met the following criteria: (1) references scales or outcome measures for measuring contraversive lateropulsion or pusher behavior toward an individual's involved side following hemispheric stroke; (2) addresses at least 1 clinimetric property of that scale; (3) includes patients diagnosed with Davies'5 definition of pusher syndrome; (4) published in English, and (5) corresponds to a full manuscript. Articles were excluded that examined patients more than 1 year poststroke, were animal studies, or that examined ipsilesional lateropulsion, lateral medullary syndrome, or Wallenberg syndrome. Patients more than 1 year poststroke were excluded to ensure consistency and secondary to lateropulsion frequently resolving by 6 months.7,10,11 Editorials, letters, or case reports were excluded.

Study Selection

Two authors (SR and EW) independently performed a title screen followed by abstract review in PubMed. Two different authors (MM and VH) completed full-text reads, excluding articles that did not meet inclusion criteria. A third author (RK) resolved discrepancies throughout the selection process. The same screening process was repeated in CINAHL and Embase.

Data Extraction

The authors read each of the studies included in the final analysis for the purposes assessment of methodological quality and quality of the measurement properties of each outcome tool. Two authors (SR and EW) extracted relevant data related to study population, outcome measures implemented, and clinimetric property assessment.

Quality Assessment

Methodological quality was assessed using the COSMIN 4-item ordinal scale designed for systematic reviews, wherein each measurement property has specifically outlined questions serving to rank a property as “poor,” “fair,” “good,” or “excellent.”23,24 Each study was scored individually for the properties addressed within, with the exception of content validity. The authors assessed the content validity for each outcome measure using COSMIN and the information provided in each of the included studies. The poorest score for any item in a subsection represented the overall score for that section.18 Authors CC and RK independently applied the COSMIN checklist to included articles using the COSMIN Manual without prior training.18,24 Consensus was reached on final ratings. Interrater reliability was not formally assessed between the COSMIN-administering authors.


The result of the search process is summarized in the Figure. In total, 195 records were identified in PubMed and 208 in Embase and CINAHL collectively; 26 articles underwent abstract review, and 17 articles were reviewed in full. Four outcome measures were identified within 7 included manuscripts.14–17,25–27Table 2 compares the 4 scales identified. Table 3 summarizes the findings for clinimetric properties of the SCP, the Swedish Scale for Contraversive Pushing (S-SCP), M-SCP, and the BLS. Table 4 summarizes results of the COSMIN checklist quality assessment.14–17,25–27

PRISMA flow diagram.
Table 2. - Comparison of the 4 Scales Used to Identify Pusher Behavior
Scale Component Scale for Contraversive Pushing2 Swedish Scale for Contraversive Pushing3 Modified Scale for Contraversive Pushing4 Burke Lateropulsion Scale5
Postures in which patients are assessed 3 components scored in both sitting and standing:
  1. Spontaneous body posture

  2. Extension/abduction of uninvolved extremities

  3. Resistance to passive correction

Identical to SCP. Translated into Swedish with modified verbal instruction for patient cueing 4 testing positions:
  1. Sitting

  2. Standing

  3. Sitting transfer

  4. Standing transfer

5 testing positions:
  1. Supine rolling

  2. Sitting

  3. Transferring

  4. Standing

  5. Walking

Scoring Spontaneous body posture (0, 0.25, 0.75, 1 = severe tilt); extension/abduction of uninvolved extremities (0, 0.5, 1 = performed spontaneously at rest); resistance to passive correction (0, 1 = resistance occurs) Identical to SCP Each scored 0-2 = pushes continuously with force enough to fall if not supported, abducts uninvolved arm and/or leg spontaneously, even at rest
Total maximum = 8
Scoring based on degree and point of onset of resistance to passive correction (sit, stand) or degree of pushing evident (supine, transfer, walking). 0-3 scale except for standing (0-4)
Total maximum = 17
0 = no PB, ≥2 = PB
Total maximum score = 6; 0 = no tilt, extension, resistance
6 = severe tilt, resistance, extension
Cutoff scores indicative of Pusher syndrome Not definitively established
Original suggestion of >1 on each section2
Later recommended that the cutoff score be >0 in each section6
Total score ≥3 Total score ≥3 ≥2 points have been used inconsistently as the cutoff score. Recent studies have recommended >27–9
MCID/MDC Not yet established Not yet established Not yet established Not yet established
Abbreviations: MCID/MDC: Minimal Clinically Important Difference/Minimal Detectable Change; PB: Pusher Behavior; SCP: Scale for Contraversive Pushing.

Table 3. - Summary of Populations, Outcome Measure Exposed, and Measurement Properties Studied for Each Included Article
Study Population Exposure Outcome
D'Aquila et al17 85 patients (50 female, 35 male) with strokea (19 ± 2 d from initial evaluation) collected using convenience sampling from the inpatient rehabilitation unit at Burke Rehabilitation Hospital in White Plains, New York. Location of stroke: 35 right hemisphere, 33 left hemisphere, 4 bilateral hemispheres, 15 brainstem, cerebellum, or other combination Patients were evaluated using the BLS.
  • Inter-rater and intrarater reliability using ICC

  • Concurrent validity, using r s, of BLS compared with the FM-B subscore, admission FIM mobility subscore, discharge FIM mobility subscore, and length of rehabilitation hospital stay

Interrater reliability with an expert:
  • ICC = 0.93b

Intrarater reliability:
  • ICC = 0.94b

Concurrent validity:
  • FM-B: r s = −0.57b

  • IE FIM mobility: r s = −0.56c

  • DC FIM mobility: r s = −0.58c

  • Length of rehabilitation stay: r s = 0.60c

Baccini et al14 26 patients (11 female, 15 male; mean age 72.2 ± 10.0 y) with first-time stroke (19.0 ± 7.8 d from stroke onset) exhibiting postural asymmetry. Side of lesion: 19 right-sided, 7 left-sided Patients were evaluated using the SCP Calculated:
  • Diagnostic accuracy with cutoff scores of ≥1 and >0

  • Interrater reliability for total SCP score and each of the 3 sections (Section A: spontaneous body posture, Section B: use of nonparetic extremities, and Section C: resistance to passive correction of tilted posture)

  • Internal consistency (Cronbach α)

  • Criterion validity of SCP by comparing diagnosis of Pusher syndrome against clinical examination. Utilized Cohen's κ coefficient for intertester agreement

Diagnostic accuracy:
  • Abnormal test ≥ 1: 73.1%

  • Abnormal test > 0: 96.2%

Interrater reliability:
  • SCP total: ICC = 0.97

  • Section A: ICC = 0.94

  • Section B: ICC = 0.93

  • Section C: ICC = 0.94

Internal consistency:
  • Cronbach α: 0.92

Pearson coefficients:
  • Section A/SCP total: 0.91b

  • Section B/SCP total: 0.96b

  • Section C/SCP total: 0.91b

  • Section A/Section B: 0.86b

  • Section A/Section C: 0.71b

  • Section B/Section C: 0.81b

Criterion validity:
  • Subscores ≥1: κ = 0.50, SE = 0.10, SN 58.8%, SP 56.3%

  • Subscores > 0: κ = 0.92, SE = 0.06, SN 94.1%, SP 90%

Lagerqvist and Skargren16 All patients with stroke admitted into acute stroke unit at Rehabilitation South, The County Hospital of Kalmar, Sweden between January 2004 and January 2005 that exhibited probable pushing after strokea: 19 patients total (9 female, 10 male; 66-95 y old). Hospital stay varied from 6 to 61 d with a median of 15 d.
8 patients exhibited right hemiplegia and 11 patients exhibited left hemiplegia
Patients evaluated using the M-SCP
  • Interrater reliability using κ and r s

  • Concurrent validity comparing to BBS and S-COVS

  • Degree of pushing and functional ability from initial evaluation to discharge

Interrater reliability for total score:
  • κ = 0.51 at IE and κ= 0.73 at DC

  • r s = 0.82 at IE and r s = 0.94 at DC

Concurrent validity of BBS:
  • r s = −0.52 at IE and −0.49 second evaluation

Concurrent validity of S-COVS:
  • r s = −0.43 IE and −0.45 second evaluation

Correlation between BBS and S-COVS:
  • r s = 0.90, both evaluations

Baccini et al15 105 patients (51 female, 54 male; 70.6 ± 11.2 y old) with first-time strokea admitted to 1 of 2 inpatient rehabilitation hospitals (Villa Fiorita Hospital, Prato, Italy, and Piero Palagi Hospital, Florence, Italy) between January 2006, and December 2006. Subjects met the following 4 criteria: first stroke, monolateral strokea, stroke ≤ 30 d from onset, and presence of unilateral impairment of limbs, impaired balance and motricity on FM. Location of stroke: 49 right, 56 left-sided stroke. 17 subjects were diagnosed as having contraversive lateropulsion on the basis of clinical examination Patients evaluated using the SCP
  • Diagnostic agreement between SCP and clinical diagnosis based on 3 suggested cutoff scores:

    • Criteria 1: total SCP score >0

    • Criteria 2: score of >0 in each of the 3 SCP sections

    • Criteria 3: score of ≥1 in each of the 3 SCP sections

  • Validity of SCP with the Barthel Index, the balance subscore of the FM-B, and the LIND-MOB with Spearman rank correlation coefficient

Diagnostic agreement between SCP and clinical diagnosis based on suggested cutoff scores:
  • Criteria 1: 54.3%, Cohen κ = 0.212

  • Criteria 2: 98.1%, Cohen κ = 0.93

  • Criteria 3: 94.3%, Cohen κ = 0.75

Construct validity (balance and function)
  • Barthel Index: r s = −0.63

  • FM-B: r s = −0.67

  • LIND-MOB: r s = −0.60

Criterion validity:
  • Criterion 1: SN 100%, SP 46%, PPV 26%, NPV 100%

  • Criterion 2: SN 100%, SP 98%, PPV 90%, NPV 100%

  • Criterion 3: SN 65%, SP 100%, PPV 100%, NPV 94%

Hallin et al26 22 patients (10 female, 10 male; 64 ± 9.6 y old) with hemiparesis after stroke (in the acute phase with signs of pushing behavior) admitted to the neurological stroke unit at Sahlgrenska University Hospital in Gothenburg, Sweden, from March 2005 to February 2006. Location of strokea: 15 right-brain lesion and 7 left-brain lesion. Type of stroke: 12 ischemic and 10 hemorrhagic Patients were evaluated using S-SCP Calculated:
  • Interrater reliability using the ICC for the variables of spontaneous body posture in sitting and standing, abduction and extension of the nonparetic extremities in sitting and standing, and the total score

  • Interrater reliability using κ for variable resistance to passive correction of tilted posture in sitting and standing

Interrater reliability:
  • SBP in sitting: ICC = 0.83d, PA = 78%

  • SBP in standing: ICC = 0.86d, PA = 82%

  • Abd, ext sitting: ICC = 0.73d, PA = 72%

  • Abd, ext standing: ICC = 0.85d, PA = 88%

  • Resistance sitting: κ = 0.61e, PA = 82%

  • Resistance standing: κ = 0.79e, PA = 93%

  • S-SCP total score: ICC = 0.84d, PA 48%

  • Pusher syndrome diagnosis: κ = 0.71e, PA = 86%

Clark et al25 Prospective, purposive sample of 43 patients with acute strokea (21 female, 22 male; 72.1 ± 6.7 y old) demonstrating lateropulsion within the Northern Health Rehabilitation Units and Geriatric Units across 2 sites in Melbourne, Australia. Mean time from stroke to assessment: 21.1 ± 7.1 d Patients were evaluated using the BLS
  • Internal responsiveness of the BLS to quantify severity of contraversive lateropulsion, which was evaluated using SRM

  • A correlation was performed to compare BLS and the PASS, which measures postural abilities, using r s

Internal responsiveness:
  • Mean change from IE to DC on the BLS was 4.7 ± 3.6, SRM = 1.30

  • Mean change between 0- to 4-week period on the BLS was 3.59 ± 2.43, SRM = 1.48

  • Mean change between 0- to 8-week period on the BLS, SRM = 2.24

Correlation of BLS and PASS:
  • 0 wk: r s = −0.61

  • 2 wk: r s = −0.72

  • 4 wk: r s = −0.76

Bergmann et al27 23 patients (6 female, 17 male; 68 ± 10.6 y old) with hemiparesis due to a unilateral strokea (1.6 ± 0.7 mo poststroke) in an inpatient stroke rehabilitation unit. Location of stroke: 19 right-sided, 4 left-sided Patients were evaluated using SCP and the BLS immediately before and after 3 different interventions.
  • SN and SP of the BLS were calculated using the SCP as the reference standard

  • Agreement between diagnoses of pusher behavior compared with the χ2 test and κ

  • Responsiveness was assessed by comparing the changes detected between the scales between and across administrations

Criterion validity of BLS with the SCP:
  • SN: 100%, SP: 67%

Agreement between diagnoses of pusher behavior based on SCP (cutoff > 0) and the BLS:
  • χ2 = 54.26; κ = 0.564, SE = 0.062

Differences detected post-intervention between the SCP and BLS:
  • χ2 = 19.15 with moderate agreement (κ = 0.50, SE = 0.10)

  • BLS may be more responsive to small changes than the SCP. Out of 69 cases, change was detected in 29 patients on the BLS and 19 patients on the SCP. 16 of those cases showed change on both scales

Abbreviations: Abd, ext, abduction and extension; BBS, Berg Balance Scale; BLS, Burke Lateropulsion Scale; DC, discharge; FIM, Functional Independence Measure; FM, Fugl Meyer; FM-B, Fugl-Meyer Assessment Balance subscore; ICC, intraclass correlation coefficient; IE, initial evaluation; LIND-MOB, subscore of the mobility section of the motor assessment chart developed by Lindmark and Hamrin; m-SCP, modified Scale for Contraversive Pushing; NPV, negative predictive value; PA, percent agreement; PASS, Postural Assessment Scale for Stroke Patients; PPV, positive predictive value; rs, Spearman rank order correlation coefficient; SBP, spontaneous body posture; S-COVS, Swedish Physiotherapy Clinical Outcome Measure; SCP, Scale for Contraversive Pushing; SE, standardized error; SN, sensitivity; SP, specificity; SRM, standardized response mean; S-SCP, Swedish Scale for Contraversive Pushing.
aStroke type not specified between ischemic or hemorrhagic.
bP < 0.001.
cP < 0.0001.
dP < 0.05.
eP < 0.01.

Table 4. - Summary of COSMIN Checklist Findings Based on the Most Frequent Quality Rating in Each Categorya
Lateropulsion Outcome Measure Internal Consistency Reliability Cross-Cultural Validity Criterion Validity Responsiveness Content Validity
Scale for Contraversive Pushing Poor Poor No studies Mixed—poor to excellent Poor Poor
Swedish Scale for Contraversive Pushing No studies Poor Poor No studies No studies Poor
Modified Scale for Contraversive Pushing No studies Poor No studies Poor No studies Poor
Burke Lateropulsion Scale No studies Mixed—poor to good No studies Mixed—poor to good Mixed—poor to fair Poor
aFor measurement properties with mixed evidence, a range of quality ratings is provided.

Scale for Contraversive Pushing


The SCP was developed by Karnath et al28 and is based on Davies'5 definition of pusher syndrome. The English translation of the SCP was published in 2000.29 The original German language version was later published in 2001.28 The SCP assesses patients in both seated and standing positions, assessing 3 components of contraversive lateropulsion: (1) symmetry of spontaneous body posture, (2) the use of nonparetic extremities (leg or arm) to push by abduction and extension thrust, and (3) the degree of resistance to passive correction of the tilted posture. The individual is considered to exhibit lateropulsion if each component has a minimum score of 1; however, a modified cutoff criterion of more than 0 has been frequently used to improve sensitivity and specificity (Table 2).7,15,25,30

Clinimetric Properties

Three articles that analyzed the SCP were included in the systematic review.14,15,27 A study by Baccini et al14 examined reliability, internal consistency, and criterion validity using clinical diagnosis by an expert as the reference standard (Table 3).14 Testers had moderate agreement using the original criterion for diagnosing contraversive lateropulsion (each component >1); however, a modified cutoff criterion (>0 on each SCP component) improved agreement between the expert's clinical diagnosis and the SCP score.14 Because of a sample size under 30 for the study, the COSMIN checklist shows poor-quality assessment for reliability, internal consistency, and criterion validity.18

Beyond their examination of psychometric properties of the SCP, Baccini et al15 also examined diagnostic agreement between SCP and clinical diagnosis with 3 cutoff criteria. The more conservative criterion (each component >1) yielded 6 false-negative results; however, the modified criteria (>0 on each SCP component) resulted in 2 misdiagnoses.18 Moderate construct validity was observed between the SCP component and total scores with the Barthel Index, the balance subscore of the Fugl-Meyer Assessment Scale, and the subscore of the mobility section of the motor assessment chart developed by Lindmark and Hamrin.31 The modified cutoff criteria (each component >0) yielded the highest sensitivity and positive predictive values.15

A 2014 study by Bergmann et al27 provides the first investigation in which multiple tools for contraversive lateropulsion evaluation were directly compared. When comparing the SCP to the BLS, 22.5% of cases resulted in inconsistent classification, with more false negatives observed with the SCP. These tools demonstrated moderate diagnostic agreement on the basis of the modified criterion (cutoff >0) and on detecting pre- and postintervention changes.27 The COSMIN checklist yielded a result of poor reliability secondary to small sample size (Table 4).

The cross-cultural validity of the SCP has not yet been considered in the translation from the original language of German into English. A formal description of the translation process including the number of translators, their expertise, and whether the process included forward and backward translation was not provided.4,32 Baccini et al14,15 did not describe which translation was used in their studies performed in Italy.

Clinical Practicality

The original SCP did not provide a detailed guide for its administration or verbal instructions that should be provided. Karnath et al33 later clarified how to test the amount of extension and abduction of the nonparetic extremities. Findings from Baccini et al14,15 and Bergmann et al27 suggest an a posteriori change in the criterion for diagnosing contraversive lateropulsion (each component <0 instead of <1). Despite improvements in the sensitivity and consistency of classification with the SCP, the BLS appears to be more responsive. Future study requires larger sample sizes.

Swedish Scale for Contraversive Pushing


Hallin et al26 translated the SCP into Swedish, resulting in the S-SCP. The S-SCP is equivalent to the original scale, with the exception of modified verbal instruction for patients to when assessing resistance to passive correction of tilted posture.

Clinimetric Properties

Good interrater reliability was reported for the total S-SCP score and for each of the 6 postures assessed (Table 3).26 A sample size under 30 led to a poor-quality rating by the COSMIN checklist.26

Clinical Practicality

Hallin et al26 note that the variables assessed in sitting posture had lower reliability than variables assessed in standing, postulating that contraversive lateropulsion manifests to a greater extent in stance. During standing, persons exhibiting lateropulsion have a reduced base of support, increased postural control demands, and increased opportunity to push with the nonparetic lower extremity. Although this is the only study to report the difference in reliability between these tasks, all of the outcome measures identified for lateropulsion include both sitting and standing components. Lateropulsion analysis across multiple tasks may contribute to a better understanding of the underlying mechanism, more accurately identifying the incidence of contraversive lateropulsion, and better classifying individuals based upon severity. Further study of the internal consistency, responsiveness, and validity is warranted to better recommend substitution of the S-SCP for the original SCP in Swedish-speaking settings.

Modified Scale for Contraversive Pushing


The M-SCP was developed from a modified version of the S-SCP. The modified scale further expanded the tasks examined to 4 sections: static sitting, static standing, sitting transfer, and standing transfer. All sections are scored between 0 and 2, with a score of 0 indicating no symptoms and 2 indicating that the condition fully meets the definition of contraversive lateropulsion.5,16

Clinimetric Properties

Lagerqvist and Skargren16 report the M-SCP to have moderate to excellent agreement for interrater reliability. The creators of the M-SCP were also involved in the data collection for the study, potentially introducing bias in the administration of the assessment.16 Low-to-moderate concurrent validity measurements observed comparing the M-SCP with the Berg Balance Scale and the Swedish Physiotherapy Clinical Outcome Measure (S-COVS). COSMIN review found the quality of the reliability and criterion validity assessment to be poor secondary to small sample size.

Clinical Practicality

Further study is needed to recommend M-SCP use in English-speaking countries including internal consistency, responsiveness, and cross-cultural validity assessment on a larger clinical population with a descriptive English translation process.

Burke Lateropulsion Scale


The construct of the BLS is based on Davies'5 definition of lateropulsion and measures the amount of resistance offered by the patient when sustaining or changing position.17,25,27 Performance is assessed during the execution of 5 tasks: rolling, sitting, standing, transferring, and walking. Each component score reflects the amount of resistance encountered by the examiner when correcting tilted posture and reflex equilibrium responses. Component scores range 0 to 3, with the exception of standing, which ranges from 0 to 4. Scores for rolling in supine, walking, and transferring are based on the severity of pushing sensed by the examiner. The sum score ranges from 0 to 17, with 0 indicating no lateropulsion. A total score of 2 or more has been recommended as the cutoff score to indicate the presence of lateropulsion; however, a cutoff score of more than 2 shows better agreement between the BLS and the SCP. 13,27,34

Clinimetric Properties

D'Aquila et al17 examined the concurrent validity and the inter- and intrarater reliability of the BLS (Table 3). Moderate correlations between the BLS and additional outcome measures suggest that lateropulsion is related to constructs of functional status and balance. Reliability and criterion validity each scored good on the COSMIN checklist secondary to a sample size of 85 (Table 4).

Clark et al25 reported the BLS to be highly responsive at 4 and 8 weeks poststroke, highlighting that the BLS may be a sensitive tool to measure reductions in contraversive lateropulsion over time. The sample size was moderate; however, responsiveness received a COSMIN score of poor because standardized response mean is considered an inappropriate assessment of responsiveness.18,24

As previously described, Bergmann et al27 used both the BLS and the SCP to evaluate contraversive lateropulsion. Criterion validity of the BLS was calculated using the SCP as the reference standard. Caution is advised in interpreting the reported results, as there is no established gold standard. Moderate diagnostic agreement was found between the 2 outcome measures as well as for postintervention changes (Table 3). Results suggest that the BLS is more responsive to small changes compared with the SCP. The BLS is the only measure to receive ratings greater than poor for reliability and responsiveness.

Clinical Practicality

The BLS has a precise guide for administration and reportedly satisfactory clinimetric properties; however, some studies were rated with COSMIN as poor quality secondary to small sample sizes.25,27 Studies suggest that the BLS is more responsive than the SCP and has greater sensitivity for identifying contraversive lateropulsion.25,27 The wider score range of the BLS (0-17) compared with the SCP (0-6), including the assessment of the graduated angle where resistance to passive alignment correction begins, may better enable identification of small changes.30 Strong interrater and intrarater reliability is reported for the BLS in a wide range of individuals with and without contraversive lateropulsion.17 Internal consistency has yet to be assessed. The BLS includes the assessment of rolling, a movement that is easy to assess in early inpatient stays, as well as more functional tasks including transfers and walking.5 Examining a wider array of functional tasks may enable clinicians to better plan interventions when working with patients who have severe contraversive lateropulsion. Moderate negative correlations observed with the Postural Assessment Scale for Stroke Patients (PASS) suggest that clinicians should consider administration of the BLS in conjunction with other balance or postural control outcome measures that have well-established psychometric properties and are validated in the stroke population.


The aim of this study was to update and improve upon the systematic review by Babyar et al13 because of the lack of a well-established gold standard outcome measure. This was to be accomplished by (1) identifying additional scales for classifying contraversive lateropulsion, (2) providing newfound data regarding clinimetric properties for scales previously identified, and (3) creating an updated resource that can make recommendation for use of the most appropriate tool in clinical practice.

This revised search yielded 3 new articles and 1 additional outcome measure to add to the findings of the previous review.25–27 Limited results reveal a persistent lack of quality literature related to contraversive lateropulsion, perhaps because of failure to consistently identify the condition clinically, as well as the fact that it typically resolves within 1 year poststroke.10 This review raises questions about the cross-cultural validity of the SCP, which was not addressed by Babyar et al.13 The S-SCP has a thorough description of the translation process and modified verbal instruction, whereas the SCP lacks formal description of the translation from the original German text to English.4 The language of these outcome measures may significantly influence patient performance, as it influences the way in which clinicians encourage patients to alter postural alignment and the amount of resistance needed to move the patient into correct alignment. The SCP provides a single translated instruction to give to patients: “I will move your body sideways, please permit this movement.” 4 The extent to which the clinician may paraphrase outside of this instruction remains unclear. The BLS offers no specific verbal instruction to the client, but does more specifically describe testing positions to minimize variability. Individuals with native-level proficiency should perform all future translations, which should be fully disclosed.

Despite unevaluated clinimetric properties and mostly poor-quality assessment ratings (primarily due to small sample sizes), we recommend the BLS for early detection of contraversive lateropulsion in English-speaking countries.13 The BLS appears to be the most responsive of the lateropulsion scales and has the potential to be a more sensitive instrument to measure deviations in postural control in the context of multiple activities. The BLS assesses movements and postures, which may be more likely to be viewed early in hospital stays (ie, rolling), prior to assessing more advanced movements that may be impaired secondary to medical instability, decreased staff support, and decreased lower extremity strength.27 Unlike the SCP, the BLS includes a walking assessment, making it more functionally relevant later in an individual's recovery.

Although the original description Davies5 used to define contraversive lateropulsion remains relevant, clinicians should consider implementation of the BLS as soon as deviations in postural vertical are suspected following stroke. Identifying the presence of contraversive lateropulsion may help clinicians, particularly those in acute care environments, more accurately establish prognosis, set goals, and better formulate an appropriate plan of care, as contraversive lateropulsion has such a significant potential to alter rate of recovery and increase hospital length of stay.2,7,11,12 Early assessment will more objectively reveal presence of contraversive lateropulsion, quantify its severity, enabling clinicians to better plan interventions (ie, rehab tech assistance) and objectively quantify improvement in symptoms over time. The BLS includes more comprehensive list of functional tasks than the SCP, which are all likely to be included within a clinician's initial examination. The BLS administration would likely add only a few minutes to any clinical examination. Clinician-scientists should feel more confident designing and implementing interventional research on individuals with quantifiable contraversive lateropulsion using this tool alone.

Contraversive Lateropulsion can significantly limit a clinician's ability to treat patients with stroke without the assistance of another clinician because of safety concerns with transfers, standing, and ambulation.35 Delayed or absent functional training may thereby increase length of hospital stay and necessitate more intensive therapy.36,35 Identifying treatments that maximize functional recovery and minimize physical demand on therapists will help improve patient prognoses, promote safety, and minimize burden of care.

The etiology of contraversive lateropulsion remains poorly understood; however, it is clear that its manifestation relates to vision, somaesthetic input processing, and verticality perception, suggesting that consideration of constructs not evaluated in current lateropulsion outcome measures is warranted.30,36 With a range of affected regions of the brain, multiple subtypes of contraversive lateropulsion may exist, making identification of subtle changes and examination of various postures more valuable. The correlation between contraversive lateropulsion and subjective postural vertical should be further investigated along with function and balance to better validate all constructs comprising lateropulsion.6,34 We suggest that future studies concomitantly implement both the BLS and a well-established outcome measure for an associated construct (ie, balance) to improve clinical applicability.

Several limitations exist within the literature. Type of lesion (eg, ischemic and hemorrhagic) and stroke severity relating to lesion size and motor, sensory, and perceptual status were not provided in any included study. Reported duration from the stroke onset to time of assessment was wide: 19 ± 7.8 days to 1.6 ± 0.7 months (Table 3).14,27 Only 1 included study used a sample size greater than 100 and several lacked a heterogeneous population (Table 3).15 Moderate relationships were observed between measures of contraversive lateropulsion and several measures that assess balance and function; however, there is currently no clear gold standard with which to compare these scales. A few of the studies included in this review used assessors who also helped develop the scales, potentially biasing reliability measurements.16,17,26 Similar to Babyar et al,13 we question whether the presence or severity of contraversive lateropulsion can be assumed without a complete composite score, such as when a patient cannot be formally tested in a given subsection of a scale due to reasons unrelated to contraversive lateropulsion such as cardiovascular, physical, or cognitive impairments. In addition, not all statistical methods to assess clinimetric properties were considered appropriate according to the COSMIN checklist standards.25

There were limitations in the way this updated systematic review was conducted. Two individuals independently performed the title/abstract screen, and 2 other individuals performed full-text reviews. Potential for bias may have been reduced, and consistency may have been improved if the same individuals performed all portions of the search and screening process.


The BLS is recommended as the preferred tool to evaluate contraversive lateropulsion in English-speaking countries. The BLS is the strongest scale available, as it includes the widest array of functional testing positions and has less uncertainty with regard to cross-cultural validity in comparison to other existing scales. The BLS should be implemented in inpatient settings as soon as contraversive lateropulsion is suspected, to quantify severity, help clinicians plan interventions, and track change over time. Further studies are warranted with larger, more heterogeneous populations to improve the quality of examined measurement properties of these scales. Formal translations of contraversive lateropulsion scales should fully disclose the specific instructions to be given to the patient during the assessment.


We thank Jeannine Bergmann for her assistance in reviewing the article and Leila Ledbetter, MLIS, for her assistance in developing the search strategy.


1. Brandt T, Dieterich M. (2000). Perceived Vertical and Lateropulsion: Clinical Syndromes, Localization, and Prognosis. Neurorehabilitation and Neural Repair, 14, 1–20.
2. Pedersen PM, Wandel A, Jorgensen HS, Nakayama H, Raaschou HO, Olsen TS. Ipsilateral pushing in stroke: incidence, relation to neuropsychological symptoms, and impact on rehabilitation. The Copenhagen Stroke Study. Arch Phys Med Rehabil. 1996;77(1):25–28.
3. Gress-Heister M, Letzel C, Heister E. Pusher-syndrome—hypotheses on the pathogenesis, diagnosis and therapeutical approaches. Neurol Rehabil. 2000;6(2):59–67.
4. Karnath HO, Broetz D. Understanding and treating “pusher syndrome.” Phys Ther. 2003;83(12):1119–1125.
5. Davies P. Steps to Follow: A Guide to the Treatment of Adult Hemiplegia. New York, NY: Springer-Verlag; 1985.
6. Pérennou DA, Mazibrada G, Chauvineau V, et al. Lateropulsion, pushing and verticality perception in hemisphere stroke: a causal relationship? Brain. 2008;131(9):2401–2413.
7. Danells CJ, Black SE, Gladstone DJ, McIlroy WE. Poststroke “pushing”: natural history and relationship to motor and functional recovery. Stroke J Cereb Circ. 2004;35(12):2873–2878.
8. Lafosse C, Kerckhofs E, Troch M, et al. Contraversive pushing and inattention of the contralesional hemispace. J Clin Exp Neuropsychol. 2005;27(4):460–484.
9. Premoselli S, Cesana L, Cerri C. Pusher syndrome in stroke: clinical, neuropsychological and neurophysiological investigation. Europa Medicophysica. 2001;37(3):143–151.
10. Karnath HO, Johannsen L, Broetz D, Ferber S, Dichgans J. Prognosis of contraversive pushing. J Neurol. 2002;249(9):1250–1253.
11. Santos-Pontelli TEG, Pontes-Neto OM, de Araujo DB, Santos AC, Leite JP. Persistent pusher behavior after a stroke. Clinics. 2011;66(12):2169–2171.
12. Babyar SR, Peterson MG, Reding M. Time to recovery from lateropulsion dependent on key stroke deficits: a retrospective analysis. Neurorehabil Neural Repair. 2015;29(3):207–213.
13. Babyar SR, Peterson MG, Bohannon R, Perennou D, Reding M. Clinical examination tools for lateropulsion or pusher syndrome following stroke: a systematic review of the literature. Clin Rehabil. 2009;23(7):639–650.
14. Baccini M, Paci M, Rinaldi LA. The scale for contraversive pushing: a reliability and validity study. Neurorehabil Neural Repair. 2006;20(4):468–472.
15. Baccini M, Paci M, Nannetti L, Biricolti C, Rinaldi LA. Scale for contraversive pushing: cutoff scores for diagnosing “pusher behavior” and construct validity. Phys Ther. 2008;88(8):947–955.
16. Lagerqvist J, Skargren E. Pusher syndrome: reliability, validity and sensitivity to change of a classification instrument. Adv Physiother. 2006;8(4):154–160.
17. D'Aquila MA, Smith T, Organ D, Lichtman S, Reding M. Validation of a lateropulsion scale for patients recovering from stroke. Clin Rehabil. 2004;18(1):102–109.
18. Mokkink LB, Terwee CB, Patrick DL, et al. The COSMIN checklist for assessing the methodological quality of studies on measurement properties of health status measurement instruments: an international Delphi study. Qual Life Res. 2010;19(4):539–549.
19. Moher D, Shamseer L, Clarke M, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev. 2015;4:1.
20. Mokkink LB, Terwee CB, Gibbons E, et al. Inter-rater agreement and reliability of the COSMIN (COnsensus-based Standards for the selection of health status Measurement Instruments) checklist. BMC Med Res Methodol. 2010;10:82.
21. Mokkink LB, Terwee CB, Patrick DL, et al. The COSMIN study reached international consensus on taxonomy, terminology, and definitions of measurement properties for health-related patient-reported outcomes. J Clin Epidemiol. 2010;63(7):737–745.
22. Jewell DV. Guide to Evidence-Based Physical Therapist Practice. 3rd ed. Sudbury, MA: Jones & Bartlett Learning; 2015.
23. Terwee CB, Mokkink LB, Knol DL, Ostelo RW, Bouter LM, de Vet HC. Rating the methodological quality in systematic reviews of studies on measurement properties: a scoring system for the COSMIN checklist. Qual Life Res. 2012;21(4):651–657.
24. Mokkink LB, Terwee CB, Patrick DL, et al. COSMIN checklist manual. checklist manual v9.pdf. Published 2012.
25. Clark E, Hill KD, Punt TD. Responsiveness of 2 scales to evaluate lateropulsion or pusher syndrome recovery after stroke. Arch Phys Med Rehabil. 2012;93(1):149–155.
26. Hallin U, Blomsterwall E, Svantesson U. Clinical Assessment Scale for Contraversive Pushing, interrater reliability of a Swedish version. Adv Physiother. 2008;10(4):173–177.
27. Bergmann J, Krewer C, Riess K, Muller F, Koenig E, Jahn K. Inconsistent classification of pusher behaviour in stroke patients: a direct comparison of the Scale for Contraversive Pushing and the Burke Lateropulsion Scale. Clin Rehabil. 2014;28(7):696–703.
28. Karnath HO, Brotz D, Gotz A. Clinical symptoms, origin, and therapy of the “pusher syndrome”. Nervenarzt. 2001;72(2):86–92.
29. Karnath HO, Ferber S, Dichgans J. The origin of contraversive pushing: evidence for a second graviceptive system in humans. Neurology. 2000;55(9):1298–1304.
30. Krewer C, Riess K, Bergmann J, Muller F, Jahn K, Koenig E. Immediate effectiveness of single-session therapeutic interventions in pusher behaviour. Gait Posture. 2013;37(2):246–250.
31. Lindmark B, Hamrin E. Evaluation of functional capacity after stroke as a basis for active intervention. Presentation of a modified chart for motor capacity assessment and its reliability. Scand J Rehabil Med. 1988;20(3):103–109.
32. Karnath HO, Ferber S, Dichgans J. The neural representation of postural control in humans. Proc Natl Acad Sci U S A. 2000;97(25):13931–13936.
33. Karnath HO, Brötz D, Baccini M, Paci M, Rinaldi LA. Instructions for the Clinical Scale for Contraversive Pushing (SCP). Neurorehabil Neural Repair. 2007;21.4:370–371.
34. Bergmann J, Krewer C, Selge C, Muller F, Jahn K. The subjective postural vertical determined in patients with pusher behavior during standing. Top Stroke Rehabil. 2016;23(3):184–190.
35. Babyar SR, Peterson MGE, Reding M. Case-control study of impairments associated with recovery from “pusher syndrome” after stroke: logistic regression analyses. J Stroke Cerebrovasc Dis. 2017;26(1):25–33.
36. Paci M, Baccini M, Rinaldi LA. Pusher behaviour: a critical review of controversial issues. Disabil Rehabil. 2009;31(4):249–258.

contraversive lateropulsion; contraversive pushing; COSMIN; pusher behavior; pusher syndrome

Supplemental Digital Content

© 2017 Academy of Neurologic Physical Therapy, APTA.