2011 Annual Conference

Seale, Jill

To investigate the relationships between gait symmetry (temporal and spatial) and overall walking function (Classification of Walking Handicap), voluntary movement and basic mobility (STREAM), and gait velocity.

Thirty participants at least 1 year post stroke, ranging in age from 29-80 were recruited from the community. Participants were all community dwellers, although many relied on a caregiver for some amount of assistance in mobility and/or activities of daily living. All were able to ambulate at least 10 meters without assistance.

Following informed consent, the participant completed the Stroke Rehabilitation Assessment of Movement (STREAM). Participants then completed 3 walks on the GAITRite^®, walking at their comfortable pace. Based on average velocity, participants' overall walking function was classified using the Classification of Walking Handicap (CWH).

Descriptive statistics and distributions of data for all variables were examined. The data from trials two and three were averaged (trial one was considered a practice trial) to obtain average velocity, as well as average step length, stance time, and swing time for each limb. Spearman’s correlation coefficients were used to determine the level of association for all comparisons. Correlational analysis of gait symmetry (step length, stance time, and swing time ratios) and the STREAM scores, walking classification, and gait velocity were performed. Participants were categorized according to level of asymmetry (normal, mild, and severe), and secondary analyses were conducted to determine any differences in STREAM scores between groups (specifically the LE voluntary movement subscale of the STREAM) using Kruskal-Wallis followed by Mann-Whitney tests with Bonferroni adjustments.

Swing time asymmetry was the most prevalent asymmetry, found in 73.33% of the sample. Stance time had the smallest degree of asymmetry (53.33% of sample had normal stance time asymmetry) while swing time had the greatest (only 26.67% of sample had normal swing time asymmetry). Step length asymmetry significantly, inversely correlated with the CHW, all STREAM measures, and velocity (r_s=0.456-0.708). Likewise, swing time symmetry significantly, inversely correlated with all measures except STREAM basic mobility (r_s=0.427-0.474).Stance time symmetry demonstrated the least association with the dependent variables, correlating significantly with the STREAM lower extremity subscale only (r_s=0.440).Secondary analyses were completed to determine if there were significant differences in STREAM LE scores between the categories of asymmetry. For stance asymmetry, all subjects were either normal or mildly asymmetric. Therefore, a Mann-Whitney Test was performed and revealed significant differences between STREAM LE scores of the two categories (p=0.039). In the case of swing time and stance time asymmetry, subjects were normal, mildly asymmetric, or severely asymmetric. Kruskal-Wallis tests were performed to determine significant difference, followed by post-hoc analyses of Mann-Whitney tests with Bonferroni adjustments of alpha to p=0.017. For swing time asymmetry, the Krusal-Wallis revealed a significant difference (p=0.018). Post-hoc analysis indicated no significant differences between normal and mild swing asymmetry or normal and severe swing asymmetry. The only significant difference in STREAM LE scores were found between the mild and severely asymmetric samples. With step length asymmetry, a significant difference was found between groups (p=0.022), with post-hoc comparisons revealing similar results as found in swing asymmetry significant difference. The only significant difference was between the mild and severe categories of step length asymmetry (p=0.003).

The results of this study indicate that gait asymmetry is common in persons with chronic stroke. This study included only those who could walk without assistance; therefore, independence with gait does not seem to preclude substantial gait asymmetries. Several gait parameters correlated inversely with the chosen measures of asymmetry. The strongest correlation was between CWH and step length asymmetry, indicating that the higher overall walking function is associated with more symmetrical step lengths. Moderate correlations were also demonstrated between step length symmetry and the STREAM Total and STREAM Basic Mobility subscale, but only a fair association was found between step length symmetry and STREAM LE subscale. The current study, and those previously done indicate that motor impairment/recovery is related to gait asymmetry, but the lack of consistent strong correlations signify that other factors are involved in asymmetry. In fact, there were significant differences in the STREAM LE subscale scores in all three of the symmetry measures. However, post-hoc analysis revealed significant differences between the mild and severely asymmetric categories only. These results are surprising, as one would assume there would be significant differences between the normal and severely asymmetric. Although this study found that velocity and symmetries are related, assessment and understanding of gait asymmetry should be considered an important facet of gait evaluation. This work suggests that understanding the presence, type, and degree of gait asymmetry can guide therapists in developing appropriate treatment plans to address all gait deficits.

Gait symmetry including swing time, stance time, and step length is only moderately related to return of lower extremity voluntary motor movement. Results suggest that other factors contribute to gait symmetry, in addition to motor recovery; potential factors include abnormal tone, loss of flexibility, impairments in sensation. Interventions which promote recovery of voluntary lower extremity movement may positively impact gait asymmetry post stroke.