Relationship between gait profile score and clinical assessments of gait in post-stroke patients.

BACKGROUND Instrumental assessment of gait pattern in hemiparetic post-stroke patients can be difficult. The Gait Profile Score, together with Gait Variable Scores, was recently validated as a quality measure for gait analysis in children with cerebral palsy. However, use of the Gait Profile Score has not been compared with that of clinical scales in post-stroke patients. OBJECTIVE The aims of this study were: (i) to quantify functional limitations in a sample of post-stroke hemiparetic patients, using clinical functional scales and the Gait Profile Score; (ii) to determine the correlation between the Gait Profile Score and clinical outcome scales, in order to determine the relationship between balance and gait impairment. METHODS In this cross-sectional study, patients were assessed for gait functionality with the Berg Balance Scale, Trunk Impairment Scale, Functional Ambulation Category, Functional Independence Measure, and Ten-Meter Walk Test. Gait Profile Score was obtained using the 3D-Gait Analysis system. RESULTS A total of 33 post-stroke patients were included; 10 females, 23 males; 18 patients were affected on the left side; mean age 59 years; and mean time since stroke was 4.46 years. A total of 20 healthy controls, mean age 53.9 years, were also assessed. In the hemiparetic group the Gait Profile Score values for paretic and non-paretic lower limbs (10.07° ± 3.17° vs 9.26°± 3.35°, respectively) were statistically similar (p > 0.05). The Gait Profile Score showed a fair relationship with Functional Ambulation Category (p=0.017, r = -0.412), Trunk Impairment Scale (p = 0.011, r = -0.436) and Ten-Meter Walk Test (p = 0.009, r = 0.49) and good correlation with the Berg Balance Scale (p = 0.001, r = -0.561) in the hemiparetic group. The results of the 3 regression models were statistically significant. Model 1 showed that Functional Ambulation Category, Gait Profile Score and Functional Independence Measure had a statistically significant effect on the determination of the Berg Balance Scale, with an adjusted R2 of 0.72. In model 2, Functional Independence Measure was the only variable that had a statistically significant effect on Trunk Impairment Scale determination, with a model adjusted R2 of 0.46. Model 3 was in agreement with model 2, which showed that Trunk Impairment Scale was the only variable with a statistically significant effect on Functional Independence Measure determination. CONCLUSION Multivariate linear models show that the Gait Profile Score seems to be an independent linear predictor of balance performance in patients with chronic stroke, and Gait Variable Scores on the sagittal plane may help clinicians to investigate acquired compensatory strategies.