Search

Introduction: Altered foot characteristics are common in people with stroke, with a third presenting with abnormal foot posture which is associated with ambulatory difficulties. Understanding the relationship between measures of foot and ankle impairment and their association with mobility and balance outcomes is therefore important; however, poor clinimetric properties of foot and ankle measures after stroke precludes evaluation of these relationships. Therefore, this research, undertaken as part of a multicentred research project, had the following aims: Study 1: To evaluate the clinimetric properties (feasibility, test-retest reliability, and clinical relevance) of measures of foot and ankle impairments, for application in people with stroke. Study 2: To examine how these measures differ between people with stroke and normal controls; and whether they are associated with mobility and balance outcomes. Methods: In Study 1, community-dwelling people with stroke, able to walk 10 m (metres), attended two testing sessions to evaluate the clinimetric properties of different foot and ankle measures. These included: static foot posture and dynamic foot loading (peak plantar pressure, PPP, contact area, CA and centre of pressure, CP) using a plantar pressure mat; isometric muscle strength using a hand-held dynamometer (HHD); peak ankle and hallux dorsiflexion and stiffness using bespoke rigs; and ankle plantarflexion spasticity using the Tardieu scale. Statistical analysis used intraclass correlation coefficients (ICCs₍₃,₁₎), standard error of measurement (SEM) and Bland-Altman plots. In Study 2, measures identified as reliable from Study 1 were incorporated in a cross-sectional study design. Participants were recruited from acute and community neurological services in East London and North Devon. Statistical analysis tested the differences between groups and between affected limbs in people with stroke. Impairment measures were evaluated using multivariate regression analysis for their association with functional outcomes: walking speed (over 10 m); Timed Up and Go (TUAG), Forward Functional Reach Test (FFRT) and presence of falls (> 1 in the last 3 months). Results: In Study 1, 21 people with stroke tested the measures. These were found to be feasible and easy to administer, although loss of data (up to 33%) was observed. All measures had moderate to excellent test-retest reliability (coefficients 0.50‒0.98), except ankle plantarflexion stiffness (ICCs₍₃,₁₎ = 0.00‒0.11). In Study 2, there were significant differences in all measures between people with stroke (n = 180) and controls (n = 46), apart from static foot posture (p = 0.670), toe deformity (p = 0.782) and peak hallux dorsiflexion (p = 0.320). Between limb differences were identified for all measures except foot posture (p = 0.489) and foot CA (p > 0.05). Multicollinearity analysis found 10 measures appropriate for multivariate regression which identified the following R² and variance explained: 59% walking speed (R² = 0.543); 49% TUAG (R² = 0.435); 36% FFRT (R² = 0.285) and 26% for Falls Presence. Conclusion: The study demonstrated that seven foot and ankle measures of impairment after stroke were clinically feasible, reliable and associated with mobility and balance outcomes. The measures were ankle and foot isometric muscle strength, sway velocity, PPP (RFT and FFT), CA (MFT and FFT) and peak ankle dorsiflexion. These measures can now be incorporated into research to examine methods to improve the treatment of foot and ankle after stroke.