Discrimination of independent walking by spatiotemporal gait parameter variability in patients with hemiparesis following stroke.

Abstract

Gait variability in spatiotemporal parameters, a measure related to gait stability, is associated with fall risk in patients with stroke. Assessing gait variability may also be critical for independent walking in patients with hemiparesis following stroke. However, it remains unclear whether gait variability can discriminate between independent walking. This study aimed to evaluate the efficacy of gait variability indicators in discriminating independent walking in patients with hemiparesis following stroke. This cross-sectional study included 135 patients with hemiparesis following stroke (94 training and 41 validation samples). Gait variability indicators, calculated as the coefficient of variation (CV), included stride time, step time, single support time, and step length. Independent walking was assessed using the Functional Ambulation Category (FAC) and classified into independent and non-independent groups. Discriminant performance was evaluated using the area under the receiver operating characteristic curve (AUROC) and cutoff values, analyzed separately for the training and validation samples. The AUROC values for gait variability indicators were above 0.8, indicating moderate discrimination accuracy. The cutoff values were 4.32 % for stride time CV, 6.08 % for step time CV, 7.82 % for single support time CV, and 9.97 % for step length CV. The training and validation samples showed equivalent discrimination performance. Gait variability indicators demonstrated moderate accuracy in discriminating independent walking in patients with hemiparesis following stroke. These findings highlight the effectiveness of gait stability assessment for discriminating walking independence in this population.