Assessment of Agreement Between a New Application to Compute the Wisconsin Gait Score and 3-Dimensional Gait Analysis, and Reliability of the Application in Stroke Patients

Abstract

Currently, there are no computerized tools enabling objective interpretation of observational gait assessment based on Wisconsin Gait Scale (WGS), which is a reliable and well-tested tool. The solution envisaged by us may provide a practical tool for assessing gait deviations in patients with hemiparesis after stroke. The present study assessed agreement between a new application software for computerized WGS and 3-dimensional gait analysis (3DGA), and reliability of the application. The study involved 33 individuals with hemiparesis after stroke. The software was developed based on a model designed taking into account components of the WGS and incorporating auxiliary lines passing through the relevant anthropometric points on the patient's body, as well as measurements of angular values, distances and duration of the specific gait phases, which make it possible to substantiate assessment based on this scale. Series of videos were made to record gait of the qualified patients. After the gait evaluation was carried out using the app, the data were retrieved from the software. The gait assessment was performed separately by three independent examiners who reviewed the video recording using the new app twice (two weeks apart). Additionally, 3DGA was carried out for all the subjects, and the results of the app-aided assessment were compared to those acquired using 3DGA. The findings show statistically significant correlations (p < 0.05) between majority of the WGS items measured using the new app, and the relevant spatiotemporal and kinematic parameters identified by 3DGA. Agreement between the scores reported by the three examiners was high in both measurements, as reflected by Cronbach's alpha exceeding 0.8. The findings reflect very good intra-observer reliability (as reflected by kappa coefficients from 0.847 to 1) and inter-observer reliability (as reflected by kappa coefficients from 0.634 to 1) of the new application software for computerized WGS. The opportunities offered by the observational gait scale objectified through our new software for computerized WGS result from the fact that the tool provides a useful low-cost and time-effective feedback to monitor ongoing treatments or formulate hypotheses.

Keywords: 3D gait analysis; application software; gait analysis; hemiparetic gait; observational gait analysis; stroke.

FIGURE 1

A new application for computerized gait analysis. Subject with right hemiparesis after stroke. Sample assessment of a selected gait parameter in WGS. Swing phase of the affected leg - external rotation during initial swing. On the freeze frame showing the frontal view, we draw two auxiliary straight lines and measure the angle between them. One line shows the direction of gait and runs in the middle of the distance between the medial malleoli, and the other line marks the long axis of the foot (in the middle of the distance between the heads of the first and the fifth metatarsal bones or between the calcaneal tuber and the second toe). The result is compared to the healthy leg. The examiner checks the angles measured and determines the rating (the same angles in both legs, reflecting a normal condition - 1 point; greater angle on the paretic side, but lower than 45 degrees - 2 points; angle exceeding 45 degrees on the paretic side - 3 points).