Spatiotemporal Gait Parameters That Predict Gait Function Based on Timed Up and Go Test Performance in the Hemiplegic Stroke Patients

The purpose of this study was to determine which spatiotemporal gait parameters obtained during hemiplegic walking could be a predictive factor for the Timed Up and Go test (TUG). Two hundreds nine subjects who had suffered a stroke were recruited for this study. They were participated in two assessments; the TUG test and gait analysis. The relationship between the TUG test and spatiotemporal parameters was analyzed using Pearson's correlation coefficients. In addition, to predict the spatiotemporal gait parameters that correlated most with the TUG scores, we used multiple linear regression analyses (stepwise method). The results show that the normalized velocity was strongly correlated with the TUG performance (r=-.72, p<.001). Additionally, single support percentage (SSP), double support percentage (DSP), step time difference (STD), and step length difference (SLD) significantly were correlated with the TUG test. Normalized velocity, STD, DSP of affected side, and SSP of non-affected side explained 53%, 8%, 3%, 2%, of variance in the TUG test respectively. In conclusion, an increase in gait velocity and a decrease in STD would be effective indicators of improvement on the functional mobility in the stroke rehabilitation.