RESUMO
Background Stroke represents the largest cause of chronic disability resulting in muscle weakness and instability in the trunk muscles. Despite the reliable measures from isokinetic devices for upper/lower limb muscles, there is a lack of measures for trunk muscles in post-stroke hemiparesis. Objectives To investigate the reliability of the strength and endurance measures from an isokinetic dynamometer for able-bodied and post-stroke hemiparesis people. Methods The measures were taken from both groups (control/hemiparesis) performing antagonistic movements (flexion/extension) during different protocols (seated-compressed and semi-standing) to assess strength (60º/s) and endurance (120º/s). The intra-class correlation coefficient (ICC) and limits of agreement (LOA) defined the quality and magnitude of reliability on the measurements plotted with 95% confidence interval (95% CI) by Bland-Altman method. Results ICC ranged from 0.58 to 0.99, with few ICC values classified as moderate when repeated by raters during test and a single value during retest. The total work was the only variable to present LOA higher than the limits. Conclusion Results indicate acceptable reliability, in two different protocols, confirming the repeatability of the isokinetic measures for trunk muscles in able-bodied and post-stroke hemiparesis.