Reliability of elbow stretch reflex assessment in chronic post-stroke hemiparesis

Objective To establish reliability of quantitative measures of elbow joint spastic hypertonia in post-stroke hemiparesis. Methods Nine subjects with post-stroke hemiparesis (mn duration: 42 months) were tested on three separate days. Biceps brachii and brachioradialis EMG were recorded during passive ramp-and-hold extensions applied at seven speeds between 30 and 210 °/s. EMG burst duration, onset position threshold, and burst intensity were used to evaluate reflex activity. Torque at 40° of elbow flexion was used as a mechanical indicator of spastic hypertonia. Results Across speeds ICCs were consistent, means ranged between 0.63 and 0.85. Thus, relative reliability was fair to excellent for all parameters. Absolute reliability, determined using standard error of measurement expressed as a percentage of the mean score (%SEM), improved at higher speeds (≥120 °/s). Conclusions These results establish reliability of reflex and mechanical measures of elbow spastic hypertonia post-stroke. The data demonstrate greater reflex detection at high speeds, indicating greater potential to document meaningful changes in these distinct aspects of spastic hypertonia following intervention. Significance Based on findings of this study, reliability was demonstrated using four parameters of reflex EMG and torque indicating measurement consistency across sessions. These observations motivate determination of requisite effect sizes for clinical trials that evaluate treatment outcome.