Upper-extremity H-reflex measurement post-stroke: Reliability and inter-limb differences

ObjectiveTo establish test–retest reliability of flexor carpi radialis (FCR) H-reflexes in non-disabled and stroke participants. We also investigated inter-limb differences and effects of chronicity post-stroke compared to non-disabled group and age-related effects in the non-disabled group.MethodsSixteen chronic stroke and twenty-two non-disabled participants were recruited. Bilateral FCR H-reflexes were tested on two separate days by stimulating the median nerve and recording surface electromyography responses. FCR recruitment curves were plotted for H-reflex (H) and motor (M) waves and normalized as a percentage of maximal M-wave (ordinate) and motor threshold (abscissa).StatisticsIntraclass correlation coefficients [two-way mixed model-ICC (1, 2)], one-way ANOVA, Bland–Altman plots, standard error of measurement (SEM), and smallest real difference (SRD).ResultsICCs ranged from 0.55 to 0.95 (stroke) and 0.69–0.88 (non-disabled). SEM% (% of the mean) ranged from 9% to 24% (stroke) and 18–38% (non-disabled); SRD% ranged from 18% to 66% (stroke) and 6% to 50% (non-disabled). H-reflex amplitude and slope were greater in the paretic vs. non-paretic arm post-stroke (p = 0.07 and 0.01, respectively) and the paretic arm vs. non-disabled participants (p = 0.007 and 0.002, respectively). Stroke participants with longer chronicity (mean 9.4 years) revealed a significantly greater Hslp/Mslp on the paretic side compared to shorter chronicity (2.5 years; p = 0.05). Mean Hslp/Mslp was significantly greater in the young (mean 29 years) compared to the older group (62 years; p = 0.045).ConclusionsThese results establish reliability of FCR H-reflexes in stroke and non-disabled participants. SEM and SRD measurements can be used to interpret recovery patterns and longitudinal effects of therapeutic interventions.SignificanceFCR H-reflex amplitude and slope can be reliably measured and used to investigate neurophysiological mechanisms of motor recovery post-stroke.

► Flexor carpi radialis (FCR) H-reflexes can be reliably evoked in non-disabled participants and in both paretic and non-paretic arms of participants post-stroke. ► FCR H-reflexes (recruitment slope, amplitude, and threshold) are a sensitive measure of changes in spinal excitability post-stroke. ► FCR H-reflex recruitment slope and amplitude were significantly greater in the paretic arm compared to the non-paretic arm post-stroke.