Reliability of a new measure of H-reflex excitability

Objective: This study examined the intraclass reliability of different measures extracted from Hoffmann reflex (H-reflex) stimulus-response curve that are used to assess neuromuscular excitability. The following measures were compared: (1) the peak-to-peak amplitude of the H-reflex at a stimulus intensity associated with 5% of the maximum M-wave; (2) the slope of the regression line of the H-reflex stimulus-response curve; and (3) the peak of the first derivative of the H-reflex stimulus-response curve, a new measure introduced in this paper. Methods: The H-reflex was elicited in the soleus for 24 subjects (12 males and 12 females) on 5 separate days. Vibration was applied to the Achilles tendon prior to stimulation to test the sensitivity of the measures on test day 4. The stimulus intensity was gradually increased from below the threshold for an H-reflex response to above the maximum M-wave (Mmax) response. The means of 5 evoked potentials at each intensity level were used to create both the H-reflex and M-wave stimulus-response curves for each subject across test days. Determination of reliability involves the consideration of both the stability and consistency of the measures. A repeated measures analysis of variance evaluated the stability of the group means across test sessions. The consistency of scores within individuals was determined by calculating the intraclass correlation coefficient (ICC). Calculation of the 95% confidence interval of estimation was used to assess significant differences between ICCs. Results: The H-reflex measures were both stable and consistent across the first 3 test days. Achilles tendon vibration resulted in a profound reduction (59–70%) on test day 4, and then there was a return to baseline levels on test day 5. The ICC for H-reflex at a stimulus intensity associated with 5% of the maximum M-wave was 0.85. The ICC for the slope of the regression line of the H-reflex stimulus-response curve was 0.79, while it was 0.89 for the peak of the first derivative of the H-reflex stimulus-response curve. However, there was no statistical significance (P>0.05) between the 3 EMG measures of the H-reflex arc. Maximum M-wave amplitude had an ICC of 0.96 attesting to careful methodological controls. Conclusions: The peak of the first derivative of the H-reflex stimulus-response curve was shown to have comparable sensitivity and reliability as other more established measures. Significance: The first derivative of the H-reflex stimulus-response curve provides the rate of change, rather than amplitude, making it a robust measure of reflex arc excitability. The higher ICC for the first derivative offers greater statistical power, which is of practical significance.