Comparing the reliability of voluntary and evoked muscle actions

Summary: Purpose: The purpose of this exploratory study was to quantify the test-retest reliability, intertrial variability and correlations between variables calculated during voluntary and evoked muscle actions. Methods: During three separate trials of isometric leg extension muscle actions with 14 men [mean age (±SD) = 21·9 (±3·2) years; height = 179·2 (±8·0) cm; mass = 77·7 (±10·9) kg], peak torque (PT_V), time to peak torque (TPT_V), rate of torque development (RTD_V), time to peak rate of torque development (TRTD_V), electromechanical delay (EMD_V) and EMG amplitude (EMG_RMS) were quantified for voluntary muscle actions. Peak twitch torque (PT_E), time to peak twitch torque (TPT_E), rate of torque development (RTD_E), time to peak rate of torque development (TRTD_E), electromechanical delay (EMD_E), peak-to-peak M-wave (M_p-p) and M-wave area (M_area) were calculated for evoked twitches. All electromyographic measurements were recorded over the vastus lateralis. Results: Voluntary intraclass correlation coefficients (ICC) were 0·84-0·96, except for TRTD_V (0·30) and EMD_V (0·74). Evoked ICC were 0·80-0·95, except for EMD_E (0·52). Relative standard errors of measurement for TPT and TRTD were greater for voluntary than evoked, whereas the value for PT_E was greater than PT_V. There were low to moderate correlations (r = -0·41-0·51) between common voluntary and evoked variables. Conclusions: Evoked variables showed lower intertrial variability than voluntary. Overall, voluntary and evoked muscle actions provide unique, complimentary information regarding neuromuscular function that cannot be used interchangeably.