Firing Characteristics of Motor Units during the First 90 Seconds of a Fatiguing Low-Level Contraction

Abstract

The mechanisms responsible for the changes in surface electromyographic (EMG) signal amplitude and frequency content during prolonged fatiguing low-level contractions are not well understood. In addition, the normal behaviour of motor units (MU) during these contractions also remains elusive. PURPOSE: This study was designed (i) to determine decomposition-based quantitative electromyography (DQEMG), could detect changes in motor unit firing statistics that accompany fatigue in the brachioradialis (BR) and extensor carpi radialis (ECR) muscles over the first 90 seconds of sustained, low-level isometric contractions, (ii) to determine if time-dependent changes in surface EMG parameters are correlated with changes in MU firing rates, and (iii) to determine if time-dependant changes seen in the MU firing characteristics are repeatable within individuals. METHODS: Changes in mean MU firing rates were investigated in the BR (n=1 0) and ECR (n=10) muscles in healthy volunteers (mean age = 28.6 yr, SD ± 3.9) during constant-torque isometric contractions held until exhaustion. Three BR contractions were sustained at 25% of the maximum voluntary contraction (MVC) and two ECR contractions were sustained at 20% MVC. Mean MU firing rates were estimated following the decomposition of 30-second EMG data windows. Time dependant changes in firing rates were investigated using an analysis of variance with time (first, second and third 30 second epochs) included as a main effect. Pearson product-moment correlation coefficients were computed to determine if changes in MU firing rates were related to changes in surface EMG amplitude and frequency parameters (α=0.05). Intra-class correlation coefficients (ICC) were used to determine the within-subject repeatability of the MU firing rate changes, and SEMG parameters. RESULTS: Mean MU firing rate decreased within the first 90 seconds in both muscle groups, and was repeatable across trials using the DQEMG method. Significant correlations between changes in mean firing rates and surface EMG amplitude and frequency content were found in the ECR muscle but not the BR muscle. The reliability of firing rate changes within subjects was moderate in both the ECR (ICC=0.72) and BR muscles (ICC=0.59). The SEMG mean power frequency significantly decreased over contraction time in the ECR muscle, and was found to be a highly reliable measure in both muscles (ICC=0.99 and 0.98, respectively). CONCLUSIONS: Changes in MU characteristics were observed as early as 90 seconds into a low level contraction. Differences found between the two muscles may be due to contraction level or fiber type distributions.