Reliability of quantitative EMG analysis of the extensor carpi radialis muscle

This study investigated the within-subject, intra- and inter-operator reliability of quantitative electromyographic (EMG) analysis using decomposition-based quantitative electromyography (DQEMG). Needle and surface-detected EMG signals were collected during low-level isometric contractions of the extensor carpi radialis (ECR) muscle. DQEMG was used to extract needle-detected motor unit potential (MUP) trains and surface-detected MUPs (SMUPs) associated with each train. Two independent experienced operators re-decomposed and processed the MUP data on two separate occasions. One-way analyses of variance (ANOVA) were performed to identify within-subject differences (test-retest, n=6), and operator and trial differences (intra- and inter-operator, n=20) for the MUP morphological variables. The within-subject reliability, as well as the intra- and inter-operator reliability were estimated using intraclass correlation coefficients (ICCs). The 95% limits of agreement were calculated to measure within-subject and between operator agreements. MUP durations were found to be significantly different between days (p<0.05). For intra- and inter-operator reliability, a significant difference was found within and between the operators for MUP duration, and between the operators for MUP number of turns (p<0.05). SMUP morphological parameters yielded higher ICC values for both intra (0.96-0.99) and inter (0.96-0.99) operator scores when compared to MUP morphological parameters (intra-operator: 0.81-0.99; and inter-operator: 0.67-0.98). Mean motor unit (MU) firing rate was found to be a highly reliable measure for both intra- and inter-operators (0.99-0.97, respectively). The between-operator agreement was above 95% for all morphological parameters. These results concerning within-subject, intra- and inter-operator reliability and levels of agreement for quantitative motor unit analysis suggest that DQEMG provides sufficiently consistent results to allow it to be effectively used for QEMG analysis. Needle-detected MUP morphology although clinically useful in the diagnosis and monitoring of neuropathies, was not as reliable as surface MUP QEMG measures.