Comparing motor unit number estimation techniques

Abstract

It is unclear how comparable motor unit number estimates (MUNEs) are when derived from a non-invasive technique involving repetitive peripheral nerve stimulation vs. one involving volitional contractions and intramuscular recordings of single motor units (MUs). Therefore, this study examined MUNEs from MScanFit (MScan) and Decomposition-Enhanced Spike-Triggered Averaging (DE-STA). Eighteen participants (8 females, 10 males; 29.7 ± 7.1 years) sat with their right leg positioned in an isometric myograph while surface electromyography (EMG) was recorded from the tibialis anterior (TA). The MScan protocol isolated and derived the size of single MUs by repeatedly stimulating the common fibular nerve at progressively weaker currents to model a compound muscle action potential (CMAP) stimulus–response curve. For DE-STA, a concentric needle electrode was inserted into the TA, and participants performed 30-s isometric dorsiflexion contractions at 25 % of maximal voluntary torque to obtain ≥20 individual surface MU potentials (S-MUPs; i.e., single MUs extracted from the surface EMG signal based on needle-detected spikes). Both techniques used the same maximal CMAP to calculate a MUNE, yet MScan used a mathematical model to simulate the recorded CMAP stimulus–response, which was compared to the recorded scan to minimize disagreement; whereas DE-STA compared the size of the maximal CMAP to the average S-MUP. There was no difference between the MUNE calculated via DE-STA (132 ± 26 MUs) and MScan (142 ± 22 MUs; p = 0.11), and the bias (10.0 MUs) and limits of agreement (67.6 vs −47.6 MU difference) suggests that either technique may independently offer a reasonable MU estimate for the TA of young adults.