The M waves of the biceps brachii have a stationary (shoulder-like) component in the first phase: implications and recommendations for M-wave analysis.

We recently documented that compound muscle action potentials (M waves) recorded over the 'pennate' vastus lateralis showed a sharp deflection (named as a shoulder) in the first phase. Here, we investigated whether such a shoulder was also present in M waves evoked in a muscle with different architecture, such as the biceps brachii, with the purpose of elucidating the electrical origin of such afeature. M waves evoked by maximal single shocks to the brachial plexus were recorded in monopolar and bipolar configurations from 72 individuals using large (10 mm diameter) electrodes and from eight individuals using small (1 mm diameter) electrodes arranged in a linear array. The changes in M-wave features at different locations along the muscle fiber direction were examined. The shoulder was recognizable in most (87%) monopolar M waves, whereas it was rarely observed (6%) in bipolar derivations. Recordings made along the fiber direction showed that the shoulder was a stationary (non-propagating) feature, with short duration (spiky), which had positive polarity at all locations along the fibers. The latency of the shoulder (9.5 ± 0.5 ms) was significantly shorter than the estimated time taken for the action potentials to reach the biceps tendon (12.8 ms). The shoulder must be generated by a dipole source, i.e. a source created at a fixed anatomical position, although the exact origin of this dipole is uncertain. Our results suggest that the shoulder may not be due to the end-of-fiber signals formed at the biceps brachii tendon. The shoulder is not related to any specific arrangement of muscle fibers, as it has been observed in both pennate and fusiform muscles. Being a stationary (non-propagating) component, the shoulder is not reliable for studying changes in sarcolemmal excitability, and thus should be excluded from the M-wave analysis.