Effect of innervation-zone distribution on estimates of average muscle-fiber conduction velocity

Abstract

Muscle-fiber conduction velocity (MFCV) may be a diagnostic variable, but the factors affecting its estimate from surface electromyographic (SEMG) signals are still not fully understood. We investigated the effect of innervation-zone distribution on estimates of average MFCV from simulated SEMG signals. Single-fiber action potentials were simulated using a computer model that describes an idealized cylindrical volume conductor comprised of bone, muscle, subcutaneous, and skin tissues. A model describing the characteristics and control of a motor-neuron pool was used to simulate 60 s of SEMG. Five innervation-zone distributions, inclined with respect to the fiber direction, were simulated within the muscle tissue, based on data from the literature and an experimental analysis. Two parameters were varied concurrently: (1) excitation level (5%-100%, 5% increments); and (2) subcutaneous tissue thickness (4 and 10 mm). The MFCV estimates were affected by the distribution of innervation zones with bias varying with excitation level and thickness of subcutaneous tissue. The MFCV estimates ranged from 3.95 to 11.67 m/s for single-differential and from 3.99 to 5.23 m/s for double-differential recordings when all simulated motor units were assigned the same conduction velocity of 4 m/s. Amplitude and characteristic spectral frequencies were also influenced by the distribution of innervation zones. The effect of innervation-zone distribution on MFCV estimates is thus substantial and may be a significant confounding factor in experimental and clinical studies on muscles with diffuse innervation zones.