Résumé
Fatigue induced by neuromuscular electrical stimulation (NMES) is poorly understood. Although recent research gives evidence for the implication of different sites along the pathway of force production, the muscle factors susceptible to influence the response to electrically induced fatigue are still unknown. The purpose of the present study was to identify the time course of neuromuscular changes during NMES according to the muscle's capacity to generate force. Twelve healthy subjects with different force generating capacities [7 Strong (S) with mean torque during Maximal Voluntary Contraction (MVC) 88.8±1.6Nm and 5 Weak (W) with mean torque 64.4±3.2Nm], participated in an electrostimulation protocol for the triceps surae, composed of 3 series of 17 stimulation trains (4s ON - 6s OFF, pulse duration 450μs, frequency 30Hz, at maximal tolerated intensity). Neuromuscular tests were performed before, during and immediately after the protocol. Torque and EMG activity of the gastrocnemius medialis muscle were continuously recorded. Alterations in muscle's characteristics (excitability and contractile properties) were evaluated by analysis of the muscle compound action potential (M-wave) and twitch torque. Motoneuron excitability was assessed by the H reflex, expressed in absolute value and normalized to M-wave maximal amplitude (respectively Hmax and H/M). Changes in the central command were assessed by using the twitch interpolation technique and the root mean square (RMS and RMS/M) obtained during MVC. MVC significantly decreased from the first 17-train bout and throughout the protocol for both groups (from 88.8±1.6Nm vs 64.4±3.2Nm at pre to 78.8±3.3Nm vs 58.2±2.7 at post51, for S and W respectively), giving evidence of precocious neuromuscular impairments. Motoneuron excitability was not affected (no change of Hmax and H/M). Muscle contractile properties were significantly potentiated at post17 and for the rest of the protocol (42±14% for S vs 37±6% for W for Pt values at post51) for both groups. Muscle excitability was significantly altered only in S, as proved by the significant decrease in M-wave amplitude and muscle response to trains of stimulation (-2.9±3.7% and -13.4±5.9% for post51 respectively). Level of voluntary activation assessed by the twitch interpolation technique was lower for W and, although RMS/M was significantly decreased for both groups, the decrease in W was significantly more pronounced (-21.8±4.5% vs -14.5±6.2% for S). Neuromuscular fatigue can be attributed to both central and peripheral mechanisms for the S group, while for W it appears that mainly central mechanisms are involved. These observations should be taken under consideration when seeking to optimise training strategies for people with neurological disorders. The results show that improving muscle strength does not necessarily delay neuromuscular fatigue, but does change its nature.