Measurement and modeling of stimulus-evoked electromyography in lengthened and shortened muscles for spinal cord injured subjects during an electrically-elicited fatigue process

This study compares the amplitude and temporal features of stimulus-evoked electromyography (EMG) of paralyzed muscle, rectus femoris (RF), in both lengthened and shortened positions of six spinal cord injured (SCI) subjects during an electrically elicited fatigue process. The torque output and evoked EMG were fitted by hyperbolic tangent functions from which their amplitude residual levels and temporal inflection times can be extracted. Furthermore, a structural EMG model of Fuglevand et al (1992 Biol. Cybern. 67 143-53) was modified to include type I (slow twitch) and type II (fast twitch) of motor unit (MU) fibers with viable parameters obtained from paralyzed muscles to observe their amplitude and temporal changes. Our results showed that the amplitude of stimulus-evoked EMG decreased earlier in the lengthened muscle with a shorter inflection time (48.53 ± 8.7 s versus 55.13 ± 4.03 s) than that of the shortened position during 120 s of stimulation time (p < 0.05). Similarly, the peak-to-peak duration (PTPd) of the evoked EMG increased faster at an earlier time to a higher asymptotical value in lengthened muscle (2.23 ± 0.74 versus 1.77 ± 0.54), compared to that of a shortened one (p < 0.05). These observations coincided with the higher rising rate and larger final value of the temporal coefficients, i.e., longer duration, in both type I and II MUs of lengthened muscles. From the observation of all parameters, the fatigue process in lengthened muscle proceeds faster than that in shortened muscle.