Spinal cord injury (SCI) results in deleterious skeletal muscle adaptations, such as relevant atrophy and loss of force. In particular, the relevant loss of lower-limb force-generating capacity may limit functional mobility even if neuronal control was sufficient. Currently, methods of assessing maximal force-generating capacity using neuromuscular electrical stimulation (NMES) are limited in individuals who cannot tolerate higher stimulation amplitudes, such as those with residual sensation and those at risk of fracture. In this study, we examined the relationship between NMES amplitude and muscle torque exerted (recruitment curve) in order to determine whether maximal torque output can be characterized by a submaximal marker. Recruitment curves for knee extensors, knee flexors, and ankle plantarflexors were recorded from 30 individuals with motor complete SCI. NMES was delivered starting with an amplitude of 5 mA, and increasing by 5 mA for every subsequent stimulation until either the participant requested to stop the stimulation or the maximum stimulation amplitude (140 mA) was reached. Significant correlations between peak slope of the recruitment curve and peak torque for all muscle groups were found (knee extensors, r = 0.75; p < 0.0001; knee flexors, r = 0.68; p < 0.0001; ankle plantarflexors, r = 0.91; p < 0.0001), indicating that muscles that show greater peak slope of the recruitment curve tend to generate a greater peak torque. This suggests that peak slope, which was achieved at an average stimulation intensity (55.0 mA) that was 43% smaller than that corresponding to peak torque (97.4 mA), may be used as a submaximal marker for characterizing maximal torque output in individuals with SCI.
