A narrative review on the utility of paraspinal electromyography for evaluation of the effects of exoskeletons on spine load

Most studies evaluating the effect of trunk support exoskeletons on low back loading, report as outcomes either electromyography amplitudes (EMG) only, or EMG and kinematics. This narrative review aims to clarify to what extent effects of an exoskeleton on low back load can be misinterpreted when relying on EMG without correction for muscle length, contraction velocity, and abdominal muscle activation. For muscle length, we pooled model and experimental data from the literature to show how the magnitude of passive moments around the lumbar spine depends on lumbar flexion, and how this affects paraspinal EMG changes versus spine load changes when wearing a trunk support exoskeleton. Additionally, we consider the confounding effect of exoskeleton-induced changes in lifting speed and abdominal muscle activation. We conclude that in the absence of correction based on kinematics, exoskeleton-induced changes in paraspinal EMG are not informative on spine load changes. Subsequently, we show how the data summarized in this review, can be used to translate EMG changes into a crude estimate of low back load changes if external moments and lumbar flexion are known.