Spinal manipulation in acute low back pain patients is associated with reduced motor-unit firing variability without altering bilateral coordination: a single-motor-unit EMG study of the lumbar multifidus

Abstract Background The neurophysiology of high velocity, low amplitude spinal manipulation (HVLA-SM) is poorly understood. We have characterized motor unit (MU) discharge in acute low back pain (ALBP) patients after spinal manipulation and compared with pre-manipulation findings in the same subjects. Methods Fine-wire electrodes were implanted bilaterally into the deep lumbar multifidus muscle at the painful spinal level in 9 ALBP patients with pain of less than three weeks duration in March–April 2007. HVLA-SM was applied to the painful spinal level. EMG activity was recorded during manipulation and during spontaneous and voluntary force production while standing and sitting, and individual MUs were identified and their discharge characteristics described. All subjects were compared to their own data recorded before HVLA-SM for effects on gross EMG activity duration, MU discharge rate, interspike interval variability quantified as inter-quartile range for differences between consecutive inter-spike intervals (∆ISI IQR), and common drive coefficient between concurrently active MUs. Results No effects on MU discharge rate were evident in the first seconds after HVLA-SM. ∆ISI IQR was reduced by 5.25 ms (95% confidence interval 0.68–9.82; P = 0.025) after spinal manipulation compared with pre-manipulation findings, while median MU discharge rate and total duration of gross activity remained unchanged. Notably, the common drive coefficient was lower for bilateral than unilateral MU pairs (− 0.149; 95% confidence interval − 0.087 to − 0.210; P < 0.0001). This is in agreement with findings in the same ALBP population before HVLA-SM and in contrast to previous findings during spontaneous standing in pain free subjects, where common drive to uni- and bilateral pairs was not significantly different. MUs showed signs of self-sustained firing, but rotation between units was not observed. Conclusions We have established a methodological framework for future investigations using spontaneous muscle activity rather than forced movements to understand the influence of spinal manipulation on the role of the nervous system in postural support. HVLA-SM was associated with reduced MU firing variability, suggesting a shift toward more stable motoneuron output in the deep lumbar multifidus. Longitudinal recovery data are needed to discriminate whether the reduced common drive to bilateral muscle pairs represents a pre-existing trait or a pain-driven state.