Ultrasound estimated connective tissue proportion is lower after 6 months of restorative neurostimulation therapy: an exploratory analysis

Abstract Purpose To assess whether ultrasound-estimated connective tissue proportion in the human lumbar multifidus muscle is changed by restorative neurostimulation, and whether changes are associated with clinical outcomes. Methods Fifty individuals with mechanical chronic low back pain (cLBP) who were scheduled to receive a neurostimulation device to evoke contraction of the multifidus muscle via stimulation of the medial branch of the dorsal ramus at L2 were invited to participate. Connective tissue features of the multifidus and muscle cross-sectional area were extracted from transverse B-mode ultrasound images made bilaterally from L1 to S1. Measurements were made prior to implantation of the neurostimulator and at 6-months post-implantation. Self-reported back pain (numeric rating scale [NRS]), disability (Oswestry disability index [ODI]), and quality of life (EuroQol quality-of-life survey [EQ-5D-5L]) were recorded. Results Complete data were available for 45 participants. Errors in data collection precluded analysis of the remaining 5 participants. The relative proportion of connective tissue within the multifidus was lower by 6.0(13.5)% (unadjusted p = 0.003; Holm-adjusted p = 0.020) at L3 when the epimysium was excluded, and this remained significant following adjustment for cross-sectional area. There was no change at other lumbar levels. Improvements in back pain, disability, and quality-of-life were not linearly correlated with the amplitude of changes in muscle structure. Conclusion The ultrasound-estimated proportion of connective tissue within the multifidus muscle at L3 was lower after 6 months of restorative neurostimulation. This is the location of the largest bulk of multifidus muscle that is innervated by the L2 nerve root stimulated by the device. The absence of linear association between the amplitude of changes in connective tissue proportion and clinical outcomes is not unexpected and larger studies will be required to investigate more complex relationships.