Background & Aims

Restorative neurostimulation for chronic mechanical low back pain relies on stimulation of the L2 medial branch of the dorsal ramus to elicit episodic tetanic contractions of the lumbar multifidus muscle in order to restore neuromuscular control of the lumbar spine. Quantification of metabolic activity, heterogeneity using PET uptake measurements and radiomic based texture analysis on 18F FDG-PET/CT provide a unique opportunity to visualize and quantify changes in the segmental and entire lumbo-sacral length of the multifidus and paraspinal muscles along with activity in brain pain matrices in response to restorative neurostimulation therapy.

This is first prospective study performing radiomic based CT and PET texture analysis and FDG uptake of multifidus muscle to quantify changes in muscle heterogeneity and uptake in patients with chronic back pain stimulated with Reactiv8 neurostimulation system.

Methods

This single center study (Clin Trials Gov NCT04327817 and Rec 20/LO/0740) recruited 8 patients to undergo three 18F FDG-PET/CT at baseline, 6 months and 1 year following multifidus stimulator implant. 18F-FDG-PET CT was acquired on GE-Discovery 710 PET system with a 128 slice CT (approx. 250 MBq i.v). Baseline scans were performed with patients at rest while the six-months scan were obtained immediately after a 30 minute multifidus stimulation session. Metabolic activity within the brain and multifidus was quantified using maximum standardized uptake value (SUVmax). Low dose CT texture analysis (CTTA) of the multifidus muscle was performed to examine changes in heterogeneity within the multifidus muscles (using entropy as a parameter reflecting irregularity at pixel resolution).

Results

Analysis of regions of interest in the superficial and deep multifidus showed a statistifically significant increase in PET-FDG uptake (SUVmax) at L3, L4 and L5 levels when comparing baseline uptake (n=8) vs uptake at 6 months post-implant (n=8). Analysis showed that there was also a statistically significant increase in SUVmax in the superficial multifidus when comparing baseline to 1 year post implant at the L3, L4 and L5 levels – in the case of the deep multifidus muscle there was a statistically significant increase in SUVmax when comparing baseline to 1 year post implant at the L3 and L4 levels.
CTTA (entropy) (p<0.001 and p=0.004) decreased between post-stimulation at 6 and 12 months with baseline scans within the deep multifidus muscle from L1 to L5 level. Increased metabolic activity and heterogeneity below the stimulated spinal level, suggests that there is extensive muscle activation in the lumbar region, consistent with the electrophysiological findings. (1)

Conclusions

This is the first human study elucidating the mechanism of action for multifidus stimulation using 18FDG-PET/CT uptake and texture analysis. Although the stimulation target is L2 Dorsal ramus, first time in humans a significant increase in PET-FDG uptake (SUVmax has been demonstrated in L3-L5 suggesting the muscle activation). We hypothesize that these changes may be attributable to specific aspects of the restorative mechanism.

References

1. Kang YM, Choi WS, Pickar JG. Electrophysiologic evidence for an intersegmental reflex pathway between lumbar paraspinal tissues. Spine 2002;27:56–63.

Presenting Author

Kavita Poply

Poster Authors

Gaurav Bajaj

MBBS, BSc

Barts Health NHS Trust

Lead Author

Kavita Poply

Barts Health NHS Trust

Lead Author

Athar Haroon

Lead Author

Theresa Wodehouse

Barts NHS Trust

Lead Author

Balaji Ganeshan

Lead Author

Raymond Endozo

Lead Author

Ben Goss

Lead Author

Vivek Mehta

Barts Health NHS Trust

Lead Author

Topics

  • Evidence, Clinical Trials, Systematic Review, Guidelines, and Implementation Science