Lumbar multifidus muscle morphology on MRI after restorative neurostimulation for chronic low back pain: a single centre case series using the Goutallier classification.

study typeSingle Centre case series.

objectivesRestorative neurostimulation of the lumbar multifidus muscle is a novel therapy for chronic low back pain (CLBP). The effects of this therapy on the MRI appearances of the human multifidus are unknown. Here we describe changes in patients after restorative neurostimulation therapy.

methodsA single centre case series. Patients who had lumbar MRI scans before and after restorative neurostimulation therapy were identified. Demographic data, history of prior lumbar spinal surgery, time between MRI scans, duration of and clinical response to restorative neurostimulation, and indications for the MRI scan were retrieved. Using the index and post-stimulation T2 weighted axial MRI images the multifidus was graded using the Goutallier classification, bilaterally at the L3/4, L4/5, and L5/S1 disc space levels. The findings were compared to control CLBP patients.

resultsFifteen patients were identified (mean age 58 ± 12 yrs; 10 female and 5 male). The mean time between the two MRI scans was 31.8 months (95% CI 24.9–38.7 months) and the mean duration of restorative stimulation was 17.9 months (95% CI 13.1–22.7 months), range 13–39 months. Of the 90 muscle gradings 61 (68%; 95% CI 57–77%) were unchanged between the two scans; 24 (27%; 95% CI 18–37%) were worse by one or less points on the Goutallier Classification, and five (6%; 95%CI 2–12%) were better by one or less points. The pattern of these changes at the L3/4, L4/5, and L5/S1 levels was almost identical. Patients with prior lumbar spinal surgery (n = 7) had significantly worse baseline Goutallier scores than those (n = 8) not having had surgery (2.0 ± 0.9 v 1.6 ± 0.6; p = 0.038).There were no significant differences in changes in Goutallier scores between (i) clinical responders and non-responders to restorative neurostimulation; and (ii) those who did or did not have prior lumbar surgery. Neither duration of restorative neurostimulation nor time between index and second MRI significantly influenced changes in Goutallier scores. Findings in control CLBP patients (n = 15) were almost identical to those having restorative neurostimulation.

conclusionsUsing the Goutallier classification serial MRI changes in the multifidus muscle following restorative neurostimulation are like those seen in control CLBP patients. Further studies using additional methods of muscle analysis are needed to provide insights about mechanisms underlying the benefits of multifidus restorative neurostimulation therapy.