Coronal Spinal Deformity Does Not Predict Clinical Response to Thoracic Spinal Cord Stimulation

Coronal spinal deformity is common among patients undergoing spinal cord stimulation (SCS) and may complicate lead placement and programming. Whether coronal alignment influences clinical response to SCS remains poorly characterized. We evaluated the association between thoracic and lumbar coronal deformity and one-year SCS outcomes.

We performed a retrospective cohort study of adults undergoing thoracic SCS implantation at a single center. Thoracic and lumbar Cobb angles were measured on perioperative radiographs. The primary outcome was one-year responder status, defined as ≥50% reduction in Numeric Rating Scale (NRS) pain score from baseline. Associations between Cobb angles and responder status were evaluated using univariate and multivariable logistic regression. The spatial relationship between curve apex and lead position were examined. Sensitivity analyses used alternative clinically anchored responder definitions on the basis of minimal clinically important difference thresholds and stricter NRS responder thresholds.

Among 118 patients with thoracic SCS systems, thoracic Cobb angles were available for 102 patients and lumbar Cobb angles for 113 patients. Thoracic and lumbar Cobb angles were not associated with responder status when modeled continuously (thoracic: OR 1.44 per 5°, 95% CI 0.86-2.43; lumbar: OR 1.04 per 5°, 95% CI 0.79-1.37) or when dichotomized at clinically relevant thresholds. The position of the curve apex relative to the lead did not modify the association between thoracic Cobb angle and response. In adjusted analyses (n = 96), neither thoracic nor lumbar Cobb angle independently predicted responder status. Exploratory sensitivity analyses suggested possible geometry-dependent effects for select responder definitions, but overall conclusions were unchanged.

Coronal deformity severity was not associated with short-term response to thoracic SCS in this cohort and should not be used to exclude otherwise appropriate candidates from SCS. Exploratory findings raise the possibility that lead or curve geometry may be relevant in select contexts, although any such effects appear modest in typical clinical populations.