Purpose: Determine the prevalence of complications in neuromuscular scoliosis surgery and to identify risk factors. We hypothesized that patients with smaller pre-operative curve magnitudes would have lower complication rates.

Method: Our Pediatric Orthopaedic Spine Database identified a cohort of 151 consecutive patients with neu-romuscular scoliosis who underwent corrective surgery between 1992 and 2005 and had a minimum of 2 years of follow-up. Twenty-two patients (15%) were excluded; 20 patients with a diagnosis of myelodysplasia and two due to death during the follow-up period. Preoperative, operative, and postoperative factors were analyzed for any association with major complications and length of stay (LOS) utilizing stepwise logistic and multiple regression. Only factors with p-values < 0.05 remained in the analysis models. Odds ratios were calculated for significant dichotomous variables and receiver operator characteristic (ROC) curves were created for significant continuous variables.

Results: There were 79 male and 50 female patients with a mean age at surgery of 13.4 years (range, 6 to 21 years). Eighty-seven patients (68%) underwent posterior spinal fusion (PSF) with segmental spinal instrumentation (SSI). and 42 patients (32%) underwent anterior spinal fusion (ASF) and PSF with SSI. Mean follow-up was 46.4 months (range, 24 to 251 months). There were 45 major complications in 37 patients (29% prevalence). Non-ambulatory status (p < 0.05) and a high preoperative Cobb angle (p < 0.01) were associated with an increasing prevalence of major complications. Non-ambulatory patients were almost four times more likely to have a major complication (OR of 3.8, p < 0.05) in comparison to ambulatory patients. A pre-operative Cobb angle ≥ 60 degrees (p < 0.01) was the most accurate predictor for an increased risk for major complication. Patients undergoing PSF with SSI only or combined ASF and PSF with SSI on the same day who sustained one major (p < 0.05) or two minor (p < 0.01) complications had a significantly increased LOS.

Conclusion: The prevalence of complications following surgery for neuromuscular scoliosis is high. Non-ambulatory status and higher preoperative Cobb curve magnitude are directly associated with an increased risk for major complication and indirectly associated with increased LOS.