Risk Factors for Proximal Junctional Kyphosis Following Surgical Deformity Correction in Pediatric Neuromuscular Scoliosis

Study Design. Single-center retrospective cohort analysis. Objective. The aim of this study was to evaluate risk factors associated with the development of proximal junctional kyphosis (PJK) in pediatric neuromuscular scoliosis (NMS). Summary of Background Data. PJK is a common cause of reoperation in adult deformity but has been less well reported in pediatric NMS. Methods. Sixty consecutive pediatric patients underwent spinal fusion for NMS with a minimum 2-year follow-up. PJK was defined as >10 degrees increase between the inferior end plate of the upper instrumented vertebra (UIV) and the superior end plate of the vertebra two segments above. Regression analyses as well as binary correlational models and Student t tests were employed for further statistical analysis assessing variables of primary and compensatory curve magnitudes, thoracic kyphosis, proximal kyphosis, lumbar lordosis, pelvic obliquity, shoulder imbalance, Risser classification, and sagittal profile. Results. The present cohort consisted of 29 boys and 31 girls with a mean age at surgery of 14 +/- 2.7 years. The most prevalent diagnoses were spinal cord injury (23%) and cerebral palsy (20%). Analysis reflected an overall radiographic PJK rate of 27% (n = 16) and a proximal junctional failure rate of 7% (n = 4). No significant association was identified with previously suggested risk factors such as extent of rostral fixation (P = 0.750), rod metal type (P = 0.776), laminar hooks (P = 0.654), implant density (P = 0.386), nonambulatory functional status (P = 0.254), or pelvic fixation (P = 0.746). Significant risk factors for development of PJK included perioperative use of halo gravity traction (38%, P = 0.029), greater postoperative C2 sagittal translation (P = 0.030), decreased proximal kyphosis preoperatively (P = 0.002), and loss of correction of primary curve magnitude at follow-up (P = 0.047). Increase in lumbar lordosis from post-op to last follow-up trended toward significance (P = 0.055). Conclusion. Twenty-seven percent of patients with NMS developed PJK, and 7% had revision surgery. Those treated with halo gravity traction or with greater postoperative C2 sagittal translation, loss of primary curve correction, and smaller preoperative proximal kyphosis had the greatest risk of developing PJK.