Study design: A retrospective study of 61 patients with cerebral palsy (CP) and neuromuscular scoliosis treated by either a combined anterior-posterior spinal arthrodesis or a posterior-only arthrodesis with the unit rod.
Objectives: Compare coronal and sagittal plane radiographic outcomes in patients undergoing either a combined anterior-posterior spinal fusion (PSF) or a posterior-only fusion with the unit rod for neuromuscular scoliosis in patients with CP.
Summary of background data: Although an anterior release before posterior spinal arthrodesis is commonly done for larger and stiffer neuromuscular curves, it is unclear whether or not an all-posterior construct produces similar correction in pelvic obliquity as that seen with an anterior-posterior spinal fusion.
Methods: Sixty-one consecutive children with CP and scoliosis were treated at a single institution between 1991 and 2003 with PSF using the unit rod with an anterior release (group A: 19 patients; average = 14.4 years) or without an anterior release (group B: 42 patients; average = 13.7 years). Side-bending, AP, and lateral radiographs were used to assess various sagittal and coronal plane parameters at baseline, after surgery, and at 2 years. RESULTS.: Before surgery, group A had larger major curves (91 degrees A vs. 72 degrees B; P = 0.001), less flexible major curves (21% A vs. 40% B; P = 0.01), with greater pelvic obliquity (26 degrees A vs. 19 degrees B, P = 0.02) than group B. In the subset of patients with a more severe preoperative pelvic obliquity (>20 degrees ), percent correction in pelvic obliquity was equivalent between groups A (71%) and B (74%, P = 0.91). With respect to coronal and sagittal plane radiographic outcomes, there were no significant group differences in major curve correction (58% A vs. 60% B), but group A trended toward greater % correction from preop bending films. At most recent follow-up, there were no differences with respect to loss of curve correction (7.6 A vs. 8.1 degrees B, P = 0.80). The rate of major complications was 26% for both groups, but group A patients had significantly longer operative times.
Conclusion: We demonstrate that excellent correction in severe pelvic obliquity can be achieved in smaller, more flexible curves using an all-posterior PSF, and in larger, less flexible curves using an anterior release with PSF.