The Role of Cross-Links in Posterior Spinal Fusion for Cerebral Palsy-Related Scoliosis.

STUDY DESIGN Retrospective review of a multicenter, prospective database. OBJECTIVE Our aim was to compare complication rates and maintenance of radiographic correction at 2 years after posterior spinal fusion (PSF) with or without cross-links in patients with cerebral palsy (CP)-related scoliosis. SUMMARY OF BACKGROUND DATA Cross-links are frequently used in PSF to correct scoliosis in patients with CP because they are thought to increase the stiffness and torsional rigidity of the construct. METHODS We reviewed the records of patients with CP who underwent primary PSF with or without cross-links between August 2008 and April 2015. Inclusion criteria were minimum follow-up of 2 years, availability of complications data (implant failure, surgical site infection, revision), and pre- and postoperative measurements of the major curve (measured using the Cobb method). The 256 patients included in this analysis had a mean age of 14.1 ± 2.7 years. Ninety-four patients had cross-links (57% using one cross-link; 43% using two cross-links) and 162 patients did not have cross-links. P < 0.05 was considered statistically significant. RESULTS The two groups did not differ significantly with regard to sex, age at surgery, preoperative menarche status, Gross Motor Function Classification System level, major curve magnitude, pelvic obliquity, kyphosis, and lordosis angles. There were no significant differences between groups in the correction achieved or the maintenance of correction at 2 years for the major curve, pelvic obliquity, kyphosis, or lordosis (all P > 0.05). Complication rates were similar between the cross-link group (16%, N = 15) and the non-cross-link group (14%, N = 22). CONCLUSION At 2 years after PSF to treat CP-related scoliosis, patients had no significant differences in the degree of correction achieved, the maintenance of correction, or the rate of complications between those whose fusion constructs used cross-links and those whose constructs did not. LEVEL OF EVIDENCE 3.