Study design: Retrospective review.
Objective: The aim of this study was to determine whether major postoperative complications ("complications") are associated with 2-year improvements in Caregiver Priorities and Child Health Index of Life with Disabilities (CPCHILD) scores after scoliosis surgery, and whether complications and preoperative characteristics predict 2-year improvements in CPCHILD Total score.
Summary of background data: Spinal arthrodesis can halt the progression of spinal deformity in patients with cerebral palsy (CP)-related scoliosis. However, these patients are prone to postoperative complications.
Methods: Using a multicenter CP registry, we identified 222 patients aged ≤21 years who underwent spinal fusion from 2008 to 2015 and had ≥2-year follow-up. We compared CPCHILD score improvement between 71 patients who had 1 or more complications ("complications group") versus 151 who did not ("no-complications group"). Complications were deep infections, thromboembolic events, and cardiopulmonary, gastrointestinal, and neurologic complications. Multiple linear regression was used to identify predictors of 2-year postoperative CPCHILD score improvement (alpha = 0.05).
Results: At 2-year follow-up, the complications group had similar mean improvement in CPCHILD score across all domains compared with the no-complications group (P > 0.05). When stratifying by complication type, deep infection was associated with less improvement in CPCHILD Comfort and Emotions (P = 0.02), Quality of Life (P < 0.01), and Total (P = 0.04) scores. When controlling for Gross Motor Function Classification System subcategory, age, and body mass index, only preoperative CPCHILD Total score and postoperative deep infection (F[4, 176] = 14; P < 0.0001; R = 0.24) predicted 2-year improvement in CPCHILD Total score. Higher preoperative Total score and postoperative deep infection independently predicted less improvement in Total score.
Conclusion: Postoperative deep infection and higher preoperative CPCHILD Total score independently predicted less improvement in CPCHILD Total score. Other major postoperative complications were not associated with differences in 2-year postoperative improvements in CPCHILD scores across all domains.
Level of evidence: 3.