Postoperative Cervical Deformity in 215 Thoracolumbar Patients With Adult Spinal Deformity

Study Design. Retrospective review of prospective multicenter database. Objective. Quantify the incidence of new onset cervical deformity (CD) after adult spinal deformity surgery of the thoracolumbar spine, identify predictors of development, and determine the impact on outcomes. Summary of Background Data. High prevalence of residual CD has been identified after surgical treatment of adult spinal deformity. Development of new onset CD is less understood and its clinical impact unclear. Methods. A total of 215 patients with complete 2-year follow-up and full-length radiographs met inclusion criteria. CD was defined by T1 slope minus Cervical Lordosis (CL) more than 20 degrees, C2-C7 sagittal vertical axis more than 40 mm, or C2-C7 kyphosis more than 10 degrees. Univariate analysis was performed using t tests or tests of proportion. Multivariate logistic regression was used to determine independent predictors of new onset CD. The impact of CD on health-related quality of life and satisfaction was measured using repeated measures mixed models or logistic regression as appropriate, accounting for potential confounders. Results. The overall rate of CD at 2 years after surgery was 63%. Univariate analysis revealed that patients who developed new onset CD postoperatively had higher incidence of diabetes (7.35% vs. 1.28%, P = 0.05), increased preoperative C2-C7 sagittal vertical axis (P = 0.04) and C2 slope (P = 0.038), and smaller diameter rods used at surgery (P = 0.032). Independent predictors of new onset CD at 2 years included: diabetes (odds ratio, 10.49; P = 0.046) and increased preoperative T1 slope minus cervical lordosis (odds ratio, 1.08/degrees; P = 0.022). Ending instrumentation below T4 was a negative predictor (odds ratio, 0.31; P = 0.019). Patients with and without CD experienced improvements in 2-year 36-Item Short Form Health Survey (P = 0.0001), Oswestry Disability Index (P = 0.0001), and Scoliosis Research Society (P = 0.0001). Rates and overall improvement were similar. CD was not associated with decreased satisfaction (P = 0.28). Conclusion. A total of 47.7% of patients without preoperative CD developed new onset postoperative CD after thoracolumbar surgery. Independent predictors of new onset CD at 2 years included diabetes, higher preoperative T1 slope minus cervical lordosis, and ending instrumentation above T4. Significant improvements in health-related quality of life scores occurred despite the development of postoperative CD.