Risk Factors for Complications Associated With Growing-Rod Surgery for Early-Onset Scoliosis

Study Design. A retrospective multicenter study. Objective. To identify risk factors for postoperative complications associated with growing-rod (GR) surgery for early-onset scoliosis (EOS). Summary of Background Data. Results and complications of GR surgery for EOS have not been adequately studied. Methods. We evaluated clinical and radiographical results from 88 patients with EOS who underwent GR surgery in 12 spine centers in Japan. The mean age at the time of initial surgery was 6.5 +/- 2.2 years (range, 1.5-9.8 yr) and the mean follow-up period was 3.9 +/- 2.6 years (range, 2.0-12.0 yr). Risk factors for postoperative complications were analyzed using binomial multiple logistic regression analysis. We considered the potential factors of sex, age, number of rod-lengthening procedures, whether a pedicle screw foundation was used, the uppermost level of the proximal foundation and lowermost level of the distal foundation, Cobb angles of the proximal thoracic, main thoracic, and lumbar curves, and the kyphosis angles in the proximal, main thoracic, thoracolumbar, and lumbar spine. Kaplan-Meier analysis was used to determine the complication-free survival rate of GR surgery as a function of the number of surgical procedures. Results. Complications affected 50 of the patients (57%) and were associated with 119 of 538 surgical procedures, with 86 implant-related failures (72%), 19 infections (16%), 3 neurological impairments (3%), and 11 other complications. The most frequent implant-related failure was dislodged implant (71%) and 95% of the dislodgements occurred at the proximal foundation. Kaplan-Meier analysis demonstrated a linear decrease in complication-free rates as the number of rod-lengthening procedures increased. Binomial multiple logistic regression analysis found the following significant independent risk factors: 6 or more rod-lengthening procedures (odds ratio [OR], 6.534), an increase of every 20 degrees in the proximal thoracic Cobb angle (OR, 3.091), and an increase of every 25 degrees in the lumbar lordosis angle (OR, 2.607) in the preoperative condition. Conclusion. Increases in the upper thoracic scoliotic curve, thoracic kyphosis, and number of rod-lengthening procedures are positively associated with an increased risk of complications after GR surgery for EOS.