Risk factors for predicting complications associated with growing rod surgery for early-onset scoliosis

Objective To identify risk factors for postoperative complications associated with growing rod surgery for early-onset scoliosis (EOS). Methods A total of 55 consecutive patients underwent growing rod surgery for EOS were examined from database. Data included age at initial surgeries, sex, diagnosis, body mass index (BMI), duration of follow-up, initial and final measure of major curve, T2–5, T5–12, T10–L2, and T12–S1 kyphosis angles, levels and type of instrumentation, total number of surgeries, number of rods inserted, number of lengthenings, lengthening intervals and rod location were studied. Risk factors for postoperative complications were analyzed using binomial multiple logistic regression analysis. Results Postoperative complications were associated with 37 of 272 procedures (14%) and affected 23 patients (42%). Complications included 25 implant-related failures (66%), 4 alignment complications (11%), 4 infections (11%), 1 neurological impairment (3%), 3 respiratory problems, 2 gastrointestinal problems, 1 urinary problem, and 1 dural tear. The most frequent implant-related failure was dislodged implant (76%) and 92% of the dislodgements occurred at the proximal foundation. Binomial multiple logistic regression analysis demonstrated that curve magnitude in last follow-up (OR: 1.042; P =0.036), duration between growing-rod lengthening procedures (OR: 1.121; P =0.003) and duration of follow-up (OR: 1.079; P =0.001) maintained its significance in predicting likelihood of postoperative complications. Conclusion The occurrence of postoperative complications in growing rod surgery for EOS is most likely multifactorial and is related to curve magnitude in last follow-up and duration between growing-rod lengthening procedures.