Comparison of Growing Rod Instrumentation Versus Serial Cast Treatment for Early-Onset Scoliosis.

Abstract

A comparison of 2 methods of early-onset scoliosis treatment using radiographic measures and complication rates. To determine whether a delaying tactic (serial casting) has comparable efficacy to a surgical method (insertion of growing rod instrumentation [GRI]) in the initial phase of early-onset deformity management. Serial casts are used in experienced centers to delay operative management of curves of surgical magnitude (greater than 50°) in children up to age 6 years. A total of 27 casted patients from 3 institutions were matched with 27 patients from a multicenter database according to age (within 6 months of each other), curve magnitude (within 10° of each other), and diagnosis. Outcomes were compared according to major curve magnitude, spine length (T1-S1), duration and number of treatment encounters, and complications. There was no difference in age (5.5 years) or initial curve magnitude (65°) between groups, which reflects the accuracy of the matching process. Six pairs of patients had neuromuscular diagnoses, 11 had idiopathic deformities, and 10 had syndromic scoliosis. Growing rod instrumentation patients had smaller curves (45.9° vs. 64.9°; p = .002) at follow-up, but there was no difference in absolute spine length (GRI = 32.0 cm; cast = 30.6 cm; p = .26), even though GRI patients had been under treatment for a longer duration (4.5 vs. 2.4 years; p < .0001) and had undergone a mean of 5.5 lengthenings compared with 4.0 casts. Growing rod instrumentation patients had a 44% complication rate, compared with 1 cast complication. Of 27 casted patients, 15 eventually had operative treatment after a mean delay of 1.7 years after casting. Cast treatment is a valuable delaying tactic for younger children with early-onset scoliosis. Spine deformity is adequately controlled, spine length is not compromised, and surgical complications associated with early GRI treatment are avoided.