Serial casting as a delay tactic in the treatment of moderate-to-severe early-onset scoliosis

BACKGROUND: Serial casting can cure mild infantile idiopathic scoliosis. Its use in delaying surgery in older children and those with larger curves or syndromes is poorly defined. METHODS: A review of a single center's experience with casting was performed. Patients were included if they had a syndromic, neuromuscular, or congenital scoliosis or were older than 2.5 years with an idiopathic scoliosis measuring >50 degrees. RESULTS: A retrospective review was performed on 29 patients meeting all inclusion criteria. Of these, 12 were idiopathic and 17 were nonidiopathic curves. Average age at first cast was 4.4±2.1 years, and 3.0±1.8 cast changes were performed over 1.4±1.1 years. Patients were transitioned to a brace and followed up for 5.5 years (range, 2.2 to 11.4 y). The main thoracic Cobb angle before casting was 68.8±12.3 degrees, which corrected to 39.1±16.4 degrees in a cast. Cobb angle after cast removal was 60.9±18.4 degrees, which increased to 76.3±24.0 degrees at final follow-up. T1-T12 height increased to 1.1±2.6 cm during the treatment period (P=0.05). There were 5 minor complications. Fifteen patients (51.7%) required surgical treatment for their scoliosis at most recent follow-up and an additional 7 patients (24.1%) were delayed until a definitive anterior/posterior spinal fusion could be performed. Surgery was delayed 39±25 months from the first cast. Growing rods were required in 8 patients (27.6%). The patients who ultimately underwent surgical intervention (SG) were more likely to have a larger postcasting residual main thoracic Cobb angle than those who did not require surgery [NS; 69.5±14.6 degrees (SG) vs. 51.6±17.9 degrees (NS), P=0.007] and had a greater progression of their curves after cast removal [20.9±13.5 degrees (SG) vs. 9.4±11.0 degrees (NS), P=0.02]. CONCLUSIONS: Serial casting is a viable alternative to surgical growth sparing techniques in moderate-to-severe early-onset scoliosis and may help delay eventual surgical intervention. Although a cure cannot be expected, an average of 39 months of delay was achieved in this patient cohort and 72.4% have avoided growing spine surgery. LEVEL OF EVIDENCE: Level IV, case series.