Selective anterior fusion of thoracolumbar/lumbar curves in adolescents: When can the associated thoracic curve be left unfused?

Study Design. A retrospective multicenter study was conducted to investigate patients with a major thoracolumbar/lumbar adolescent idiopathic scoliosis and an associated minor thoracic curve treated with an anterior instrumentation and fusion of the lower curve. Objective. To establish criteria for determining when such curves can be successfully treated by an anterior only procedure of the lower curve with acceptable spinal balance and residual thoracic curve. Summary of Background Data. Anterior spinal instrumentation techniques have been proved effective for the management of isolated thoracolumbar/lumbar scoliosis with small compensatory thoracic curves. The success of a selective anterior fusion when the associated thoracic curve had some structural changes in a small preliminary study was variable and was the stimulus for this study. Methods. A multicenter study involved 49 adolescent patients with a major thoracolumbar/lumbar curve in which the associated minor thoracic curve measured between 30° and 55°. In all the patients, the apical vertebra of the lower curve lay outside the midsacral line, and the thoracic apical vertebra fell outside a line dropped from the center of C7. Multiple radiographic parameters were evaluated. The Risser sign, height, weight, onset of menses, and closure of the triradiate cartilages were studied to access the patients' maturity. All the patients were observed at least 2 years. Patients were considered to have a satisfactory result if the thoracic curve at the final follow-up assessment measured 40° or less, if balance and sagittal alignment were reasonable, and if additional procedures were not required. Results. At final follow-up assessment, two groups emerged. Group 1 (n = 43) had satisfactory results. The preoperative thoracic curve in this group averaged 40° and 26° after surgery. The lumbar curve averaged 56° before surgery and 22° after surgery. Group 2 (n = 6) had unsatisfactory results. The average thoracic curve was 49° before surgery 54° after surgery, whereas the lumbar curve averaged 59° before surgery and 27° after surgery. Three of these patients underwent posterior thoracic instrumentation and fusion. Conclusions. Statistical analysis showed that a successful surgical outcome was dependent on both the structural changes in the thoracic curve and the patient's maturity. The thoracolumbar/lumbar-thoracic (TL/L:T) Cobb ratio in combination with the degree of the thoracic curve on lateral bending was the best predictor among the structural indexes. Of 44 patients with a TL/L:T Cobb ratio of 1.25 or greater and/or a thoracic curve, which bent out to 20° or less, 42 had a satisfactory result. The best predictor among the maturity indexes was closure of the triradiate cartilages. Of 43 patients in whom the triradiate cartilages were closed, 42 had satisfactory results. When this data is combined, the outcome for the thoracic curve can be reasonably predicted.