Is there a need for anterior release for 70-90° thoracic curves in adolescent scoliosis?

Large and stiff thoracic scoliotic curves in the adolescent represent a classic indication of anterior release followed by posterior instrumentation. However, third-generation segmental spinal instrumentations have shown increased correction of thoracic curves. Indication for an anterior release may therefore not be required even in large and stiff thoracic curves. The objective of the study was, therefore, to analyze retrospectively the results of third-generation segmental posterior instrumentation in large and stiff thoracic curves and to compare our results with the current literature of anterior release followed by posterior instrumentation. An independent observer, who had not participated in any of the case, reviewed our electronic database of adolescent scoliosis surgery (Scolisoft) with the following query: thoracic curves, Cobb angle between 70degrees and 90degrees and posterior surgery only. He was able to identify 19 patients whose thoracic curves were measured between 70degrees and 90degrees. Out of these, four had convex-side bending Cobb angle values of less than 45degrees and were not included in the study, as they were judged too flexible. Fifteen patients (aged 11-18 years, mean 13.6 years) with thoracic scoliosis were left for the study (average Cobb angles 78.5degrees with a flexibility index of 32.5% (range, 19-42%). The mean follow-up period was 32 months (range 18-64 months). Classic parameters of deformity correction were analysed. The average operative time was 314 min and the mean total blood loss was 1,875 ml. Average level of instrumented vertebrae was 12 (Range, 10-14). Postoperatively, the thoracic Cobb angle was measured at 34.8degrees (range, 25-45degrees), which represents a correction rate of 54% (range, 40.0-67.1%) and remained unchanged at the last follow-up (35degrees). Patients with thoracic hypokyphosis improved from an average 11degrees to 18degrees. There were three complications (one excessive bleeding, one early infection and one late infection). One case showed an add-on phenomenon at the last follow-up. Coronal balance was improved from 1.8 cm (Range 0-4cm) down to 0.75 cm (range 0-2.5 cm). Shoulder balance was improved from 1.3 cm (range 0-4cm) down to 0.75 (0-2.5 cm). All patients reported satisfactory results except the patient with an adding-on phenomena. In the literature, most of the results of anterior thoracoscopic release and posterior surgery give a percentage of Cobb angle correction similar or inferior to our series for an average initial Cobb angle of less magnitude. Therefore, with adequate posterior release, and the use.of third-generation segmental instrumentation there is no need for anterior release even for curves in the 70 -90degrees range.