Apical vertebral derotation in the posterior treatment of adolescent idiopathic scoliosis: myth or reality?

INTRODUCTION: Direct apical vertebral rotation represents an important goal of posterior surgery for thoracic adolescent idiopathic scoliosis (AIS), so as to obtain a better cosmetic effect and to avoid posterior thoracoplasty. However, the real effectiveness in correction of vertebral rotation, using posterior only procedures, is still open to debate. The aim of the present study is to compare the correction of axial apical rotation obtained with direct rotation procedure versus simple concave rod rotation, in patients treated by posterior fusion for thoracic AIS using pedicle screw-only construct.
MATERIALS AND METHODS: A retrospective review was performed on a total of 62 consecutive patients (one single institution, three different surgeons) affected by AIS, who had undergone a posterior spinal fusion with pedicle screw-only instrumentation between January 2005 and April 2008 at the reference center. All cases presented a main thoracic curve (Lenke type 1 and 2). The angle of rotation (RAsag) of the apical vertebra was measured from the preoperative and last follow-up axial CT. According to the derotation procedure, two groups were identified: a direct vertebral rotation group (DR group; n = 32 patients) and a simple concave rod rotation group (No-DR group; n = 30 patients). There were no statistical differences between the two groups, in terms of age, Risser's sign, curve patterns, Cobb main thoracic (MT) curve magnitude and flexibility, extension of fusion, offset measurements on the coronal plane and sagittal preoperative contour.
RESULTS: All 62 patients were reviewed at an average follow-up of 3.7 years (range 2.5-4.2 years). The DR group compared to the No-DR group showed a significantly better final correction of apical vertebral rotation (DR 63.4 % vs. No-DR 14.8 %; p < 0.05) and a greater final correction (61.3 vs. 52.4 %; p < 0.05) with better maintenance of the initial correction (-1.7° vs. -1.9°; ns) of the main thoracic curve. Concerning the coronal balance, there was the same aforementioned trend of better results in the DR group, with less final apical MT vertebra translation (DR 2.2 cm vs. No-DR 4.1 cm), greater overall change (preop-final) of lower instrumented vertebra (LIV) coronal tilt (-14.9° vs. -11.1°; p < 0.05); the final global coronal balance (C7-S1) resulted quite better in DR group, but without a significant difference. The T5-T12 kyphosis angle was quite similar in both group before surgery (DR 16.8° vs. No-DR 17.5°) and was little lower at final follow-up evaluation in direct vertebral rotation group (14.5° vs. 16.5°). The T10-L2 sagittal alignment angle was similar in each group before surgery (12.5° in DR vs. 11.8° in No-DR), and at the latest follow-up averaged 5.3° versus 8.2°, respectively. Lumbar lordosis was similar in each group before surgery (DR -42° vs. No-DR -44.1°) and at the final follow-up evaluation (-45.9° vs. -43.2°). At the latest follow-up, SRS-30 and SF-36 findings were similar between the two groups. The complication rate was higher in the simple concave rod rotation group (13.3 vs. 9.3 %), related in two cases to thoracoplasty, which was never utilized in direct rotation patients.
CONCLUSIONS: The direct vertebral rotation obtained significantly better final results, when compared to simple concave rod rotation, both concerning correction of apical vertebral rotation and magnitude of MT curve. On the other hand, the DR group presented a little reduction in T5-T12 kyphosis at follow-up, in comparison with concave rod rotation procedure. Both procedures were found to be satisfying from patients' perspective. Nevertheless overall complication rate was higher in the simple concave rod rotation group, related mainly to thoracoplasty (2 cases), which was never necessary in direct rotation patients.