Efficacy and Safety of Direct Vertebral Rotation in the Surgical Correction of Scoliosis: A Meta-Analysis.

Abstract

To compare outcomes of scoliosis correction with direct vertebral rotation (DVR) and without DVR (no direct vertebral rotation [N-DVR]) through meta-analysis. MEDLINE and Embase databases were searched from the earliest available date of indexing through April 10, 2018, for studies evaluating outcomes of DVR and N-DVR in scoliosis. Two researchers performed the data extraction independently. Any discrepancies were resolved by a consensus. Seven comparative studies were identified. There were no significant differences between DVR and N-DVR for apical vertebral translation, thoracic kyphosis, lumbar lordosis, coronal balance, sagittal balance, complication rate, and preoperative major Cobb angles. DVR was more effective than N-DVR for correction of thoracic Cobb angle (P = 0.02, weighted mean difference = -3.46° [95% confidence interval, -5.57°, -1.35°]), lumbar Cobb angle (P < 0.0001, weighted mean difference = -4.37° [95% confidence interval, -6.31°, -2.42°]), and apical vertebral rotation (P < 0.0001, weighted mean difference = -3.28° [95% confidence interval, -4.85°, -1.72°]). This meta-analysis showed that postoperative thoracic Cobb angle, postoperative lumbar Cobb angle, and postoperative apical vertebral rotation were better with DVR compared with N-DVR. Further large multicenter studies will be necessary to substantiate these results.