Abstract
BACKGROUND: Radiological outcomes between anterior and posterior approach in Lenke 5C curves were still controversial. Meta-analysis on published articles to compare fusion segments and radiological outcomes between the two surgical approaches was performed. METHODS: Electronic database was conducted for searching studies concerning the anterior versus posterior approach in Lenke 5C curves. After quality assessment, data of means, standard deviations, and sample sizes were extracted. RevMan 5.3 was adopted for data analysis. RESULTS: Seven case-control studies involving 308 Lenke 5C AIS patients were identified in the meta-analysis. No significant differences were noted in correction rate of thoracolumbar/lumbar curve (95 % CI -6.02 to 4.32, P = 0.75) and incidence of proximal junctional kyphosis (95 % CI 0.12 to 7.19, P = 0.94) of final follow-up, in change values of thoracolumbar/lumbar curve (95 % CI -3.28 to 7.19, P = 0.46) and thoracic kyphosis (95 % CI -4.10 to 0.13, P = 0.07). The anterior approach represented a significant shorter fusion segments compared to posterior approach (95 % CI -1.72 to -0.71, P < 0.00001). The posterior approach obtained a larger increasing Cobb angle of lumbar lordosis than the anterior approach (95 % CI -6.06 to -0.61, P = 0.02). CONCLUSIONS: The anterior and posterior approach can obtain comparable coronal correction, change values of thoracic kyphosis, and incidence of proximal junctional kyphosis. The anterior approach saves approximate one more fusion segment, and the posterior approach can obtain a larger increasing Cobb angle of lumbar lordosis, from preoperation to final follow-up. TRIAL REGISTRATION: The article type of this study is meta-analysis and prospective registration is not required.