RESUMO
BACKGROUND: One- and two-level osteotomies have been used to treat thoracolumbar kyphosis in patients with ankylosing spondylitis (AS). However, the effectiveness and safety of the 2 methods have not been systematically compared, and few studies have reported on which method might be more suitable for treating AS. METHODS: We performed a systematic literature search. All comparative studies of one- and two-level osteotomy for the treatment of thoracolumbar kyphosis caused by AS were included. Efficacy was determined by the radiographic outcomes, including global kyphosis, sagittal vertical axis, and lumbar lordosis and the Oswestry disability index. The complication rates were used to assess the safety. We analyzed the data using Review Manager software. RESULTS: A total of 10 studies were included and used for the meta-analysis. We pooled the studies according to the type of data they had reported to evaluate the radiographic correction and incidence of complications. We found that two-level osteotomy provided advantages compared with one-level osteotomy for spinal correction, including global kyphosis and lumbar lordosis. However, the correction was dependent on the degree of preoperative kyphosis. In addition, we found no differences between the 2 groups in the correction of the sagittal vertical axis or improvement in the Oswestry disability index. However, we did find differences in the incidence of neural complications, dural tear, and operative site infection. CONCLUSIONS: Our findings have shown that one-level osteotomy is safer than two-level osteotomy. For AS patients with nonsevere kyphosis, one-level osteotomy will be sufficient for spinal sequence correction and maintenance of balance, and two-level osteotomy will not be needed. For AS patients with severe kyphosis, two-level osteotomy can be recommended for better orthopedic effects; however, a careful surgical technique is required.