Purpose: We aimed to provide evidence for clinical choice of surgical approach in treating spinal tuberculosis, including anterior, posterior and combined approaches (combined anterior and posterior approach).
Methods: A literature search up to June 2015 was performed on PubMed, Embase, Cochrane library, CNKI, Wanfang and Weipu database. Weighted mean differences (WMDs) or risk radios (RRs) and their 95 % confidence intervals (CI) were calculated.
Results: Total 26 studies with 2345 spinal tuberculosis adults were analyzed. Results showed advantages of posterior approach compared with anterior approach in operation time (WMD = 20.91; 95 % CI: 9.05-32.76), blood loss (WMD = 72.32, 95 % CI: 13.87-130.78), correction of angle (WMD = -2.47; 95 % CI: -4.04 to -0.90) and complications (RR = 1.78; 95 % CI: 1.21-2.60), and compared with combined approach in operation time (WMD = -82.76; 95 % CI: -94.38 to -71.14), blood loss (WMD = -263.63; 95 % CI: -336.85 to -190.41), hospital stay [(WMD = -4.60; 95 % CI: -5.10 to -4.10) and complications (RR = 0.36; 95 % CI: 0.23-0.58]. Meanwhile, significantly larger correction of angle (WMD = -2.25; 95 % CI: -4.35 to -0.14; P = 0.04) and less loss of correction (WMD = 3.97; 95 % CI: 2.22-5.72) were found when compared combined approach with anterior approach. However, combined approach had significantly longer operation time (WMD = -41.92; 95 % CI: -52.45 to -31.38) and more blood loss (WMD = -102.18; 95 % CI: -160.45 to -43.91) than anterior approach.
Conclusion: Posterior approach has better clinical outcomes than anterior or combined approach for spinal tuberculosis. However, individual assessment of each case should be considered in the clinical application of these surgical approaches.
Keywords: Anterior approach; Combined anterior and posterior approach; Meta-analysis; Posterior approach; Spinal tuberculosis.