We compared the perioperative results and complications associated with PLIF and TLIF, and collected evidence for choosing the better fusion method. A literature survey of the MEDLINE and EMBASE databases identified 7 comparative observational studies that met our inclusion criteria. Checklists by Cowley were used to evaluate the risk of bias of the included studies. A database including patient demographic information, perioperative results, and complications was established. The summary odds ratio and weighed mean difference with 95% confidence interval were calculated with a random-effects model. We found that PLIF had a higher complication rate (P <0.00001), and TLIF reduced the rate of durotomy (P = 0.01). No statistical difference was found between the two groups with regard to clinical satisfaction (P = 0.54), blood loss (P = 0.14), vertebral root injury (P = 0.08), graft malposition (P = 0.06), infection (P = 0.36), or rate of radiographic fusion (P = 0.27). The evidence indicated that PLIF required longer operative time (P = 0.03). The evidence indicated that TLIF could reduce the complication rate and durotomy. Neither TLIP nor PLIF was found superior in terms of clinical satisfaction or radiographic fusion rate. PLIF might result in longer time in surgery.