Objective To compare the clinical efficacy between one-stage combined posterior and anterior approaches (PA-approach) and simple posterior approach (P-approach) for lower lumbar tuberculosis so as to provide some clinical reference for different surgical procedures of lower lumbar tuberculosis. Methods A retrospective analysis was made on the clinical data of 48 patients with lower lumbar tuberculosis treated between January 2010 and November 2014. Of them, 28 patients underwent debridement, bone graft, and instrumentation by PA-approach (PA-approach group), and 20 patients underwent debridement, interbody fusion, and instrumentation by P-approach (P-approach group). There was no significant difference in gender, age, course of the disease, and destructive segment between 2 groups (P>0.05). The operation time, blood loss, bed rest time, visual analogue scale (VAS) and complication were recorded and compared between 2 groups; American Spinal Injury Association (ASIA) grade was used to evaluate the nerve function, Bridwell classification and CT fusion criteria to assess bone fusion, erythrocyte sedimentation rate (ESR) to evaluate the tuberculosis control, and Oswestry disability index (ODI) to estimate lumbar function. Results The operation time, blood loss, and the bed rest time of the P-approach group were significantly less than those of the PA-approach group (P<0.05). Iliac vessels rupture was observed in 1 case of the PA-approach group and sinus tract formed in 2 cases of the P-approach group. The patients were followed up 13-35 months (mean, 15.7 months) in the PA-approach group and 15-37 months (mean, 16.3 months) in the P-approach group. At last follow-up, common toxic symptom of tuberculosis disappeared and the ASIA scale was improved to grade E. The VAS score and ESR at 1 year after operation and last follow-up, and ODI at last follow-up were significantly improved when compared with preoperative ones in 2 groups (P<0.05), but there was no significant difference between the 2 groups (P>0.05). During follow-up, no internal fixation broken, loosening, or pulling was found. Bridwell bone fusion rates were 89.29% (25/28) and 80.00% (16/20) respectively, and CT fusion rates were 96.43% (27/28) and 90.00% (18/20) respectively, showing no significant difference between the 2 groups (P>0.05). Conclusion Both one-stage PA-approach and simple P-approach could obtain good clinical efficacy. The PA-approach should be selected for patients with anterior-vertebral destroy, presacral or psoas major muscles abscess, and multiple vertebral body destroy, while P-approach should be selected for patient who could gain a good debridement evaluated by imaging before operation, especially for patients with middle-vertebral body destroy, block the iliac blood vessels and old patients.