Objective To compare microsurgery lumbar discectomy (MSLD) via posterior approach with traditional open discectomy by fenestration for single-level lumbar disc protrusion in terms of methodology and therapeutic effect. Methods From January 2001 to January 2008, 230 patients with single-level lumbar disc protrusion were randomized into two groups. In group A, 114 patients underwent MSLD, including 77 males and 37 females aged 15-76 years old (average 41 years old); the duration of the disease ranged from 6 months to 28 years (average 51 months); the lumbar disc protrusion involved L4,5 level in 52 cases, and L5 - S1 level in 62 cases; there were 50 cases of lumbar disc protrusion on the left side, 54 onthe right side, and 10 of the central type; preoperative JOA score was 6-18 points (average 11.8 points). In group B, 116 patients underwent traditional posterior open discectomy by fenestration, including 78 males and 38 females aged 14-78 years old (average 42 years old); the duration of the disease ranged from 8 months to 26 years (average 52 months); the lumbar disc protrusion involved L4,5 level in 56 cases, and L5 - S1 level in 60 cases; there were 53 cases of lumbar disc protrusion on the left side, 52 on the right side, and 11 of the central type; preoperative JOA score was 5-19 points (average 12.3 points). No significant difference was evident between two groups in terms of general information (P gt; 0.01). Parameters of operative time, volume of blood loss during operation, length of operative incision, length of hospital stay after operation, and total medical cost of single disease were analyzed. Therapeutic effect was assessed by postoperative JOA score during follow-up period. Results The operative time was (40 ± 9) minutes in group A and (47 ± 11) minutes in group B. The volume of blood loss during operation was (26 ± 5) mL in group A and (60 ± 6) mL in group B. The length of operative incision was (2.6 ± 0.8) cm in group A and (5.6 ± 0.5) cm in group B. The length of hospital stay after operation was (4.0 ± 2.6) days in group A and (8.0 ± 2.9) days in group B. The total medical cost of single disease was (5 500 ± 1 800) ￥ in group A and (6 300 ± 1 500) ￥ in group B. Significant difference was evident between two groups in terms of the above parameters (P lt; 0.01). The incisions in two groups all healed by first intention.No compl ications such as wrong orientation, nerve root injury, cauda equina injury, and infection occurred. The follow-upperiod was 12-37 months (average 26 months) for 102 patients of group A and 12-35 months (average 24 months) for 98 patients of group B. The JOA score 12 months after operation was 21-28 points (average 24.8 points) in group A and 22-27 points (average 25.2 points) in group B, showing a significant difference when compared with preoperative score (P lt; 0.01), and no significant difference between two groups (P gt; 0.01). Conclusion Two methods have similar cl inical outcomes, but MSLD has merits of minimal invasion, less blood loss, shorter operative time, shorter length of hospital stay, and lower medical cost. It is one of ideal minimally invasive operations for single-level lumbar disc protrusion．