Objective To analyze the stabil ity and cl inical outcomes of arthroscopic anterior cruciate l igament (ACL) reconstruction with γ irradiated patellar tendon allograft compared with autograft. Methods From January 2004 to October 2007, 69 patients undergoing arthroscopic ACL reconstruction were prospectively randomized consecutively into two groups: group A (autograft, n=36) and group B (γ irradiated allograft, n=33). In group A, there were 30 males and 6 females with an average age of 30.1 years, including 30 cases of simple ACL rupture and 6 cases of ACL rupture with medial accessory l igament injury; ACL rupture was caused by sports in 28 cases, by traffic accident in 5 cases, and by others in 3 cases; and the time from injury to operation was 1.4 months on average. In group B, there were 26 males and 7 femaleswith an average age of 32.5 years, including 27 cases of simple ACL rupture and 6 cases of ACL rupture with medial accessory l igament injury; ACL rupture was caused by sports in 27 cases, by traffic accident in 4 cases, and by others in 2 cases; and the time from injury to operation was 1.5 months on average. There were no significant differences in general data between two groups (P gt; 0.05). The same arthroscopic technique was used in all ACL reconstructions done by the same surgeon. The cl inical outcome was evaluated and compared by general conditions, pivot shift test, Lachman test, KT-2000 arthrometer testing, Daniel’s one-leg hop test, International Knee Documental Committee (IKDC) scoring, Lysholm knee scoring scale, and Tegner activity score. Results All patients were followed up for 39.5 months (group A) and 37.6 months (group B). In group A, patella fracture occurred in 1 case and anterior knee pain in 2 cases postoperatively. No compl ication occurred in group B. The hospital ization times in groups A and B were (15.6 ± 2.4) days and (15.5 ± 1.5) days, respectively, showing no significant difference (P gt; 0.05). The operation time of group A was longer than that of group B and the fever time of group A was shorter than that of group B, showing significant differences (P lt; 0.05). At the final follow-up, there were significant differences (P lt; 0.05) in Lachman test and the pivot shift test between two groups, between pre- and post-operation; there were no significant differences (P gt; 0.05) in Daniel’s one-leg hop test, the IKDC, Lysholm, and Tegner activity scores between two groups, however, there was a decreased trend in the functional and activity levels in group B. And there was significant difference between pre- and post-operation (P lt; 0.05). At the final follow-up, the differences between normal side and affected side were (2.4 ± 0.6) mm in group A and (5.5 ± 3.6) mm in group B, showing significant difference (P lt; 0.05). There was significant difference in tibial advancement between pre- and post-operation (P lt; 0.05). Conclusion The functional and activity level of the knee after ACL reconstruction with autograft and γ irradiated patellar tendon allograft were similar, but anterior and rotational stabil ity of the involved knee decreases significantly in the group with γ irradiated patellar tendon allograft.