Objective To evaluate knee stability after double-bundle and single-bundle anterior cruciate ligament reconstruction. Methods Randomized controlled trials (RCTs) and quasi-randomized controlled trials (quasi-RCTs) were collected from MEDLINE (1966 to October 2007), OVID (1950 to October 2007), The Cochrane Library (issue 4, 2007) and China Academic Journals Full-text Database (1979 to October 2007). The quality of included trials was assessed. Data analyses were performed with The Cochrane Collaboration’s RevMan 4.2.10 software. Results One RCT (quality B) and 5 quasi-RCTs (quality C) involving 426 patients met the inclusion criteria. Meta-analyses showed significant differences between the two operative procedures in terms of anterior stability (WMD –0.75, 95%CI –1.14 to –0.37, P=0.000 1) and rotational stability [RR 1.38, 95%CI 1.17 to 1.61, Plt;0.000 1]. Conclusion The double-bundle anterior cruciate liagament is superior to single-bundle anterior cruciate ligament in terms of anterior stability and rotational stability. Since the included trials were small and of poor quality, more high-quality, large-scale randomized controlled trials are required.