Bone-patellar tendon-bone autografts versus hamstring autografts for reconstruction of anterior cruciate ligament: meta-analysis.

OBJECTIVES To compare bone-patellar tendon-bone autografts with hamstring autografts for reconstruction of the anterior cruciate ligament. DATA SOURCES Medline, WebSPIRS, Science Citation Index, Current Contents databases, and Cochrane Central Register of Controlled Trials. Review methods All randomised controlled trials reporting one or more outcome related to stability (instrumented measurement of knee laxity, Lachman test, or pivot shift test) and morbidity (anterior knee pain, kneeling test, loss of extension, or graft failure). Study quality was assessed by using a 5 point scale. Random effect models were used to pool the data. Heterogeneity in the effect of treatment was tested on the basis of study quality, randomisation status, and number of tendon strands used. RESULTS 24 trials of 18 cohorts (1512 patients) met the inclusion criteria. Study quality was poor for nine studies and fair for nine studies. The weighted mean difference of the instrumented measurement of knee laxity was 0.36 (95% confidence interval 0.01 to 0.71; P = 0.04). Relative risk of a positive Lachman test was 1.22 (1.01 to 1.47; P = 0.04), of anterior knee pain 0.57 (0.44 to 0.74; P < 0.0001), of a positive kneeling test 0.26 (0.14 to 0.48; P < 0.0001), and of loss of extension 0.52 (0.34 to 0.80; P = 0.003). Other results were not significant. CONCLUSION Morbidity was lower for hamstring autografts than for patellar tendon autografts. Evidence that patellar tendon autografts offer better stability was weak. The poor quality of the studies calls into question the robustness of the analyses.