A meta-analysis of stability after anterior cruciate ligament reconstruction as a function of hamstring versus patellar tendon graft and fixation type.

ABSTRACT

PURPOSE: Four-strand hamstring graft (4HS) is stronger than 10-mm bone-patellar tendon-bone graft (BPTB) and has equal tunnel pullout strength, but is believed by some to produce lower rates of stability after anterior cruciate ligament reconstruction (ACLR). Our purpose was to test the hypothesis that 4HS ACLR with modern fixation would produce equal or greater stability than BPTB ACLR. TYPE OF STUDY Meta-analysis. METHODS: A computer search was used to find all published reports of ACLR series using HS and/or BPTB. Inclusion criteria were minimum 24-month follow-up, stratified presentation of arthrometric stability data, and at least 30-lb arthrometric testing force. Twenty-four 4HS, 8 2-strand hamstring (2HS), and 32 BPTB series met these criteria and were subdivided into groups according to fixation type. We used the International Knee Documentation Committee classification of a side-to-side instrumented Lachman test difference of < or = 2 mm as normal stability, and > 5 mm difference as abnormal stability. Series with at least 80% normal and at most 3% abnormal stability were designated as high-stability. Meta-analytic methods were used to determine group level differences. RESULTS: Total 4HS had a higher normal stability rate than total BPTB 77% versus 66%, P < .001; and lower abnormal stability 4.4% versus 5.9%, P = .029. The 4HS ACLR using the EndoButton (Smith & Nephew Endoscopy, Andover, MA) and second-generation tibial fixation (EB2-4HS) had higher normal stability (80%) and lower abnormal stability (1.7%) than all other subgroups, including BPTB with 2 interference screws (70% normal, 5.0% abnormal) P < .001; 84% of the series in the EB2-4HS group were high-stability series. No more than 33% of the series from any other group were high-stability. CONCLUSIONS: The recent literature would suggest that 4HS ACLR produces higher stability rates than BPTB, that 4HS stability rates are fixation dependent, that aperture fixation offers no stability advantage, and that EndoButton with second-generation tibial fixation produces consistently high stability rates. LEVEL OF EVIDENCE Level IV.