Relationship of Graft Type and Vancomycin Presoaking to Rate of Infection in Anterior Cruciate Ligament Reconstruction: A Meta-Analysis of 198 Studies with 68,453 Grafts.

BACKGROUND

Infection is a devastating complication in anterior cruciate ligament reconstruction (ACLR) surgery. Given the rarity of infection, pooling individual studies via meta-analysis can allow more meaningful evaluation of factors influencing infection rates. We aimed to determine the relationship of graft type and vancomycin graft presoaking to bacterial infection rates following ACLR.

METHODS

A systematic literature search was conducted on PubMed, Ovid MEDLINE, Embase, and CENTRAL (Cochrane Register of Controlled Trials). Included articles were those reporting on primary arthroscopic or open ACLR procedures, using hamstring (HT) or bone-patellar tendon-bone (BPTB) autografts or allografts of any type, with regard to the outcome of infection (deep infection or septic arthritis). Meta-analyses were performed to estimate the overall infection rates in ACLR surgery according to graft type and to examine the effect of presoaking grafts in vancomycin on infection rates.

RESULTS

We identified 306 bacterial infections in 68,453 grafts across 198 studies. The overall estimated ACL graft infection rate in our meta-analysis was 0.9% (95% confidence interval [CI] = 0.8% to 1.0%). HT autografts were associated with a higher infection rate (1.1%, CI = 0.9% to 1.2%) than BPTB autografts (0.7%, CI = 0.6% to 0.9%) and allografts (0.5%, CI = 0.4% to 0.8%) (Q = 15.58, p < 0.001). Presoaking HT autografts in vancomycin reduced infection rates to 0.1% (CI = 0.0% to 0.4%) (Q = 10.62, p = 0.001).

CONCLUSIONS

Infection following ACLR remains a rare but serious complication. HT autografts are associated with higher infection rates than other graft types. Presoaking HT autografts in vancomycin reduces infection rates by an estimated tenfold.

LEVEL OF EVIDENCE

Therapeutic Level IV. See Instructions for Authors for a complete description of levels of evidence.