Effect of graft selection on the incidence of postoperative infection in anterior cruciate ligament reconstruction.

BACKGROUND: Knee joint infection is a potentially devastating complication of anterior cruciate ligament (ACL) reconstruction. There is a theoretical increased risk of infection with the use of allograft material. HYPOTHESIS: An allograft ACL reconstruction predisposes patients to a higher risk of bacterial infection. STUDY
DESIGN: Cohort Study; Level of evidence, 3.
METHODS: All primary ACL reconstructions performed at our institution between January 2002 and December 2006 were reviewed; 3126 total procedures were identified. A retrospective medical record review was performed to determine the incidence of infection, offending organism, time after surgery until presentation, infection treatment, and graft salvage as an outcome of graft choice.
RESULTS: Of the 3126 ACL reconstructions, 1777 autografts and 1349 allografts were performed. Eighteen infections were identified (0.58%). Infections occurred in 6 of the 1349 allografts (0.44%), 7 of the 1430 bone-patellar tendon-bone (BPTB) autografts (0.49%), and 5 of the 347 hamstring autografts (1.44%). Five grafts were removed because of graft incompetence or loosening: 3 hamstring tendon, 1 BPTB, and 1 allograft. The most common organism isolated was Staphylococcus aureus. Hamstring tendon autograft had an increased incidence of infection compared with both BPTB autograft and allograft (P <.05), with a trend toward a more common need for graft removal (P = .09). Allograft reconstructions were equally likely to have graft salvage as autograft reconstructions.
CONCLUSION: Hamstring tendon autografts have a higher incidence of infection than BPTB autografts or allografts. The use of allograft material in ACL reconstructions does not increase the risk of infection or the need for graft removal with infection.