An important objective of anterior cruciate ligament (ACL) registries is to detect and report early graft failure and revision surgery after ACL reconstruction.

To investigate surgical variables and identify predictors of revision surgery after ACL reconstruction.

Prospective cohort study; Level of evidence, 2.

This prospective cohort study was based on data from the Swedish National Knee Ligament Register during the years 2005 through 2011. Eight surgical variables were investigated: graft selection, graft width, single-bundle or double-bundle techniques, femoral graft fixation, tibial graft fixation, injury-to-surgery interval, injuries to menisci, and injuries to cartilage. The primary endpoint was the 2-year incidence of revision surgery. Relative risks (RRs) and 95% confidence intervals (CIs) were calculated and adjusted for confounders by use of multivariate statistics.

A total of 13,102 patients were included (5541 women [42%] and 7561 men [58%]; P < .001). Hamstring tendon autografts accounted for 90% (11,764 patients) of all reconstructions, of which 96% were performed with a single-bundle technique (11,339 patients). Patellar tendon autografts accounted for the remaining 10% (1338 patients). At index reconstruction, observed injuries to menisci and cartilage were common (40% and 28%, respectively). The overall 2-year incidence of revision surgery was 1.60% (women, 1.57%; men, 1.63%; P = .854). Patients with metal interference screw fixation of a semitendinosus tendon autograft on the tibia had a significantly reduced risk of early revision surgery (RR = 0.32; 95% CI, 0.12-0.90; P = .031).

Metal interference screw fixation of a semitendinosus tendon autograft on the tibia was an independent predictor of significantly lower 2-year incidence of revision surgery. Graft selection, graft width, a single-bundle or a double-bundle technique, femoral graft fixation, the injury-to-surgery interval, and meniscus injury were not predictors of early revision surgery.