LAUSR

To elucidate the relationship between graft tunnel position and knee laxity in the cases of double-bundle ACL reconstruction. Total of 132 cases were included. Femoral and tibial tunnels were evaluated by quadrant method on 3D-CT. As additional reference of tibia, the distances from medial tibial spine to the tunnel center (DMS) and from Parsons’ knob to the tunnel center (DPK) were evaluated; %DMS/ML and %DPK/AP were calculated (ML and AP: mediolateral and anteroposterior width of tibial plateau). Preoperative and postoperative (1 year from surgery) stabilities were evaluated by Lachman and pivot-shift procedures. If there was ≥ 2 mm side-to-side difference, the subject was defined as having anterior knee laxity (AKL); if the pivot-shift phenomenon was observed with IKDC grade ≥ 1, there was rotatory knee laxity (RKL). Multiple logistic regression analysis was conducted with the prevalence of AKL or RKL as the dependent variable and with tunnel positions as the independent variables. Overall, 21 subjects (15.9%) showed AKL, and 15 subjects (11.4%) showed RKL. Those with postoperative laxity showed higher %DMS/ML and higher femoral position than those without laxity. Regarding posterolateral bundle, logistic regression model estimated that %DMS/ML was associated with the prevalence of AKL (B = 0.608; p < 0.001) and RKL (B = 0.789; p < 0.001); %high-low femoral tunnel position (B =  − 0.127; p = 0.023) was associated with that of RKL. There was the risk of residual knee laxity in ACL-reconstructed knee when tibial tunnel shifted more laterally or higher femoral tunnel was created with regard to posterolateral bundle. III.