Prediction of secondary ACL injury in female athletes using 2D video-based measurements obtained during dynamic tasks: a retrospective case-control study.

Abstract

To determine if two-dimensional (2D) video-based angular measurements obtained during dynamic tasks predict secondary anterior cruciate ligament (ACL) injury in female athletes post-ACL reconstruction (ACLR). Female athletes post-ACLR underwent 2D video assessment during six tasks (step down, drop jump, lateral shuffle, deceleration, triple hop and side-step-cut) before returning to sport. Reinjury status was determined via survey after returning to sport (N=345). Non-contact ACL reinjuries (n=23) were matched to non-injured athletes (n=57). Five 2D video-based surrogates of known risk factors for primary ACL injury (trunk-tibia angle, thigh angle, trunk lean, pelvis tilt, frontal plane projection angle (FPPA)) were measured from 2D videos across tasks. Reinjury cut-offs for the 2D angles were determined using receiver operating characteristic curves, and composite scores were calculated for each 2D surrogate, with higher scores indicating a high-risk movement across more tasks. Logistic regression, adjusted for confounders, assessed ACL reinjury risk for each composite score. Stepwise regression identified the strongest 2D predictor. Four of five 2D surrogate composite scores independently predicted ACL reinjury after adjusting for confounders. For each additional task in which an athlete demonstrated a high-risk movement, the odds of reinjury increased: trunk-tibia angle composite score (representing increased knee extensor relative to hip extensor moments; OR 1.75; 95% CI 1.24 to 2.47; p=0.001), thigh angle composite score (representing increased vertical ground reaction forces; OR 1.69, 95% CI 1.14 to 2.50, p=0.01), FPPA composite score (representing increased knee valgus moments; OR 2.29, 95% CI 1.13 to 4.65, p=0.02) and pelvis tilt composite score (representing increased pelvis drop; OR 1.85, 95% CI 1.16 to 2.95, p=0.01). The trunk-tibia angle composite score was selected as the strongest predictor. 2D surrogates of known three-dimensional risk factors for primary ACL injury can predict secondary ACL injury. These findings support developing a movement assessment using 2D video-based measurements to inform return to sport decisions and identify at-risk athletes following ACLR.