Knee position at the moment of bone bruise could reflect the late phase of non-contact anterior cruciate ligament injury rather than the mechanisms leading to ligament failure.

Purpose: The aim of the present study was to trace knee position at the time of bone bruise (BB) and investigate how much this position departed from the knee biomechanics of an in vivo flexion–extension. Methods: From an original cohort of 62 patients, seven (11%) presented bicompartmental edemas and were included in the study. 3D models of bones and BB were obtained from MRI. Matching bone edemas, a reconstruction of the knee at the moment of BB was obtained. For the same patients, knee kinematics of a squat was calculated using dynamic Roentgen sterephotogrammetric analysis (RSA). Data describing knee position at the moment of BB were compared to kinematics of the same knee extrapolated from RSA system. Results: Knee positions at the moment of BB was significantly different from the kinematics of the squat. In particular, all the patients' positions were out of squat range for both anterior and proximal tibial translation, varus–valgus rotation (five in valgus and two in varus), tibial internal–external rotation (all but one, five externally and one internally). A direct comparison at same flexion angle between knee at the moment of BB (average 46.1° ± 3.8°) and knee during squat confirmed that tibia in the former was significantly more anterior (p < 0.0001), more externally rotated (6.1 ± 3.7°, p = 0.04), and valgus (4.1 ± 2.4°, p = 0.03). Conclusion: Knee position at the moment of Bone bruise position was out of physiological in-vivo knee range of motion and could reflect a locked anterior subluxation occurring in the late phase of ACL injury rather than the mechanism leading to ligament failure. Level of evidence: Level IV [ABSTRACT FROM AUTHOR]