Kinematics measured during a short arc quadriceps knee extension exercise were compared in the knees of functionally unstable ACL-deficient patients, these patients' uninjured knees, and uninjured control subjects' knees. Cine phase contrast dynamic magnetic resonance imaging, in combination with a model-based tracking algorithm developed by the authors, was used to measure tibiofemoral kinematics as the subjects performed the active, supine posture knee extension exercise in the terminal 30 degrees of motion. Two determinants of tibiofemoral motion were measured: anterior/posterior location of the tibia relative to the femur, and axial rotation of the tibia relative to the femur. We hypothesized that more anterior tibial positioning, as well as differences in axial tibial rotation patterns, would be observed in ACL-deficient (ACL-D) knees when compared to uninjured knees. Multifactor ANOVA analyses were used to determine the dependence of the kinematic variables on (i) side (injured vs. uninjured, matched by subject in the control group), (ii) flexion angle measured at five-degree increments, and (iii) subject group (ACL-injured vs. control). Statistically significant anterior translation and external tibial rotation (screw home motion) accompanying knee extension were found. The ACL-D knees of the injured group exhibited significantly more anterior tibial positioning than the uninjured knees of these subjects (average difference over extension range ¼ 3.4 AE 2.8 mm, p < 0.01 at all angles compared), as well as the matched knees of the control subjects. There was a significant effect of interaction between side and subject group on A/P tibial position. We did not find significant differences in external tibial rotation associated with ACL deficiency. The changes to active joint kinematics documented in this entirely noninvasive study may contribute to cartilage degradation in ACL-D knees, and encourage more extensive investigations using similar methodology in the future. ß