Summary
A physics-based computational knee model revealed that a cohort of young, female athletes suffering first-time, non-contact ACL rupture (n=60) experienced 19% greater ACL force compared to uninjured controls during a simulated pivot shift, which highlights the role of unfavorable knee mechanics in ACL injury.
Abstract
Authors:
Andrew Pearle, Mitchell Wheatley, Jacob Zeitlin, Jacob Hirth, Thomas Wickiewicz, Danyal Nawabi, Bruce Beynnon, Carl Imhauser
Introduction
Unique geometric features of tibiofemoral anatomy (e.g., lateral tibial slope, femoral notch width) are related to risk of ACL injury. Therefore, physics-based computational models that can simulate the complex interactions of knee geometries may help identify those at heightened risk of noncontact ACL injury. Thus, our study had three objectives: 1) to create athlete-specific, 3D, physics-based computational knee models; 2) to estimate ACL forces during a simulated pivot shift loading sequence using these computational models; and 3) to characterize the relationship between predicted ACL force and risk of first-time, noncontact ACL injury.
Methods
Previously collected MRI data from 60 female athletes suffering first-time, noncontact ACL rupture and 60 sex-, age-, and sport-matched controls from the same team were utilized. Imaging was obtained on contralateral, uninjured knees of injured athletes, and the same side of matched controls. To accomplish our first objective, volumetric representations of tibiofemoral bone, articular cartilage, and menisci were segmented and incorporated into a physics-based computational modeling pipeline. Next, we applied a sequence consisting of 100 N of compression, an 8 Nm valgus moment, and a 30 N anterior-directed force to simulate a pivot shift maneuver. At the peak applied load, we recorded the model-based prediction of ACL force. These ACL forces were compared between cases and controls using a paired Student’s t-test (a = 0.05). The discriminative ability of ACL force to identify injured versus uninjured athletes was assessed using the area under the receiver-operating characteristic curve (AUROC). Logistic regression was used to quantify the risk associated with increased ACL force, expressed as an odds ratio (a = 0.05).
Results
Model predictions of ACL force were, on average, 22.5 N (95% CI: 6.7 ¬– 38.3 N) greater in injured athletes than in their uninjured teammates (mean +/- SD: 119.0 +/- 41.0 N vs. 96.5 +/- 42.7 N; p=0.003). Using ACL force to classify athletes as either injured or uninjured had an AUROC of 0.65 (0.56 – 0.74). The odds ratio for each one-SD increase in ACL force was 1.73 (1.18 – 2.55; p=0.005), indicating a 73% increase in likelihood of ACL injury for each one-SD increase in force.
Conclusion
Our physics-based computational knee model revealed an association between increased ACL force and risk of ACL injury in female athletes. Notably, the model predicted 19% greater ACL forces in injured athletes, on average, compared to uninjured controls. The use of athlete-specific 3D computational knee models allows for a detailed, targeted investigation of the relationship between knee biomechanics and ACL injury risk in subpopulations (i.e., young, female athletes) at the greatest risk of ACL rupture. Further refinement of this model is likely necessary to improve predictive accuracy such as incorporating additional risk factors (e.g., knee laxity) or loading conditions (e.g., Lachman exam, cutting maneuver). The association between heightened ACL force and injury risk highlights the importance of addressing unfavorable knee mechanics (i.e., elevated ACL force) to enhance injury prevention strategies for female athletes.