Towards an algorithm for the prediction of non-contact anterior cruciate ligament injuries

Fickling, Shaun Dean

Towards an algorithm for the prediction of non-contact anterior cruciate ligament injuries

Master Thesis

2015

Publisher

University of Cape Town

Department

Division of Biomedical Engineering

Faculty

Faculty of Health Sciences

Abstract

Background: The anterior cruciate ligament (ACL) of the knee is one of the most frequently injured ligaments in the body. 70% of ACL injuries are sustained without any direct contact to the knee, during the early stance phase of a rapid deceleration movement. Females have a significantly greater risk of injury than males participating in the same activities. In the years following injury, ACL deficient individuals are likely to experience lasting joint pain, functional instabilities and the onset of osteoarthritis. The best practice model for management of ACL injuries is a continued emphasis on prevention, which is currently limited by an incomplete understanding of how the injuries occur. Hypothesis: Body biomechanics occurring during the terminal swing phase of a dynamic deceleration movement can predict the resulting weight acceptance phase ACL loading in both ligament bundles. This will further the understanding of the sequence of events that result in non - con tact ACL injuries. Methods: For a preliminary feasibility study, a musculoskeletal model was developed in OpenSim incorporating both anteromedial (AMB) and posterolateral (PLB) bundles of the ACL. Motion capture data of female soccer players (n = 10, mean age = 19.60 ± 1.49 years) performing unanticipated side - step cutting movements were recorded. Instantaneous, three dimensional joint angles and angular velocities at the mid - swing stage of the side - step were selected as the independent variables. The dependent variables were the maximum stance - phase AMB and PLB strains. Multiple pairwise correlation analyses were used to quantify linear relationships between these variables. To evaluate the overall potential to predict ACL strain, a best subsets linear regression model was implemented using only the significantly correlated independent variables. Each ligament bundle was analysed independently. Results: Hip internal rotation at the mid - swing stage explained 79.1% (95% CI: 59.9% - 98.2%) of the variance in maximum stance - phase anteromedial bundle strain (p = 0.0006). Mid - swing knee varus position and knee valgus velocity combined explained 83.3% (95% CI: 69.2% - 97.3%) of the variance in maximum stance - phase posterolateral bundle strain (p = 0.0019). Conclusions: Swing - phase body kinematics during a side - step movement can provide meaningful predictive information as to the future strain in both bundles of the ACL. They are thus useful components in understanding and exploring elements of the inciting e vent, particularly a kinematic "sequence of no return" that directly precedes the injury. The results validate continued research in this area, where the iv relationships identified in this preliminary investigation can guide the development of a priori hypotheses for future studies to be completed at higher levels of evidence. Clinical Relevance: A more comprehensive understanding of the variables that result in non - contact ACL injuries will allow for the design and implementation of more effective preventative measures. For example, knowledge of the "sequence of no return" could be used in sophisticated statistical systems to predict ACL injury events in real - time. This could be used to trigger an active knee brace to apply external support to the knee, pre venting damage to the ligament. The long - term outcome of this project is to move towards reducing the risk and incidence of ACL injuries and the associated negative effects, preserving knee - vitality and ensuring quality of life for athletes and active individuals.

Keywords

Biomedical Engineering

Reference:

Collections

Masters

Full item page