Summary

The present study provides an algorithm to objectively detect the injury risk biomechanics in ACL-reconstructed football players when tested on the field, with a dedicated real-time automatic report for clinicians

Abstract

Movement biomechanics has become crucial in the rehabilitation and return to sport (RTS) after anterior cruciate ligament (ACL) injury in football (soccer). Recent studies have identified clear injury risk patterns associated with an ACL injury. These patterns involve whole-body mechanics and are likely to occur during cutting movements, such as the ones occurring during pressing, deceiving actions, and decelerations. Due to the high risk of ACL re-injury, especially in young populations, data collected in ecological environments has been therefore advocated to improve injury risk pattern detection. Algorithms to test football-specific cut maneuvers tasks captured on the field are therefore crucial to detecting dangerous biomechanical patterns.
The aim of the present study was to provide a practical tool to assess the ACL injury risk during the RTS continuum through sport-specific biomechanical testing.
68 competitive football (soccer) players were enrolled (47 healthy, 21 ACLR, mean age 16.3±2.7 years) were enrolled. All the ACLR players were cleared for RTS (>14 months after ACL surgery). Data collection was held in a football pitch equipped with artificial turf. The players performed pre-planned 90° changes of direction and unplanned football-specific changes of direction. The football-specific change of direction consisted of a cut with an opponent in ball possession, to simulate a football-specific defensive pressing pattern, the most typical ACL injury risk situation. Kinematics was collected through 8 wearable inertial sensors (MTw Awinda, Xsens) placed on lower body and trunk through a validated workflow. Joint kinematics for ankle, knee, hip, pelvis, and trunk in the three anatomical planes was computed for the cut foot stance. The two-tailed Student’s t-test within the Statistical Parametric Mapping was used to inspect the differences (p<0.05) in joint kinematic waveforms between healthy and ACLR players (injured limb).
Healthy and ACLR players differed in hip and knee rotation during the entire cut phase (p<0.045), hip and knee flexion at initial foot contact (p<0.001), and pelvis-trunk tilt and rotation (p<0.036). An algorithm to determine the risk of knee loading based on the dangerous movement patterns was provided. The algorithm simultaneously tests multiple biomechanical risk factors based on quantitative thresholds belonging to three areas: dynamic valgus collapse, sagittal knee loading, and trunk-pelvis imbalance. The algorithm therefore indicates the presence of a movement performed with a dangerous pattern in real time during the RTS continuum and can describe the macro-areas of risk to target a neuromuscular intervention. A graphical interface with an automatic report for doctors and patients is provided based on the computation.
Differences between healthy and ACLR players persisted after RTS clearance: ACLR players showed stiffer strategy, greater hip rotation, and trunk tilt than healthy players. The algorithm provided can support clinicians in RTS continuum decisions through objective on-field measurements and can upgrade previous tools to inspect ACL injury risk based on in-lab measures only. The strengths of this tool rely on the simultaneous assessment of well-known ACL injury risk factors, the objective metrics based on football-specific field data, and real-time feedback to clinicians.