Can Knee Extensor Loading Deficits Be Diagnosed Using Variables from Force Plates during Squats Post-ACLr?

Abstract

The persistence of knee extensor moment (KEM) deficits across rehabilitation post-ACL reconstruction may be attributed to difficulty detecting and addressing deficits clinically. During squats, large KEM deficits are present along with much smaller differences in joint or segment angles, making them difficult to detect visually. Data outputs from force plates and video technology may improve clinical identification. Investigate the accuracy of three approaches to estimate KEM deficits using outputs from the force plate and two-dimensional (2D) position data to identify individuals with surgical limb KEM deficits during a bilateral squat. Forty individuals post-ACLr performed bilateral squats. Kinematic and ground reaction force (GRF) data were collected via 3D motion capture system and triaxial force plates. Surgical limb deficits were calculated using limb symmetry index (LSI: nonsurgical/surgical) at peak knee flexion. Gold-standard KEM was calculated using inverse dynamics. Vertical GRF was considered alone, and together with the center of pressure (COP) position to estimate deficits in GRF, and GRF-COP Approaches. Vector Approach used the product of vertical GRF and moment arm calculated using vertical and anterior/posterior GRF, COP, and knee position in trigonometric equations. Separate linear regression and intraclass correlation coefficient (2,k) examined concurrent validity between gold-standard LSI and LSI from GRF, GRF-COP, and Vector Approaches. Specificity and sensitivity (LSI threshold ≥0.85) determined diagnostic accuracy. GRF, GRF-COP, and Vector LSI's predicted gold-standard LSI ( R2 = 0.56, 0.74, and 0.85, respectively) with intraclass correlation coefficient (2,k) 0.49, 0.92, and 0.96; sensitivity of 72.2%, 94.4%, 100%, and specificity of 100%, 50%, 100%, respectively. Vertical GRF alone is not adequate to detect KEM deficits. Additional force plate and 2D position data (COP position and moment arm) can strengthen predictive ability and diagnostic accuracy, supporting translation of these variables into commercial products to enhance clinical assessments.