Proximal sesamoid bone (PSB) fractures in racehorses are likely fatigue fractures that occur due to repetitive loads and stress remodeling. The loading circumstances that may induce damage in the PSBs are not well understood. The goal of this study was to determine in three-dimensions, PSB motions relative to the opposing metacarpal condyle during simulated mid-stance loads. Seven equine cadaveric forelimbs were axially loaded in a material testing system to simulate standing and mid-stance walk, trot, and gallop load conditions (1.8-10.5 kN). Joint angles were determined by tracking the positions of bone-fixed kinematic markers. Internal-external rotation, abduction-adduction, and flexion-extension of each PSB relative to the third metacarpal condyle were compared between loads and between PSBs using an ANOVA with Tukey-Kramer post hoc tests for pairwise comparisons. The medial PSB rotated externally and the lateral PSB apex abducted during limb loading. Medial PSB external rotation was significantly greater at the gallop load condition than at the walk or trot load conditions. The medial and lateral PSB motions observed in this study are consistent with location of fatigue damage and fracture configurations frequently seen in medial and lateral PSBs from Thoroughbred racehorses. Specifically, medial PSB external rotation is consistent with the development of an abaxial subchondral medial PSB lesion that is reported in association with medial PSB transverse fracture and lateral PSB abduction is consistent with axial longitudinal fracture of the lateral PSB.

Medial proximal sesamoid bones (PSBs) from Thoroughbred racehorses that did (Case) or did not (Control) experience unilateral biaxial PSB fracture were evaluated for bone volume fraction (BVF), apparent mineral density (AMD), tissue mineral density (TMD), and microdamage in Case fractured, Case contralateral limb intact, and Control bones. A majority of Case bones had a subchondral lesion with high microdamage density, and low BVF, AMD, and TMD. Lesion microdamage and densitometric measures were associated with training history by robust linear regression. Exercise intensity was negatively related to BVF (0.07 ≤ R² ≤ 0.12) and positively related to microcrack areal density (0.21 ≤ R² ≤ 0.29) in the lesion; however, in an undamaged site, the relationships were opposite in direction. Regardless of location, TMD decreased with event frequency for both Case and Control, suggesting increased bone remodeling with exercise. Measures of how often animals were removed from active training (layups) predicted a decrease in TMD, AMD, BVF, and microdamage at regions away from the lesion site. A steady‐state compartment model was used to organize the differences in the correlations between variables within the data set. The overall conclusions are that at the osteopenic lesion site, repair of microdamage by remodeling was not successful (e.g., lower bone mass, increased damage, and lower mineralization) but that in regions away from the lesion remodeling successfully controlled damage (e.g., higher bone mass, less microdamage, and lower mineralization).

The ability to quantify equine kinematics is essential for clinical evaluation, research, and performance feedback. However, current methods are challenging to implement. This study presents a motion capture methodology for horses, where three-dimensional, full-body kinematics are calculated without instrumentation on the animal, offering a more scalable and labor-efficient approach when compared with traditional techniques. Kinematic trajectories are calculated from multi-camera video data. First, a neural network identifies skeletal landmarks (markers) in each camera view and the 3D location of each marker is triangulated. An equine biomechanics model is scaled to match the subject’s shape, using segment lengths defined by markers. Finally, inverse kinematics (IK) produces full kinematic trajectories. We test this methodology on a horse at three gaits. Multiple neural networks (NNs), trained on different equine datasets, were evaluated. All networks predicted over 78% of the markers within 25% of the length of the radius bone on test data. Root-mean-square-error (RMSE) between joint angles predicted via IK using ground truth marker-based motion capture data and network-predicted data was less than 10 degrees for 25 to 32 of 35 degrees of freedom, depending on the gait and data used for network training. NNs trained over a larger variety of data improved joint angle RMSE and curve similarity. Marker prediction error, the average distance between ground truth and predicted marker locations, and IK marker error, the distance between experimental and model markers, were used to assess network, scaling, and registration errors. The results demonstrate the potential of markerless motion capture for full-body equine kinematic analysis.

Focal bone lesions are often found prior to clinically relevant stress-fractures. Lesions are characterized by low bone volume fraction, low mineral density, and high levels of microdamage and are hypothesized to develop when bone tissue cannot sufficiently respond to damaging loading. It is difficult to determine how exercise drives the formation of these lesions because bone responds to mechanical loading and repairs damage. In this study, we derive steady-state rate constants for a compartment model of bone turnover using morphometric data from fractured and non-fractured racehorse proximal sesamoid bones (PSBs) and relate rate constants to racing-speed exercise data. Fractured PSBs had a subchondral focus of bone turnover and microdamage typical of lesions that develop prior to fracture. We determined steady-state model rate constants at the lesion site and an internal region without microdamage using bone volume fraction, tissue mineral density, and microdamage area fraction measurements. The derived undamaged bone resorption rate, damage formation rate, and osteoid formation rate had significant robust regression relationships to exercise intensity (rate) variables, layup (time out of exercise), and exercise 2–10 months before death. However, the direction of these relationships varied between the damaged (lesion) and non-damaged regions, reflecting that the biological response to damaging-loading differs from the response to non-damaging loading.