Prospective Clinical Study Assessing Serum Biomarkers for Musculoskeletal Disease in 2- to 3-Yr-Old Racing Thoroughbreds

The use of serum biomarkers provides useful information on the pathophysiology of musculoskeletal disease (intra-articular fragmentation, injury to a tendon or ligamentous structure, incomplete or complete non-articular fracture, and periostitis) and may be of use as a diagnostic tool.

1. Introduction

The use of peripheral blood to assess musculoskeletal disease has been a goal researchers have been working toward for over a decade. The use of serum "biomarkers" have gone through many studies to validate their usefulness in horses [1]. To date, however, only one other study has assessed their usefulness in clinical cases [2]. This study showed significant differences in horses with dorsal metacarpal disease compared with controls but no difference in horses that went on to sustain a fracture in this region. This study was designed to build on previous work by the investigators [3,4] that showed promising results of biomarkers for the identification of intra-articular pathology in a controlled clinical or experimental setting. This study represents the real world application of six serum biomarkers for the detection of musculoskeletal injuries.

2. Materials and Methods

Two- or 3-yr-old Thoroughbred racehorses were entered into the study when they arrived at Thoroughbred race tracks in southern California. Each month, a lameness examination was performed by a study veterinarian, peripheral blood was collected, and serum was stored. Horses were removed from the study when they were out of training for >30 days for any reason or they were enrolled in the study for 10 mo. Only horses that sustained a single musculoskeletal injury and had completed at least 2 mo in the study were analyzed. For the purposes of this study, a musculoskeletal injury was considered one of the following:intra-articular fragmentation (IAF), injury to a tendon or ligamentous structure (TL), incomplete or complete non-articular fracture (ICF), and periostitis (BS).

Serum samples were analyzed for glycosaminoglycan (GAG), type I and II collagen (Col short), type II collagen synthesis (CPII), type II collagen content (Col CEQ), aggrecan synthesis (846), osteocalcin (OC), as a marker of bone formation (CTX), and as a marker of bone degradation as previously reported [3].

Statistical analysis was performed using the SAS statistical software package, version 8e. Non-parametric measures of assumption were tested using a χ 2 analysis. All outcome variables that were concentrations were log-transformed (natural log) to meet assumptions of normality. When direct comparisons were made, a least squares means procedure was used, and p < 0.05 was considered significant. The presence of a musculoskeletal injury and type of injury were both assessed as main and interaction effects, with the horse acting as a random effect.

3. Results

Ninety-two of the 238 horses entering the study were not analyzed because they did not complete at least 2 mo in the study or were diagnosed with multiple lesions. Of the 145 horses, 74 (51%) were considered to sustain a musculoskeletal injury during the study, with the remaining 71 (49%) horses acting as control (CNT) horses. No significant difference in the proportions was observed in the control or injured horses by age, gender, or type of lesion sustained. Of the injured horses, 23 (31%) sustained an IAF, 18 (24%) sustained a TL, 12 (16%) sustained an ICF, and 21 (29%) were diagnosed with BS.

When a comparison was made between samples collected before and after injury within the same horse, the following observations were made (Table 1) a significant increase in GAG levels for TL and BS horses, a significant increase in type II collagen synthesis was noted for IAF horses, a significant increase was measured for ICF horses with respect to the amount of type I and II collagen degradation and osteocalcin was decreased significantly with IAF, BS, and TL injuries.

4. Discussion

The results of this study indicated that biomarkers can be used as a diagnostic aid in clinical musculoskeletal injuries as well as provide insight to the pathogenesis of disease processes. Each of the injuries studied here had a unique biomarker pattern after injury, suggesting further study on predictability of injury is warranted. Furthermore, the fact that the biomarkers were indeed significantly altered in clinical cases despite the confounding effects present in the "real world" is promising.

This project was funded by the Grayson Jockey Club Research Foundation.