Analysis of stress fractures associated with lameness in Thoroughbred flat racehorses training on different track surfaces undergoing nuclear scintigraphic examination

REASONS FOR PERFORMING STUDY: There is limited information regarding the impact of training track surface on the occurrence of stress fractures. OBJECTIVES: To evaluate the impact of training track surface on the proportion of long bone and pelvic stress fractures associated with lameness in Thoroughbred horses in flat race training undergoing nuclear scintigraphic examination. STUDY DESIGN: Retrospective study. METHODS: Scintigraphic examinations of Thoroughbred flat racehorses were evaluated from 2 hospitals (hospital A [Toronto Equine Hospital], 2003–2009, and hospital B [George D. Widener Hospital for Large Animals, School of Veterinary Medicine, University of Pennsylvania], 1994–2006). Horses admitted to hospital A trained at a single track, at which the main training surface changed from dirt to synthetic on 27 August 2006. Two distinct populations existed at hospital B: horses that trained on dirt (numerous trainers) and those that trained on turf (single trainer). All scintigraphic images were evaluated by a blinded reviewer. Fisher's exact test and logistic regression were used when appropriate, and significance was set at P<0.05. RESULTS: When reviewing 528 scintigraphic examinations from hospital A (257 dirt and 271 synthetic, numerous trainers), there was a greater proportion of stress fractures detected in scintigraphic examinations from horses training on a synthetic surface (31.7%) in comparison to scintigraphic examinations from horses training on a dirt surface (23.0%) at an earlier point in time (P = 0.03). There was a greater proportion of hindlimb/pelvic and tibial stress fractures diagnosed in horses from the synthetic surface‐trained group than from the dirt‐trained group at hospital A (P<0.04 and P = 0.03, respectively). CONCLUSIONS: This study provides evidence that training surface may affect the proportion of stress fractures diagnosed, but other factors, such as training philosophy, appear to be important. Future prospective investigations to fully elucidate the relationship between training track surface and the proportion of stress fractures and other nonfatal musculoskeletal injuries are warranted.