18 F-sodium fluoride positron emission tomography of the racing Thoroughbred fetlock: Validation and comparison with other imaging modalities in nine horses.

BACKGROUND: Early and accurate detection of stress remodelling in racehorses is of utmost importance to prevent catastrophic injuries. Current imaging techniques have limitations in assessing early changes predisposing to catastrophic breakdowns. Positron emission tomography (PET) using 18 F-sodium fluoride (18 F-NaF) is a sensitive method for the detection of early bone turnover and may improve early recognition of subtle injuries. OBJECTIVES: To validate the clinical use of 18 F-NaF PET in Thoroughbred racehorses, to assess the value of PET in the detection of bone lesions and to compare PET results with findings of other advanced imaging modalities, clinical examination and pathology. STUDY DESIGN: Experimental exploratory study. METHODS: Twenty fetlocks from nine Thoroughbred racehorses were imaged using 18 F-NaF PET, computed tomography (CT) and scintigraphy. Five fetlocks were also imaged with magnetic resonance imaging and four fetlocks were also examined histologically. Imaging findings were independently reviewed by three board certified radiologists. Imaging, clinical and histopathological findings were correlated. RESULTS: PET imaging was well-tolerated by all horses. PET detected focal areas of 18 F-NaF uptake in instances where other imaging modalities did not identify abnormalities, in particular in the proximal sesamoid bones. Maximal standardised uptake values could be measured to quantify the activity of lesions. Areas of 18 F-NaF uptake corresponded to regions of increased vascularity and increased osteoblastic activity. MAIN LIMITATIONS: Limited number of cases. CONCLUSIONS: 18 F-NaF PET imaging of the Thoroughbred fetlock is feasible and compares favourably with other imaging modalities in detecting stress remodelling in Thoroughbred racehorses. PET appears to be a beneficial imaging modality when used for early detection of stress remodelling in an effort to prevent catastrophic musculoskeletal injuries in this population of horses.

In vitro biomechanical comparison of load to failure testing of a canine unconstrained medial compartment elbow arthroplasty system and normal canine thoracic limbs.

Elbow dysplasia, primarily affecting the medial compartment, is the most common cause of lameness in the thoracic limb. Elbow arthroplasty is an option for end stage or severely affected patients. The purpose of this study was to compare ex vivo axial load to failure of an implanted novel elbow arthroplasty system to control limbs. The partial arthroplasty is a medial compartmental, unconstrained system, intended to allow conversion to total arthroplasty. We hypothesized that there would not be any significant difference between implanted and controlled limbs when loaded to failure. Six pairs of medium mixed breed canine cadaveric thoracic limbs were prepared for comparison of failure loading of control and implanted limbs. Axial compression was performed using a mechanical testing system. Failure loads were normalized to bodyweight. The mean normalized failure load (N/kg) for the implanted limbs and control limbs were 2.47 (range: 1.62-3.38) and 2.68 (range: 2.25-3.25), respectively. An implanted to control ratio of 0.93 ± 0.19 was calculated. The difference between paired control and implanted limbs in normalized failure loading was not significant (p = 0.38). There were not any differences noted in the yield load (p = 0.30), stiffness (p = 0.62), or energy (0.58). Failure modes were recorded. We concluded that the differences between implanted and control limbs in supra-physiologic axial load to failure were not significant.

Effects of trial repetition, limb side, intraday and inter-week variation on vertical and craniocaudal ground reaction forces in clinically normal Labrador Retrievers.

OBJECTIVES: To document the contributions of trial repetition, limb side, and intraday and inter-week measurements on variation in vertical and craniocaudal ground reaction force data. METHODS: Following habituation, force and time data were collected for all four limbs of seven Labrador Retrievers during sets of five valid trot trials. Each set was performed twice daily (morning and afternoon), every seven days for three consecutive weeks. A repeated measures analysis of variance was used to determine the effects of limb, trial, intraday, and inter-week factors on ground reaction force data for the thoracic and pelvic limbs. RESULTS: Of the four factors evaluated, variation due to trial repetition had the largest magnitude of effect on ground reaction forces. Trial within a set of data had an effect on all craniocaudal, but not vertical, ground reaction force variables studied, for the thoracic limbs. The first of five trials was often different from later trials. Some thoracic limb and pelvic limb variables were different between weeks. A limb side difference was only apparent for pelvic limb vertical ground reaction force data. Only pelvic limb craniocaudal braking variables were different between sets within a day. DISCUSSION AND CLINICAL SIGNIFICANCE: When controlling for speed, handler, gait, weight and dog breed, variation in ground reaction forces mainly arise from trial repetition and inter-week data collection. When using vertical peak force and impulse to evaluate treatment, trial repetition and inter-week data collection should have minimal effect of the data.