| dc.description.abstract | Objective – To evaluate whether the ability of infrared spectroscopic methods to classify osteoarthritic and normal equine joints is improved by the removal of the cellular components of synovial fluid samples.
Method – Following approval from the Massey University Animal Ethics Committee, 16 untrained New Zealand Thoroughbred fillies (2 to 3 YO) were obtained. The horses were randomly allocated in two groups of eight animals (treatment and control). At day 0, arthroscopy was performed on one middle carpal joint of each treatment horse under general anaesthesia and an 8-mm osteochondral fragment was created (“Chip”), mimicking naturally occurring equine osteoarthritis (OA). The control horses underwent the same arthroscopic surgery, but no cartilage or subchondral bone damage was induced, serving as non-arthritic controls (“Sham”). Starting at day 14, horses were exercised on a high-speed treadmill for 5 consecutive days, followed by 2 days rest each week until the end of the study. Synovial fluid samples were collected weekly aseptically from both mid-carpal joints of each horse. From each sample, an aliquot was drawn (Neat) and the remaining fluid was centrifuged at 1000 g for 10 minutes, the supernatant was separated (Supernatant). For IR spectroscopy dry films were made for each sample and IR spectra were recorded in the mid-IR range. Briefly, principal component analysis (PCA) was performed after pre-processing the data and detecting outliers. Chip joints were compared with contralateral joints in the same horse and with sham joints. With each comparison, the optimal number of components were determined, and these found for neat and supernatant synovial fluids were compared by Wilcoxon matched pairs tests. Differences were considered significant for p < 0.05.
Results – Both the comparisons of the Chip joints versus the contralateral joints and the Chip joints versus the Sham joints showed that the percentage of explained variance was significantly higher for the analysis of neat synovial fluid spectra. Also, the neat samples showed good to moderate group separation in both comparisons while the supernatant samples showed poor group separation.
Conclusion and Clinical Relevance – The findings of this research seem to suggest that the cellular component of the synovial fluid does contribute in the development and diagnosis of OA. In order to create IR spectroscopic methods to detect OA, the model used in this study requires further development to improve it’s diagnostic accuracy. | |
