Metabolomic analysis of synovial fluid from Thoroughbred racehorses diagnosed with palmar osteochondral disease using magnetic resonance imaging
Summary Background Palmar osteochondral disease (POD) is a common cause of lameness in competition horses. Magnetic resonance imaging (MRI) is the most sensitive diagnostic modality currently available, however it may not be financially or logistically practical for routine screening of POD. There i...
Gespeichert in:
Veröffentlicht in: | Equine veterinary journal 2020-05, Vol.52 (3), p.384-390 |
---|---|
Hauptverfasser: | , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Summary
Background
Palmar osteochondral disease (POD) is a common cause of lameness in competition horses. Magnetic resonance imaging (MRI) is the most sensitive diagnostic modality currently available, however it may not be financially or logistically practical for routine screening of POD. There is increasing interest in the use of metabolomics for diagnosis prior to progression to irreversible damage.
Objectives
To determine metabolite levels in synovial fluid (SF) of horses with a clinical diagnosis of POD based on diagnostic analgesia and MRI, with the hypothesis that metabolomic profiles differ between diseased and healthy joints.
Study design
Prospective clinical study.
Methods
Synovial fluid was collected from metacarpo/tarsophalangeal joints (MC/TPJ) of 29 horses (n = 51 joints), including 14 controls (n = 26) and 15 cases (n = 25), the latter with lameness localised to the MC/TPJ and MR changes consistent with POD (n = 23). Spectra were produced using 1H‐nuclear magnetic resonance (NMR) spectroscopy and analysed.
Results
Twenty‐five metabolites were recognised associated with various biosynthetic and degradation pathways. The metabolite abundances within the controls demonstrated increased variability compared with the clinical group. The low level of variance between the spectra of the two groups was explained by five principal components. Cross‐validation of the cohort demonstrated modest separation of predictive power (R2 = 0.67; Q2 = 0.34). Although statistical significance was not achieved, the most influential metabolites were glucose and lactate.
Main limitations
The modest sample size and variation in signalment, background and presenting condition of the controls may have impacted the discriminative power of the constructed models. The lack of matched controls, differences in time of fluid collection and freezing times may have also reduced accuracy when representing metabolite profiles.
Conclusions
This study identified and quantified metabolites present in MC/TPJ SF of clinical cases with POD. |
---|---|
ISSN: | 0425-1644 2042-3306 |
DOI: | 10.1111/evj.13199 |