Use of principle component analysis to quantitatively score the equine metabolic syndrome phenotype in an Arabian horse population.
Authors: Lewis Samantha L, Holl Heather M, Long Maureen T, Mallicote Martha F, Brooks Samantha A
Journal: PloS one
Summary
# Editorial Summary Equine metabolic syndrome (EMS) presents a diagnostic challenge because it comprises overlapping physical and endocrinological signs—obesity, insulin dysregulation, and predisposition to secondary inflammation—that can be difficult to distinguish from pituitary pars intermedia dysfunction (PPID) and may not manifest obviously in mildly affected individuals. Lewis and colleagues applied principal component analysis to blood and physical parameters (body measurements, plasma insulin, leptin, ACTH, glucose, and lipids) from an Arabian horse population to develop a quantitative scoring system that could characterise EMS phenotypes without relying on fixed reference ranges. Two rotated principal components explained 46.3% of variation in the dataset and enabled hierarchical clustering into three distinct groups—healthy, PPID-suspect, and EMS-suspect—corresponding well to traditional diagnostic categories; notably, owner-assigned body condition scores were approximately one point lower than researcher assessments on the nine-point scale, highlighting awareness gaps about healthy equine weight. The researchers successfully created simplified proxies using commonly measured variables that predicted EMS or PPID status with up to 93.4% accuracy, making breed-specific quantitative assessment feasible for field practitioners. This approach is particularly valuable for identifying at-risk horses with subclinical or atypical presentations and provides a foundation for future genomic and metabolomic studies to elucidate EMS aetiology.
Read the full abstract on PubMed
Practical Takeaways
- •Owner perception of body condition significantly underestimates obesity risk; use objective scoring systems alongside owner assessment to identify at-risk horses
- •A breed-specific quantitative scoring system combining physical measurements and blood work (insulin, leptin, ACTH, glucose, lipids) may improve early detection of EMS in Arabian horses before severe clinical signs develop
- •Simplified field-applicable diagnostic protocols using common variables can effectively risk-stratify horses for metabolic syndrome without complex testing
Key Findings
- •Principal component analysis identified two factors explaining 46.3% of variation in EMS phenotype parameters in Arabian horses
- •Owner body condition assessments were one full score lower than researcher assessments on a 9-point scale, indicating perception bias about healthy equine weight
- •Hierarchical clustering using two PCA factors successfully separated horses into three diagnostic categories: Healthy, PPID-suspect, and EMS-suspect
- •A reduced set of commonly used diagnostic variables can estimate up to 93.4% of composite EMS and PPID risk scores in Arabian horses