Evaluating hormonal levels in plasma and saliva as temperament indicators in horses.

Overview

• This study examined whether hormone levels in horse plasma and saliva—specifically cortisol, oxytocin, and serotonin—can serve as indicators of horse temperament.
• The research focused on correlations between plasma and saliva hormone levels and explored links between these hormones and temperament traits in horses, aiming to identify non-invasive biomarkers for assessing temperament.

Background and Importance

• Assessing temperament is key for the welfare, management, and safety of horses.
• Hormones such as cortisol, oxytocin, and serotonin are implicated in stress and social behavior, making them potential biological markers of temperament.
• Non-invasive sampling methods, like saliva collection, are preferred over blood draws for routine temperament assessments in horses.

Study Design and Methods

• Sample: 40 horses were studied, with detailed temperament-hormone association analyses conducted on a subset of 20 horses from the Jeonju cohort.
• Hormones measured: Concentrations of cortisol, oxytocin, and serotonin were analyzed in both plasma and saliva using enzyme-linked immunosorbent assays (ELISA).
• Statistical analyses included:
  • Correlation analysis (Pearson’s r) between plasma and saliva hormone concentrations.
  • Bland-Altman analysis to evaluate agreement and individual variability between plasma and saliva measures.
  • Spearman’s rank correlations to examine associations between hormone concentrations and temperament traits in the Jeonju subset.
  • Correction for multiple comparisons using Benjamini-Hochberg false discovery rate adjustment.
• Comparison of hormone levels was also made between horses with regular human handling versus minimal handling to assess influence of management context on hormone levels.

Key Findings

• Correlations between plasma and saliva hormone levels:
  • Cortisol showed a moderate positive correlation (r = 0.441), indicating reasonable agreement between plasma and saliva.
  • Oxytocin showed a stronger positive correlation (r = 0.587), but Bland-Altman analysis revealed considerable individual variability despite this correlation.
  • Serotonin showed a weak correlation (r = 0.133) and concentration-dependent bias, suggesting poor agreement between plasma and saliva measures.
• Associations between hormones and temperament traits:
  • Several nominal (unadjusted) associations were identified in the Jeonju subset.
  • However, none of these associations remained statistically significant after correcting for multiple testing, indicating insufficient evidence to link hormone levels confidently with temperament traits.
• Effect of human handling:
  • Horses regularly handled by humans exhibited lower cortisol levels and higher oxytocin levels compared to minimally handled horses.
  • These results might reflect the impact of management style on hormone levels but are confounded by other contextual factors and thus should be interpreted cautiously.

Conclusions and Implications

• Salivary cortisol emerged as the most promising non-invasive indicator for assessing stress-related aspects of horse temperament due to its moderate correlation and acceptable agreement with plasma levels.
• Oxytocin and serotonin measurements in saliva showed limitations:
  • Oxytocin had strong correlation but high individual variability, complicating its use for individual-level assessments.
  • Serotonin demonstrated poor plasma-saliva agreement, suggesting limited utility as a saliva biomarker.
• Current findings suggest caution in applying oxytocin and serotonin measures for temperament assessment without further validation.
• Future research with larger sample sizes and improved analytical validation is needed to clarify the relationship between these hormones and temperament traits and to confirm the reliability of saliva as a diagnostic fluid.
• This study lays groundwork for non-invasive hormone monitoring to improve horse welfare and management but highlights the complexity of hormone-behavior relationships.