Changes in equine complete blood count parameters and inflammatory indices with inflammation

Background: Early detection of inflammation using a common test could be a valuable tool for preserving horse health. Hypothesis: This observational study aimed to identify variation in complete blood count (CBC) parameters and inflammatory indices with inflammation, and to develop reference intervals (RIs) from horses with known inflammatory states, hypothesizing that RIs developed from horses with normal acute phase protein concentrations would improve the ability to predict inflammation. Methods: Concentrations of serum amyloid A (SAA) and haptoglobin from clinically healthy horses and horses under veterinary care (n=199) were used to categorize horses as inflamed (SH+; n=48) or not inflamed (SH-; n=150). Non-inflamed RIs were created using 95 % RIs from horses classified as SH- (non-inflamed). Differences in median values of CBC parameters and inflammatory indices were determined using the Kruskal-Wallis test with Dunn's post-hoc analysis and a Bonferroni correction. Predictive value was determined using ROC analysis, and odds ratios were calculated to determine the value of industry and non-inflamed RIs in detecting inflammation. Significance was set at P ≤ 0.05. Results: All three inflammatory indices revealed greater median values in SH+ horses and predictive value greater than all other metrics (AUC=0.64, 0.64, 0.66, respectively). Horses falling above or below all three inflammatory index RIs (SII, SIRI, and AISI) were 6.32, 4.54, and 7 times more likely to be SH+. Additionally, a greater number of non-inflamed RIs yielded increased likelihood of SH+ categorization than industry RIs. Conclusion: These results suggest inflammatory indices and non-inflamed RIs may be a valuable tool to aid in the detection of inflammation in horses.