Plasma endotoxin in horses presented to an equine referral hospital: Correlation to selected clinical parameters and outcomes

Abstract

Endotoxaemia is frequently presumed on the basis of clinical signs in horses with colic. Measurements of plasma endotoxin (LPS) are rarely made in clinical cases and there is little information on the correlations between this variable, clinical variables and outcomes. To measure LPS levels in plasma of horses presented to the Philip Leverhulme Equine Hospital on admission and daily for up to 4 days and to relate LPS levels to selected clinical parameters, such as heart rate and packed cell volume, and outcomes. Blood samples were collected and stored at -20°C prior to assay of the plasma using a validated kinetic chromogenic Limulus amoebocyte lysate (LAL) assay. Clinical parameters and outcome variables were collected from hospital records. Associations were determined by Chi-squared test and logistic regression analysis. Daily blood samples were collected from 234 horses. LPS was detected in 26.5% of the study population and in 29% of those horses presented for colic. Horses providing samples with detectable LPS were more likely to die whilst in the hospital than those that did not (P = 0.045). Horses presenting with colic were more likely to have detectable LPS in their plasma than noncolic cases (P = 0.037), although LPS was detected in the plasma of 8 out of 42 noncolic horses. A horse that did not meet the study definition of clinical endotoxaemia was 10 times less likely to provide a positive LPS sample (OR 0.10, 95% CI: 0.05-0.22). The proportion of horses providing samples with detectable LPS was similar to other studies. LPS was detected in the minority of horses presented with colic. Increased levels of LPS positively correlated with packed cell volume and with risk of mortality in colic cases.