RESEARCH COMMUNICATIONS OF THE 28th ECVIM-CA CONGRESS

1) Haptoglobin was measured using an in‐house enzymatic method adapted for the ABX Pentra in 20 pairs of residual stored samples obtained 1 hour apart before the administration of trilostane. The coefficient of variation between the haptoglobin measurements was low (median=2.3%, range 0–7.9%), even when there were clinically significant differences between the pre‐trilostane cortisol concentrations (median=12.3%, range 0–61.2%, 8 pairs of samples greater than 20%)., 2) Cases of HAC that had received a consistent dose of trilostane for more than two weeks were blood sampled and clinical scored using an owner questionnaire, from which cases were then categorised as being unwell, controlled, or under‐controlled. Unwell dogs were excluded from further analysis, regardless of the control of their HAC. Haptoglobin was measured as part of our normal monitoring using a commercially available immunoturbidimetric assay (Avacta Laboratories, Wetherby). There were 44 samples, obtained from 19 dogs. One dog was noted as being particularly anxious before 2 sampling points and its pre‐trilostane cortisol concentrations were excluded from all subsequent analysis. Mann Whitney tests demonstrated that there were significant differences between both the pre‐trilostane cortisol (p=0.036) and haptoglobin (p=0.012) of dogs classified as being controlled and those that were uncontrolled. There was a significant correlation between all the haptoglobin concentrations and the owners score (r=0.52, p<0.001) whereas pre‐trilostane cortisol concentrations were not significantly correlated with the score (r=0.27, p=0.13). ROC analysis suggested that a previously established reference range of haptoglobin (0.1‐3.0g/l) could serve as a useful target range for therapeutic monitoring. Similar analysis for pre‐trilostane cortisol was consistent with the current therapeutic range (40‐138nmol/l).