Performance characteristics of an enzyme-linked immunosorbent assay performed in milk for the detection of liver fluke (Fasciola hepatica) infection in cattle.

Aim: To determine the performance characteristics of an enzyme-linked immunosorbent assay (ELISA) for the detection of antibodies against liver fluke (Fasciola hepatica) in bovine milk.

Methods: Serum and milk from liver fluke infected and non-infected cattle was assayed in a commercially available enzyme-linked immunosorbent assay. Serum test results were used to determine the "gold standard" infection status of cattle and milk ELISA results assessed by ROC analysis.

Results: ROC analysis suggested changes to the ELISA protocol, arriving at milk dilutions assayed considerably higher than those suggested by the manufacturer. With those changes, the ELISA performed with high sensitivity and specificity, 95 and 98.2%, respectively, for individual bovine milks (relative to sera). For bovine tank milks, sensitivity was lower, with bulk milks only testing positive if 60% or more of cattle milking in the herd were infected.

Conclusions: The analysis of the ELISA's performance when used on individual bovine milks demonstrated high sensitivity and specificity. ROC analyses optimised the assay conditions and cut-off point suggested by the manufacturer for this commercial diagnostic assay. This would help with the identification and control of fasciolosis, enabling simpler sample collection.