A Bayesian evaluation of two dip-stick assays for the on-site diagnosis of infection in calves suspected of clinical giardiasis

The objective of the present study was to evaluate two commercially available dip-stick assays for the diagnosis of Giardia infections in faecal samples from calves suspected of clinical giardiasis. The dip-stick assays provide an on-site and hence quick alternative to laboratory diagnosis. A three-test Bayesian model was used, including the test results of the Coris Giardia strip (Coris Bioconcept, Gembloux, Belgium), the Speed®Giardia or BVT dip-stick (Bio Veto Test/Virbac, La Seyne sur Mer, France), and the Meridian immunofluorescence assay (IFA: Meridian Diagnostics Inc., Cincinnati, OH). In total, 421 faecal samples were examined with the three diagnostic assays between October 2008 and November 2009. Overall, the number of positive samples was markedly higher using the IFA compared to both dip-stick assays, resulting in a high sensitivity (se: 88%, with a 95% probability interval (PI) 60–99%) compared to the Coris dip-stick assay (se: 28%; PI: 16–41%) and the BVT dip-stick assay (se: 26%, PI: 16–35). The specificities of all the three assays were very high (IFA sp: 94%, PI: 90–99%; Coris sp: 92%, PI 86–98%; BVT sp: 93%, PI 88–98%). A positive diagnosis by the dip-stick assays was significantly correlated with a higher cyst excretion level, as measured by IFA. The majority (76%) of the positive samples in the present study contained less than 5000 cyst per gram of faeces, even though all these animals displayed clinical symptoms of diarrhea potentially due to Giardia. The low level of cyst excretion in these samples might in part explain the poor sensitivity of both dip-stick assays. Although multiple samplings might be an option to increase the sensitivity of the dip-stick assays, the laboratory based IFA seems at current to be the best option for clinical diagnosis of Giardia in calves.