Abstract

In the framework of an echinococcosis surveillance and control programme in Cyprus, a commercial enzyme-linked immunosorbent assay (ELISA) (CHEKIT ECHINOTEST) designed for the detection of Echinococcus granulosus and E. multilocularis coproantigens was used in 1997-2000 for the investigation of large numbers of dogs. Most of the animals originated from areas where approximately 0.2% of the dogs had been found to be infected with E. granulosus in previous (1993-1996) arecoline surveys. The sensitivity of the coproantigen test was 83%, as determined in 35 dogs naturally infected with this cestode species. The specificity was 98% in 97 randomly selected dogs from Cyprus, but it was reduced to 80% in a group of 50 dogs, infected with Taenia spp. A total of 6551 dogs (mainly of rural origin) was examined, including three large groups (N: 2928, 1761 and 1800) from the Government Control Area (GCA) in southern Cyprus and a small group (N: 62) from the Non-Government Control Area (NonGCA) in the northern part of the island. Among the dogs from the GCA, 184 (2.8%) tested positive for coproantigen; coproantigen prevalences were 2.6, 4.9 and 1.1% in these three groups, and of 62 dogs from the NonGCA 8.1% were positive. The calculated true prevalences of E. granulosus in the dog population of the GCA ranged between 0.0 and 3.58%. The predictive values of the test, based on a 0.2% prevalence, was >99.9% for negative results, but very low (7%) for coproantigen-positive results. However, the relatively small number of coproantigen-positive dogs can be treated with praziquantel or the results can further be confirmed by arecoline purging.