The aim of this study was to develop, perfect and validate the PCR (polymerase chain reaction) technique using mitochondrial (mt) and ribosomal (ITS-2) DNA for the accurate identification of Dicrocoelium dendriticum in molluscs and ants, the first and second intermediate hosts, and their early detection. The first primers that were designed amplified a 169 pb mtDNA fragment of D. dendriticum permitted the detection of a single D. dendriticum metacercaria from the Formica rufibarbis and Formica pratensis abdomen, as well as the detection of the brainworm in the head of the ants collected in tetania. Although these primers did not amplify Dicrocoelium chinensis DNA and permitted detected D. dendriticum in the molluscs, they did not discriminate Brachylaimidae metacercariae found in the same mollusc. The PCR that was designed to amplify a 93 bp fragment of the ITS-2 is D. dendriticum specific as it did not amplify D. chinensis, Brachylaimidae and other trematodes. This technique is very sensitive since it permitted the detection of D. dendriticum in the molluscs from the first day post-infection, the brainworm in the head of the ants and only 1 D. dendriticum metacercaria from the abdomen of the ants. Both techniques are important, mainly the latter.

A large field study has been performed in the Cochabamba region of Bolivia with the aim of comparing the polymerase chain reaction (PCR) with other diagnostic methods for Chagas' disease. The amplification of Trypanosoma cruzi-specific kinetoplast DNA sequences in blood samples was compared with classical serological methods, specific IgM detection and direct parasite visualization for 268 school children in a single village where Chagas' disease transmission is active. Of 113 children positive by classical serology or buffy coat examination, 106 were detected by PCR (sensitivity: 93·8%). We did not observe any significant difference of PCR sensitivity between initial (IgM and/or buffy coat positive) and indeterminate stage (only IgG positive) patients. Among the remaining 155 children unconfirmed as chagasic (who were either only IgM positive, IgG-, IgM-, and buffy coat-negative) only 1 case was PCR positive. This case may be due to DNA contamination, or to a very recent infection not detected otherwise, or to specific immune depression. These results show that PCR is a very sensitive parasitological test for Chagas' disease in active transmission regions. The future follow-up of the possibly infected patients who were only IgM-positive should clarify the interest of PCR and IgM tests in the detection of starting infections.