The diagnosis of acute infection with Trypanosoma cruzi, the protozoan parasite that causes Chagas' disease, is generally made by detecting parasites by microscopic examination of fresh blood. Although highly specific, this approach often lacks sensitivity. Several years ago, PCR assays for the detection of T. cruzi were described, but the sensitivities and specificities of these tests have not yet been defined precisely. In the present study, we first compared the sensitivities of PCR methods that differ in sample processing as well as in the target sequences that are amplified. Then, we challenged eight mice with T. cruzi, and on 31 days over a 380-day period, we compared the ability of the PCR method with the highest sensitivity to detect parasites in blood with that of microscopic examination. During the acute phase of the infections, parasites were detected on average 3.9 days earlier by the PCR method than by microscopy. Furthermore, the infected mice were consistently positive by the PCR method during the chronic phase, while parasites were intermittently detected by microscopic examination during that period. Overall, among the 248 comparisons, in 84 the PCR method was positive and no parasites were seen by microscopic examination, whereas the reverse was true in only 1 case, a difference that is highly significant. These findings suggest that this approach should be in patients suspected of having acute Chagas' disease. Moreover, the higher sensitivity of the PCR method observed in both the acute and chronic phases of the T. cruzi infections in the mice that we studied indicates that this approach should be useful in evaluating experimental drugs in T. cruzi-infected laboratory animals.