This study compared the prevalence of trypanosome infections estimated by PFR-loop-mediated isothermal amplification (LAMP) with conventional polymerase chain reaction (PCR) tests. One hundred forty eight cattle blood samples were collected from Robanda village, Mara region, Tanzania in April 2008. In conventional PCR, four sets of primers, specific for the detection of Trypanosoma sp., Trypanosoma brucei rhodesiense, Trypanosoma vivax, and Trypanozoon, as well as a modified LAMP were used. Conventional PCR detected no infection or up to 8, 1, and 3 infections with Trypanosoma congolense savannah, Trypanozoon, and T. vivax, respectively, whereas LAMP detected additional 44 Trypanozoon positive cases. Our results clearly indicate that the prevalence of Trypanozoon spp. in cattle in Robanda village estimated by PFR-LAMP (30.4%) was significantly higher than the estimates by PCR assays (0.6-2%). As such, future studies should target epidemiological surveys of Trypanozoon and T. brucei rhodesiense infections in possible reservoir animals by LAMP to further elucidate the actual prevalence of these parasites.