Parasitological diagnosis of canine visceral leishmaniasis: is intact skin a good target?

The objective of this study was to evaluate intact skin of seroreactive dogs as a possible target for the parasitological confirmation of canine visceral leishmaniasis (CVL). For this purpose, 394 dogs identified in serological surveys carried out in the metropolitan region of Belo Horizonte were studied. Blood was collected from all animals for serology and a tissue sample was obtained from two sites for parasitological diagnosis. Skin obtained from the ear and scapular region was simultaneously analyzed in 247 animals and lesion samples and ear skin were analyzed in 147 dogs. Leishmania parasites were isolated from 310 (78.7%) animals, and all isolates were identified as Leishmania chagasi. Simultaneous isolation from two sites was possible in 240 of the 310 animals, including ear and scapular skin in 151/247 (61.1%) and ear skin and skin lesions in 89/147 (60.5%). Ours results suggest that intact skin is one of the main target sites for the parasitological confirmation of CVL in seroreactive dogs.

Accuracy of an ELISA and indirect immunofluorescence for the laboratory diagnosis of American tegumentary leishmaniasis.

We compared the accuracy of ELISA and indirect immunofluorescence (IIF) using Leishmania braziliensis and L. major-like antigens and antigens from the Bio-Manguinhos kit for serological diagnosis of American tegumentary leishmaniasis (ATL). Cut-off values were defined by the area under the receiver-operating characteristic curve. For ELISA, statistical analyses revealed better accuracy [95.7% sensitivity, 100% specificity, 100% positive predictive value (PPV), 97.5% negative predictive value (NPV)] and reliability [intraclass correlation coefficient (ICC): 0.940] for L. braziliensis antigen compared with L. major-like antigen (78.7% sensitivity, 82.8% specificity, 73.3% PPV, 86.6% NPV, ICC: 0.833). ELISA optical density values obtained for both antigens were higher in mucosal forms of ATL. For IIF, sensitivity and specificity were 81.5 and 86.2%, respectively, for the L. braziliensis antigen, compared with 95.4 and 77.7% for the L. major-like antigen and 75.4 and 89.2% for the Bio-Manguinhos kit. No difference in the specificity of the IIF test was observed between antigens, whereas sensitivity differed between the L. braziliensis and L. major-like antigens and the Bio-Manguinhos kit. Parallel ELISA and IIF testing increased sensitivity, irrespective of the antigen employed, and serial testing increased overall specificity. These results support the recommendation that ELISA employing L. braziliensis antigen be used as a diagnostic tool for suspected cases of ATL in L. braziliensis-endemic areas.