The diagnosis of tegumentary leishmaniasis (TL) presents problems by the variable sensitivity and specificity of the tests, and the biological samples used are also invasive. Here, ELISA experiments were performed using paired TL patient urine and serum samples in reaction against the recombinant LiHyS protein, a predicted B cell epitope and parasite antigenic extract (SLA). Two hundred and five paired samples were used, which were provided by TL patients, healthy controls and patients with Chagas disease, leprosy, malaria or HIV-infected. An urine-based ELISA showed sensitivity values of 100%, 92.1%, and 82.5%, when rLiHyS, peptide and SLA were used, respectively
and specificity of 100%, 87.6%, and 79.5%, respectively. A serum-based ELISA showed sensitivity values of 100%, 99.3%, and 81.5%, using rLiHyS, peptide and SLA, respectively, and sensitivity of 100%, 96.5%, and 72.2%, respectively. In both cases, rLiHyS presented the better performance to diagnose TL by using patient serum and urine.