RESUMEN
Leishmaniasis is a parasitic disease of medical importance widely distributed around the world. Several methods are available for diagnosis but molecular approaches are highly recommended. To improve the sensitivity of an existing hsp20 gene based-PCR protocol to detect Leishmania parasites, primers were redesigned to amplify a shorter fragment using a new PCR variant (PCR-hsp20S). In this study, we aimed at characterizing the performance of the new method on cutaneous clinical samples and compare it with the former PCR-hsp20. The analytical sensitivity of the PCR-hsp20S was evaluated using DNA dilutions (100-0.1 pg) from Leishmania donovani and resulted in the detection of 10 fg of parasitic DNA, the equivalent to 0.05 parasite genome. For the diagnostic evaluation, a panel of 127 human clinical samples was used to calculate the parameters of sensitivity, specificity, accuracy, and positive and negative predictive values of the PCR-hsp20S. Diagnostic sensitivity was 94% (CI, 89.1-99.7%) and the specificity of 100% (CI, 98.6-100%). The same panel was also evaluated with the PCR-hsp20 to calculate the agreement between both molecular assays and to compare their performances. While both hsp20-based PCRs showed a good agreement coefficient (kappa index = 0.6), the performance of the novel variant, PCR-hsp20S, was significantly higher in terms of sensitivity (P = 0.0001) allowing the accurate detection of a higher number of Leishmania-positive clinical samples. We endorse the use of the PCR-hsp20S over the former protocol for the detection of Leishmania parasites from cutaneous clinical samples. In addition, as an improved sensitivity was achieved with the new method merely through the amplification of a shorter gene fragment, this investigation constitutes an experimental proof of this concept.