Comparison of real-time PCR and conventional PCR with two DNA targets for detection of Leishmania (Leishmania) infantum infection in human and dog blood samples.

Zoonotic visceral leishmaniasis (VL) is endemic in northwestern Iran. Real-time PCR, conventional PCR, and the direct agglutination test (DAT) were used to diagnose Leishmania infantum infection in blood samples from 100 domestic dogs and 100 humans. Based on clinical evaluation, 82 humans and 72 dogs from the endemic area were categorized as having asymptomatic infection, DAT positive with no clinical signs of VL, or symptomatic infection, DAT positive with at least one sign of VL. Eighteen human samples containing no Leishmania antibodies (DAT(-)) and 28 dog DAT(-) sera from non-endemic areas with no history of VL constituted negative controls. All 46 DAT(-) samples were also negative by Dipstick rK39. Bone marrow material was used for parasitological examinations in symptomatic VL, and peripheral blood samples were used for detection of L. infantum infection using conventional PCR and real-time PCR in non-symptomatic subjects. Two DNA targets (ITS1 kDNA) were used for conventional PCR. L. infantum antibodies in sera were detected by DAT. Parasitemia was measured by real-time PCR targeting kDNA using Taqman Assay. All 72 (100%) symptomatic (38/38) and asymptomatic (34/34) dog DAT(+)samples, 45 of 48 (93.8%) symptomatic human DAT(+) samples, and 32 of 34 (94.1%) human asymptomatic cases were identified by real-time PCR. The mean (59.19 vs 12.38 parasite equivalents/mL of blood) and median (16.15 vs 1 parasite equivalents/mL of blood) ranges of parasitemia were higher in dogs than in humans (P<0.05). The highest agreement was obtained between real-time PCR and DAT (99% in dogs and 95% in humans). Sensitivity of 100% and 93.9%, specificity of 96.4% and 100%, positive predictive values of 98.6% and 100%, and negative predictive values of 100% and 78.3% were found by real-time PCR for dog and human samples, respectively.