Accuracy evaluation of diagnostic methods for visceral leishmaniasis in adult patients with and without HIV infection: Clinical management implications

Visceral leishmaniasis (VL) poses a serious health threat, particularly when untreated, necessitating accurate diagnosis. While the gold-standard method involves identifying amastigotes in bone marrow aspirate (BMA), this procedure is invasive and occasionally contraindicated. Additionally, when VL is associated with HIV infection the serologies accuracies could be affected. This study aims to evaluate and compare diagnostic methods for VL in patients with and without HIV coinfection. We enrolled prospectively 127 consecutive adult VL patients, 48 (37.8%) of whom had HIV coinfection, in Brazil's Midwestern region, where VL is endemic. Parasitological examination served as the reference standard for accuracy analysis, with index tests including immunofluorescent antibody test (IFAT), immunochromatographic test with rK39 protein (rK39-ICT), and blood polymerase chain reaction (PCR). Specificity assessment involved 430 healthy blood donors from the same endemic area. Ninety-two patients had parasitologically confirmed VL. Among HIV-uninfected patients, rK39-ICT exhibited sensitivity comparable to PCR (93.6%; 95% CI: 83.6–100 vs. 97.8%; 95% CI: 93.6–99.2, respectively) and superior to IFAT (71.1%; 95% CI: 57.9–84.3). However, in HIV-infected patients, rK39-ICT sensitivity was notably lower than PCR (40.0%; 95% CI: 22.5–57.5 vs. 97.4%; 95% CI: 92.5–98.9) and similar to IFAT (67.5%; 95% CI: 52.9–82.0). Combining two serological tests in parallel identified 82.1% of parasitologically confirmed VL cases, with a negative likelihood ratio significantly lower than either test alone. No test achieved a specificity of 90%, and there were no significant differences in specificity observed among the index tests. The positivity rate of parasitological examination in the 127 VL patients was higher in HIV-infected compared to HIV-uninfected patients, 91.3% (95% CI: 83.2–99.4) versus 67.6% (95% CI: 56.9–78.3), respectively. These findings underscore the necessity of accounting for HIV infection when choosing VL diagnostic methods. Although rK39-ICT provides reliable results in HIV-uninfected patients, BMA examination remains crucial for accurate diagnosis in individuals with HIV/AIDS. In cases where bone marrow aspiration is contraindicated, employing IFAT and rK39-ICT in parallel could be considered, as the occurrence of both positive results is uncommon in healthy individuals from endemic areas.