A Diagnostic Algorithm Based on Urine Antigen and Blood Polymerase Chain Reaction Testing for Visceral Leishmaniasis...

The diagnosis of visceral leishmaniasis (VL) in HIV patients relies on tissue aspiration, which is potentially life-threatening. Because antibody-based tests are unreliable for this group, Leishmania blood polymerase chain reaction (PCR) and urine antigen tests can serve as noninvasive alternatives. Using data from the PreLeisH cohort (ClinicalTrials.gov [NCT03013673]), the diagnostic accuracy of these two tests was determined individually, and an algorithm was designed for targeted tissue aspiration. Of the 490 PreLeish participants, 51 underwent diagnostic procedures for VL 118 times at different follow-up time points when clinically suspected. Of these, 48 (94.1%) patients with 102 (84.4%) procedures had complete data, including tissue aspiration, Leishmania kinetoplast DNA blood PCR, and urine antigen (KAtex [Kalon Biological Ltd., Guildford, United Kingdom]) results. The diagnostic performance of KAtex and blood PCR testing was assessed individually against tissue aspiration, and sequential and parallel algorithms were developed. Visceral leishmaniasis was confirmed via tissue microscopy in 55/102 (54%) episodes. The sensitivity and specificity of blood PCR testing were 98% and 64%, respectively; the sensitivity and specificity of KAtex testing were 93% and 81%, respectively. A sequential algorithm using KAtex first and restricting PCR to KAtex-negative suspects limited PCR testing to 42 (41.2%) episodes and tissue aspiration to 14 (13.7%) episodes, with sensitivity, specificity, and positive predictive value of 100%, 81%, and 86%, respectively. The parallel testing algorithm exhibited comparable performance but requires PCR testing for all suspects. The sequential KAtex and PCR testing algorithm, combined with targeted tissue aspiration, met the WHO target product profile, except for specificity. This approach supports efficient resource use and reduces the need for invasive procedures.