Prospective evaluation and comparison of the direct agglutination test and an rK39-antigen-based dipstick test for the diagnosis of suspected kala-azar in Nepal.
BACKGROUND A firm diagnosis of visceral leishmaniasis (kala-azar) requires demonstration of the parasite in organ aspirates or tissue biopsy samples. The aim of this prospective study was to assess the diagnostic usefulness of non-invasive testing for antibody to the leishmanial antigen K39 by means of antigen-impregnated nitrocellulose paper strips adapted for use under field conditions. METHODS One drop of peripheral blood is applied to the hitrocellulose strip. Three drops of test buffer (phosphate-buffered saline plus bovine serum albumin) are added to the dried blood. The development of two visible bands indicates presence of IgG anti-K39. 323 consecutive patients with suspected kala-azar referred to two specialist units in India, and 25 healthy controls, provided fingerstick blood samples for the test. Spleen aspirates were taken from 250 patients. FINDINGS Kala-azar was confirmed by microscopy of spleen-aspirate smears in 127 patients. The K39 strip test was positive in all 127; the estimated sensitivity was therefore 100% (95% CI 98-100). Four patients had positive strip tests but negative aspirate smears; all four responded to treatment for leishmaniasis. 217 individuals, including the 25 healthy controls, 73 patients with malaria or tuberculosis, and 119 spleen-aspirate-negative patients who had presumed malaria or cirrhosis (79) or no final diagnosis (40), had negative strip-test results. None of the 119 aspirate-negative patients developed evidence of kala-azar during 3-6 months of follow-up. The estimated specificity of the strip test was 98% (95-100; 217/221). INTERPRETATION Detection of anti-K39 by immunochromatographic strip testing is a rapid and non-invasive method of diagnosing kala-azar, which has good sensitivity and specificity and is well suited for use in field conditions.