Evaluation of four rapid diagnostic tests for canine and human visceral Leishmaniasis in Colombia

BACKGROUND: Leishmaniasis caused by different species of Leishmania affect 98 countries worldwide. Visceral Leishmaniasis (VL) is the mortal clinical presentation of the disease that causes the dead to more than 90% of the patients who suffer it. The diagnosis of VL is made by the direct observation of the parasite in bone marrow, spleen and/or liver aspirates that requires complex proceedings. Therefore, serum samples are submitted to Indirect Immunofluorescence to identify the presence of anti-Leishmania antibodies. Despite the variability in the diagnostic performance of the Immunochromatographic Tests (ICTs), there are many evidences that suggest that ICTs can be used for epidemiological screening. However, in Colombia there are not any evidence about the performance of the ICTs for VL diagnosis, both for human and canine serum samples. Therefore, this study evaluated the diagnostic performance of 4 ICTs for VL (2 ICTs in human sera and 2 ICTs in canine sera) in samples from endemic areas of Colombia. METHODS: We selected a total of 156 human serum samples (82 positive and 74 negative for VL) and 126 canine serum samples (71 positive and 54 negative) diagnosed by in house Indirect Immunofluorescence (IIF). The samples were submitted to the ICTs following the manufacturers’ instructions. Statistical analysis was performed to evaluate the diagnostic performance of each ICT in comparison with the IIF. PCR for HSP70 gene and sanger sequencing was performed in samples with negative results for both ICTs. RESULTS: The sensitivity (S) of both ICTs for human samples (Ad-bio Leishmania IgG/IgM Combo Rapid Test and Kalazar Detect™) was 91.5% and specificity (E) were 93.2 and 89.2% respectively, while for the ICTs tested on canine samples (Kalazar Detect™ Rapid Test, Canine and DPP® CVL rapid test) we found S values between 82.9 and 85.7% and E values between 79.6 and 92.6%. We found L. infantum by PCR and sequencing in 2 human samples, and L. braziliensis and L. amazonensis in canine serum samples that were negative by both ICTs. CONCLUSIONS: We conclude that both tests evaluated on human samples have a similar diagnostic performance, while the Kalazar Detect™ Rapid Test, Canine showed a better diagnostic performance than the DPP® CVL rapid test evaluated on canine samples. Also, we suggest that it is necessary to design tests with antigens of the circulating strains to increase its diagnostic utility.