Detection of mosquitos infected with plasmodium falciparum using the ELISA test

The feasibility and efficiency of the enzyme-linked immunosorbent assay (ELISA) for the detection of P. falciparum-infected mosquitos were studied. The relative lack of reproducibility of the test may be compensated for by the application of a correction factor that is based on the mean optical density (OD) of a positive reference sample. The corrected OD of every plate permits comparison between plates and between surveys. The critical value of positivity in the reaction is determined by using the distribution of optical densities in negative mosquitos; this value is chosen in such a way that 99.9% of the non-infected mosquitos have an OD below it (specificity = 99.9%). The sensitivity is determined by using a population of mosquitos experimentally infected and found positive by dissection of the salivary glands. This sensitivity, which varies with respect to the average yield of sporozoites in infected mosquitos, lies between 83% and 90%. The level of detection with the ELISA is fixed at a minimum of 1200 sporozoites per mosquito. The study investigated the variations in sensitivity and specificity and the efficiency of the test with respect to the transmission, the choice of the critical level of positivity, and the pooling of mosquitos. The efficiency of the test in the field was investigated in a longitudinal study of the dynamics of P. falciparum transmission in south-west Burkina Faso in 1985. The immunological sporozoite index (ISI) followed the same course of development as the parasitological sporozoite index (ISP) and was at its peak when the dissection showed a maximum prevalence; it was zero when the dissection revealed no sporozoite-positive gland whatsoever. In the first part of this study, intact field-caught mosquitos were tested in the ELISA. The ISI always gave higher percentages than the ISP when the latter was positive. On the other hand, the ISI was zero when the ISP was also zero, which indicates excellent specificity of the immunological test. In the second part, the mosquitos were tested in the ELISA after removal of the abdomen. Under these conditions the ISI was very similar to the ISP. The authors consider the differences observed between ISI and ISP in the first part to be partly due to the detection in some mosquitos of maturing oocysts containing sporozoite antigen. These are necessarily false positives, but the immunological index of mature oocysts is too small to explain the importance of the discrepancy on its own. The authors discuss the possibility of an underevaluation of the parasitological results and stress the importance of the size of the samples for a more precise immunological prevalence. It therefore seems that the ELISA is a specific technique whose sensitivity at least matches that of dissection.