Comparison between microscopic examination, ELISA and quantitative buffy coat analysis in the diagnosis of falciparum malaria in an endemic population.

Monoclonal antibody-based ELISA and QBC (quantitative buffy coat analysis) were tested in two endemic areas with low and high incidence of malaria in Kanchanaburi Province, West Thailand with annual parasite incidence in 1992 of 119 and 5 per 1,000 population, respectively. The numbers of individuals positive by thick blood film examination (TBF) for P. falciparum with or without P. vivax, and P. vivax only were 82 and 69, respectively. The detection limit of ELISA was 10 parasites/10(6) red blood cells (RBC) (0.001% parasitemia). Of 1,095 individuals involved in the study at the beginning of the study, ELISA showed sensitivity, specificity, positive predictive value and negative predictive value of 78.1%, 94.9%, 72% and 98.1%, respectively. Nine of 18 (50%) TBF-positive but ELISA-positive individuals had parasitemia of less than 10 parasites/10(6) RBC. High and low incidence areas did not affect the validity of our result. Regression analysis showed good correlation between log parasitemia and ELISA percent OD increase (Y = 0 + 64.9*logX, r = 0.65), and agreement between TBF and ELISA results was 95.9%. In a fortnightly follow-up, in 82 TBF-positive individuals, both ELISA and TBF positive rates correlatively declined with agreement of 96.3%. With samples taken on the first day of the study, the TBF and QBC results were also correlated with agreement of 95.8% for P. falciparum, 95.6% for P. vivax. During 8 week follow-up involving altogether 191 samples, agreement between TBF and QBC results were 87.4% for P. falciparum. QBC detected more cases with P. falciparum infections but detected smaller number of cases with P. vivax infections.