Numerical Distributions of Parasite Densities During Asymptomatic Malaria

Abstract

Background: Asymptomatic parasitemia is common even in areas of low seasonal malaria transmission, but the true proportion of the population infected has not been estimated previously because of the limited sensitivity of available detection methods.

Methods: Cross-sectional malaria surveys were conducted in areas of low seasonal transmission along the border between eastern Myanmar and northwestern Thailand and in western Cambodia. DNA was quantitated by an ultrasensitive polymerase chain reaction (uPCR) assay (limit of accurate detection, 22 parasites/mL) to characterize parasite density distributions for Plasmodium falciparum and Plasmodium vivax, and the proportions of undetected infections were imputed.

Results: The prevalence of asymptomatic malaria as determined by uPCR was 27.5% (1303 of 4740 people tested). Both P. vivax and P. falciparum density distributions were unimodal and log normal, with modal values well within the quantifiable range. The estimated proportions of all parasitemic individuals identified by uPCR were >70% among individuals infected with P. falciparum and >85% among those infected with P. vivax. Overall, 83% of infections were predicted to be P. vivax infections, 13% were predicted to be P. falciparum infections, and 4% were predicted to be mixed infections. Geometric mean parasite densities were similar; 5601 P. vivax parasites/mL and 5158 P. falciparum parasites/mL.

Conclusions: This uPCR method identified most infected individuals in malaria-endemic areas. Malaria parasitemia persists in humans at levels that optimize the probability of generating transmissible gametocyte densities without causing illness.

Keywords: PCR; Plasmodium falciparum; Plasmodium vivax; asymptomatic parasitemia; malaria.

Figure 1.

Erythrocytes containing dividing forms of Plasmodium vivax circulate freely, in contrast to Plasmodium falciparum, in which erythrocytes containing dividing forms are sequestered. This correction factor was applied to convert densities of peripheral blood P. vivax genomes to densities of parasites in relation to parasite age.

Figure 2.

A, Histogram showing Plasmodium vivax and Plasmodium falciparum densities. B, The kernel density function based on densities of 30–30 000 000 parasites/mL estimated for each species. C, The kernel density function based on all parasite densities estimated for each species. The arrows show approximate lower limits of detection by rapid diagnostic tests (RDTs) and polymerase chain reaction (PCR) analysis of capillary blood specimens (5 µL).

Figure 3.

The predicted geometric distributions of Plasmodium falciparum and Plasmodium vivax densities. Results are shown for the mean values (m) and standard deviations (s) of parameters (m,s) of (2.9,2.2) for P. falciparum and (3.3,1.6) for P. vivax.

Figure 4.

Histogram of the imputed parasite densities for Plasmodium falciparum and Plasmodium vivax. The height of the bars is scaled so that the sum of the heights equals 1. Results are shown for the mean values (m) and standard deviations (s) of parameters (m,s) of (2.9,2.2) for P. falciparum and (3.3,1.6) for P. vivax.

Figure 5.

Relationship between parasite density and risk of fever in this study.