Quantifying Effect of Geographic Location on Epidemiology of Plasmodium vivax Malaria

Andrew A. Lover

and Richard J. Coker

Author affiliations: Saw Swee Hock School of Public Health, National University of Singapore, Singapore (A.A. Lover, R.J. Coker); London School of Tropical Medicine and Hygiene, Bangkok, Thailand (R.J. Coker)

Abstract

Recent autochthonous transmission of Plasmodium vivax malaria in previously malaria-free temperate regions has generated renewed interest in the epidemiology of this disease. Accurate estimates of the incubation period and time to relapse are required for effective malaria surveillance; however, this information is currently lacking. By using historical data from experimental human infections with diverse P. vivax strains, survival analysis models were used to obtain quantitative estimates of the incubation period and time to first relapse for P. vivax malaria in broad geographic regions. Results show that Eurasian strains from temperate regions have longer incubation periods, and Western Hemisphere strains from tropical and temperate regions have longer times to relapse compared with Eastern Hemisphere strains. The diversity in these estimates of key epidemiologic parameters for P. vivax supports the need for elucidating local epidemiology to inform clinical follow-up and to build an evidence base toward global elimination of malaria.

The malaria parasite Plasmodium vivax, which received limited research attention for a number of decades, has moved onto the global health agenda for 2 key reasons. First, compared with the P. falciparum parasite, which also causes malaria, the P. vivax parasite is more difficult to eliminate because it has a broader geographic range and, unlike P. falciparum, has dormant liver stages. Second, P. vivax malaria has reemerged in previously malaria-free temperate regions, including Greece, Corsica, the Korean Peninsula, central China, and Australia (1–5). Moreover, increasing evidence indicates that P. vivax infections can be severe and fatal (6,7).
A large body of epidemiologic and clinical data supports the existence of discrete strains within each Plasmodium species. Much of the reliable data are from patients who were infected in temperate climates (study centers in the United States, United Kingdom, and Europe), and the parasites were from various locales. Tropical vivax strains generally cause a larger number of closely spaced relapses, and temperate strains have generally evolved to cause a long incubation period, enabling the survival of the parasite as dormant hypnozoites during colder months (8,9). Malariologists have long recognized that the incubation period for malaria varies by strain and geographic latitude (10).
Recent studies of temperate P. vivax strains in South Korea support these observations and suggest that chemoprophylaxis of infected patients may contribute to long latency (11,12). However, these observational studies could not estimate the time from infection to relapse or relapse periodicity because exact infection times were unknown. In a related primaquine dosing study, the relapse rate for 3 vivax strains was compared, but relapse time was not examined (13).
Numerous P. vivax classification schemes have been suggested, including temperate/tropical, temperate/subtropical/temperate, and northern/southern/Chesson-type. These schemes were suggested on the basis of observed clinical characteristics, but quantitative data to support these distinctions are sparse (14–16). Recent molecular and entomologic data suggests that P. vivax may consist of 2 subspecies, 1 in the Old World/Eastern Hemisphere (P. vivax vivax) and the other in the Americas (P. vivax collins) (17); however, research with other isolates has not confirmed these results (18). These 2 strains/subspecies show remarkable differences in their infectivity to different Anopheles spp. mosquitoes; however, the effect of these differences on the epidemiology of malaria in humans has not been demonstrated. In addition, at least 2 other subspecies, P. vivax hibernans and P. vivax multinucleatum, produce disease that is similar to currently circulating strains from the Korean Peninsula (19,20).
A large body of historical data exists from the deliberate laboratory infection of 2 populations: 1) patients receiving malariotherapy (intentionally induced malaria) for neurosyphilis and related disorders; and 2) healthy prisoners who participated in malaria drug testing trials. We used the experimental infection data from these controlled settings to explore and quantify the relationship between P. vivax parasite origin and characteristics of malaria caused by these subpopulations.