Japanese encephalitis virus in Bangkok: factors influencing vector infections in three suburban communities.

An unexpected outbreak of Japanese encephalitis (JE) in Bangkok in 1985 led us to investigate the vector ecology of urban JE from January 1986 to June 1987 at three suburban sites that displayed a wide range of factors imputed to influence JE transmission. Culex tritaeniorhynchus Giles and Cx. gelidus Theobald, suspected vectors, comprised 71-96% of all mosquitoes collected by CO2-baited CDC traps at the three sites. Mean of mosquito abundance per two trap-nights per month ranged from 28 to 5,728 mosquitoes at the sites of lowest and highest abundance, respectively. Cx. tritaeniorhynchus yielded more JE isolates (n = 16) than Cx. gelidus (n = 7), but the minimum infection rates of the two species (number of JE isolates per 1,000 mosquitoes tested; MIR, 0.17 and 0.47, respectively) were comparable and covaried with vector abundance. Moreover, the proportion of sentinel pigs that had JE antibodies generally increased proportionately with vector abundance at the sites. Vector abundance was high in monsoon (May-October), moderate in transition (March-April and November-December), and low in dry (January-February) seasons. Mosquitoes collected in monsoon seasons yielded 96% of the JE isolates, whereas 4 and 0% of the isolates were obtained from transition and dry season collections, respectively. More pigs seroconverted in monsoon and transition seasons than in dry seasons. Indices of JE transmission activity (vector abundance, pig seroconversions, and MIRs) increased proportionately with rainfall. Despite higher indices at the site of greatest vector abundance than elsewhere, the risk of human infection appeared greatest at the site with moderate vector abundance because of its greatest human population density.