West Nile virus infection rates in pooled and individual mosquito samples.

The detection of West Nile virus (WNV) in mosquitoes by real-time RT-PCR provides valuable information on the epidemiology of the virus and identifies mosquito species that are potential vectors. Testing sets of pooled mosquitoes of the same species is logistically the easiest and most cost-effective approach for WNV testing; however, little information is available on how the results of small pooled sets relate to those of testing individual mosquitoes. During the 2002 outbreak, we compared pooled and individual samples of two mosquito species (Culex pipiens and Culex restuans) collected from three Health Unit regions in Ontario, Canada. Significantly more Cx. restuans were infected with WNV compared to Cx. pipiens. We show that with pool sizes of five individuals both MIR (minimum infection rates) and MLE (maximum likelihood estimation) values were acceptable in estimating infection rates.