An estimation of potential vector control effect of gravid mosquito trapping in Fort Worth, Texas.

Author:Lee, Joon-hak


West Nile virus (WNV) first appeared in the United States in 1999 and since then has spread rapidly to the entire country. In Texas, the virus was first detected in 2002 and has become endemic, with occasional outbreaks and epidemics.

WNV is a public health concern in north central Texas, and local governments in the region play an essential role in WNV surveillance and response. Since the 2006 WNV outbreak, the number of human WNV cases has not been high enough to maintain public attention on WNV and accordingly, entomological WNV surveillance activities decreased until 2012, when the biggest WNV epidemic occurred in the region. During this interepidemic period, many entomological WNV surveillance pro grams in the region were eliminated or had limited activities due to an additive impact of low WNV activities and global economic hardship. The 2012 WNV epidemic, however, has increased public awareness on the unpredictable nature of WNV activities and enhanced public demand for a better and sustainable WNV monitoring and response system.

Integrated vector management (IVM) is the current best practice for WNV prevention and control, and entomological surveillance is an essential component of IVM. The surveillance provides real-time information on WNV activities for the public and also enables public health professionals to make a scientific judgment for appropriate public health intervention methods and the level necessary to control WNV

Entomological surveillance in the southern U.S., including Texas, focuses on the southern house mosquito, Culex quinquefasciatus Say. This mosquito species is a primary vector for WNV in the Southwestern U.S. and the principal vector in Texas (Andreadis, 2012). This WNV vector is the predominant species in urban areas where the use of CDC Gravid Traps is more effective in vector mosquito sample collection compared with other sampling tools (Lee & Kokas, 2004; Reiter, Jakob, Francy, & Mullenix, 1986). This trap is designed to collect gravid Cx. mosquitoes, particularly mosquitoes in the Cx. pipiens complex, which includes Cx. quinquefasciatus. Testing gravid WNV vector mosquitoes offers a higher sensitivity to detect WNV than non-gravid vectors. In addition, capturing gravid WNV vector mosquitoes prohibits them from ovipositing eggs that could increase the vector population. This innate vector control effect of entomological surveillance activities, however, has not been recognized.

Thus, our study explored a way to quantify the underappreciated mosquito control effect of mosquito surveillance activities on the primary WNV vector and estimated potential WNV vector control effect of the 2013 entomological surveillance activities in Fort Worth, Texas.


Life table characteristics of Cx. quinquefasciatus in previous studies were used to generate a model to predict the female progeny size of a gravid female adult mosquito. We conducted a literature search for available peer-reviewed information in PubMed and tabulated the identified information for review and comparison. From the literature review, a three-factor model was constructed: the total number of eggs laid, the proportion of eggs to become adult stage (emergence), and the proportion of females in the newly emerged mosquito population. In selecting a value for each factor, experimental conditions of each study were examined for similarity to natural conditions, in particular, host availability during the natural feeding time of Cx. quinquefasciatus and rearing temperatures that are similar to ambient temperatures during the summer in the Dallas-Fort Worth Metroplex. In addition, a conservative or generally accepted value was chosen if multiple options were available.

The total number of eggs laid by a gravid female mosquito over its lifetime was estimated by the linear model that was derived from the published information (Suman et al., 2011). The linear model was generated in SAS version 9.3 and the equation of this model was used to estimate the total number of eggs produced by a gravid female Cx. quinquefasciatus with a particular longevity.

Mosquitoes were collected weekly in CDC Gravid Traps with grass infusion to monitor WNV activities in Fort Worth, Texas, in 2013. The data were collected from the beginning of May to the end of October. Each month the number of Cx. quinquefasciatus was counted and for comparison, monthly abundance of the mosquito population was described as the number of female Cx. quinquefasciatus per trap night.

To obtain a representative proportion of gravid female Cx. quinquefasciatus in the city, a total of 4,274 female mosquitoes was collected from July 2 through August 6, and individually observed for mature eggs in the abdomen under a stereo microscope.

Monthly longevity of the mosquito population was estimated with an assumption that mosquito population abundance in a particular month is determined by longevity of parental mosquito population in the previous month--the longer lifetime a female has, the more offspring she produces. We estimated the longevity of Cx. quinquefasciatus based on our literature review and used this for the expected lifetime of the parental mosquito population in the previous month (X - 1) of a particular month (X) when abundance of a mosquito population was highest. Expected lifetime of the parental mosquito population in other months was estimated by multiplying the highest longevity to the ratio of the population abundance in a particular month to the highest...

To continue reading