Fine Scale Spatiotemporal Clustering of Dengue Virus Transmission in Children and Aedes aegypti in Rural Thai Villages

Abstract

BackgroundBased on spatiotemporal clustering of human dengue virus (DENV) infections, transmission is thought to occur at fine spatiotemporal scales by horizontal transfer of virus between humans and mosquito vectors. To define the dimensions of local transmission and quantify the factors that support it, we examined relationships between infected humans and Aedes aegypti in Thai villages.Methodology/Principal FindingsGeographic cluster investigations of 100-meter radius were conducted around DENV-positive and DENV-negative febrile “index” cases (positive and negative clusters, respectively) from a longitudinal cohort study in rural Thailand. Child contacts and Ae. aegypti from cluster houses were assessed for DENV infection. Spatiotemporal, demographic, and entomological parameters were evaluated. In positive clusters, the DENV infection rate among child contacts was 35.3% in index houses, 29.9% in houses within 20 meters, and decreased with distance from the index house to 6.2% in houses 80–100 meters away (p<0.001). Significantly more Ae. aegypti were DENV-infectious (i.e., DENV-positive in head/thorax) in positive clusters (23/1755; 1.3%) than negative clusters (1/1548; 0.1%). In positive clusters, 8.2% of mosquitoes were DENV-infectious in index houses, 4.2% in other houses with DENV-infected children, and 0.4% in houses without infected children (p<0.001). The DENV infection rate in contacts was 47.4% in houses with infectious mosquitoes, 28.7% in other houses in the same cluster, and 10.8% in positive clusters without infectious mosquitoes (p<0.001). Ae. aegypti pupae and adult females were more numerous only in houses containing infectious mosquitoes.Conclusions/SignificanceHuman and mosquito infections are positively associated at the level of individual houses and neighboring residences. Certain houses with high transmission risk contribute disproportionately to DENV spread to neighboring houses. Small groups of houses with elevated transmission risk are consistent with over-dispersion of transmission (i.e., at a given point in time, people/mosquitoes from a small portion of houses are responsible for the majority of transmission).

A new paradigm for Aedes spp. surveillance using gravid ovipositing sticky trap and NS1 antigen test kit

“…The study in Thailand using RT-PCR to detect the dengue virus in mosquitoes also showed a positive association between infected Ae. aegypti and dengue-infected children [68]. That study demonstrated the occurrence of an infected mosquito prior to the reporting of the index case(s).…”

Temporal Correlations Between Mosquito-Based Dengue Virus Surveillance Measures or Indoor Mosquito Abundance and Dengue Case Numbers in Mérida City, México

“…Improved methods for detection of DENV RNA or antigen in mosquitoes (Bangs et al 2007, Chao et al 2007, Chen et al 2010, Tan et al 2011, Muller et al 2012, Voge et al 2013) provide new opportunities for mosquito-based DENV surveillance, and studies focusing entirely or in part on dengue patient premises have shown promise with regards to the potential for linking DENV infections in Ae. aegypti to human dengue cases in space and time (Pinheiro et al 2005, Urdaneta et al 2005, García-Rejón et al 2008, Guedes et al 2010, Yoon et al 2012). Despite a substantial effort – including monthly indoor backpack aspiration mosquito collections from a large number of homes (typically >125 per month) scattered over ~30 neighborhoods in Mérida City, and with nearly 5,800 Ae.…”

Assessing the epidemiological effect of wolbachia for dengue control

“…Active surveillance of human infections can be efficiently achieved using geographical cluster sampling around dengue index cases. 25,26 Here, ‘index case’ refers to the laboratory-diagnosed clinical dengue case that initiates a cluster investigation within a geographically restricted area around the home of a person with a documented DENV infection. Geographical cluster investigations could be used to compare the fine-scale spatial signature of DENV transmission in areas with and without Wolbachia (Figure 2).…”

“…aegypti mosquitoes is challenging because of the low infection rates (typically ~0.1%) in the adult females across the population, although infection rates can be higher in locations of geographical cluster investigations. 25 Because mosquitoes that test positive for virus are not necessarily infectious, the proportion of DENV-infected mosquitoes does not directly translate into an estimate of virus transmission unless virus disseminated from the mosquito midgut or in saliva is also assayed, and even this approach is limited by the sensitivity of assays and variation of in vitro saliva collections. Nonetheless, a successful intervention is expected to reduce the incidence of viremic and infectious humans and, therefore, similarly reduce the incidence of DENV infection in mosquitoes in areas where Wolbachia infection predominates.…”