Controlling Dengue Fever: An Environmental Approach

Dengue fever, a mosquito-borne virus, is an acute and untreatable disease that has plagued the tropical region of the world for centuries. But the past several decades have seen an alarming rise in cases, with incidence increasing thirty-fold in the past fifty years. Now with 50 to 100 million cases each year and over 2.5 billion people at risk of infection, dengue has become the target of increased research efforts aimed at prevention.

The Dengue Fever Virus

Similar to yellow fever and malaria, dengue is transmitted by the bites of infected mosquitoes. Fortunately, an individual’s first exposure to the virus is typically asymptomatic, with only five percent developing a more serious illness that shares characteristics with a severe flu. Following this initial infection, individuals develop an acquired immunity, but this immunity is no savior. In contrast, a subsequent infection by another of the four dengue serotypes – a distinct subspecies of the same virus – causes the body to mount a massive immune response. This response greatly increases the chance of the disease developing into dengue hemorrhagic fever, a fatal condition characterized by extreme internal bleeding.

This increased risk of severe complications from subsequent infections is compounded by the epidemic nature of dengue, which flares up every two to three years and infects significant portions of at-risk populations. The rising mortality rate from this disease, the risk of wider transmission, and the lack of effective treatments have spurred efforts to develop a vaccine. Yet, the existence of multiple serotypes and the risk of harmful immune responses have become seemingly insurmountable hurdles and have stalled further development of vaccine solutions. Thus, current efforts to control dengue have focused on vector-based approaches.

An epidemiological map of regions at risk of dengue fever transmission along with known epidemics from the past decade. Courtesy of WHO, Denguenet.

Vector Control

The mosquito Aedes aegypti is widely recognized as the major vector for dengue transmission. Primarily a mosquito of urban areas, Ae. aegypti typically breeds in domestic water-filled artificial containers, such as water tanks or bathing buckets. Because dengue outbreaks often occur in rapidly developing urban areas with limited water infrastructure and coincide with population bursts of Ae. aegypti, vector-control efforts target these species. These efforts focus on disrupting the mosquito’s habitats – artificial water containers – either by complete removal or by treating the water with particular insecticides or biological control agents. Such efforts have met with some success, especially during epidemic outbreaks. Curiously however, the virus consistently reappears, often within a year or two.

Seeking to explain this resurgence, Eliza Little – a joint master’s degree student with the Schools of Forestry and Public Health – has investigated the existence of a secondary vector that carries the virus during inter-epidemic periods, especially just after vector-control efforts are put in place.

Aedes aegyptiis, the primary vector for dengue fever, picking up and transmitting the virus through human bites. Courtesy of Wikipedia.

The Second Vector

While screening for potential vector candidates, Little discovered that Aedes mediovittatus – a mosquito in the same genus as Ae. aegypti – had been shown to feed on humans and was capable of transmitting dengue in laboratory conditions. And importantly, Ae. mediovittatus was found to transmit the virus vertically – from the mother mosquito to its offspring – at a far greater rate than Ae. aegypti, suggesting that it could act as a reservoir for dengue during inter-epidemic periods.

Unlike the primary vector, Ae. mediovittatus is not an urban dweller and instead is a native tree hole mosquito, residing close to or within arboreal zones and breeding in natural water containers. This difference in habitat combined with its ability to transmit and maintain dengue under laboratory conditions led Little to posit that “areas where Ae. aegypti and Ae. mediovittatus co-occur may be at higher risk for dengue persistence between epidemics.” Yet, there remained a problem: up until now, no efficient method has been developed for identifying the co-occurrence of the two vectors especially in specific developed areas over wide regions. Thus working at the interface of ecology, environmental health, and human health outcomes, Little set out to expedite the process of localizing the two vectors through the use of cutting-edge satellite mapping techniques and environmental analysis.

GIS Screening and Environmental Survey

Selecting the area of Patillas, Puerto Rico because it contains both vector species and experiences frequent dengue outbreaks, Little identified nine study sites that represented high densities of urban development at various altitudes. She then set out 90 BG-sentinel mosquito snares around the sites that captured mosquitoes using a non-toxic lure and a negative air-current trap. Mosquitoes were collected for three days, after which the entire catch was examined for the presence of either or both species of mosquito vectors. Finally, using satellite imagery of the region and specific sites obtained from an ultra-high resolution global satellite sensor, Little conducted an analysis of “fine-scale environmental drivers.”

The satellite data was analyzed using ENVI (Environment for Visualizing Images) and sites were extensively broken down by urban structure density and degree of vegetation. These images were combined with the mosquito population data and a statistical model to create a predictive map of areas of co-occurrence. After validating the map, Little demonstrated that the model could be used to accurately predict the rates of co-occurrence of the two vectors and thereby identify zones at high risk of consistent dengue epidemics.

Little now hopes that that these models can be generally applied to create risk maps for epidemic-prone regions, with which public health officials could create targeted interventions to mitigate and even halt the cycle of dengue outbreaks. Further, she indicates that such risk maps could be easily constructed for developing countries and therefore guide urban development to avoid the risk of serious dengue epidemics.

Future Control Efforts

When asked what she envisions for the future of dengue fever control, Little readily admits that “eradication is unlikely without vaccination” but suggests that controled efforts focused on high-vector densities of Ae. aegypti during epidemics and Ae. mediovittatus during inter-epidemic periods may greatly reduce the burden of the disease worldwide.

She is quick to add that there remain questions that even the most sophisticated models cannot yet answer – such as the heterogeneous nature of the population infected even during large epidemics – and that large-scale change will not be accomplished until socioeconomic, behavioral, and other drivers are incorporated into dengue control efforts.

Nevertheless, Little remains optimistic about the future of dengue control. When asked why she chose to focus on this particular disease, she answers, “Dengue presented itself as a unique disease that lies at the epicenter of so many changes: from urbanization to globalization to climate change. I find it really interesting that you have mosquitoes that are adapting to urban life and human behavior.” Her take on the big picture: “I really want to know if we can change human behavior by exploring what harms them.”

About the Author
Sunny Kumar is a junior Molecular, Cellular, and Developmental Biology major in Saybrook College. He conducts research in Professor Zhong’s lab investigating the role of genes in neurogenesis and asymmetric stem cell division as well as the Cappello lab investigating hookworm epidemiology.

The author would like to thank Eliza Little for her time and excellent research.

Further Reading
Troyo A, Fuller DO, Calderon-Arguedas O, Solano ME, and Beier JC (2009). Urban structure and dengue incidence in Puntarenas, Costa Rica. Singapore Journal of Tropical Geography 30:265-282.