Date Published: February 13, 2015
Publisher: Public Library of Science
Author(s): Katherine L. Anders, Le Hong Nga, Nguyen Thi Van Thuy, Tran Van Ngoc, Cao Thi Tam, Luong Thi Hue Tai, Nguyen Thanh Truong, Huynh Thi Le Duyen, Vu Tuan Trung, Duong Thi Hue Kien, Marcel Wolbers, Bridget Wills, Nguyen Van Vinh Chau, Nguyen Dac Tho, Cameron P. Simmons, Scott B Halstead. http://doi.org/10.1371/journal.pntd.0003528
Abstract: BackgroundDengue control programs commonly employ reactive insecticide spraying around houses of reported cases, with the assumption that most dengue virus (DENV) transmission occurs in the home. Focal household transmission has been demonstrated in rural settings, but it is unclear whether this holds true in dense and mobile urban populations. We conducted a prospective study of dengue clustering around households in highly urban Ho Chi Minh City, Vietnam.MethodsWe enrolled 71 index cases with suspected dengue (subsequently classified as 52 dengue cases and 19 non-dengue controls); each initiated the enrollment of a cluster of 25–35 household members and neighbors who were followed up over 14 days. Incident DENV infections in cluster participants were identified by RT-PCR, NS1-ELISA, and/or DENV-IgM/-IgG seroconversion, and recent infections by DENV-IgM positivity at baseline.Principal Findings/ConclusionsThere was no excess risk of DENV infection within dengue case clusters during the two-week follow-up, compared to control clusters, but the prevalence of recent DENV infection at baseline was two-fold higher in case clusters than controls (OR 2.3, 95%CI 1.0–5.1, p = 0.05). Prevalence of DENV infection in Aedes aegypti was similar in case and control houses, and low overall (1%). Our findings are broadly consistent with household clustering of dengue risk, but indicate that any clustering is at a short temporal scale rather than sustained chains of localized transmission. This suggests that reactive perifocal insecticide spraying may have a limited impact in this setting.
Partial Text: Dengue represents a large and growing public health problem throughout the tropical and sub-tropical world, with nearly 100 million clinical cases estimated to occur annually . Dengue results from infection with one of four serotypes of dengue virus (DENV-1–4), transmitted between humans primarily by Aedes aegypti mosquitoes. Ae. aegypti are highly efficient vectors due to their preference for human blood meals, and for breeding, feeding and resting in and around domestic areas . Despite considerable research, the most advanced dengue vaccine candidate has only intermediate efficacy , making control of the mosquito vector an ongoing feature of dengue prevention and epidemic response for the foreseeable future.
This study of fine-scale clustering of dengue in urban HCMC found no excess risk of DENV infection among household members and neighbors of confirmed dengue cases, compared with equivalent contacts of controls, during the two-week period following enrollment of the index case. We did however observe a two-fold higher prevalence of DENV-reactive IgM, suggestive of recent DENV infection, in case clusters compared with control clusters. Overall, these results are broadly consistent with clustering of dengue risk around households, but indicate that any clustering is at a short temporal scale rather than sustained chains of localized transmission. These findings are relevant when considering the most effective way to deliver dengue control interventions in resource-limited endemic settings, and for understanding the spatial distribution of dengue risk in different settings.