A dengue-2 epidemic causing dengue hemorrhagic fever (DHF) occurred in the contiguous border cities of Matamoros, Tamaulipas (Mexico), and Brownsville, TX, in 2005. In December, we conducted a household-based epidemiologic survey to determine the incidence and seroprevalence of dengue infection among Matamoros and Brownsville residents and to identify risk factors associated with infection. Antibodies to dengue were measured in 273 individuals. The estimated incidence of recent dengue infection was 32% and 4% among Matamoros and Brownsville participants, respectively. The estimated prevalence of past dengue infection was 77% and 39% among Matamoros and Brownsville participants, respectively. The Breteau index was 28 in Matamoros and 16 in Brownsville, reflecting an abundant winter population of Aedes mosquitoes. Discarded waste tires and buckets were the two largest categories of infested containers found in both cities. Our results underscore the risk for epidemic dengue and DHF in the Texas-Mexico border region.

Early diagnosis of dengue is challenging because the initial symptoms are often non-specific, viraemia may be below detectable levels and serological tests confirm dengue late in the course of illness. Identifying dengue early in the clinical course could be useful in reducing dengue virus transmission in a community. This study analyzed data from 145 laboratory-positive and 293 laboratory-negative dengue cases in Puerto Rico to define the early clinical features of dengue infection in children and adults and to identify the clinical features that predict a laboratory-positive dengue infection. Among children, rash and age were independently associated with laboratory-positive dengue infection. Rash in the absence of cough had a positive predictive value of 100% and a negative predictive value of 82.4% as a paediatric dengue screen. Among adults, eye pain, diarrhoea and absence of upper respiratory symptoms were independently associated with laboratory-positive dengue infection. No useful early predictors of dengue infection among adults were found. Using clinical features may promote earlier identification of a subset of paediatric dengue patients in Puerto Rico. Laboratory confirmation is still necessary for the accurate diagnosis of dengue infection.

From June 2005 to May 2006, a clinic-based enhanced surveillance system for dengue was implemented in a Puerto Rican municipality to provide a population-based measure of disease incidence and clinical outcomes. We obtained demographic and clinical information from suspected cases and performed serologic and virologic testing. We used World Health Organization (WHO) criteria to classify cases and applied a simplified case definition for severe dengue illness. There were 7.7 laboratory-positive cases of dengue per 1,000 population. The highest incidence, 13.4 per 1,000, was among 10 to 19 year olds. Of the 156 laboratory-positive cases, three patients (1.9%) met WHO criteria for dengue hemorrhagic fever, and 30 patients (19.2%) had at least one severe clinical manifestation of dengue infection. Our data suggest that in a community with endemic dengue, enhanced surveillance is useful for detecting symptomatic infections. Furthermore, the simplified case definition for severe dengue may be useful in clinic-based surveillance.