Is routine dengue vector surveillance in central Brazil able to accurately monitor the Aedes aegypti population? Results from a pupal productivity survey.

OBJECTIVE: To assess how well the Aedes aegypti infestation rapid survey, Levantamento Rapido de Indice para Aedes aegypti (LIRAa), is able to accurately estimate dengue vector densities and target the most important breeding sites in Goiania, Brazil. METHODS: In February 2009, a pupal productivity survey was conducted in 2 districts of the city of Goiania, central Brazil. The results were compared to those of LIRAas conducted in the same districts during the months before and after the pupal productivity survey. RESULTS: In the pupal productivity survey, 2,024 houses were surveyed and 2,969 water-holding containers were inspected. Discarded small water containers most frequently contained immature Ae. aegypti. The most pupal-productive containers were elevated water tanks, roof gutters and water holding roofs. Combined, these three containers accounted for 70% of all pupae. In the two districts where the pupal productivity survey was conducted, the house indices were 5.1 and 4.6 and the Breteau indices were 5.9 and 6.0. In contrast, the two LIRAs conducted in the same two districts resulted in an average house index of 1.5 and Breteau index of 2.5, with discarded items identified as the most frequently infested container type. CONCLUSION: Both the LIRAa and the pupal productivity survey identified discarded items as being most frequently infested with immature stages of Ae. aegypti, but the pupal productivity survey showed that elevated containers produced the greatest proportion of Ae. aegypti pupae (a proxy measure of adult vector density) and that the values of the Stegomyia indices were substantially underestimated by LIRAa. Although both surveys differ considerably in terms of sampling method and manpower, in the case of this study the LIRAa did not accurately identify or target the containers that were the most important to adult mosquito production