Sentinel surveillance of influenza and other respiratory viruses, Brazil, 2000-2010.

There are scanty data on the epidemiology of influenza and other respiratory viruses in South America and Brazil. The aim of this study was to summarize the data from the Brazilian surveillance system of influenza and other respiratory viruses and discuss the patterns of viral circulation. The system is based on detecting cases of influenza-like illness in sentinel sites and weekly collection of five nasopharyngeal secretions samples, which are processed in state public health laboratories for respiratory viruses by indirect immunofluorescence assay. Data from 2000 to 2010 were described over time, by region, gender, and age group, and an analysis of Spearman correlation was performed between monthly influenza detection and rainfall and temperature data in two state capitals with the highest number of positive samples, one from the northeast region (Maceió) and other from the southern region (Curitiba). There were 3,291,946 visits for influenza-like illness; of these, 37,120 had samples collected and 6421 tested positive: 1690 (26%) influenza A, 567 (9%) influenza B, 277 (4%) parainfluenza 1, 571 (9%) parainfluenza 2, 589 (9%) parainfluenza 3, 742 (12%) adenovirus, and 1985 (31%) respiratory syncytial virus. Overall, increased activity of respiratory syncytial virus was observed from March to June, preceding the peak of influenza activity, from May to August, but with regional differences. In Maceió, there was a weak correlation between temperature and influenza detection (ρ=0.05), but a moderate positive correlation between rainfall and influenza detection (ρ=0.36). In Curitiba, a high correlation was observed between the decrease in temperature and rainfall and the increase in influenza detection (ρ=-0.83 and -0.78 respectively). These data are important to guide public health control measures as the best time for influenza vaccination and use of antivirals.