Prevalence of co-infection between respiratory syncytial virus and influenza in children

Background: Respiratory syncytial virus (RSV) and influenza have varying degree of seasonal overlap. Objective: To determine the prevalence of co-infection of RSV and influenza compared to the prevalence of those infections independently when both are in season. Methods: This was a retrospective cross-sectional study of children evaluated between July 2010 and June 2013 for viral respiratory infection using multiplex PCR. Seasonality was defined retrospectively as weeks when > 2% of the total annual positive tests were obtained and was calculated for influenza A, influenza B, and RSV independently. Periods of overlapping seasonality of RSV and influenza A and RSV and influenza B were identified. The expected incidences of co-infection were modeled as the product of the incidences of the individual viruses. Results: 13,664 specimens were sent for PCR during the study period. Over all 3 seasons, RSV overlapped with influenza A and B for 22 and 18 weeks, respectively; in 2011-12, RSV overlapped with neither influenza A nor B. Based on modeling, there were 6-7 fold fewer cases of RSV/influenza co-infection observed than expected: RSV/influenza A 77 vs. 12, (p <= 0.001; RSV/influenza B 76 vs. 11 (p <= 0.001). Conclusions: The observed incidence of co-infectivity of RSV and influenza was significantly less than the expected incidence even when both were co-circulating. In light of these data, it may be reasonable to forgo rapid influenza testing or empiric antiviral treatment for children whom rapid RSV testing is positive and who are at low risk of influenza-related complications, especially in times of antiviral therapy shortages. (C) 2016 Elsevier Inc. All rights reserved.