2627. Dynamics of Respiratory Viral Co-infections: Predisposition for and Clinical Impact of Viral Pairings in Children and Adults

Background The clinical relevance of respiratory viral co-infections is unclear. Few studies determine epidemiology and impact of specific co-infection pairings. Here we assess the dynamics of respiratory viral co-infections, determine any predisposition for specific pairings to occur and evaluate resulting clinical impact on hospitalization. Methods We reviewed respiratory viral panel results collected at The Cleveland Clinic between November 2013 to Jun 2018. Monthly prevalences, mono-infections and co-infections of 13 viral pathogens were tabulated. Employing a mathematical model which utilized each individual virus’ co-infection rate and prevalence patterns of concurrent circulating respiratory viruses, we calculated an expected number of occurrences for 132 viral pairing permutations. Expected vs observed co-infection occurrences were compared using binomial tests. For viral pairings occurring at significantly higher prevalence than expected, logistic regression models were used to compare hospitalization between patients with co-infection to ones with mono-infection. Results Of 30,535 respiratory samples, 9,843 (32.2%) samples were positive for at least 1 virus and 1,018 (10.82%) were co-infected. Co-infections occurred in 18% of pediatric samples and only 3% of adult samples (P < 0.001). Adenovirus C (ADVC had the highest co-infection rate (68.3%) while influenza B had the lowest (10.07%). Using our model, ADVC – rhinovirus (HRV), RSVA - HRV, and RSVB - HRV pairings occurred at significantly higher prevalence than expected (P < 0.05). In children, HRV-RSVB co-infection were significantly less likely to be hospitalized than patients with HRV mono-infections (ORmono/co = 2.3; 95% CI 1.1 to 4.7; P = 0.028). Additionally, HRV - ADVC co-infected children were less likely to be hospitalized than either HRV (ORmono/co = 3.3; 95% CI 1.6 to 6.8; P < 0.001) or ADVC (ORmono/co = 1.9; 95% CI 1.1 to 3.2; P = 0.024) mono-infected children. Regardless of the infecting virus, children were less likely to be hospitalized than similarly-infected adults. Conclusion Respiratory viral co-infections are largely a pediatric phenomenon. Select viral pairings occur more often than predicted by our model, many of which are associated with altered severity of resultant disease. Disclosures All authors: No reported disclosures.