Predictors of Hospitalization and Superinfection in Viral Respiratory Tract Infections Between Influenza and Paramyxoviruses: The SUPERFLUOUS Study.

Abstract

Background: Viral respiratory tract infections (VRTIs) are among the most common diseases, but the risks of superinfection for different virus species have never been compared.

Methods: Multicenter retrospective study conducted among adults who tested positive for VRTIs with reverse-transcription polymerase chain reaction. We compared characteristics between influenza (A or B) and paramyxoviruses (respiratory syncytial virus, parainfluenza virus types 1 and 3, and human metapneumovirus) and identified predictors of superinfection and hospitalization.s.

Results: Five hundred ninety patients had VRTI, including 347 (59%) influenza and 243 paramyxovirus infections with comparable rates of superinfections (53% vs 60%). In multivariate analyses, the predictors of superinfections were age >75 years (adjusted odds ratio, 2.37 [95% confidence interval, 1.65-3.40]), chronic respiratory disease (1.79 [1.20-2.67]), and biological abnormalities, including neutrophil count >7000/µL (1.98 [1.34-2.91)], eosinophil count <50/µL (2.53 [1.61-3.98], and procalcitonin level >0.25ng/mL (2.8 [1.65-4.73]). The predictors of hospitalization were age >75 years old (adjusted odds ratio, 3.49 [95% confidence interval, 2.17-5.63]), paramyxovirus infection (2.28 [1.39-3.75]), long-term use of inhaled corticosteroids (2.49 [1.13-5.49]), and biological abnormalities, including neutrophil count >7000/µL (2.38 [1.37-4.12)] and procalcitonin level >0.25ng/mL (2.49 [1.23-5.02]). Kaplan-Meier survival curves showed that influenza-infected patients had a higher mortality rate than those with paramyxovirus infections (8.9% vs 4.5%, respectively; P = .02).

Conclusions: Our study revealed a high rate of superinfection (56%), not related to viral species. However influenza virus was associated with a poorer prognosis than paramyxoviruses, pleading for a broader and large-scale vaccination of individual at risk of VRTIs.