Factors and outcomes associated with influenza virus loads in hospitalized patients

Abstract

Annual influenza outbreaks constitute a substantial economic and public health burden accounting for 3-5 million hospitalizations and up to 645 000 deaths worldwide. Two influenza A subtypes (H1N1pdm09 and H3N2) and two influenza B lineages (B/Yamagata and B/Victoria) circulate around the globe causing seasonal outbreaks. In this study, we aimed at performing clinical and virologic characterization of patients hospitalized with severe influenza disease during the 2018-2019 and 2019-2020 seasons. We enrolled 1654 patients within 72 hours of being hospitalized with acute respiratory symptoms and who had symptoms consistent with influenza within the seven days before admission. The study was approved by IRB. Demographic and clinical data were collected from individual patients along with nasopharyngeal (NP) and oropharyngeal (OP) swabs. The samples were screened for influenza A, B and their subtypes/lineages using reverse transcriptase real-time PCR (RT-PCR) following RNA extraction. The agreement between sample types was assessed. The viral titers were compared for patient demographics, respiratory symptoms, patient health characteristics, and clinical outcomes. NP and OP samples had minor agreement for influenza A (kappa=.464, p<.001) and moderate agreement for influenza B (kappa =.732, p<.001). NP swabs had better sensitivities in detecting both influenza A (91.2% vs. 70.4%) and B (83.3% vs. 61.5%) viruses and yielded higher viral loads compared to OP. Higher viral load was associated with asthma for influenza B/Victoria, and runny nose for influenza A and A/H1N1. A lower viral load, however, was found to be associated with ARDS (Acute Respiratory Distress Syndrome) and bacterial-co-infection for influenza A, and oxygen supplementation for A/H1N1. In conclusion, NP and OP swabs are complementary in detection influenza, however, with limited resources NP swabs are preferred. Further studies with a more appropriate sample size are needed to validate the relationship between viral load and asthma as well as vaccination status.