Outcomes of Patients With Hospital-Acquired Influenza

Abstract

Background: Hospital-acquired influenza (HA flu) lacks a consensus definition. However, it is known to be associated with increased inpatient morbidity and mortality. Objective: To describe the clinical course of HA flu in a cohort population. Methods: A retrospective cohort study was conducted at a tertiary-care adult and pediatric teaching hospital. Patients with HA flu during 3 seasons, 2016 through 2019, were identified from medical record information based on timing of the onset of signs and symptoms and positive virologic testing >72 hours after admission. Influenza infection was confirmed by multiplex respiratory PCR, influenza A/B PCR, or direct fluorescent antibody tests. Chart review was performed to abstract patient demographics and comorbidities, length of stay, testing, and timing to antiviral administration as well as diagnosis of pneumonia, coinfections, and 30-day mortality. Escalation of care during hospitalization was defined as a new requirement of supplemental oxygen, invasive or noninvasive ventilation, and transfer to an intensive care unit. Results: During the 3 flu seasons, 132 patients were identified with HA flu; 76 (58%) were women, 6 (4.6%) were aged <18 years, and 126 (95.4%) were adults. Annually, HA-flu patients accounted for 5%–7.8% of all patients hospitalized with laboratory-proven influenza. The median duration between hospitalization and positive flu test was 15 days, and the median length of stay after influenza diagnosis was 6 days. Antiviral treatment was received by 96% of the patients. In total, 41 patients (31%) showed radiographic evidence for pneumonia. Coinfection with either a viral or bacterial pathogen was identified in 25% of the cases. In addition, 26% of the patients experienced an escalation of care, and 20 patients (15%) were transferred to the intensive care unit after HA flu diagnosis. Furthermore, 4 deaths (3%) were attributed to influenza during their hospitalization. Conclusions: HA flu was a frequent cause for escalation in care and was associated with a mortality rate substantially higher than is typically seen in community-based populations with influenza. Coinfection was mostly related to bacteremia and pneumonia, yet not all pneumonias had an associated microbiological diagnosis other than influenza, and there was no significant association between coinfection and mortality. Future work should explore more precise definitions for HA flu as well as its complications.Funding: NoneDisclosures: None