Excess morbidity and mortality associated with seasonal influenza in patients with atrial fibrillation

Patients with atrial fibrillation (AF) may face an elevated risk of influenza-related complications, including cardiovascular and respiratory adverse outcomes. However, limited knowledge exists regarding the excess burden of hospitalizations and mortality due to influenza activity among patients with AF at a population level.

We sought to estimate the excess number of deaths and hospitalizations associated with influenza activity among individuals with AF in Denmark.

Data on weekly number of deaths and hospitalizations among patients with AF in Denmark were collected from the nationwide registries. Weekly estimates of influenza circulation were based on the proportion of positive influenza samples analyzed at all hospitals in Denmark during the study period. We used a time-series linear regression model to correlate the weekly estimates of influenza circulation with the weekly number of deaths and hospitalizations to estimate the annual excess number of deaths and hospitalizations associated with influenza circulation among patients with AF in Denmark. The model also incorporated data on weekly mean temperature in Denmark and restricted cubic spline terms to account for seasonal changes and trends over time. Confidence intervals were derived using block bootstrapping.

During the study period encompassing 8 influenza seasons from 2010-2018, an annual mean of 141,746 patients were living with AF. An annual mean of 25,181 samples were tested for influenza at all Danish hospitals with a mean proportion of positive samples of 7.9% (Figure 1). Based on our model, influenza activity was associated with an annual excess of 298 all-cause deaths (95% CI 126-564 deaths) and 86 cardiovascular deaths (95% CI 45-204 deaths) corresponding to 2.3% of all all-cause deaths (95% CI 1.0-4.3% deaths) and 1.8% of all cardiovascular deaths (95% CI 1.0-4.3% deaths) in patients with AF in Denmark (Figure 1). Influenza activity was also associated with an annual excess of 387 hospitalizations for pneumonia or influenza (95% CI 148-667 hospitalizations for pneumonia or influenza) corresponding to 5.5% of all hospitalizations for pneumonia or influenza (95% CI 2.3-8.8%) among patients with AF.

Based on the results of our model, approximately 2.3% of all deaths and 5.5% of all hospitalizations for pneumonia or influenza may be attributed to influenza activity among patients with AF in Denmark, suggesting substantial morbidity and mortality associated with influenza in patients with AF.Figure 1.