Acute myocardial infarctions, strokes and influenza: seasonal and pandemic effects

Abstract

The incidence of myocardial infarctions and influenza follow similar seasonal patterns. To determine if acute myocardial infarctions (AMIs) and ischaemic strokes are associated with influenza activity, we built time-series models using data from the Nationwide Inpatient Sample. In these models, we used influenza activity to predict the incidence of AMI and ischaemic stroke. We fitted national models as well as models based on four geographical regions and five age groups. Across all models, we found consistent significant associations between AMIs and influenza activity, but not between ischaemic strokes and influenza. Associations between influenza and AMI increased with age, were greatest in those aged >80 years, and were present in all geographical regions. In addition, the natural experiment provided by the second wave of the influenza pandemic in 2009 provided further evidence of the relationship between influenza and AMI, because both series peaked in the same non-winter month.

Fig. 1.

The raw time-series data for the incidence of acute myocardial infarction (AMI) (top), ischaemic stroke (middle), and influenza (bottom) from January 1998 to September 2007. All units reflect the weighted frequency of discharges as reported from the Nationwide Inpatient Sample (NIS).

Fig. 2.

Attributable risk measures for influenza activity on acute myocardial infarction incidence for age groups ⩽65, >65, >70, >75, and >80 years. Note that the risk increases with age.

Fig. 3.

Attributable risk measures for influenza activity on regional acute myocardial infarction incidence for age groups ⩽65, >65, >70, >75, and >80 years. Note that the risk increases with age within each region. (a) Northeast, (b) Midwest, (c) South, (d) West.