Laboratory-confirmed influenza infection and acute myocardial infarction among United States senior Veterans

Abstract

Background Previous studies established an association between laboratory-confirmed influenza infection (LCI) and hospitalization for acute myocardial infarction (AMI) but not causality. We aimed to explore the underlying mechanisms by adding biological mediators to an established study design used by earlier studies. Methods With data on biomarkers, we used a self-controlled case-series design to evaluate the effect of LCI on hospitalization for AMI among Veterans Health Administration (VHA) patients. We included senior Veterans (age 65 years and older) with LCI between 2010 through 2015. Patient-level data from VHA electronic medical records were used to capture laboratory results, hospitalizations, and baseline patient characteristics. We defined the “risk interval” as the first 7 days after specimen collection and the “control interval” as 1 year before and 1 year after the risk interval. More importantly, using mediation analysis, we examined the role of abnormal white blood cell (WBC) and platelet count in the relationship between LCI and AMI to explore the thrombogenic nature of this association, thus potential causality. Results We identified 391 hospitalizations for AMI that occurred within +/-1 year of a positive influenza test, of which 31 (31.1 admissions/week) occurred during the risk interval and 360 (3.5/per week) during the control interval, resulting in an incidence ratio (IR) for AMI admission of 8.89 (95% confidence interval [CI]: 6.16–12.84). In stratified analyses, AMI risk was significantly elevated among patients with high WBC count (IR, 12.43; 95% CI: 6.99–22.10) and high platelet count (IR, 15.89; 95% CI: 3.59–70.41). Conclusion We confirmed a significant association between LCI and AMI. The risk was elevated among those with high WBC or platelet count, suggesting a potential role for inflammation and platelet activation in the underlying mechanism.