Prior influenza vaccination was associated with roughly half the excess risk of acute myocardial infarction and stroke following breakthrough influenza in a nationwide Danish study.
The self-controlled case series study, published in Eurosurveillance, included 1,221 patients aged 40 years or older with a first-ever hospital admission for acute myocardial infarction or stroke within a 1-year observation window surrounding a PCR-confirmed influenza infection. Researchers linked national registry data using unique personal identifiers assigned to all Danish residents.
Primary findings
Cardiovascular events clustered in the 7 days following a positive influenza test, with an adjusted incidence rate ratio of 3.5 for acute myocardial infarction and stroke combined. The risk was higher for acute myocardial infarction (incidence rate ratio [IRR], 4.7) than for stroke (IRR, 2.9).
Among 1,231 influenza episodes, 610 (50%) occurred in patients who had received an influenza vaccine at least 14 days prior to testing positive within the same influenza season.
When stratified by vaccination status, the adjusted IRR was 4.7 in unvaccinated episodes vs 2.4 in vaccinated episodes, indicating that prior vaccination was associated with about half the excess cardiovascular risk. This difference reflected a statistically significant interaction between vaccination status and postinfection risk.
Although vaccinated episodes showed lower risks for both outcomes, differences between groups were not statistically significant when acute myocardial infarction and stroke were analyzed separately.
Timing of risk
The highest risk occurred in the first 3 days following infection (IRR, 5.2), then declined, returning to baseline by days 15 to 28. Vaccinated patients had lower risk through day 7, with no difference observed in later periods.
Study population
The median age of patients was 75 years; 54% were male. Strokes accounted for 65% of events and acute myocardial infarction for 35%. Within 30 days of admission, 8% of patients had died.
Most infections were caused by influenza A (77%). Among vaccinated episodes, 94% involved inactivated vaccines, including a small proportion of high-dose formulations.
Study design
The self-controlled case series design compared each patient’s risk during a predefined 7-day postinfection period with their own baseline risk during other times, inherently controlling for time-invariant factors such as comorbidities and socioeconomic status. Models were adjusted for calendar month to account for seasonal variation.
Additional analyses
Subgroup and sensitivity analyses showed consistent directional findings, with lower excess risk in vaccinated episodes.
Negative control analyses showed no association between influenza and retinal detachment or upper limb fracture. A separate analysis using Campylobacter infection showed increased cardiovascular risk but no modification by influenza vaccination, supporting the specificity of the main findings.
Limitations
The study used specimen collection date rather than symptom onset, which may have underestimated risk. It could not account for yearly variation in vaccine effectiveness and likely underrepresented milder influenza infections. Residual confounding from co-circulating pathogens or environmental factors could not be excluded.
The authors declared no competing interests.
Source: Eurosurveillance