Is influenza associated with acute myocardial infarction (AMI), and does the influenza vaccination protect against it?
Evidence indicating a relationship between influenza and increased risk of AMI are abundant, but remain controversial given their observational nature. The authors used a case-control study design of Australian adults 40 years and older, who were evaluated during three influenza seasons between 2008 and 2010. They compared laboratory evidence of influenza through the collection of nasopharyngeal swabs and serologies among inpatients with AMI (cases) to outpatients without AMI (controls) before and after adjusting for several factors including influenza vaccination.
Of 559 participants, 34/275 (12.4%) were cases and 19/284 (6.7%) were controls. A total of 276 (49.4%) were vaccinated for influenza. During unadjusted analysis, the likelihood of influenza was much higher among cases with AMI than controls without AMI (odds ratio [OR], 1.97; 95% confidence interval [CI], 1.09-3.54). After adjustment, however, influenza was no longer significantly associated with AMI (OR, 1.07; 95% CI, 0.53-2.19). In contrast, influenza vaccination was significantly associated with a lower risk for AMI even after adjustment (OR, 0.55; 95% CI, 0.35-0.85). The authors estimated a vaccine effectiveness of 45% (95% CI, 15%-65%), which indicates the risk reduction of AMI potentially attributable to vaccination.
The authors concluded fairly strongly that their findings of the “protective” nature of influenza vaccination should renew discussions of expanding its use in younger populations who are at risk for AMI (e.g., 50-65 year olds).
This is a hot topic of discussion, and the paper has received a lot of attention in the lay press as the Northern hemisphere readies itself for another season of influenza and influenza vaccination. The paper is presented well, and the authors’ review of earlier evidence provides a nice summary. However, I am troubled by a lot of its conclusions and believe the authors have overstretched a bit. Most concerning to me is the lack of association between influenza and AMI after adjustment for confounding factors. The authors state that this might be because their study was underpowered, but the adjusted OR was very close to 1, suggesting that not even a trend was apparent. If this is the case, I wonder how the authors can write that influenza vaccination is ‘protecting’ against AMI since the entire mechanism of its action in protection would be by limiting influenza. My worry is that the authors have confused ‘association’ with ‘causality.’ That is, I do not deny that influenza vaccination is strongly associated with fewer AMIs. But this can be due to a whole host of reasons probably related to a ‘healthy behavior’ effect, as patients who get vaccinated are likely to be quite different than those who do not. I agree with their conclusion that the benefits of influenza vaccination need to be ‘further explored,’ but I am not certain that this study adds that much to this already controversial debate.
Brahmajee K. Nallamothu, M.D., F.A.C.C. (Disclosure)
Prevention/Vascular, General Cardiology, ChronicAngina