The IAMI Study: The Forgotten Influenza Vaccine Trial

There are studies that change practice because they should. There are studies that change practice because they are misinterpreted and thought to prove something they don’t. There are also studies that should change practice but, for some reason, do not. Dr. Folmsbee returns to Sensible Medicine today to discuss a 2021 study which has had little impact. He considers whether we are missing an important intervention or not.

I fall into Dr. Folmsbee’s "Camp One.” I think this was a well-done study that was positive despite being terminated prematurely. That said, I think Kuma’s discussions of the different ways of interpreting the study and why the results have not been integrated into our practice are instructive.

Adam Cifu

Not surprisingly, observational studies have shown that influenza is a risk factor for worsening cardiac disease. A RCT published in JAMA in 2020 did not find a benefit in giving high dose vs. standard dose to high cardiovascular risk patients. The Influenza Vaccination After Myocardial Infarction (IAMI Study), published in Circulation in 2021, sought to determine if there is a benefit to giving the influenza vaccine within 3 days of a cardiac event.

The study was a randomized, double-blind, intent to treat, placebo-controlled trial that included patients who had a STEMI or NSTEMI and had coronary angiography. Patients were randomized to the influenza vaccine or placebo which was given within 72 hours of the cardiac event. The trial was conducted in 8 countries (Sweden, Denmark, Norway, Latvia, UK, Czech Republic, Bangladesh, and Australia). Participants were excluded if they had already received influenza vaccine for the season or intended to be vaccinated.

The primary endpoint was composite of all-cause death, MI or stent thrombosis at 12 months after randomization.

The study was stopped early due to the decrease in influenza related illness during the COVID pandemic. 6696 patients were screened; 1439 (22%) already had been vaccinated or intended to be vaccinated; 1202 declined (18%). Ultimately, 2571 (38%) were randomized. The patients had an average age 59, 18.2% were women, 53% had STEMI, 45% had NSTEMI and 22% had DM. 74% of patients had an intervention with PCI while 23% had medical treatment only.

The primary composite end point occurred in 67 (5.3%) participants in the vaccine arm vs 91 (7.2%) in the placebo arm (HR 0.72, CI 0.52 - 0.99). The secondary outcome of all-cause mortality was 2.9% vs 4.9% (hazard ratio, 0.59 [95% CI, 0.39–0.89]; P=0.010) and CV death was 2.7% vs 4.5% hazard ratio, 0.59 [95% CI, 0.39–0.90]; P=0.014).

Pause for a second, that is a 40% relative reduction in not only CV death but all-cause death. A 2% absolute risk reduction. As effective as beta blockers post MI.

How do we interpret that in the setting of the trial stopping early? Sometimes, studies are stopped early because the intervention group is doing better than the placebo and it is considered unethical to continue the study. When this happens, it probably exaggerates the benefits of the intervention. The IAMI study was stopped because of external factors. If this study had been run the planned amount of time would the primary results have been less significant or even not significant?

If we are confident in the results of this study, then we should be pushing the vaccine with the same rigor as we do statins and aspirin. Of course, most cardiologists would recommend the influenza vaccine, but I contend that our universal recommendation for flu vaccines has negatively affected how aggressively we recommend the vaccine in this population (if we believe these results). The new ACS guidelines give the influenza vaccine a Class I, LOE A recommendation. When I talk to my colleagues in the CICU however, the vaccine is not recommended as aggressively as statins, ASA or even cardiac rehab. Shouldn’t the strong guideline recommendation along with impressive mortality benefit justify a payer pushing for a metric? Or at the very least a EMR pop-up?

I can think of two distinct camps when it comes to the appraisal of IAMI.

Camp One - Flu Vaccine works!

No further studies are necessary, we have observational and prospective trial data that shows influenza vaccine is effective. It is also overwhelmingly plausible (sorry RFK Jr). In the US, we also universally recommend the flu vaccine to everyone over six months of age. At this point it would be unethical to do any more studies, and we need to focus efforts on increasing uptake of the vaccine. The benefit seen in IAMI study is on par with other interventions we use routinely. More carrots (and sticks) are in order.

Camp Two - We need to focus our vaccination efforts

We should be wary of the IAMI study as it had to be stopped early and the primary outcome was atypical and confidence interval was wide. While the mortality benefit (both in all cause and CV death) is impressive, it is not plausible. In addition, given the current state of vaccine hesitancy, the US’s strategy of brute force universal recommendation is not working (less than 50% of adults got vaccinated in the 2023-2024 season). We should consider reevaluating our current universal recommendation to not only rebuild trust but to focus efforts on more vulnerable populations.

However, I suspect most fall into a third camp.

Camp Three - Who Cares - We already recommend the flu shot for everyone.

We need to break free from Camp Three.

If we join Camp One, we need to embrace these findings and push the flu shot more rigorously. Perhaps the flu shot needs to be pushed along with aspirin as they are being wheeled to the cath lab. Maybe we need more EMR nudges, aggressive metrics and influenza vaccination should be added to admission orders sets for ACS?

If we join Camp Two, we need to advocate for creative ways to reproduce and confirm the findings and better clarify when and who the flu shot should be given.

Either way, given the growing distrust found in our patients, we need to demand more from not only our health system but also science.

Caspian Kuma Folmsbee is a primary care provider in Chicago. He publishes at Kuma’s Substack.

Photo Credit: Mike Baumeister