Coffee consumption had a complex relationship with arrhythmias, a randomized trial with real-time electrocardiography patch monitoring showed.

Premature atrial contractions weren’t caused by coffee consumption, whereas premature ventricular contractions (PVCs) were, reported Gregory Marcus, MD, of the University of California San Francisco, at the American Heart Association (AHA) virtual meeting.

PVCs were 54% more likely on coffee-drinking days (P=0.001), with more than a doubling in risk with one or more cups of joe (RR 2.20, 95% CI 1.24-3.92 after adjustment for day of week). Those with genetics expected to yield the fastest caffeine metabolism showed the highest relative risk of PVCs, with significant trends for both CYP12 and polygenic risk scores.

Atrial arrhythmias weren’t elevated overall during periods randomized to allow coffee. Supraventricular tachycardia was actually reduced 12% per coffee drink compared with no coffee (P=0.028).

“Overall, the acute yet everyday physiologic effects of coffee are complex,” Marcus said. “These data add to the growing evidence that those with supraventricular tachycardias or atrial fibrillation…or at risk for those diseases should not necessarily avoid coffee.”

It’s a common question in practice for which there hasn’t been a clear answer to give patients, as data have been conflicting, noted AHA press conference study discussant Sana M. Al-Khatib, MD, MHS, of Duke University Medical Center in Durham, North Carolina.

However, the relatively young, normal body-weight population studied without prior arrhythmias is “really not representative of the average patient that we see in clinical practice,” she cautioned.

Small numbers didn’t allow for a good look at non-sustained ventricular tachycardia. Given that limitation along with use of surrogate endpoints, “I think it’s going to be critically important for future researchers to try to validate these results and look at harder endpoints during a longer follow-up of time,” Al-Khatib said.

Press conference moderator Elaine Hylek, MD, MPH, of Boston University, who introduced her comments with the caveat that she is biased as a coffee drinker, agreed that, despite the small sample size of 100 patients, she was “somewhat reassured that it didn’t appear to induce any atrial arrhythmias. If anything, it seemed like individuals were somehow increasing their physical activity. Hopefully that would translate to better cardiovascular health, weight, blood pressure.”

Coffee consumers averaged 1,058 more steps per day (P=0.0010) but got 36 less minutes sleep per night (P<0.001) after adjusting for day of week, with each additional drink linked with 587 additional steps per day and 18 fewer minutes of sleep (both P<0.001).

With regard to the PVCs, “we don’t have to get too, too worried” for normal hearts in healthy individuals, Hylek suggested.

The trial enrolled 100 coffee-drinking adults and outfitted them with Fitbit Flex 2 devices, continuous electrocardiography with the wearable Zio patch, and a continuous glucose monitor. Participants were randomly assigned to consume at least one coffee (defined as one shot of espresso or a cup of regular caffeinated coffee) or as much as they wanted or to avoid all caffeine for 2 weeks in a 2 consecutive days-on, 2 days-off schedule.

“One of the biggest challenges was finding individuals willing to go without coffee,” Marcus noted. “We recognize that that could introduce some bias,” if people most prone to coffee-induced problems declined to enroll.

However, he pointed out that adherence among those who did enroll was “excellent” by all measures, which included participants pressing the button on the Zio patch each time they drank coffee, answering text messages about their actual coffee consumption on the prior day, reimbursement of coffee receipts, and electronic monitoring of their phone location to show when they were in a coffee shop.

Prior studies that have generally concluded no association or lower risk of atrial fibrillation with coffee consumption have relied on longitudinal and observational data from self-report prone to confounding, Marcus noted. The UK Biobank study for example suggested lower risk of supraventricular tachycardia, consistent with the trial findings. There hadn’t been much PVCs data, and what was available was “very conflicting,” he added.

Observational studies had also linked coffee with a reduction in mortality and in diabetes, but the CRAVE trial turned up no signal of effect on glucose.

“This study was really designed to detect acute effects,” Marcus noted. “There may be chronic effects that are different. It’s not just that you add up all the acute effects and that leads to the chronic effect. Rather, there may be some cumulative process that leads to the outcomes observed.”

“For example, regular physical activity more commonly, perhaps that leads to a reduction on overall mortality perhaps in diabetes, whereas our data would suggest that these observations that coffee is associated with a lower risk of diabetes long-term does not appear to be an acute effect of caffeine on insulin sensitivity,” he stated.