Chirag Barbhaiya, MD: Apple Heart Study Reaction

The Apple Heart Study has been a topic of discussion in the cardiology community ever since results were released at the American College of Cardiology (ACC) 2019 Annual Scientific Session in New Orleans earlier this month. The study, which enrolled more than 400,000, has been the subject of both praise and ridicule.

Chirag Barbhaiya, MD, cardiac electrophysiologist and assistant professor of medicine at NYU Langone Health and NYU School of Medicine, discussed the positive and negative aspects of one of the most talked about cardiology studies in recent memory.

MD Mag: What was your reaction to the results of the Apple Heart Study?

Barbhaiya: So, the Apple Watch study is a very exciting study in that I think it established a proof of concept of a new way of doing research in cardiovascular medicine. The fact that you can enroll 400,000 patients like that is mind-blowing for someone who is interested in or acquainted with the cardiovascular literature but all that said that there are there are some potential challenges that having this data brings up. So, the population enrolled in the Apple Heart Study is younger, healthier, more affluent than the US population at large and so while on its face the numbers reported of sensitivities and specificities in the high 90s, when you kind of take a Bayesian approach to the statistics — meaning that you look at the pretest probability of these people having a rhythm problem — and then superimpose the diagnostic accuracy of the watch you actually end up with a tremendous number of false positives and false negatives.

So, in an analysis that I read the effective sensitivity of the Apple Watch for Afib is on the order of 50% and the positive predictive value — meaning that for every patient that is identified as potentially having Afib with an Apple Watch — only about 35 percent actually end up having Afib. So, while on the one hand it is great to be able to pick up disease in patients where it was previously unrecognized, on the other hand it is going to be a significant burden on the health system to have all these people who are now coming in because their watch told them that they have a heart rhythm problem that requires consultation with a physician and primary care is going to be on the front lines of this — seeing everyone that comes in with these watch-related alerts. The other piece of this is that even for people that have proven Afib, we don't know that what we're picking up is the same disease in terms of severity and risk of the symptomatic patient that is kind of traditionally been seen and so one of the questions that comes up is: What is clinically significant Afib? Is it 30 seconds? Which is the kind of guideline that is used in Afib ablation-related clinical trials. Is it 6 minutes? Which there is some data on suggesting that your stroke risk starts to go up if you have an atrial arrhythmia that lasts 6 minutes or more, or is it 24 hours?

The best study that has looked at this was the ASSERT trial that looked at implantable monitors and the stroke risk related to various durations of Afib and there it seems that if you don't have more than 24 hours of Afib your stroke risk doesn't really pick up. Now, this is all very preliminary data, not validated in a randomized fashion, there's no therapeutic protocols that have been have tried out related to this stuff but we're really in uncharted territory. So, that makes it very exciting as someone who works in the field and is trying to help figure this stuff out but in the meantime we have to deal with the patients that are showing up with their watches, with their arrhythmia alerts, and figuring out what best to do with them.