Heart Rate Trackers & Atrial Fibrillation

[SunsetRunner]

I have persistent atrial fibrillation. My cardiologist and I wanted to see if it was worthwhile to use a heart rate tracker in between my 24 hour Holter monitor tests. My results thus far with the Inspire HR compared with the Holter results for the same period of time for the same exercise regimen have been disappointing. Differences of 20% and more that I have experienced suggests that the device is not useful.

I was curious about why such a large discrepancy should occur when I read reviews before I purchased the device that indicated that the accuracy was quite good (+/- 5%) even when compared to chest straps. After a bit of research I found a recent study published in the European Journal of Medicine on 21 October 2019. The objective was to determine the heart rate accuracy of the Fitbit Charge HR and Apple Watch Series 3 compared with Holter monitoring in an ambulant patients.

There were 32 patients, 26 of which had persistent atrial fibrillation and 6 with normal sinus rhythm. The patients were randomly allocated in two groups to one of the devices. Each group had 3 patients with normal heart beats. At total of 53,288 heart rate values were recorded. A +/- 10-beat different between Holter-HR and the Apple & Fitbit-HR was used as a clinically relevant range to establish the accuracy of heart rate estimation of the two devices.

In the normal sinus rhythm patients the two devices performed satisfactorily compared to the Holter tests. The Apple Watch was within the +/- 10 beat difference on 95.2% of the readings and the Fitbit on 92.2% of the readings. This suggests a reasonably accurate result.

However, in the case of the atrial fibrillation patients 76.5% of the Apple Watch readings were within the +/- 10% threshold and only 56.1% of the Fitbit readings. You can find the abstract of the study here.

I am well aware that Fitbit states “The accuracy of the data collected and presented through the Fitbit Service is not intended to match that of medical devices or scientific measurement devices.” I am not suggesting that they should. However, if they are offering a device that measures heart rate it should be accurate enough within some reasonable variance otherwise why have it as a measurement at all.

I believe it would be useful for Fitbit to acknowledge in their advertising that it may not be to patients with atrial fibrillation benefit to buy their devices that track heart rate, but it is unlikely they will. This would save time in ordering, testing and returning the devices.

I would however like to hear from others who have atrial fibrillation as to their experience with their devices.

[JohnnyRow]

I thought with AFib, there was not consistent ventral pulse, so how could it be measured.

What am I missing?

[SunsetRunner]

Are you referring to ventricular fibrillation when you say "ventral pulse"? The study is about atrial fibrillation, the condition I have. Ventricular fibrillation causes blood pressure to plummet within seconds and often leads to death. Pulse is triggered by electrical activity in the specialized cells located in the right atrium, called the SA node. They cause the contractions in the atria and ventricles that results in a heart beat. If the electrical activity becomes erratic it leads to afib.

If I misunderstood your statement please let me know.

[SunsetRunner]

Let me try this again. I think you are saying that atrial fibrillation affects the ventricular heart rate so how can it be measured. You said it clearly, I just didn't read it carefully.

The Holter is essentially a portable ECG machine and my cardiologist tells me that he can determine the heart rate even if Afib is present. It complicates the measurement but it can be done. It has to do with counting the spikes (I think they are called QRS complexes which equate to the ventricular rate) on the ECG and multiply by some factor. Wrist monitors simply measure changes in blood volume and cannot possibly be accurate if Afib is present.

I wanted to try the Fitbit device even though he was sceptical. He was right.

[Johnmeik · Accepted solution]

This is correct. Measuring changes in blood volume will never give an accurate reading of afib. Afib  has a variable rate but also a variable cardiac output, hence the constantly changing volume bring measured. It can only give you a trend which in itself is useful. 

[SunsetRunner]

Many thanks for the explanation.

[LZeeW]

@SunsetRunner   I want to point out some issues with the linked abstract.  

• The study used a Charge HR and an Apple Watch 3.  The Charge HR was released in early 2015 and the Apple Watch 3 was released in Sept 2017.  The two devices aren't technologically equivalent.  Technology needs to be equivalent for a study to be meaningful.  Likewise, they are not technologically current.
• The study was released as an abstract.  This means that it hasn't undergone peer review.  Journals print hundreds of abstracts each year.  Only a small fraction make it to peer reviewed publication.

I don't know if this abstract rises to the level of garbage in = garbage out.  But it's not great.

Fitbit's best heart rate algorithms are in the Versa 2.  I don't know if they were included in the Charge 4.  I'm confident that the heart rate tracking ability of the Inspire HR is better than the Charge HR.  And @Johnmeik has the right answer as to how this stuff works.

You cardiologist should know that when you look at a 24 hour graph on your app, any heart rate value is an average heart rate over a five minute interval.  You cardiologist must also compare apples to apples when using Holter data.

Does this help?

[SunsetRunner]

I understand what you are saying about the the two devices not being technologically equivalent but what I took away from the study was the fact that people with afib aren't getting a true measurement of their heart rate with these devices.

Thanks for your input and it was helpful.

[Not-Fit-Bit]

I have atrial flutter as of last week and my Fitbit Charge 2 HR didn't pick it up at all.  I felt breathless so I used an Omron heart rate tracker and my wife's apple watch and they both said my heart rate was 138 BPM.  My Fitbit said 71 BPM.  If I had relied on the Fitbit, I would have gone undiagnosed and risked a blood clot.  I spoke to Fitbit and they did nothing but told me "some issues may happens and that is why we have these channels to try some steps in order to see if the issues are fixable".  They had no interest in resolving the problem and didn't see it as a major issue at all.  No solution was offered.  I am buying an Apple watch nd throwing the Fitbit in the bin.

[SunsetRunner]

I wouldn't recommend "relying" on any of these devices from any company for
detecting accurate blood pressure readings. I trashed mine already. I'm
sure you have seen your family doctor already. But, if you want to track
your BP over time get a good quality cuff heart rate monitor.

All the best