The Reveal AF Study. AF is like speeding... sooner or later you'll be caught.

Thanks to Ken and others for the great discussion on AF in another thread, I started looking into the prevalence of undiagnosed AFIB. One study showed it to be about 13% but this longer study shows it to be 40% when a long enough time line is used.

It's like speeding, we do it often, maybe every day but rarely get caught. But when someone is looking closely, turns out it happens quite often.

Sort of like Bob Loblaw "Why should you got to jail for a crime someone else noticed?"

[www.youtube.com]

Results A total of 446 patients were enrolled; 233 (52.2%) were male, and the mean (SD) age was 71.5 (9.9) years. A total of 385 patients (86.3%) received an insertable cardiac monitor, met the primary analysis cohort definition, and were observed for a mean (SD) period of 22.5 (7.7) months. The detection rate of AF lasting 6 or more minutes at 18 months was 29.3%. Detection rates at 30 days and 6, 12, 24, and 30 months were 6.2%, 20.4%, 27.1%, 33.6%, and 40.0%, respectively. At 18 months, AF incidence was similar among patients with CHADS2 scores of 2 (24.7%; 95% CI, 17.3-31.4), 3 (32.7%; 95% CI, 23.8-40.7), and 4 or greater (31.7%; 95% CI, 22.0-40.3) (P = .23). Median (interquartile) time from device insertion to first AF episode detection was 123 (41-330) days. Of patients meeting the primary end point, 13 (10.2%) had 1 or more episodes lasting 24 hours or longer, and oral anticoagulation therapy was prescribed for 72 patients (56.3%).

Conclusions and Relevance The incidence of previously undiagnosed AF may be substantial in patients with risk factors for AF and stroke. Atrial fibrillation would have gone undetected in most patients had monitoring been limited to 30 days. Further trials regarding the value of detecting subclinical AF and of prophylactic therapies are warranted.

[jamanetwork.com]

It's like a silent killer. A few years back, I listened to a cardiologist discuss AF and its treatment over time. He stated that doctors, as a group, thought they had done well when few patients returned to them after five or more years. It wasn't until some bright bulb thought to do a study of many cases that he realized they were mostly gone after five years. Gone...as in six feet under.

I've thought of that. When EPs or Cardiologists say their patients haven't returned in x amount of years, my dark side thinks some of them were fatalities. There should be a followup every year to see actual success rates. But then....do we really want to know?

My thoughts on the prevalence is that it's possible wearables will get better at detecting AF in the future and having been warned, people will take the right steps to avoid triggers and negligence in their diets and lifestyles and ending up where they will need a procedure.

I can't even recall how long it was but I wore a Polar monitor over night and kept seeing this irregular spikes in my HR while sleeping. Could those have been warning signs?

Quote
PavanPharter
I can't even recall how long it was but I wore a Polar monitor over night and kept seeing this irregular spikes in my HR while sleeping. Could those have been warning signs?

Maybe 14 or 15 years ago a UK optometrist member here sent me this collection of rhythms he'd captured on his Polar with RR (beat to beat) recording on. It was referenced in Conference Room 52a [www.afibbers.org] See if anything in here matches your memory. I can certainly recognize afib on my Polar recordings..

There are some images of afib on my Polar in this short episode from 21 Jan 2021. [www.afibbers.org]

One's heart rate will spike pretty much every time you enter REM, and certainly if you have a dream during REM. If your sleep is uninterrupted much, say you getting three or four solid hours at a time until you are awakened for any reason (crying child, dog wants out, you have to 'go'...), you will have several dream cycles each night, with most of it taking place in the final 2 hours of sleep. If you can look at your spikes, and see that they tend to concentrate near the end of the sleep period, that's a decent indicator that your HR spikes are associated with REM.

We don't know if the monitor was picking up AF, PACs, or SVT. You'd need an EKG depiction to see if there was an absence of P waves, and if the R-R intervals were varied in length.

Mark's files all look more extreme than I recall mine being. Basically, I had pretty smooth HRs but every few minutes. but not at regular intervals there would be a spike of about 15-20 BPM over the normal HR. I still have the labtop that those files were on. I'll look to see if they are still there.

Quote
gloaming
It's like a silent killer. It wasn't until some bright bulb thought to do a study of many cases that he realized they were mostly gone after five years. Gone...as in six feet under.

Thankfully, there are many wonderful people here still on the right side of the grass. Jackie, George, Ken, Susan and any other long timers.

Thanks Pavan! I just say I’m above ground when asked. Who has grass in CA during a drought and water restrictions anyways? As long as I’m breathing I’ll fight this…4 ablations, pacemaker and watchman and weekly to daily trips to the hospital this past year and a half …and the last time my gurney was transporting me from the ER to telemetry, too many people greeted me (frequent flyer I’m told) with hello Susan and asked what my cardiovert record is now (54).

Quote
gloaming
We don't know if the monitor was picking up AF, PACs, or SVT. You'd need an EKG depiction to see if there was an absence of P waves, and if the R-R intervals were varied in length.

With a Polar chest strap, you can get R-R data, if using a watch or ;more recently an app that will record RR. With these data you can certainly see a lot as shown in Mark's recordings. [www.afibbers.org] It is helpful to have an existing diagnosis, but certainly afib stands out like a sore thumb.

Wow, what a mess!

Study is interesting and opens up some real questions that I think for me personally need to be answered. My question is this. I'm assuming stroke risk is figured by some calculation that divides the number of strokes by the number of people diagnosed with afib? Study suggests to me that the denominator of this equation may be FAR LARGER than what I'm generally hearing. If that is the case, then stroke risk should be FAR LESS than the numbers being thrown around at us? Probably more nuanced than my simple idea here, but it still makes me wonder.

This is important. Are our treatment decisions (especially those for those of us with only an occasional short lived episode) built on some questionable foundations? Is using a universal set of assumptions for both those with persistent afib and those with an occasional episode good medicine?



Edited 1 time(s). Last edit at 12/15/2022 10:51AM by Kindog.

That kind of determination would lead to a false one, in my view. Which strokes should we count, and how to we determine after the fact that any one stroke of several are attributable to AF in any way? You'd need autopsies in the hope that there might be residue of a clot in, say, the atrial appendage, a dead (sorry) giveaway. I would hope the medical community doesn't just look at stroke deaths, and then divide them by the total number of people so afflicted who also happen to have AF. They might as well count prostatitis in males, breast cancer in women, or the number of cases of flu one has had over a lifetime.

Gloaming your point is fair and well taken. Still, I'm not sure how the statisticians could accurately adjust , without even having a clue as to how many undiagnosed cases exist.......and how many millions of cases will never be discovered. I'd guess the current method would be along the lines of dividing all stokes among afib patients by the total number of afib patients.
I was asymptomatic, and my own diagnosis was only discovered by an odd accident when I was almost 78 years old. Still my suggested treatment was the usual. The study suggests that there are millions like me out there, most of whom may never be diagnosed. Unless there are a great many unreported strokes going on it seems that most of these people will go through life without ever knowing they had afib, and never have a stroke. Using the 40% estimate of afib cases we'll need a hell of a lot more stokes to get to the stroke percentages that I keep hearing about. It appears to me that the estimated stroke rate for ALL (diagnosed and undiagnosed) afib cases is far too high.....and hence that many are being over treated.



Edited 4 time(s). Last edit at 12/15/2022 03:14PM by Kindog.

Kindog - stats are great for statisticians but if you're looking for a reason to not be on a DOAC, I don't think anyone here will support or encourage that.

Strokes are the ultimate black swan.

Pavan, Sorry, don't mean to suggest anything. Just the opposite....Like all of us, I'm looking for answers. My question is fair and legitimate. My medical knowledge is pretty limited and i"d guess as you imply nobody here is looking for advice from me, and I'm certainly not offering it. Still, my point seems valid. Yes a number of 40% suggests many future strokes to come, but most importantly it also suggests a far lower stroke rate than a 10% rate. Should we all be wearing event monitors?



Edited 1 time(s). Last edit at 12/15/2022 04:11PM by Kindog.

I doubt there are enough undiagnosed cases of afib out there to significantly alter the stats. Most asymptomatic afib is persistent, so that means to escape detection you have to avoid ever having a medical professional take an EKG, listen to your chest, or feel a pulse (or buy an Apple Watch). At least one of those things is likely to happen even in the most cursory visit to a doctor's office, walk-in clinic, or ER.

And while it's true that most people with untreated afib will never suffer a stroke, that's small comfort when you consider the consequences of stroke. It's one of the few medical conditions where the outcome can be far worse than death.

Carey, The point is that this study suggests a huge number of undiagnosed cases.

I guess we'd have to pose the question to the underwriters, because they have to convince their investors that the risk they undertake is manageable. Obviously, over time they would refine those estimates, but I haven't a clue what they use to do so.

I personally know a gentleman whose flutter was undiagnosed formally, and unnoticed by him, until he went for a checkup. So, it happens, and I don't think that is in dispute. Not at this forum, anyway. I would guess that there is an educated estimate by now of how prevalent the typical arrhythmias are, now getting close to a third of the way through the 21st century. This, in turn, would suggest how much of it goes both undetected and untreated. I'm not aware of jurisdictional autopsy stats that tell us of an infarct over a clot, or a stroke, or a pulmonary embolism, or whatever else could send someone under the turf in a hurry due to dislodged clots.

Here's the link to the full paper - some important points [jamanetwork.com]

Data gathered by device interrogation from patients with implanted pacemakers and defibrillators suggest that as little as 6 minutes to 6 hours of AF may double incident stroke risk,

While the primary end point in the REVEAL AF study was an AF episode lasting 6 or more minutes, many patients had prolonged cumulative periods of AF; the incidence of AF lasting 30 minutes or more in a day was greater than 40% at 30 months, and the corresponding incidence of AF lasting 6 or more hours in a day was nearly 20% at 30 months. Additionally, 10.2% of patients with 6 or more consecutive minutes of AF had at least 1 episode lasting 24 hours or longer. Both the Asymptomatic Atrial Fibrillation and Stroke Evaluation in Pacemaker Patients and the Atrial Fibrillation Reduction Atrial Pacing Trial (ASSERT)32 and the study by Israel et al33 demonstrated that even prolonged episodes of AF can be asymptomatic.

Two smaller trials24,25 have also reported on the incidence of AF in high-risk populations using ICMs. The Prevalence of Sub-Clinical Atrial Fibrillation Using an Implantable Cardiac Monitor (ASSERT-II) study (NCT01694394)24 was a multicenter trial that enrolled 273 at-risk patients without known AF; 252 (92.3%) received an ICM and completed follow-up. The incidence rate of AF 5 or more minutes was 34.4% per person-year. The Predicting Atrial Fibrillation or Flutter (PREDATE-AF) study (NCT01851902)25 was a single-center trial of patients without known AF but with a CHA2DS2-VASc score of 2 or greater. In 245 patients, the incidence of AF or atrial flutter was 22% during an average follow-up of approximately 15 months. The mean time to detection was similar to the REVEAL AF study at 141 days. There was no difference in the AF detection rates between patients with CHA2DS2-VASc scores of 4 or less vs 5 or greater. Male sex was the only significant predictor of AF. Oral anticoagulation therapy was initiated in 76%. In contrast to both the PREDATE-AF study25 and the ASSERT-II study,24 the REVEAL AF study required a minimum of 24 hours of external monitoring without AF detected prior to device insertion. This, in addition to broader inclusion criteria, may explain the lower AF detection rate in the REVEAL AF study vs the ASSERT-II study.24

This video is linked to this entire conversation. [www.afibbers.org]

Quote
Kindog
Carey, The point is that this study suggests a huge number of undiagnosed cases.

What study are you referring to?