EKG Machine errors?

I haven't been here for a while, but I used to be here a lot!

In 2007, I had aortic valve replacement with a bovine valve. The doctor did a maze procedure because I had experienced some paroxsymal afib. He also closed off my atrial appendage. All was good until recently.

Last year, at age 72, they put me on BP meds because I had a high BP episode. Metoprolol, fursomide and lisinopril. Ever since, my blood work has been wacky. The cause of the BP episode was never found, but a pulmonologist ruled out any pulmonary issues.

Then I had an ekg in February. The machine said afib. The cardiologist sent me to an electro guy. His machine said afib. My resting heart rate was in the low 60's. When I had afib prior to the maze, I could always tell. I didn't feel any afib and they were telling me I was in afib with a HR in the 80's.

They did a rush ablation, or at least now I think that. I did have 2 episodes of afib I felt in the first month after the ablation.

In May, I had a routine followup with a different cardiologist in a different state. His machine said afib, but he said it was a run of PACs.

So now I'm paranoid. The Kardia device frequently tells me its inconclusive or possible afib, even with a HR in the 70's or 80's. My smart watch hasn't detected any afib except the 2 episodes I felt.

The BP meds make me feel crappy.

Is it possible to be in afib with a resting HR in the low 60's and a HR in the 70s-80's when the Kardia says possible afib or the doctor's ekg machine says afib?

Its so frustrating.

You can have a low afib heart rate - in the 80's or lower.

I don't know what the maze might do to the ECG. Generally NSR should have a "p" wave. On a Kardia, a "Lead II" presentation has a better shot at showing a p wave. Basically hold the Kardia with your right thumb on the contact and put the left contact on a lower part of the left side of your body. I tend to put it below and to the left of my navel. The left knee or thigh will work. Pompon talks about the V4 position, which is on the chest, below the heart. See here. Make sure both contacts of the Kardia and the skin touching it are damp or wet for good conduction.

The other hallmark of afib is the variability of the beat rate or beat time length. If you have a subscription or are grandfathered in to their subscription, they will make a heart rate vs time graph for you on the PDF of the ECG. I have examples of NSR & afib in this post. The Kardia HR vs time plot is squirrely as it is on the right hand Y axis and increases going down, almost in logarithmic format. If you don't have a subscription, you can make the graph yourself. Just take a plot that Kardia thinks is afib and figure out the time for each beat, or just the number of small boxes between each beat. Then plot these, either in a spreadsheet or by hand. As you can see in my linked example. NSR has minor (a couple of beats per minute) changes in beat rate, while afib has big changes. In my post here, you can see how PAC's differ from afib. PAC's or PAC's with a compensatory beat following still have a regular pattern, where afib is just random. I'm not sure Kardia does a good job telling the difference. I have no idea about a doctor's ECG machine in automated analysis.

Quote
67walkon
Is it possible to be in afib with a resting HR in the low 60's and a HR in the 70s-80's when the Kardia says possible afib or the doctor's ekg machine says afib?

George has already answered correctly, so let me add one tidbit for you....

Most doctors, and particularly cardiologists, completely ignore the diagnostic info ECGs print. It's totally unreliable. And any doctor can identify afib with a brief glance or even just by feeling your pulse. It's probably the easiest arrhythmia of all to identify. Even I can identify it with 99% accuracy just by feeling your pulse or seeing your ECG/Kardia reading. And actually, the Kardia is more reliable than an ECG machine. When it says afib, it's almost always right.

With some time spent looking at various readouts from ECG prints, you can soon see immediately if you are in NSR or in AF. You don't really need to look at a normal readout, in fact. All you have to do is to look for obvious distances between any two R waves, the tall spikes. if the distance varies quite apparently, then look if there are discernible P wave blips shortly after the T wave blips that take place near halfway between QRS intervals. (This might make your head spin, but I also have to reread a few times, try to get it into my head, and then have three or four example readouts handy. Suddenly, it pops, and just like learning to spot deer in the forest, your eyes will immediately see the differences on an ECG).

Quote
Carey
I And actually, the Kardia is more reliable than an ECG machine. When it says afib, it's almost always right.

Although it can confuse Afib and runs of PACs. I have had a Kardia Afib reading and then sent it to the Natale team and they said that it was actually PACs.

Quote
Daisy
Although it can confuse Afib and runs of PACs. I have had a Kardia Afib reading and then sent it to the Natale team and they said that it was actually PACs.

Yes, I have seen it make that mistake a few times, and I'll bet that's the same thing that fooled the ECG machine.

It is my understanding that Kardia will not give you are reading of “aFib ” but will only give you a reading of “possible aFib.” Maybe not a technically important distinction and one made for liability issues or whatever but one that is still made—I think. Maybe the subscription model makes a definitive statement of “aFib”vs possible “aFib.” I don’t know. Is it then fair to say that if Kardia says “possible aFib” that you are in fact in “aFib.” Splitting hairs here I think but just want to be clear in how I interpret the Kardia when discussing this with my EP or cardiologist if the need arises.

Quote
Mark
It is my understanding that Kardia will not give you are reading of “aFib ” but will only give you a reading of “possible aFib.” Maybe not a technically important distinction and one made for liability issues or whatever but one that is still made—I think. Maybe the subscription model makes a definitive statement of “aFib”vs possible “aFib.” I don’t know. Is it then fair to say that if Kardia says “possible aFib” that you are in fact in “aFib.” Splitting hairs here I think but just want to be clear in how I interpret the Kardia when discussing this with my EP or cardiologist if the need arises.

I'm grandfathered in to the subscription since I've had a device since 2014. It does give an afib determination. Prior to the introduction of the subscription in July of 2020, all I ever got was possible afib.

Runs of PACs, when they're quite long, may look like AFib.
I've lots of ECG tracings with long runs of totally random PACs (a dozen or so), followed by normal beats (half a dozen or so), then followed by a new run of PACs, and so on...
I'm used to call this phenomenon "unsustained AFib". It may happen and quickly disappear ; but it can be the start of a true AFib episode... or the end of another one.



Edited 1 time(s). Last edit at 07/27/2023 08:28AM by Pompon.

I'm inclined to agree with Pompon, but based solely on my own 'sample'. After my first ablation, I developed PACs, often sustained and intermittent. They showed up at 13 weeks in the post-ablation Holter. A lot of them. Within 8 weeks I was getting AF, but the ECG folks at the ER said it was actually flutter. My Samsung Galaxy watch said, "AF detected' when I used it at about the same time.

As most of us already know, it really takes a 10=12 lead ECG, and then read by a skilled reader, to determine what's what. In my case, during the flutter episode over this past New Years, the ER physician sought my permission to administer adenosine to see what the ECG showed with more discrete information than it was capable of reading without the adenosine's influence on my heart. As soon as I began to feel awful, which the team prepared me for, the gentleman said, 'There it is...flutter, and it's a 4:1 block.'

All this to mean that the best equipment used the most effectively will give veridical information that MUST BE ACQUIRED in order to safely perform restorative or ameliorating measures on the patient's behalf.

Kardia is pretty good at calling afib, but not always. An ep will trump Kardia and in fact they don't even look at what their machine says. Even someone like myself with no medical training can trump Kardia with an afib call after a short learning curve. What can sometimes throw Kardia off are ectopics. Same with Apple Watch.

The problem is with GP's and some cardiologists -- especially when they look at a single lead like the Kardia. Many of them cannot read an ekg at all and rely on their office machine for determinations. So unless you really trust the doctor, double check with an ep.

That said, even ep's can differ interpreting some single leads. I've shown the same Kardia single lead to four ep's and got determinations of aflutter, SVT and afib! So sometimes it's as much of an art as anything else.

Jim

Yikes. I would think that an ECG readout would be highly definitive...albeit with variance due to conductance, the machine, the leads and where they are placed, etc. I was most happy to learn that my EP accepted my Samsung Watch's readout sent to him. It was clear, and clearly narsty!

Quote
gloaming
Yikes. I would think that an ECG readout would be highly definitive...albeit with variance due to conductance, the machine, the leads and where they are placed, etc. I was most happy to learn that my EP accepted my Samsung Watch's readout sent to him. It was clear, and clearly narsty!

The electrodes are not always applied just where they should. This may lead to wrong interpretations.

Quote
gloaming
Yikes. I would think that an ECG readout would be highly definitive...albeit with variance due to conductance, the machine, the leads and where they are placed, etc. I was most happy to learn that my EP accepted my Samsung Watch's readout sent to him. It was clear, and clearly narsty!

It's not simply the clarity of the readouts. In certain cases, even ep's have a hard time distinguishing say atach from aflutter and even afib, especially with a single lead. But in the majority of cases, an afib call is not a hard one.

Quote
mjamesone
That said, even ep's can differ interpreting some single leads. I've shown the same Kardia single lead to four ep's and got determinations of aflutter, SVT and afib! So sometimes it's as much of an art as anything else.

Quote
gloaming
Yikes. I would think that an ECG readout would be highly definitive...albeit with variance due to conductance, the machine, the leads and where they are placed, etc. I was most happy to learn that my EP accepted my Samsung Watch's readout sent to him. It was clear, and clearly narsty!

There can also be context. For myself, if I see a steady, high rate (>100) rhythm in the midst of afib, I assume it has converted to flutter. My experience is I have to get the rhythm to revert to afib before it will convert to NSR as the flutter is very stable. Now a lot of time for these data, I'm just looking ar R to R rate over time, not a Kardia or other single lead ECG as I generally don't record ECG's for longer than 30 seconds.

There is apparently an increase in cardiologists relying on what the machine says rather than exercising their expertise to read the print out.

I’m mildly concerned that the cardiologist mis diagnosed me. And that the EP relied on the machine rather than his own interpretation.

And I’m worried they have messed me up worse than I was before they did the ablation. And I’ll likely never know.

This is all confusing and frustrating.

Is there any preclusion to your requesting copies of pre and post and taking them to, say, a university professor in the field, even if you have to send them electronically across the country? Would Natale take a look if asked? I'd like to know it's a 'no', personally, and would go to some lengths to find out.

You clearly need some peace of mind. You could have it by next Friday if you took steps tomorrow.

Quote
67walkon
There is apparently an increase in cardiologists relying on what the machine says rather than exercising their expertise to read the print out.

I’m mildly concerned that the cardiologist mis diagnosed me. And that the EP relied on the machine rather than his own interpretation.

Who says there's an increase in cardiologists relying on machine interpretations? I very much doubt that so could you please show us what you base that statement on?

And no way an EP relied on a machine interpretation. An EP has been interpreting ECGs day in and day out for years before they earned the right to call themself an EP. They don't need a machine to interpret for them. It's as hard for them to interpret an ECG as it is for you to interpret the English language.

I've lots and lots of pro grade 12-lead ECG printouts showing NSR around 60 BPM, and saying (in French) something like "septal ischemia, date unknown". My cardiologists and EPs just told me I should not worry about that.
My heart is perfectly normal. Plumbary is good. There are quite some electrical issues and my doctors know them. The machine does not.