Mar 06 2026 This Week in Cardiology

0:00 You're listening to this week in cardiology from the heart.org Medscape Cardiology. This podcast is intended for health care professionals only. Any views expressed are the presenter's own and do not necessarily reflect the views of WebMD or Medscape. >> Hi everyone, this is John Mandola from the heart.org Medscape Cardiology and this is this week in cardiology for March 6, 2026. This week some really interesting listener feedback. Urgent AF ablation, AF ablation as a stroke reducing therapy, ILR loop recorder accuracy, and heart failure management in the setting of serious disease. So, first is some listener feedback. I received a nice email about my coverage of the provocative JAMAMA internal medicine

0:47 letter on patient preferences for statin therapy. This was a letter in which authors from Japan surveyed people on what their smallest worthwhile difference would be for accepting a daily statin prescription. [sighs] I was struck by the findings because when the authors surveyed about 500 adults from the US and Japan, they found that nearly one in four individuals would refuse statins even if they could get risk reduction from 20% to zero. and that the smallest worthwhile difference for patients is essentially three times more than the actual risk reduction found in trials. I took from these observations that most people tend to be minimizers. They want a lot more bang for their buck in risk reduction if they were to accept a daily statin. Well, internist and

1:35 primary care doctor Dr. Kuma Foby from Rush University in Chicago wrote to me to say that number one, the survey only included people who have never been on a statin and that might select for minimizers. But more importantly, Dr. Foby wrote, "This sort of discussion probably hits different when it's done in a personalized setting with physician patient continuity. While someone might decline a medication when answering questions on a survey, if given the option from a doctor they have known and trusted for years, the results would likely be different. Suddenly, small effect sizes could feel larger. Okay, I really like this comment because it highlights the concept of external validity of evidence, doesn't it? That

2:20 is how we generalize evidence from carefully controlled studies which are always special circumstances to everyday practice. And a survey is certainly a special circumstance. Dr. FY's comment makes perfect sense. If you are responding to a survey, you may be more apt to be purely quantitative, but you're in the office of a trusted professional. You might feel differently. Now, let me tell you an example. I felt this firsthand in my urologist office. I was there for a specific issue that turned out to be nothing, thank goodness. But while I was there, I got the full court press on PSA screening for prostate cancer. Now, I've studied screening. I've written about screening. This includes PSA screening. I knew the data. And I had decided that

3:06 no doctor ever would check my PSA. But but I trusted my urologist. He was clearly an expert. He was also kind and professional. And he spoke from his point of view of caring for oodles of people with bad prostate cancer. He gave me his best case and I can't believe it, but I relented and let him take a PSA on me. Gosh darn it. And thankfully it was okay. But that will be the last PSA I get. But this experience underscores Dr. Foly's excellent point that an online questionnaire likely does not recapitulate a conversation with a trusted professional. That said, I still think statin drugs should be available over the counter without a prescription in an aisle that has a risk calculator and people can

3:51 decide for themselves. All right, next topic is urgent AF ablations. Jackie P has published an observational study looking at the use and safety of urgent AF ablation in the US. That is an AF ablation done in patients hospitalized for something other than AF. Now if you care for patients with AF the idea may have entered your brain too that is a patient comes in with something say pneumonia hypotension heart failure VTE and they're found to be an aphib and you think crap this aphib is new it's causing trouble we're going to pursue rhythm control perhaps with drugs like amiotarone procanomide sodalol doetylolide then you think wait if we're

4:37 going to do all that why not just take the patient down to the EP lab and do a quick AF ablation while he or she is here. An observational study first author Amnite Sandu published in Jackie P suggests perhaps you should resist that urge. They used an AIB ablation regist registry which compiled cases from 2016 to 2023 and compared urgent versus elective AF ablation cases. They had 140,000 patients who had had their first AF ablation. About 2% or 2700 cases were labeled as urgent. 98% were elective and so they had two comparator groups. Urgent patients had higher rates of comorbid conditions such as diabetes, coronary disease, heart failure, which you'd expect. Urgent AF ablation was

5:24 more often used among black patients and those presenting to the procedure for an urgent AF ablation were more often in atrial fibrillation. Hospitals that did more AF ablation also did more urgent AF ablation. And when they looked at the seven-year time period, the number of urgent AF ablations were increasing over time from 0.5% to 2% at the end of the study. And the main result, which is not surprising, was that the adjusted procedure related complication rate was significantly higher with urgent AF ablation compared with elective 4.9 versus 2.4%. 4% that p value was highly significant and notable is that these are just inpatient complications. The registry is likely an underestimate

6:11 because it can't see complications that occur after discharge things like stroke, bleeding, pneumonia, etc. And the authors concluded the only thing that they could conclude that urgent AF ablation has increased over the past decade. It's done in sicker patients and it has a much higher complication rate even when risk adjusted. Okay, my comments. The first thing to say is that this is a good use of observational data. Science tells us what we can do. Trials tell us what we should do. And registries tell us what we are actually doing. And none of this data should be surprising. Patients in whom a doctor is tempted to do ablation during a hospitalization for something else will surely be inherently sicker and thus they're going to have a higher complication rate. This study's nearly

6:56 5% complication rate for urgent AF ablation is strikingly high in my mind. Now I'm just going to move to Mandola opinion. Here it is. Doing urgent AF ablation is almost always a bad idea. If the patient is in the hospital for something else, it's best to treat that something else and bring the patient back for an elective AF ablation after you've seen him in clinic and things are stable. I can think of only a handful of situations where I have done an urgent AF ablation and that is one thing that pops into my mind is when they're in a left atrial flutter of some sort and it's recalcitrant and it's causing heart failure and you just can't get them out of it. And these slow flutters can be especially bad because if a flutter is

7:42 slow in the atrium, they are often fast in the ventricle due to onetoone conduction. But most often you can bridge the patient with some drugs. Uh for instance, if they're really sick with heart failure, you can use amiotarone and a cardio version and then you can bring these patients back. And I call this getting that patient out of a hole, then doing ablation later when things are more stable. Recall that at least in the US AF ablation is done with general anesthesia and we should respect the dangers of that when patients are acutely medically ill. The observations in this paper are not surprising, but I think it was a worthwhile exercise to show us the dangers of urgent AF ablation. The authors don't tell us not to do urgent AF ablation, but the data

8:27 surely suggests that we should be cautious and in almost all cases resist the urge. All right, a little more on AF ablation. AFablation is probably not a good therapy for stroke reduction. Jamma Neurology has published the results of the stabled RCT from 45 sites in Japan. The clinical question is what to do with a patient with AF who's had eskeemic stroke. Do you give DOAC alone and in this case they use or DOAC plus AF ablation? Now adding AF ablation would diminish AF episodes and maybe that would reduce eskeemic stroke going forward. This trial was conducted in 2018 to 2021. So it's pretty old ablation technology

9:13 was only 250 patients randomized one to one. These were older patients age 72, 25% female and split nearly half into paroxismal and persistent AF. The Chad's vast score was high because they've had prior stroke 4.5. The trial did exclude patients with severe stroke. Median follow-up was very good at nearly four years. The primary endpoint was a composite of recurrent eskeemic stroke, systemic embolism, allcaused death or heart failure hospitalization. So they're really catching all important endpoints. The primary composite endpoint occurred in 22 patients in both groups. Uh so the hazard ratio 1.11 clearly not significant. Major bleeding was three in

9:59 the standard arm, eight in the ablation arm. That hazard ratio was three times greater in the ablation arm. But again the confidence intervals were 0.79 to 11. So not significant eskeemic strokes occurred in 14 and versus 10 patients and these were obviously not significantly different. And the author's conclusion reads quote in patients with atrial fibrillation in the recent stroke history standard therapy plus catheter ablation did not significantly reduce the risk of the primary composite endpoint. However, they write the observed event rate was lower than anticipated, suggesting that the study was underpowered to detect clinically meaningful differences. All right, so my comments, this is a bit like the ACE AF trial where a rhythm control strategy a in this case AF

10:47 ablation is being tested not so much as an AF episode reducer, but a hard outcome reducer. Recall that in East AF it was early rhythm control versus rate control and early rhythm control wasn't just for control of AF. It was to reduce uh important outcomes and it was the same in this trial. In fact, the authors of the stabled RCT don't even provide results on AF burden at two arms. And I think the use of AF ablation to reduce outcomes, especially in clinically ill patients, this case prior stroke is a reasonable question because intuitively you would think that reducing AF burden with ablation would add to the DOAC efficacy and the topline results in this

11:33 trial are null. No statistical differences were noted. But this is a big butt. This is one of the situations where I would not not conclude that ablation is an ineffective reducer of stroke, heart failure, or death because this trial was massively underpowered, which was evident in the calculations of their sample size. They estimated a 13% incidence of the primary endpoint, but then assumed that ablation would reduce events by 50%. 50%. That is way too optimistic because the control arm would be on DOAC which would severely reduce stroke rate. Death is unlikely to be affected by either arm and since these patients did not have heart failure to begin with, you wouldn't expect heart failure hospitalization to be much different either.

12:20 What's more, they had a lot of crossover and these pragmatic ablation versus no ablation trials are always going to have crossover and this makes it even harder to find signal from noise. For instance, 16 of the 124 patients in the standard arm had ablation and 19 patients in the ablation arm did not have ablation crossing over to the standard arm. For the incidence of stroke, the hazard ratio is 0.75, which is better for the DOAC arm, but the conference intervals went from 0.33 to 1.70. So, standard therapy could have been 66% better or 70% worse than ablation. I'm really surprised that a trial like this gets approved by the IRBs because you could have predicted wide conference

13:06 intervals and you could have predicted that the trial would have been uninformative from the beginning. I would say that if a stroke is due to aphib and presumably it was in this trial because DOAC was the main intervention in the standard arm and I would also think that emolic strokes were the main strokes because less than 10% of patients were taking antipllet drugs then it's going to be very hard to improve on doac therapy at using aphib as a stroke reducer because doaxs are so effective. So I would say in the end that no practice changes occur with this trial. AF ablation remains an intervention to improve quality of life by reducing AF burden. This is true for patients with and without a history of stroke. If you want to show AF ablation reduces stroke,

13:51 you're going to need five to five to 10 times more patients. And that's really not going to be feasible. Not because there aren't this many patients there are, but because many patients with a history of stroke and aphib will likely have symptoms and they'll want their symptoms reduced. Now, one warning I want to say about this, my friends. Do not be duped by observational studies showing that those who get AF ablation have lower stroke rates. The authors cite a Swedish Health Registry study of more than 300,000 patients that finds that those who have ablation have lower stroke incidents. These are non-random comparisons and they're flawed by selection bias wherein healthier patients get ablated versus notablated. All right, loop recorders. Let's talk

14:37 implantable loop recorders or ILRs. These devices can be handy for long-term monitoring for arrhythmias. They can detect AF. It can tell us AF burden. But perhaps their greatest use, and the reason I most often use them is for the diagnosis of infrequent but severe syncopy. Say a patient has what I call stone cold syncopy. I mean, they just out stone cold, but it happens only once every few months. You want to know the diagnosis. But if you do a two-eek monitor, it's unlikely to capture the event. But poorly recognized is that ILR recordings are often not perfect. So this is a study from multiple authors published in Jackie P and they authors set out to evaluate and compare the accuracy of currently used loop

15:23 recorders from Medronic, Boston, Biotronic, and Abbott. The database included a huge number of ECG verified episodes from 6,700 patients. For the study, the authors selected a random sample of just,40 patients. Now, the indications for ILR were the usual things like cryptogenic stroke, atrial fibrillation, syncopy, palpitations ventricardia. [snorts] The most common reason for alerts was AF. And for AF, Boston Scientific had the highest positive predictive value, followed by Abbott and Metronic, which had similar and moderate positive predictive values, whereas Biotronic for this one had the lowest positive predictive value. We're talking 0.73 at best and 0.23

16:13 at worst. Also for AIB, false positive burden overall was shockingly high at 51%. Biotronic had the highest false positive rate at 89% whereas other devices had comparable accuracy with false positive rate rates of like 39, 35 and 32%. Now pause there and just forget the comparisons amongst companies for a moment and consider that at least one in three AIB alerts are not AIB. For brada cardia, Boston Scientific demonstrated the highest positive predictive value at 0.96. Metronic had the lowest at 0.36. Pause detection accuracy was lowest for Abbott. Their positive predictive value

16:59 and I hope this isn't a misprint is 0.01 and 80% of this is due to undersensing. The highest pause detection accuracy was Boston Scientific at 0.70. So the authors concluded that ILRS demonstrate variable accuracy and arhythmia detection with a substantial burden of false positive alerts across all vendors despite AI based and other algorithmic enhancements. Further improvement of alert algorithms to reduce clinic burden is needed and I would say that is so. So my comments, I don't think we need to get too bogged down on devicetodevice comparisons, although some are clearly worse for specific issues. The key point in in this report is that loop recorders, like

17:46 so many things in medicine, require attention to detail. Yes, these loop recorders are super easy to insert, like 2 minutes. Yes, they're very lucrative, not only for the initial implant, but it's recurring income every 30 days. So my take of this is to spend a few moments in with proper insertion. Use an oblique angle over the heart. Then don't just put it in and then put glue on it. Make sure there's a good Rwave signal. Then when reviewing the tracings in clinic, take a moment to review the electrograms to exclude these false positives. [snorts] And I strongly urge programming a longer AF detection time than what comes out nomally or out of the box. One device I use is set out of the box to detect six minutes of AIB. This is

18:32 ridiculous. No one cares about six minutes of AIB. I usually change it to 1 to two hours because if you're detecting six minutes of API, it's going to be much more likely to detect PAC's while the patient's out walking or doing some other exercise. And I I wonder I mean I don't know what you all think, but maybe people with arhythmia knowledge should be the one inserting these monitors. All right, last topic is heart failure therapy when there is a serious disease like cancer. The European Heart Journal has published results of a trial called impat emphatic empatic. The goal was to assess typical heart failure therapy in patients with signs of both progression of heart failure and advanced cancer. Now, you might not

19:19 think such a trial is necessary because most doctors would forego heartfire therapy so as to maximize comfort in a patient with 1 to 6 months life expectancy. But you would be wrong. So strong as the push to use heartfire therapy. I see people who should be eating cheeseburgers eating in tresto instead. Well, German authors led by a group at Charity Hospital in Berlin decided to study it in an RCT form. It's a small modest study but it's worth talking about. Patients had to have stage 4 solid tumors and low life expectancy and already be taking paliotative care. They also had to have functional limitations from heart failure and at least two cardiovascular risk criteria. For instance, heart rate

20:04 greater than 70, NT probnp greater than 600, a high troponin, uh low EF, uh LV mass loss greater than 15%, transparent saturation less than 20% or just heft and 93 patients were then randomized to get heart failure therapy which could include sacubetrol, valart, empa ivabberdine ferra kovoxy malttos or placebo in a double blind setting. The placebo here is a nice feature because it's always better than one group getting active treatment, a pill or procedure, and the other group gets nothing. Uh, placebo arms avoid subtraction anxiety bias. So, good on them. The primary endpoint was also a strong endpoint, hierarchical endpoint, days alive and able to wash oneself.

20:51 That's a new one for me. Uh second composite on the hierarchy was ability to walk four meters three and self-reported global assessment of subjective well-being during the 30-day placeboc controlled phase. So the results the primary endpoint did not differ between groups. The win ratio was 0.95 uh with conference intervals going from 0.57 to 1.58 a p value of 0.8. Overall mortality was 32% at 30 days and that did not differ between groups. The heart failure arm did uh get a reduced probn. It increased the EF by 2.9% and a very small improvement in the uh

21:39 patient reported global assessment. So the authors include con concluded in a population with advanced cancer re receiving specialized paliotative care and high early mortality optimized heart failure therapy did not improve patients self-care ability. Now my comments I realize that this trial will not win any of these authors the Nobel Prize for medicine but kudos to them for gathering empirical data and kudos to the European Heart Journal for publishing it. If I've said this once, I've said it a thousand times on this podcast. Guideline directed heartfire therapy was proven beneficial in trials that enrolled relatively healthy ambulatoratory patients, mostly recruited from heartfire clinics, not

22:27 hospital beds, not cancer centers. The primary problem that patients in heartfire trials have is a bad ventricle. They are sick from LV dysfunction. But many patients with these other diseases are sick with LV dysfunction. For these latter patients, we would be wise to avoid therapies that don't improve quality of life, especially when time is limited. I see frail patients with multimorbid conditions who are unfortunate enough to get an echo cardiogram and then they receive all manner of nonbeneficial heart failure interventions. The emphatic trial enrolled cancer patients, but I'd extend the null trial results to patients with dementia, CKD, COPD, frailty, and other lifelimiting

23:13 conditions. For these patients, ignore the shadows that you see on the echo cardiogram and give these patients only things that improve their quality of life. Things like hamburgers, French fries, ice cream, not secubitrol valart. So that's it for this week in cardiology. As always, I'm grateful that you listened. Thank you. And remember, if you [snorts] like this podcast, please take the time, give us a rating, write us a review. If you find something that uh you don't agree with, send me a note and we can do a listener feedback. Until next week, this is John Mandola from the heart.org Medscape Cardiology. >> You're listening to This Week in Cardiology [music] from theart.org Medscape Cardiology. This podcast is

23:59 intended for health care professionals only. Any views expressed are the presenter's own and do not necessarily reflect the views of WebMD or Medscape.