Jun 13 2025 This Week in Cardiology

Medscape

a day ago

Please note that the text below is not a full transcript and has not been copyedited. For more insight and commentary on these stories, subscribe to the This Week in Cardiology podcast , download the Medscape app or subscribe on Apple Podcasts, Spotify, or your preferred podcast provider. This podcast is intended for healthcare professionals only. In This Week's Podcast
For the week ending June 13, 2025, John Mandrola, MD, comments on the following topics: Listener feedback on cardiac sarcoidosis, SA node ablation, surgical AF ablation, chronic kidney disease protection, and recruiting for clinical trials.
One of the best parts of this podcast is the listener feedback I receive. This week, David Birnie, MD, the division chief of cardiology at the University of Ottawa and an academic electrophysiologist, active in cardiac sarcoid research, sent me feedback on a paper from Dutch and American authors that I covered 2 weeks ago on May 30.
The Mostly Dutch paper studied ~1500 patients with biopsy-proved established sarcoidosis, most of which were non-cardiac. The question was how to risk stratify for an implantable cardioverter-defibrillator (ICD). The two ways were using either professional society recommendations or cardiac magnetic resonance imaging (CMR). CMR was grouped into 4 categories: one high-risk category and three low-risk categories. High-risk category was 'pathology-frequent LGE.'
The three main findings were that if either professional society criteria or CMR indicated a high risk, the rate of ventricular tachycardia (VT) at 5 years was high — about 1 in 3 patients had VT.
The second finding was that CMR had the better C-statistic compared with risk criteria from professional societies. The C-statistic was quite good at 0.86.
The third and most sobering finding was that ICDs treat sudden death and being wrong is fatal. And yet the prediction missed some patients.
And this is what Dr Birnie writes about. That is…the nuance of risk stratifying for an ICD in patients with suspected or documented cardiac sarcoidosis (CS).
One of the most common ways that CS presents is with atrioventricular (AV) block at surprisingly young ages. When a 75-year-old presents with AV block it's surely due to aging. But when a 50-year-old presents with AV block, you should think about CS.
Birnie writes, in an email:
Specifically in CS, patients who present with advanced conduction system disease and metabolically active disease seem to be at very significant subsequent risk of VT and VF, even with normal EF.
He cited a study from Circulation EP 2018, which showed that AV block, even with a normal EF, had 5-year VT rates of nearly 10%.
It is not clear why that is, and research is ongoing and Birnie says it may partly actually be related to steroids. He cited one of his papers in JCE . And the title of that paper was: 'Treatment with corticosteroids was associated with an increase in ventricular arrhythmia burden in patients with clinically manifest cardiac sarcoidosis: Insights from implantable cardioverter-defibrillator diagnostics.'
He adds that most of us in the field regard this subgroup as an exception to any sort of late gadolinium enhancement (LGE) cutoff or pattern 'rule.'
Indeed, due to increased awareness and CMR and PET, we are increasingly diagnosing these patients earlier and earlier and there are reports where the PET scan is positive and the MRI is actually negative (Eur J Nucl Med Mol Imaging. 2016 Feb;43(2):259-269). That is, the disease is picked up so early in the process that scar has not yet formed.
These findings led Dr Birnie to strongly recommend that all patients in this sub-group are recommended to have an ICD regardless of LGE.
He and colleagues are currently updating the Heart Rhythm Society guidelines on all of this, hoping to be published early in 2026, co-chaired by myself and Jordana Kron. He also wanted me to plug a cardiac sarcoid conference that he's hosting Aug 24-25 in Ottawa.
Thanks to Dr Birnie for updating me and TWIC listeners.
Pulsed field ablation (PFA) is white hot in the EP world, So it had to happen.
Heart Rhythm Case Reports , which is a really great journal, has published a case report of a group who failed to resist the urge to mess with the sinus node. Yes, a group of ablation doctors report the first case of using PFA to ablate the sinus node in a patient with purported inappropriate sinus tachycardia (IST).
I don't usually cover case reports, but I want to mention this report because of the seriousness and danger involved.
First some background on IST. IST has been around since I was training. I know this because we tried ablating sinus nodes in the 1990s and it went terribly. It either did not work, or injured the phrenic nerve or it caused symptomatic junctional rhythm.
I tried it a handful of times earlier in my practice — and it still did not work. Many of my colleagues think ablation of the sinus node doesn't work because of the difficult anatomy of the sinus node, which is a) not in one place, b) often epicardial, and c) in close proximity to the phrenic nerve. And anatomy may indeed be problematic, but the far more important reason that ablation is a bad idea for IST is that the cause of IST is unknown and likely not cardiac. And this is a problem.
If the sinus node is normal but receiving disordered neural input, then ablating the normal structure is not going to work. It's sort of the atrial fibrillation (AF) ablation problem on steroids. Yes, there are a few variants of AF clearly a forme fruste of atrial tachycardia from the pulmonary veins (PVs), as was described by the Bordeaux group 25 years ago, most AF occurs because of a mixture of triggers and substrate. We ablate the PVs and sometimes it works.
Ablating the sinus node (SN) for IST is even worse, because IST is surely caused by something extracardiac. Perhaps an autoimmune condition, more often seen in women, and surprisingly, women who work in the health field.
Anyways, the case report involved using PFA to ablate the SN. The patient already had a pacer because of an overzealous previous SN ablation attempt. The pacer was no longer pacing because the IST had recurred, which is often the case.
The authors tell us that it worked, heart rate (HR) was reduced, and the 34-year-old women felt better with atrial pacing and had persistent junctional rhythm at 90 days — 90 days.
If I had reviewed this case, I would have strongly urged the editors to not publish it. One reason is that the women had morbid obesity, so I question the diagnosis of IST. Morbid obesity can surely cause regular sinus tachycardia (ST) due to a combination of deconditioning and genetic causes. In other words, the ST may not have been inappropriate. Second issue is that we only get 90 days of follow-up. Who knows whether the HR will come back.
We also get no discussion of the long-term risks of a 34-year-old with a pacemaker. This, my friends, should be avoided in the most strenuous ways. When she is my age, she will have 30-year-old leads. Then what happens when she gets infection at a generator change. Or there is lead failure.
I want to also mention some of those who are advocating a combined surgical–hybrid thorascopic ablation approach wherein a surgeon does epicardial linear ablation to isolate the SN. I will cite a non-random comparison study from 2022 where the surgical approach compared favorably to the endocardial ablation approach — that we know does not work. This study included authors who were employees of AtriCure, the company that makes the linear ablation device.
It was not randomized, not blinded, and included no sham procedures. The control arm of endocardial ablation we know does not work.
My take of this space and condition is that research should be focused on the causes of IST. I warn my colleagues not to run amok with PFA. I understand the urge to do something aggressive for these patients, because IST causes a lot of symptoms. Care of these patients requires tons of empathy, honesty about uncertainty, patience, and reassurance that it will likely improve with time. IST is similar in some ways to postural orthostatic tachycardia syndrome (POTS) and things like supine exercise often works. But not right away.
PFA makes it easier to destroy cardiac tissue, that's good in the left atrium (LA) because we can avoid atrioesophageal fistula. But IST should infuse us with great caution. To be honest, I am not even sure there is enough equipoise to even propose a randomized controlled trial (RCT) including ablation as an arm.
The first thing to say is that surgical AF ablation, especially concomitant ablation, holds great promise. The patient is already being operated on, so the extra risks are minimized, plus these patients often have significant structural heart disease.
The problem in this space is the limited data, multiple approaches, and varied patient presentation. We simply don't have CABANA-like data for surgical AF ablation. Probably the best compilation of the data is a meta-analysis from the group of Richard Whitlock, first author, Graham McClure in EP Europace in 2018 — yes 7 years ago. They meta-analyzed 23 RCTs and found that surgical AF ablation did reduce AF compared to no ablation, but did not affect stroke or mortality. The key finding though was that only one of the trials was considered low risk of bias.
So we need data on surgical AF ablation. Not only because it's potentially very helpful, but also because it can be incomplete, or, what I call willy-nilly ablation to add an extra CPT code. This is especially bad because it can lead to LA flutters or poor atrial transport function.
A recent retrospective, observational comparison study published by a number of academic surgeons in the Annals of Thoracic Surgery has stirred some conversation online. It's stirred conversation because the group who received surgical ablation had a significant survival advantage.
The data source came from Medicare. Patients who had ablation at the time of coronary artery bypass grafting (CABG), which also includes hospital and surgeon volume as well as the surgeon's preference for doing ablation. Some do it a lot; others do it very little. That is a hint.
The primary endpoint was all-cause mortality, which I often like because it is free of ascertainment bias — alive or dead. But it is not a great AF ablation endpoint because exactly zero non-HF trials of AF ablation show mortality reduction. That's because reducing AF burden is unlikely to reduce death, because of competing causes of death. That too is a hint.
This study had a twist, though, in the comparison, which is necessary, because literally everyone knows that healthier patients are more likely to get concomitant ablation. The authors did something else to try and sort out the obvious confounding.
They did an instrument variable analysis wherein the authors used surgeon preference for ablation as the "instrumental variable" (IV variable). Here's how it works: Some surgeons frequently perform ablation (> 40% of their cases)
Some surgeons rarely perform ablation (< 5% of their cases)
A patient's treatment depends partly on which surgeon they happen to see, and this creates — from a patient perspective — a "quasi-randomization." Then the authors do two parallel analyses: both require propensity matching techniques. As-treated analysis, which is a direct comparison of patients who got ablation vs. those who didn't. This is the traditional and highly biased approach.
Surgeon-preference analysis: a patient with AF is sort of randomized in which surgeon operates on them. And if surgical AF ablation reduces mortality, surgeons who do more often should have lower mortality.
This approach is interesting because if both analyses show similar results, it somewhat strengthens confidence that the findings are real rather than due to selection bias. And that's exactly what they found: As-treated analysis : 4.40 months survival advantage
Surgeon-preference analysis: 4.96 months survival advantage
The concordance between led the authors to conclude:
'Our findings support current guidelines recommending for surgical ablation during CABG in patients with atrial fibrillation and it highlights that ablation is currently underused in contemporary practice.'
I am very warm to the possibility that concomitant surgical AF ablation may be a good thing to do for some patients with coronary artery disease (CAD) and CABG and AF.
But this study does not support anything, nor does it suggest AF ablation is underused.
Despite doing an IV analysis, which sounds great, this is still a non-randomized comparison. There is likely selection bias that cannot be overcome by matching. For instance, surgeons who do more AF ablation may see different types of patients, they may work at a better hospital and do better follow-up. These are things that randomization takes care of.
That's not the main reason I don't put any weight on this study: the main reason is that the authors found no difference in the incidence of stroke in the two groups with either the as-treated or surgeon preference analysis. So, I ask, if the ablation of AF is not good enough to reduce stroke, how does it reduce death? I don't see a mechanism for reducing death — especially given all the other competing risks. What's more, the Whitlock meta-analysis of 23 RCTs showed no reduction in death with surgical ablation.
So, your prior belief has to be extremely pessimistic and therefore it would take strong evidence to change. About 100 times stronger data than this. What's more, the mortality curves separate (by only 4 months) after 5 years. That seems a) like a minimal difference and b) separation over that long a time seems more related to healthier patients getting surgical ablation or better health systems.
I want to continue my friendships with the cardiac surgeons. They do important work. But I would also suggest as humbly as possible that they need to improve when it comes to evidence generation.
To me, there seems a huge need for proper RCTs of standardized AF ablation techniques. Endpoints should not be all-cause death. It should be AF burden and stroke incidence.
Given the increasing prevenance of AF and CAD, I would think this would be a fantastic place for the new NIH leadership to invest money in.
I realize that TWIC is a cardiology podcast, but we have yet another big randomized controlled trial of finerenone, the newest mineralocorticoid receptor antagonist, or MRA.
As an EP doctor in the Southeastern United States, are unhealthiest geography, I see many patients for AF who also have type 2 diabetes (T2D) and chronic kidney disease (CKD). It goes with the territory. If you are an AF patient obese enough to have T2D, you likely have some CKD.
I scratch my head these days, because in T2D there are so many drug classes available. Which one should we use, or should we use both—as if the drugs were generic $4 per month tablets.
New England Journal of Medicine published the CONFIDENCE trial last week. I discuss it because it provokes me to think about science questions vs marketing efforts.
Here is the deal. Patients with T2D who were taking an ACE inhibitor or ARB and had CKD as defined by an increased urinary albumin-to-creatinine ratio. I know, cardiologists don't use this number. Just think about it as enough CKD to spill protein, which is bad, and correlates with CV outcomes.
The study treatment arms were finerenone alone, empagliflozin alone, or the combination of the two.
The primary outcome was change in the urine albumin-to-creatinine (UACR) ratio. About 800 patients were randomized in the three arms.
The main result was expressed in relative risk reduction: At day 180, the UACR ratio results were: Finerenone alone had a 32% reduction
Empagliflozin alone: 29% reduction
And the combination therapy had a 52% reduction from baseline
So, the reduction with combination therapy was 29% greater than that with finerenone alone and 32% greater than that with empagliflozin alone; both were highly statistically significant.
Adverse events and drug discontinuation were similar, though there was a decrease in the estimated glomerular filtration rate (eGFR) of greater than 30% at day 30 in two times more patients in the combination group, though GFR improved at study end when the drugs were stopped.
In addition: 'the frequency of hyperkalemia, was relatively lower by approximately 15 to 20% with combination therapy.'
For the optimistic or maximizer-docs, the previous model was sequential initiation of these two drug classes. This trial shows that simultaneous initiation worked better.
But there were many critical appraisal issues — so many in fact, that it makes me wonder whether this was a marketing exercise.
The most obvious issue is the endpoint. While we can say that cholesterol levels, or blood pressure (BP), or creatinine, these are surrogate markers that most patients don't care about. The UACR ratio is even more esoteric and likely even less of a surrogate marker than LDL or BP.
The point of kidney protection strategies is to lower kidney outcomes — new dialysis, even doubling of the creatinine, or CV events. I realize UACR ratio is easier and may be an earlier sign, but I will quote the authors own words to underscore the uncertainty:
Quote: 'We speculate that, taken together, these data suggest that the reductions observed with combination therapy with empagliflozin and finerenone will probably correlate with meaningful reductions in the risk of progression of chronic kidney disease.'
I mean, that is truly remarkable in the NEJM. We speculate, probably . Wow.
Then there is the matter of a 6-month trial. These were 67-year-old patients How long are we supposed to use these two classes of meds? Forever? Really?
Then there is the matter of the 44-day screening period before randomization. They screened 1600 patients and excluded half. Why? They don't say. Highly selected patients, though, is not nefarious — lots of trials pick the perfect patient. But I mention the screen to randomized information because it speaks to the generalizability of this data.
Finally, and I sound like a broken record, whenever finerenone is used in a trial, there should be an arm for spironolactone or eplerenone. That way, we would have an idea of the incremental value, if any, of the new nonsteroidal finerenone.
In the end, I think CONFIDENCE was a marketing exercise. It does not change practice, and 6-month data of a very surrogate endpoint hardly meets any regulatory standard.
Two UK Cardiology Trials Struggle to Recruit Patients
Journalist Sara Freeman reports from the meeting of the British Cardiovascular Society meeting in Manchester, which I was lucky enough to be asked to speak last year, on a session describing difficulty in recruiting patients for two important clinical trials.
This is an important topic. I can add a third trial that is also noting slow enrollment.
In the UK, the CRAAFT-HF trial compares surgical vs catheter ablation in patients with HF. This is a hugely important question. For multiple reasons: the first is that despite my friendship with Nassir Marrouche, Christian Sohns, and Philipp Sommer, the principal investigators of CASTLE and CASTLE HTx, the numbers of HF patients in trials of AF ablation are few. The event rates are low, and the evidence for AF ablation in HF is limited.
CRAAFT-HF would add to that evidence base. What's more, patients with HF often have atrial structural disease making them less likely to be pulmonary vein isolation (PVI) responders. And surgical AF ablation could also remove the left atrial appendage epicardially, a maneuver proven effective in the LAAOS III trial. I think surgical AF ablation could beat catheter-based ablation.
The main problem with recruitment is the strict entry criteria. Indeed, this is a problem in HF and AF, as the chief weakness of the CASTLE trials was the highly selective entry criteria.
The second UK trial having problems is the BRITISH trial of using scar on CMR to determine ICD implantation in patients with non-ischemic cardiomyopathy (NICM). BRITISH has a really brave trial protocol.
Patients with NICM LVEF < 35% who have a scar on CMR would be randomized to either ICD or CRT-D vs implantable loop recorder (ILR) or CRT-P. (Basically no backup ICD). This is exactly what we want to know. It's a beautiful study question.
The third trial that I traveled to Berlin to speak about this winter is the bravest of them all: The PROFID trial, which seeks to replicate the MADIT-II trial 20 years after. MADIT-II was ischemic CM with ejection fraction < 30% and the ICD reduced mortality with a number needed to treat of < 25. The idea behind PROFID is that sudden death rates in HF have fallen, medical therapy has improved, and the delta between the ICD arm and medical therapy arm is surely less. Well, PROFID also has had slow enrollment, for a couple of reasons: one is that there aren't that many terrible ischemic CM patients out there anymore, thanks to the IC community and rapid percutaneous coronary intervention (PCI) during ST-elevation myocardial infarction (STEMI). The second issue is that, at least in Germany, if you are a doc who randomizes a patient to PROFID and they are randomized to the medical arm, you lose reimbursement for the ICD.
I highlight this topic mostly for my EP colleagues. While PFA holds all the excitement, we have stalled in terms of new major evidence development. PFA is nice, especially for patients, but if we are to maintain are stature as a profession, we need to support those who seek to answer important questions, like CRAAFT-HF, BRITISH, and PROFID.