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 November 7, 2025, John Mandrola, MD, comments on the following topics: Listener feedback on non-culprit PCI in STEMI, a major cardiac result in patients on hemodialysis, news on GLP-1 agonists, a dubious stroke trial, and an AHA preview.
Listener Feedback
I received some pretty harsh feedback on my coverage of the nonsignificant iMODERN trial of fractional flow reserve (FFR)-guided non-culprit percutaneous coronary intervention (PCI) at the time of STEMI. In my discussion, I made the case that the nonsignificant results were entirely consistent with the mountain of evidence that PCI does not reduce outcomes in stable disease over medical therapy.
What did not make sense to me was the guideline-support of non-culprit PCI. Why would PCI of a stable lesion (over medical therapy alone) not improve outcomes in every setting except STEMI. A right coronary artery (RCA) total occlusion in an inferior MI that coexists with a left anterior descending artery (LAD) lesion; it means that the RCA lesion is the acute lesion and LAD is the chronic lesion. How could PCI of that LAD improve outcomes when no other trial of stable lesions has shown benefit to PCI over medical therapy?
iMODERN found no benefit for doing complete vs culprit-only revascularization.
To support my thesis, I appraised the PRAMI, COMPLETE, and FULL REVASC trials. PRAMI was underpowered; the COMPLETE trial was positive but it was driven only by a myocardial infarction reduction (and of course it's difficult to measure periprocedural MI at the time of a STEMI); and FULL REVASC was a negative trial.
An academic interventional cardiologist wrote:
It is unclear to me why you write off the results of COMPLETE, which is by far the largest and most definitive trial in this space (4000 patients). The composite hard endpoint of death or MI was reduced and this signal continued to diverge over the course of follow up. The comment about periprocedural MI to discount the results would seem irrelevant.
My response to that comment is I disagree. Over and over again, periprocedural MI seems very relevant. In COURAGE, and of course in the EXCEL trial. Also, I would ask my colleague, since COMPLETE is so definitive, and large, with its 4000 patients, why didn't such a profound reduction of MI also reduce cardiovascular or all-cause death? That it didn't makes me wonder about a) the importance of MI reduction as an outcome, see David Brown paper on surrogacy in CV trials in JAMA-IM, and b) perhaps periprocedural MI is more important than he thinks.
His second criticism was that I missed two important trials in the space: Compare-Acute and DANAMI-3-PRIMULTI. These trials, he writes, show lower composite endpoints from complete revasc.
* Compare-Acute trial: Small numbers of events, driven mostly by revascularization in the non-culprit PCI group. Unblinded. Treating cardiologist knew the anatomy.
* DANAMI-3-PRIMULTI: Another small trial with few endpoints totally driven by revascularization. Which is of course biased due to the open-label nature.
So this is why I did not include these trials in my discussion.
Finally, my colleague objected to my comment that I felt that the CULPRIT-SHOCK trial should be included in the meta-analyses of culprit-only vs complete revascularization. I feel this way, because CULPRIT-SHOCK asked the same question (culprit-only vs complete) in patients with STEMI and cardiogenic shock, a group that you would think full revascularization could help.
My colleague disagreed and cited two aspects of the trial. One was the results and the other a problem with the protocol of the trial -- namely, that it included doing chronic total occlusions (CTO). He made the point that there was obviously a much higher rate of renal replacement therapy (RRT) in this complete arm, and looking at results, it was predominantly driven by this.
He wrote:
"I can't imagine any interventionalist in their right mind trying to open up cto's/non infarct complex bifurcations in a shock patient, but that's what the study protocol mandated."
Well, I disagree again: both components of the composite primary -- death and RRT -- were statistically lower in the culprit only group.
And, yes, CTO was included in the protocol but were present in only 22% of both groups. So the argument that trial procedures drove the negative result for complete revascularization does not hold water.
Thanks for listening and commenting.
Now I will mention, by name, Dr Adrian Lonescu from Wales. I have never known anyone from Wales. Thank you, Adrian.
His message -- that I was biased against stents -- was couched in kinder words and tone.
We use stenosis severity to guide PCI and because of the partial overlap of diameter reduction with instability we stent -- by chance -- some unstable lesions. We also use FFR/iFR and, again, by chance we stent some unstable lesions.
We do not rigorously know that PCI is the right thing to do in unstable lesions, and we do not know how to identify them reliably, but there are hints. This should be addressed in multicentric RCTs.
So my objection is not against your primary argument. I object to the blanket dismissal of "full revasc" in ACS. There are nuances. I know that you pursue objectivity, and I have often heard you saying, "Not so fast -- there are nuances here!" And this is one such area, in my opinion.
My rebuttal to Adrian in Wales is that his argument does makes sense. I do not do PCI; all I can do is read the trials. But from being an interventional electrophysiologist, I understand that trials may not represent every scenario we come across. Trials give us an average effect size for patients similar to those in a trial.
That said, any time we think a scenario is outside a trial, it's on our profession to study that specific scenario.
I received a third email on iMODERN, this one in support of my view. I won't go into the details, because this person sent it as a letter to the New England Journal of Medicine that he hopes NEJM will publish.
Suffice to say that if you were to analyze the iMODERN data in a nonfrequentist way, the probability of large benefit or large harm would be quite low, and the probability of no difference would be high.
And if there is no difference, you would, obviously, favor doing fewer PCIs and stents.
Huge Cardiac News for Patients With ESRD
The American Society of Nephrology meets this weekend and there was a major study presented today and published in the NEJM. It's called PISCES. It's only been out for a few hours, so this is a very early look.
A group of investigators led by the University of Toronto presented positive (very positive) result of the PISCES trial of high-dose fish oil supplementation in patients with end-stage renal disease (ESRD).
This was a cardiac outcomes trial -- as heart disease is the leading cause of death in patients with ESRD.
The wild thing is that fish oil supplantation is highly controversial in the general population of patients: you have REDUCE-IT, with its icosapent ethyl looking positive, but the hook is that the mineral oil placebo may have been harmful to patients in the placebo arm. Then you have the STRENGTH trial of DHA/EPA combination having no effect.
Our priors therefore ought to be quite pessimistic, because a) fish oil supplementation is not clearly beneficial overall, and b) it's hard to show benefit of anything in patients with chronic kidney disease (CKD). Statins, for instance, have clear benefits in the general population but do not lower CV outcomes in ESRD. Neither do mineralocorticoid receptor antagonist (MRA) drugs.
The thing about patients on hemodialysis is that they often have lower levels of omega 3 fatty acids than people in the general population. So there is that.
PISCES studied citrus-flavored, omega-3, polyunsaturated fatty acids in four 1-g capsules containing a total of 1.6 g of EPA and 0.8 g of DHA vs the citrus-flavored corn-oil placebo. The primary endpoint was a composite of all serious cardiovascular events including sudden and nonsudden cardiac death, fatal and nonfatal MI, peripheral vascular disease leading to amputation, and fatal and nonfatal stroke.
The trial took a long time to finish, beginning in 2013 and finishing in 2019. Just over 1200 patients were randomized. Average follow-up was 3.5 years. Patients were 64 years old, about 40% female, and two thirds had known CV disease.
The rate of serious cardiovascular events was significantly lower in the fish oil group than in the placebo group (0.31 vs. 0.61 per 1000 patient-days; hazard ratio, 0.57; 95% CI, 0.47-0.70; P < .001).
The trial reports actual percentages of participants experiencing at least one event:
* Fish oil group: 20.8% (127 of 610 participants)
* Placebo group: 33.7% (208 of 618 participants)
That is a 12.9% absolute risk reduction (ARR) and number needed to treat (NNT) of 8.
Each component of the primary endpoint was statistically lower in the fish oil arm.
* CV death: 45% lower
* MI: 44% lower
* Peripheral vascular disease: 43% lower
* Stroke: 63% lower.
The Kaplan-Meier curves show benefit starting to occur after 6 months, and the benefits were increasing over time. There were no serious adverse events.
My Comments
Whenever you have such strong data, it's important to be skeptical.
On the one hand, you could say it's been hard to show that normal interventions such as statins and MRA don't seem to work in patients on dialysis. Neither do ICDs for CV prevention.
What's more, it's not a sure bet that fish oil works in the general population.
Now we have a situation where there are huge benefits -- both clinically meaningful and statistically robust. We see a clear heterogenous treatment effect.
How should we handle a situation where our priors are extremely pessimistic, but the data is extremely persuasive? It's akin to seeing a massively positive stress test in a very low-risk person with atypical chest pain.
Well, one way is to look back into some other trials, as the authors explain. There was a trial called the FISH trial, published in JAMA in 2012, fish oil vs placebo for graft patency. The fish oil did reduce graft thrombosis but did not quite reach statistical significance; however, a secondary outcome of CV events indeed was 57% lower in the fish supplement arm.
You could also posit that ESRD patients have very low levels of omega-3 fatty acids, and supplementing them may work by repleting an important element.
But in the end, my take is that the strength of this data can overcome the negative priors. Yes, of course, a confirmatory trial would be great, but PISCES took more than a decade to finish. What to do for the next 10 years? I'd say that given the lack of adverse effects, we ought to begin using these agents in patients on hemodialysis.
Obesity News
First news is a White House press conference.
In the oval office yesterday, the Trump administration announced a deal struck with drugmakers Eli Lilly and Novo Nordisk to offer their GLP-1 agonist drugs at steep discounts for certain Medicare and Medicaid patients.
The deal also includes lower prices for GLP-1 drugs for Americans who use cash to buy them through a new government website, expected to debut next year, that will allow people to buy drugs directly from companies.
For Medicare and Medicaid, semaglutide and tirzepatide will cost the insurance programs $245 a month. Americans on Medicare will only be responsible for a $50 copay, the White House said, and Americans on Medicaid generally pay little to no cost for prescription drugs.
For oral versions of the drugs, which first need to face FDA approval, the price could be as low as $149 to Medicare and Medicaid.
Novo Nordisk asked the FDA to approve an oral version of semaglutide in May, and Eli Lilly plans to submit an application for its weight-loss pill candidate to the agency by the end of the year.
Then there will be a direct-to-consumer site where patients will pay lower (but still pretty high) prices for the drugs.
This could be a huge deal, because now, patients with obesity really can't get these drugs covered unless they have diabetes, established heart disease, or sleep apnea -- which are the FDA-labeled indications.
I say could be a huge deal because this is just an announcement and putting it into action is yet another thing. Expanding access to GLP-1 drugs might have a direct impact on EP care, because so many of our patients with atrial fibrillation (AF) have obesity, and we know that weight loss is beneficial.
Amylin Agonists
The second piece of news is that The Lancet has published a phase 2 trial of a selective amylin receptor agonist called eloralintide, which is delivered via subcutaneous weekly injections.
Weight loss was dose dependent and up to 20% in the higher doses. Notably, unlike the GLP-1s, heart rate decreased not increased.
There were also reductions of all metabolic measures, such as A1c, lipids, and C-reactive protein.
There was also a lower rate of GI adverse events when compared to rates in the GLP-1 trials.
I did not know this, but there is already an amylin agonist approved. It's called pramlintide, and is used as an adjunct to mealtime insulin for the treatment of patients with type 1 and type 2 diabetes. Clinical studies in people with and without type 2 diabetes showed modest weight loss, although pramlintide has not been approved for weight management.
My Comments
I don't think there's much to say, other than this seems promising. It could be a replacement for people who can't take GLP-1 drugs. The next step, of course, is you have to show outcomes are better; weight loss is a surrogate, right? What we would want to know is does the drug reduce CV outcomes in patients with diabetes and CV disease. And is it safe in phase 3 studies. That much weight loss in patients with obesity, though, is surely encouraging -- as is heart rate is decreased rather than increased.
Class Effect of GLP-1 Agonists
EHJ Open published a post-hoc analysis of SURMOUNT-5 trial comparing tirzepatide with semaglutide in patients without diabetes. NEJM published the original trial in May. It found that tirzepatide was superior to semaglutide for weight loss (20.2% reduction with tirzepatide vs 13.7% reduction for semaglutide).
The most recent analysis, led by Professor Mamas Mamas, compared the change in predicted 10-year CV risks between the two therapies.
The impact of cardiovascular risk reduction was estimated as the projected preventable CVD events over 10 years for tirzepatide and semaglutide in the United States.
Here's how I understand it:
They first excluded patients with established disease, then they estimated a 10-year CVD risk score using BMI-based, sex-specific risk equations. These were developed from the Framingham heart study.
The model and equation uses sex, age, BMI, systolic blood pressure (SBP), treatment for hypertension, smoking, and diabetes status. It sounds like a pooled cohort equation (PCE).
They then graph the reduction in predicted risk based on the measures in SURMOUNT.
They find that tirzepatide results in greater reductions in predicted CV risk at each timepoint in the trial.
They then apply these risk reductions to NHANES data (NHANES being a longitudinal study of US adults) and they find that if all the eligible population received treatment with these two drugs, an estimated 2 million CVD events could be potentially prevented with tirzepatide over 10 years, compared with 1.15 million with semaglutide.
The study got a lot of attention online, but I don't love the analysis. First, it's funded by Lilly, the makers of tirzepatide. And it simply plugs the greater reductions in weight with tirzepatide in equations and finds predicted risks will be better with tirzepatide. And if applied to huge populations, there will be nearly a million fewer cardiac events with tirzepatide over semaglutide.
I see this as marketing for tirzepatide. This is not the way to compare two drugs. The way to compare them is to do SURMOUNT-5 with enough patients to measure outcome data.
Equations are fraught with issues. In the real world, semaglutide may perform better. There may be other effects besides weight loss. Not only that, semaglutide already has proven benefits as a cardiovascular disease-modifying agent in the SELECT trial. Semaglutide has proven benefits in patients with diabetes.
I think tirzepatide is likely to have similar benefits. But without trials, it's just an assumption. We accept that many drugs have a class effect, statins, ACE inhibitors and SGLT2 inhibitors, for instance, but we accept this class effect not because of predictions from equations, but rather trials.
Let's wait for trials before concluding the same for GLP-1 agents.
A Problematic Trial in Stroke Care
Speaking of GLP-1 drugs, JAMA-IM has published the LAMP trial of liraglutide in acute minor stroke or high risk transient ischemic attack (TIA) in patients with type 2 diabetes.
The idea is that patients with diabetes have an increased risk of stroke. Minor acute strokes and TIAs can be harbingers of big strokes. GLP-1 drugs have proven benefit in patients with diabetes. So, could GLP-1 drugs reduce the rate of recurrent stroke after a small stroke event?
LAMP was conducted in China; 636 patients were randomized within 24 hours of stroke symptoms to liraglutide or usual care -- no placebo was given.
The trial was terminated early for funding and low enrollment issues.
The results were amazing.
Within 90 days, 25 patients (7.9%) in the liraglutide group and 44 (13.8%) in the control group experienced stroke recurrence (hazard ratio, 0.56; 95% CI, 0.34-0.91; P = .02)
A significantly higher proportion of patients (modified Rankin Scale score ≤ 1) in the liraglutide group (87%) vs 78% in the control group achieved excellent functional outcomes (odds ratio, 1.95; 95% CI, 1.28-3.00; P = .002).
The Kaplan-Meier curves show separation within a day or two days or three days of liraglutide and that separation continues on.
My Comments
There are some good critical appraisal lessons here.
One is that the results seen too amazing. A nearly halving or recurrent stroke and ARR of nearly 6% is massive. And the P-value barely makes significance at .02. This tells me is that the results are fragile.
What's more, I have trouble explaining the early separation. How does a GLP-1 drug act that fast?
The trial was open label, so one group knows they get an injection and the other gets no injection. That to me is a much weaker design than a proper placebo, though it's common in structural cardiology, (eg, tricuspid valve procedures).
Early termination also has the effect of accentuating effect sizes.
So, in sum, this finding is unlikely to be replicated. And I am torn if there is even enough here to spend the money on a confirmatory trial. I sort of doubt it. But you may feel differently.
AHA Preview
The AHA meeting happens this weekend in New Orleans. I wrote a preview column of some big trials. Indeed there will be important studies.
One is ADAPT AF-DES. This is similar to AQUATIC, which found oral anticoagulation (OAC) alone was superior to OAC plus aspirin in patients with CAD and AF. ADAPT is a South Korean trial studying OAC plus clopidogrel vs OAC alone.
Another is OCEAN -- an RCT of stopping vs continuing OAC after a successful AF ablation. I really look forward to this one, because I think we overtreat many patients who have had no AF after ablation.
A third is CLOSURE-AF, an RCT of left atrial appendage occlusion (LAAO) vs best medical care in patients with AF at high risk for stroke and bleeding. The worry here is power. Only 912 patients and a non-inferiority endpoint with events expected to go in different directions. This is a problem because when thrombotic events go down and bleeding events go up you make non-inferiority easier to reach. Non-inferiority trials are supposed to show non-inferiority for efficacy, and superiority for safety. See the direct oral anticoagulant (DOAC) vs warfarin trials.
The fourth biggie is the VESALIUS trial of 12,000 patients without heart disease given evolocumab on top of statin use. Recall that the original PCSK9 trials were really secondary prevention trials. The question is how well these drugs perform as primary prevention in lower-risk patients. The question here won't likely be whether the drug works -- I'll bet it does -- it will be effect size relative to cost. You'd also want to know whether there is a real difference in death or CV death. Because the way MI is going in the modern era (that is, it's so much easier to get diagnosed with MI), a reduction in MI without CV death is not that substantial. What I mean is that an MI in 2025 is clearly not the same as one in 1995.
One trial that I missed in the preview is the SHAM-PFA trial. It's Monday. Featured science PFA ablation vs a sham. I can't wait to see what happens.
Also on Sunday, we hear the results of the meta-analysis of the modern beta-blocker trials (REBOOT, REDUCE-AMI, BETAMI, DANBLOCK, CAPITAL-RCT). Remember at ESC the meta-analysis was only patients with mildly impaired ejection fraction (EF). The vast majority of patients in these trials had preserved EF. I've predicted that the combination of these trials will likely show nonsignificant benefits, and if so I hope we get guideline changes made so that we are not compelled to give nonbeneficial drugs. I love the beta-blockers post-MI story because it is a classic example of how seminal trials can have expiration dates.