Oct 09 2026 This Week in Cardiology

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, Apple Podcasts, Spotify, or your preferred podcast provider. This podcast is intended only for healthcare professionals.Listener FeedbackI was honored to receive an excellent note from Professor Eva Prescott, who is Head of the CV research team at University of Copenhagen and chair of Clinical Practice Guidelines of ESC 2026–2028. She wrote regarding my coverage of the POET II trial of short vs standard dose antibiotics for infective endocarditis (IE). I was very critical of the endpoint of days alive without antibiotics, as well as the choice to emphasize that endpoint over the much more clinically relevant recurrent IE or bacteremia which was 3x greater in the short-course antibiotics arm. I enjoyed your critique of POET II and very much agree with your point that the hierarchy of endpoints seems upside down. Relapse of endocarditis is surely more important to patients than spending an additional two weeks on antibiotics. There is, however, another aspect that puzzled me, which I don’t think you mentioned. There is something almost tautological about showing superiority for days alive without antibiotics when the intervention is, by definition, earlier discontinuation of antibiotics. The important question is therefore what those additional antibiotic-free days actually represent for patients. This seems particularly relevant in light of the original POET trial, which established that clinically stable patients can switch from IV to oral antibiotic treatment and be discharged. Oral step-down treatment was also widely used in POET II. Thus, for many patients, the 13 additional days alive without antibiotics may not represent 13 additional days out of hospital or free from IV treatment, but rather 13 days at home without oral antibiotics versus 13 days at home taking oral antibiotics. That seems a rather modest patient-relevant benefit to weigh against an increase in relapse from 1.6% to 5.1%. It would be particularly informative to know whether the shorter strategy actually resulted in fewer days in hospital — a prespecified endpoint that does not appear to be reported.It made me think a little of The Emperor’s New Clothes by Hans Christian Andersen: once you look past the statistically significant primary endpoint, I am not sure the conclusion is quite as well dressed as it first appears.More broadly, I think POET II illustrates something you have also discussed before: Hot Line discussants at major congresses should be more willing to be truly critical — not only asking whether a trial meets its prespecified statistical endpoint, but whether that endpoint reflects a benefit that matters to patients. This would have been an excellent trial for that discussion.I have almost nothing to add to this wonderful comment. I love the tautological angle. As it’s always good to reinforce lessons from logic and philosophy.PVI-SHAM-AF Trial Back in the News PVI-SHAM-AF Trial Results Surprised the InvestigatorsDoes AF Ablation Work? That’s in Question After PVI-SHAM-AFThe Heart Rhythm journal has a point/counterpoint on the German PVI-SHAM-AF trial, whose negative results on the primary endpoint of quality of life was presented during an ESC Hotline session.DJ Lakkireddy led a number of key opinion leaders in EP in a strong critical appraisal of the trial’s conduct, results and interpretation. Dr Jared Bunch led another group of authors in defending the trial and sham controls.The Lakkireddy group titled their piece: “Why Sham Studies Are a Sham in 2026! Sham-controlled AF ablation trials as a case study.”They led with:On the morning of 30 August 2026, in a Hot Line session at the ESC Annual Sessions in Munich, cardiology learned that catheter ablation for atrial fibrillation (AF) does not improve quality of life. That is not what the trial showed. It is, however, what the press release said, what the wires reported, what the algorithm amplified, and what a patient in a waiting room somewhere will quote back to an electrophysiologist next Tuesday.I find it intensely rich that many of the authors of trials with endpoints such as nonprocedural bleeding and non-inferiority margins not tested with relative risk differences so as to declare non-inferiority when there should not have been non-inferiority, are complaining of misleading press releases and trial results.But I won’t say any more about the other studies. Let me address and rebut their criticisms directly.First, let me remind you what PVI-SHAM-AF found.PVI-SHAM-AF screened 1200 patients with symptomatic atrial fibrillation (AF) from nine centers in Germany and Poland. After more than 900 declined to participate, 262 patients were randomized 2:1 to catheter ablation or a placebo procedure.The placebo procedure included placement of two femoral venous sheaths, heavy sedation (similar to a standard ablation), and cardioversion if the patient was in persistent AF.The primary endpoint was quality of life as measured by the AF Effect on Quality of Life (AFEQT) score. This is a 0–100 scale that measures the impact of AF on four domains: symptoms, daily activities, treatment concerns, and treatment satisfaction. It’s been validated across multiple trials and is believed to capture what patients most care about. Higher scores reflect better quality of life. The authors also measured AF episodes and burden.Cardioversion was performed in 54% of patients in the ablation arm and 36% of patients in the sham arm.The principal assessment of blinding was done after the procedure, before discharge, and showed that blinding was near perfect.Surprising ResultsThe AFEQT score improved in the ablation group from 61.3 points at baseline to 81.1 points at 6 months. But it also improved in the sham arm from 59.2 points to 74.9 points. The between-group difference in change from baseline to 6 months was 2.6 and not statistically significant.Ablation reduced episodes and burden of AF. Clinically detected recurrence or AF detected during 7-day Holter monitoring occurred in 27% of patients in the ablation group vs 48% of patients in the sham procedure group. AF burden on 7-day Holter monitoring at 6 months was lower with ablation than with sham (11.6% vs 23.6%). Both differences were statistically significant.The authors concluded that despite a reduction in AF episodes and burden, catheter ablation did not demonstrate superiority over a sham procedure for improving AF-related quality of life at 6 months.In my written piece, which I will link to, I was struck by the humility that these trial results should bring all of us AF ablators. Secondly, I felt that the Berlin-led authors reminded us of the power of placebo effect. Thirdly, PVI-SHAM-AF reemphasized the fragility of single small trials, as two prior small trials had found a placebo resistant effect of AF ablation while this larger trial was negative.I never understood the pushback against this trial. It was truly trying to answer an important question. I would ask you all a question: Don’t doctors wear two hats? One hat is to help patients feel better — however much pragmatism this requires — but the other hat is one of genuine scientific curiosity. Are we not one bit curious that drawing the same circles around PVs in every patient is a goofy strategy for such a science-based field? In all other arrhythmias, we proudly and meticulously map arrhythmia circuits before ablating and in AF we ablate circles that takes 27.2 minutes to do. Aren’t we also curious about the patient who returns to your office feeling super good after your ablation and is in AF?The article by Lakkireddy et al argues that ablation fails the 3 tests that justify a sham, that is sham trials have been used for procedures with 1) unproven mechanisms, 2) a purely subjective endpoint and 3) no objective biomarker. They then say that AF ablation has a proven target, measurable proof of isolation, and rhythm monitoring.But these are ridiculous claims. AF ablation has a totally unproven mechanism; we have only theories of why pulmonary vein (PV) isolation works. For instance in the GAP-AF trial from Hamburg, where patients had PVI without and with purposeful gaps in PV isolation, 70% of patients in the complete PVI arm had reconnection but nearly 40% of these patients had no AF, while in the 30% of those with durable PVI, more than half still had AF. In addition, while we have rhythm monitor of AF episodes as a marker of success, there is a less than robust correlation between quality of life and actual rhythm episodes. So it’s an imperfect marker. Plus, some of the examples the authors list include renal denervation, which had an incredibly objective marker of blood pressure, and ORBITA, which had the objective marker of exercise time on a treadmill. So many of their arguments and examples don’t work.Lakkireddy et al then argue that the sham in the PVI-SHAM-AF trial wasn’t a placebo, because patients got sedation, cardioversion, and antiarrhythmic drugs and close follow-up. They claim the trial was actually a comparison of two rhythm control strategies.They also claim that the endpoint was the wrong instrument. That is, the AFEQT has a 4-week recall window, while AF episodes can be weeks apart, so both arms may report no symptoms. And mildly symptomatic patients have little room to improve (they call this a ceiling effect). And a single 7-day Holter monitor misses a lot of AF.This too is a nonpersuasive point because AFEQT is a widely used device. Further, if you think these were mildly affected patients go look at the EARLY-AF trial; these too were extremely low-burden patients.Lakkireddy et al claim the null result was over-read in that the confidence intervals around the 2.6 nonsignificant difference had wide CI so the trial couldn’t rule out a meaningful benefit; absence of evidence isn’t evidence of absence, they claim.Yes, the CIs are wide, but come on, if AF ablation was so great, we should have seen a large effect. And two previous smaller trials of sham controls delivered positive results, despite using cardioversion and sedation.They also suggest dubious ethics in that four sham patients had serious adverse events (AE), including a stroke. I, too, am sad that there were AE in the sham arm, but to have a proper sham trial you must mimic the procedure as close as possible. What’s more, the one stroke occurred 2 days after a procedure that included a cardioversion. This could have easily occurred after the procedure in the active arm.I obviously agree much more with the Bunch et al editorial. They write that the purpose of AF ablation is to relieve symptoms, so symptoms should be the endpoint of the trial. I did not know this, but they write that patient-reported outcomes aren’t “soft” either: AFEQT’s test–retest reliability (0.80) beats many anatomical and physician-rated measures.Their second point is that the active control is a feature not a flaw. Withholding cardioversion or drugs from someone in symptomatic AF would be unethical. The right question is whether ablation adds symptom benefit beyond good medical care plus the experience of a procedure. If ablation is transformational, it should clear that bar. I could not agree more strongly with this piece. By doing everything but the ablation you isolate its effects.Bunch et al dismiss the issue of selective recruiting, because all trials enroll healthier, less representative patients.Bunch et al also dismiss the ethics complaint in the usual way, in that a small sham control trial is a better ethical trade compared with exposing millions to the real risk of a nonbeneficial therapy.Patients enrolled in these trials simply have to delay their procedure for a short trial follow-up. There is also the benefit that a simple cardioversion would work. I don’t see the ethical dilemma.My final comment: I am glad HR published the side-by-side views. I favor the Bunch et al piece, and I commend the Berlin-led team for conducting such a robust trial.If you aren’t deeply curious about what we do, and what the doctors of the future will think of us, I feel bad for you. Because being curious about things is part of the joy of medical practice.ICDs in Hypertrophic CardiomyopathyCirculation has published a large international registry of implantable cardioverter-defibrillators (ICDs) in patients with hypertrophic cardiomyopathy (HCM). The data came from 25 HCM centers, from 1992-2024 and included nearly 3400 patients.The participants had received ICDs at a mean age of 47 (±17) years.Over follow-up, 550 patients (16%) experienced at least 1 appropriate ICD therapy (2.6%/year),Appropriate therapy for secondary prevention was 35%, while appropriate therapy for primary prevention was 15%Median time to first appropriate therapy of 4 years after ICD implantation; 16% of these received their first appropriate therapy more than 10 years after implantation, and 47% experienced multiple interventions.Independent predictors of appropriate ICD therapy included unexplained syncope, left ventricular apical aneurysms, left ventricular systolic dysfunction, and nonsustained ventricular tachycardia on ambulatory monitoring.Of the 3140 primary prevention patients, 94% survived and 6% died, including 68 due to HCM (0.3%/year), predominantly of end-stage heart failure (n = 43) or stroke (n = 9).Interesting findings were noted in regard to the ESC risk score, which purports to qualified patients with this condition for an ICD. Get this:In the patients with ICD interventions within 5 years from device implant, ESC risk scores were consistent with low risk in 37%, intermediate risk in 19%, and high risk in 44%. Although the majority of appropriate interventions occurred in patients with low- or intermediate-risk scores (56%), patients with high-risk scores had the highest 5-year rate of appropriate interventions (15.6% at 5 years; P < .001; Figure 3), whereas there was no difference in appropriate intervention rates between patients with low or intermediate scores (9.0% vs 8.7% at 5 years; P = .51).Adverse events: inappropriate shocks occurred in 10%, 5% had lead malfunction, 2% infection, and 2% other. The yearly rate of complication was 1.7% and the cumulative probability of a complication 28% by 20 years.The authors write that:these novel data demonstrate the effectiveness of the ICD initiative, which has probably favorably altered the natural history of many patients with HCM worldwide.An accompanying editorial added this:This study supporting the benefits of ICDs in patients with HCM around the globe should serve as a call to action for the cardiology community to advocate for patients with HCM to receive evidence-based, equitable cardiovascular care.My Comments While it would be best to have RCT-level data, this is good data to consider. Of course, it is not a mandatory nationwide registry, so you get the best centers. The data probably represent an accurate estimate of shock rate (the ICD benefit) but the complication rate is surely a best-case, as you’d expect higher rates at smaller community hospitals as well as the fact that at least some complications present to other hospitals and remain unknown or unmeasured.This registry also looked only at transvenous ICDs. The editorialists point out that we now have SCD and extravascular substernal ICDs. I am not sure how beneficial these will be because there isn’t enough data and I suspect inappropriate shock rates will be higher while lead-related issues may be less.What I will do now is a highly imperfect exercise, but let’s compare shock rate to AE rate. I had Claude help me out here: in the first year, the appropriate therapy rate is 3.9% vs 4.6% complication rate. More harm then benefit. But the curves cross around 5 years with appropriate therapy at 11% vs AE 10% and by 20 years it is 35% shock vs 28% AE.While you might argue that an appropriate shock is lifesaving and an AE causes morbidity, I would somewhat agree but add that not all appropriate therapies are lifesaving because ventricular tachycardia often stops on its own.I think we underappreciate the complication rate of ICDs in HCM. Nearly a third will have one by 20 years. Think about that when implanting an ICD in a 40 year old. Age 60 is right around the corner — which I can attest to.The other striking finding was that the beloved ESC risk score was just ok. While those with the highest ESC score had the highest rate of shocks, more than half of appropriate therapies in primary prevention were ESC 5-year risk of low or intermediate. A total of 37% of early events were low ESC 5-year risk, 19% intermediate, and 44% high.This is similar phenomenon to the general population of sudden death and ICDs, right? While heart failure patients with low ejection represent a high sudden cardiac death (SCD) risk — and those are the ICD eligible people — the vast majority of SCD occurs in patients with normal left ventricular ejection fraction (LVEF).Here, the high-risk patients have the highest rate of events, the absolute numbers are greatest in the low and intermediate risk. My point is that risk scores get you in the ballpark but have a pretty modest predictability.I don’t know about you, but when I look at this data, I think we should have a global RCT for primary prevention. Secondary prevention, yes, surely, an ICD seems wise, but for primary prevention, with the shock rates being quite similar to AE rates, and overall mortality low, I’d love to see every patient who gets one of these be in a pragmatic RCT. I am not sure we could show an average mortality benefit.Agenda LaunderingThe BMJ has published an interesting analysis paper on “agenda laundering” in critical care. The first author is Khalil Chaïbi. What is agenda laundering?The authors lay out two sorts of commercial influence in medicine. One is in the design and conduct of clinical trials. I go over this a lot. You know, trials like OPTION where left atrial appendage closure (LAAC) was designed to win.The second form is the process by which a commercially shaped framing becomes a premise of subsequent scientific work while its provenance recedes from view.Said another way, agenda laundering is like a quiet form of influence. A framing of a clinical question that originally came from, or was pushed by, industry gets adopted as the starting assumption of later research, guidelines, and practice. With each step, people lose sight of where that framing came from. The name borrows from money laundering: something with a commercial origin passes through enough legitimate-looking stages that it comes out looking clean.The core idea is about the premises. Ordinary conflict-of-interest concerns ask whether a sponsored study produced a biased result. Agenda laundering works one level higher, on which question gets asked in the first place. The authors’ test is whether later work keeps checking the framing, including whether it matters clinically, or simply works inside it and takes it for granted.The authors use examples in critical care, explaining how colloid-type fluids were continually used despite negative trials. They also use early dialysis as an example or biomarkers for predicting kidney disease. I would use the example in LAAC. First were the biased trials, then we get studies of cardioversion in patients with LAAC, or ablation in patients with LAAC, which just launder the original bias. Agenda laundering is also largely responsible for the totally unproven notion that LAAC is better than nothing in patients who truly cannot tolerate anticoagulation, as these patients were excluded from the seminal trials, and have never been randomized. I also see it coming in with early transcatheter aortic valve replacement (TAVR) for asymptomatic aortic stenosis. The EARLY TAVR trial is one of the most biased endpoints I’ve seen. Subtraction anxiety in the open label trial drove the result. Now we will see this normalized and further studies will come using early TAVR as the norm.There is agenda laundering present in coronary artery calcium screening and many forms of imaging and plaque analysis. I bet you all can name a few more.I would read the BMJ paper. Think about where you see agenda laundering and what you can do about it. For me, it’s being a student of the seminal trials. A true understanding of these is the antidote to being bamboozled by the therapeutic fashions of the day.