11-07-2025
Jul 11 2025 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 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 July 11, 2025, John Mandrola, MD, comments on the following topics: approval of first-generation devices, the ticagrelor controversy, ICD longevity, the PRAGUE-25 trial (one of the most important trials of the year), and some thoughts on the end of EP as a profession.
Catalin Toma, MD, who is Director of Interventional Cardiology at the University of Pittsburgh Medical Center, writes regarding what I talked about on lack of data in the invasive pulmonary embolism (PE) field.
I have been involved with PE interventions from very early on and could not agree more that we need randomized controlled trials (RCTs) to understand the role of interventional PE therapies in the high-intermediate risk and massive PE groups. However, until a few years ago the whole field of interventional PE was completely new, with virtually no expertise! Launching into RCT with the early devices, and with the lack of cumulative knowledge we have now would have led to predictable negative results. He says, 'Good thing we didn't send astronauts to the moon in Apollo1 or to Mars in the current generation of starships.'
So now we are able to do major RCTs with both catheter-directed thrombolysis and thrombectomy (HI-PEITHO is almost complete, the ATTRACT PE is close to completion, PEERLESS 1 and 2 are underway).
And to be honest, Toma adds, 'the fact that industry is engaging in these trials is commendable, especially given how popular and reimbursed things are already—this didn't happen out of goodwill, but because doctors pushed for getting data. At our center, we approach virtually every patient referred for intervention and enroll them in trials.'
Toma concludes by saying, 'I think the ethos for obtaining good clinical evidence is still here. But I am happy I learned how to ride before bombing downhill.'
I think this is a really important point. I highlight it because I recall last year having a conversation with Dr Bobby Yeh about early-generation devices and iteration. That is, if you study and then restrict first-generation devices, you might hurt innovation.
So there is a balance needed. That said, I stand by my stance that the FDA should reform this extremely problematic 510k approval pathway where devices get approved by being similar to an already approved device. This is a really problematic and too lackadaisical approval pathway, I think. More Ticagrelor Controversy
BMJ Investigation Finds More Concerns in Ticagrelor Trials
Just before my last podcast, the BMJ published another concerning report from senior editor Peter Doshi regarding irregularities in two key platelet function studies of ticagrelor.
I take the role here as a neutral journalist. As an EP, antiplatelet drugs are not something I use every day.
Let's review some basics. Ticagrelor is a huge player in the post-stent antiplatelet market. In 2022, the United States spent more than $750 million on it. Despite representing only 7% of total P2Y12 inhibitor prescriptions among Medicare beneficiaries in 2020, ticagrelor accounted for nearly two thirds of total Medicare spending on these drugs in the same year. Globally, the makers of ticagrelor, AstraZeneca, brings in more than a billion dollars in annual ticagrelor sales.
The seminal trial was a 2009 study called PLATO, comparing ticagrelor vs clopidogrel in 18,000 patients with acute coronary syndrome. The primary endpoint of CV death, myocardial infarction (MI), or stroke occurred in 9.8% of patients on ticagrelor vs 11.7% on clopidogrel. This was a 16% relative risk reduction (HR 0.84; 95% CI, 0.77-0.92). MI and CV death were reduced, as was all-cause death. PLATO also found no significant differences in major bleeding.
Despite this 'big win,' ticagrelor failed at its first review at FDA. Mainly because a subgroup analysis found that, in the US, ticagrelor patients had poorer outcomes than those randomized to clopidogrel—a 27% higher risk of the primary endpoint
Thomas Marciniak, who was an FDA medical officer, wrote a scathing review saying he had worries about the PLATO data quality. Marciniak called AstraZeneca's application 'the worst in my experience regarding completeness of the submissions and the sponsor responding completely and accurately to requests.'
However, FDA leadership did not endorse these reservations. One explanation put forth by the authors was that US patients fared better with clopidogrel because US patients had higher doses of aspirin.
Though you would not know it, because ticagrelor is so popular, there were multiple layers of controversy surrounding PLATO and ticagrelor.
One association was this: in an analysis based on data monitoring by the sponsor vs independent contract research organization monitoring, the four countries exclusively monitored by non-sponsor personnel—Georgia, Israel, Russia, and the United States—ticagrelor fared worse.
Here is another chapter: Eric Bates, a professor of medicine at the University of Michigan, wrote a concise but readable narrative review in Journal of the American Heart Association , and the striking finding is how scant few other observational or RCTs align with PLATO.
Yet another layer in the controversy: Victor Serebruany and Dan Atar wrote an editorial in EHJ in 2010 titled, 'PLATO trial, do you believe in magic?' Their major point was the massive overall and CV death reduction. There were 107 more lives saved with ticagrelor than after conventional clopidogrel (399 vs 506), representing a highly significant absolute mortality reduction (HR = 0.78; 95% CI, 0.69–0.89; P < .001). This they felt was highly surprising because it was three times more than the COMMITT trial where clopidogrel beat placebo. What's more, the mortality reduction (107 deaths) numerically exceeds the MI prevention benefit (89 events), making it a 'hitherto unmatchable achievement.'
Important to say is that PLATO authors have published multiple rebuttals, and the Department of Justice (DOJ) investigation was started but then terminated 11 months later. DOJ said: 'After an extensive investigation . . . we determined that the allegations lacked sufficient merit such that it was not in the best interests of the US to intervene in the suit.'
In my December podcast, I also noted how the non-industry funded ISAR-REACT 5 trial, comparing prasugrel to ticagrelor, so strongly favored prasugrel over ticagrelor, by a statistically significant 36% for the primary endpoint of death, MI, and stroke — and trended toward lower bleeding as well. It still perplexes me how little effect that trial had on actual usage. Prasugrel is also once a day.
The newest BMJ investigation that came out in June focused on two mechanistic studies supporting the rapid onset and potent antiplatelet effects of ticagrelor. One was called ONSET/OFFSET and the other study was called RESPOND. Both were published in Circulation .
ONSET/OFFSET studied 120 patients who got either ticagrelor, clopidogrel, or placebo. Ticagrelor achieved more rapid and greater platelet inhibition than high-loading-dose clopidogrel; this was sustained during the maintenance phase and was faster in offset after drug discontinuation.
RESPOND studied clopidogrel nonresponders and found that ticagrelor overcomes nonresponsiveness to clopidogrel, and its antiplatelet effect is the same in responders and nonresponders. Nearly all clopidogrel nonresponders and responders treated with ticagrelor will have platelet reactivity below the cut points associated with ischemic risk.
You could see how these studies could be used in marketing pamphlets along with PLATO to help explain how PLATO came out positive. Norwegian researcher Dan Atar was quoted as saying:
'The ONSET/OFFSET study, when it came out, caused incredible interest,' I remember numerous discussion groups where people were asked to interpret these findings.' And AstraZeneca was also pressing the message that the platelet studies, which found faster and greater platelet inhibition, explained PLATO.
FDA officer Thomas Marciniak, however, noted that early benefit (which would be expected with faster platelet inhibition) was not seen in PLATO. He wrote 'I would expect that patients with the very early invasive strategy would have the greatest need for good platelet inhibition, but ticagrelor fared worse [in PLATO] for short term outcomes in these patients.'
But that was not all: Enter Victor Serebruany, a pharmacologist with expertise in platelet function tests. Serebruany thought the pharmacodynamic data were too good to be true. ONSET/OFFSET's study schedule, he reckoned, was almost impossibly intense, including a requirement for six blood tests in an 8-hour period. This would require 210 mL of blood drawn for the study. The company told Serebruany that it looked into his concerns and found nothing amiss.
The BMJ sought to interview 15 investigators in the two platelet trials—and few, including the corresponding author, were responsive.
One investigator that the The BMJ could reach was Tonny Nielsen, a co-author of RESPOND and principal investigator in Denmark, according to AstraZeneca documents. But Nielsen told The BMJ in a written response, 'I did not participate in the RESPOND study,' a statement further substantiated by two of his colleagues. And yet he was listed as an author of the Circulation paper despite not being involved.
Meanwhile, The BMJ found that one investigator, Alberto Yataco, operated an active study site in Baltimore, but never became an author despite enrolling 12 patients and ordering extra test kits. Yataco could not be located for the story.
The next chapter was that the platelet tests were finicky, prone to artifacts and most platelet tests were done in single centers. However, these studies were multicenter and Doshi writes that it not clear that training and standardization were done in the studies.
Furthermore, Doshi found numerous other data integrity issues: more than 60 of 282 readings from platelet machines used in the trials were not present in FDA datasets.
There were missing baseline measurements for patients who had blood drawn multiple times. Some study sites began recording data before equipment arrived.
Finally, The BMJ 's review also found that the protocol-specified primary endpoint results for RESPOND were statistically nonsignificant ( P = .157) but were subsequently reported in Circulation as significant ( P = .005) because of an undeclared change in primary endpoint definition.
I would encourage you to read Doshi's reports in BMJ from December 2024 and this year. See what you think. My questions go back to PLATO. Why does the drug do worse in geographies where there are independent research organizations? Is this a coincidence?
Why, as Dr Bates, lays out, so little corroborating data supporting ticagrelor vs clopidogrel? Why did ticagrelor underperform so badly compared to prasugrel in ISAR-REACT?
I worry that when you put these observations together, there is worry that an accepted drug may not be what we thought. Maybe there is nothing there, but gosh, if we are wrong about ticagrelor, that would be a big thing.
Let me know what you think. I think this is a big story.
Heart Rhythm has published a simple but important study on the battery life of cardiac defibrillators. My opinion on this issue has changed. I once thought that the difference of a few years was not clinically relevant. My mind works like this. Maybe yours does too. When I have a talk that is a month or so in the distance, I don't worry one bit about it. I never plan anything a year in advance because a year seems like forever from now. So, the difference in battery life between, say, 7 and 10 years was hard to sort out.
But now that I am old enough to have dealt with nearly every possible device complication, I have come to hate generator changes—because this is where many of the complications start. If you meet an EP doctor not scared of generator changes, consider referring to a different one. Generator changes deserve your highest respect because these are hazardous interventions.
Therefore, ICDs with longer battery lives mean fewer generator changes. And this is no small issue.
Dr James Freeman and colleagues used a US remote monitoring dataset to study ICDs implanted from 2003 to 2023. They assessed the time from implant to replacement. I know; it's simple but the data are remarkable. And it turns out that device advertisements had some truth to them.
The first finding was that battery longevity declined 2-3 years for each extra lead implanted. Pause there. This is exactly why primary prevention ICDs should be one lead unless there is a real need for atrial-based pacing.
The second finding was that device manufacturer mattered. Estimated longevity could be 2-4 years longer based solely on who made it. Boston Scientific had long marketed their devices as having the longest battery life. And, indeed, this data corroborated their pitch. Boston Scientific devices outpaced their competitors in single, dual and biventricular devices. By a lot. I've seen this and I often use Boston Scientific devices for this very reason.
The third finding was that programmed outputs and the percentage of pacing mattered. It's like my bike light. If I keep it on high, I struggle to finish the 2-hour ride. I continue to believe that device follow-up and attention to detail is as important as skillful implant.
The fourth finding was that recent-generation devices have a little less of a longevity gap, though it still exists. For instance, the 'estimated' median longevity of a single-chamber Boston Scientific device implanted after 2022 was 180 months. It was only 110 months for an Abbott device. That's 5 years. You might quibble with this because it is estimated, but the authors would counter that they looked at estimated vs actual in older devices and found good agreement.
This study almost seems too simple to report on. But it's not. The choice of device manufacturer is too often set by initial costs or relationships with reps. These are both bad. Ugly even.
Battery longevity matters and Boston Scientific does better than its competitors. I don't have any financial relationships with any company, but how long an ICD lasts is hugely relevant. It's hard to think a decade on, but I can tell you now, as an old dude, it matters.
Even if a device costs a bit more, if it lasts 5 years longer, than it is a value. The problem is that in private systems in the US, payers may change often. So value from longer devices only accrues if the payer doesn't change.
Atrial fibrillation (AF) ablation is undergoing a transformation I have not seen in 20 years of doing this procedure. The advent of pulsed-field ablation (PFA) is the tale of two stories: one story is that we have a way of ablating the left atrium which is super-fast and without the worry about killing patients from atrioesophageal fistula.
This is good and it has lowered our bar of offering this procedure.
But…But….PFA, I worry, is killing EP as a field. Every time I finish a PFA case, I shake my head in disbelief. Why? Because EP was once a profession where we cared about physiology. We diagnosed and treated arrhythmias based on a deep understanding of mechanisms.
Pulmonary vein isolation (PVI) started to ruin that concept because we simply put electric fences around the PVs. The cryoballoon was worse because we simply put a balloon in the PV and froze. Now with PFA, we can destroy left atrial myocardium even faster. We isolate PVs in slightly more than 30 minutes. We can do 4 to 5 cases per day. The EP doctors, industry, and hospitals are making tons of money.
But we have not learned a darn thing about AF pathophysiology. And, worse, PFA has made AF ablation so easy, that anyone who has catheter skills could do it. Anyone. I mean it.
In the US and most of the Western world, except maybe Denmark, Norway and few other places, most patients with AF are obese. Obesity and its accompanying metabolic abnormalities surely promotes AF by creating structural and atrial disease. The Adelaide group has elegantly shown this over the past decade.
No matter, US doctors ignore the pathophysiology and take these obese patients for PVI. We say weight loss and RF management is hard, and don't deny the patients with obesity the 'benefits' of PVI.
Well, thank goodness our field has people like Pavel Osmancik, MD, in the Czech Republic to do a proper trial.
PRAGUE-25, performed in 5 centers in the Czech Republic, bravely compared a weight loss intervention (with anti-arrhythmic drugs) to ablation in patients with obesity and AF. These patients had a mean BMI of 35.
The lifestyle modification (LFM) arm included a weight management program with dietary and exercise specialists using tech support via mobile phone apps and regular in-person and telephone conversations.
Also brave was Dr. Osmancik's choice of endpoint — freedom from AF. Thank goodness he and his team also measured secondary endpoints, because this is where the real story lies. His team convinced more than 200 patients to undergo randomization. This would be a fun study to recruit patients for. Imagine the explanation: You are obese. We could ablate your AF or put you on a program of weight loss and exercise. That would be a tough, tough conversation.
As you would expect, AF ablation was better at reducing the surrogate endpoint of freedom from 30 seconds of AF. The percentage of patients with AF freedom at 12 months, 73% in the catheter ablation group vs 35% in the weight loss group. The HR was 2.8 times (95% CI, 1.9-4.0). (Please pause and note here that more than a third of patients with AF had freedom from AF without ablation).
This led to topline conclusions that AF ablation was better than weight loss for AF.
The real story is the nuanced secondary findings: Weight loss was significantly greater in the LFM group (6.4 kg vs 0.35 kg).
Glycemic control was better in the LFM. A1c decreased by 1.4 ± 4.8 mmol/L in the LFM while it increased by 2.5 ± 10.5 mmol/L in the ablation arm.
Fitness as measured by VO 2 max was also better in the LFM group.
Reduction in AF burden, which is arguably a more relevant endpoint than freedom from a 30-second episode, was numerically larger in the ablation arm, but the between group difference did not reach statistical significance.
Quality of life (QoL) improved equal degrees in both arms, which is surprising because you'd expect less AF and less AAD to associate with better QOL. Perhaps the weight loss and gain in fitness countered the less AF.
The key limitation of PRAGUE-25 was the modest weight loss of only 6 kg, which is less than Adelaide's number of 10% body weight loss, where the real AF benefits kick in.
The study came before GLP-1 agonists drugs were available and I would love to see this trial repeated with GLP-1 drugs.
I hope your conclusion of PRAGUE-25 is similar to mine. LFM may have lost on the surrogate measure of freedom from 30 seconds of AF, but if overall health of our patients is the doctor's goal, LFM clearly won. It led to more weight loss, improved glycemic measures, better fitness and similar QoL improvements.
I seriously think PRAGUE-25 argues for a conservative LFM-first approach to AF in patients with obesity. It also argues most persuasively for a larger, properly powered CABANA-like trial with LFM aided by GLP-1 drugs vs ablation.
That's a trial that could change EP for the better. My friends, if you keep destroying the LA with PFA, instead of treating the patient's more serious metabolic problems, you may make a lot of money in the short term; you may gain status in the hospital because your relative value units (RVUs) are so high; the company may make you a key opinion leader, but you are destroying the once great field of EP.
PRAGUE-25 should re-invigorate our drive to help patients improve their overall health. The patients in the lifestyle management group may have had more episodes of 30 seconds of AF, but they ended up with better health overall.
