Mar 1, 2024 This Week in Cardiology Podcast

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 March 1, 2024, John Mandrola, MD, comments on the following news and features stories: aspirin, cannabis use, left ventricular ejection fraction (LVEF) in athletes, and shared decision making (SDM) before implantable cardioverter defibrillator (ICD) implantation.

Western AF Meeting

Before I start, I want to say a few words about the Western AF congress in Park City Utah last weekend. First, thank you to my friend Nassir Marrouche for the invitation to this incredible meeting.  It’s a remarkable 2 days of lectures on atrial fibrillation (AF)

I had two assignments. One was to be on a panel discussing the topic, “Every patient with AF deserves a PVI.” It’s an obviously provocative topic. There was a spirited discussion about doing pulmonary vein isolation (PVI) and how early is better than later.

One of the main comments I made was to say that back in the early days (20 years ago) when I was learning AF ablation, I set a Google alert for the term “AF ablation.” I did this to pick up tips and tricks. And in those days, it worked. All of us were learning.

But. In recent years, these alerts mostly point me to business stories on how big the AF ablation market is and is going to be. It’s especially crazy these weeks and months because all the big industry players are marketing their new way to kill atrial myocytes with pulsed field ablation (PFA). The marketing with PFA rivals that of left atrial appendage closure.

My thinking, though, is that a sign of being more advanced as a field would be if we did fewer PVIs. That would mean we’d have done something to stop the massive increase in AF incidence. For example, if we learned more about the upstream causes of AF, we likely would need fewer PVIs.

I tried to remind my electrophysiology (EP) friends that while the PVs are easy to isolate, we don’t really know why we do it. Sure, in some cases of focal AF there is a driver in the PVs. But that’s rare.

Aspirin Is Back in the News

I like talking about aspirin (ASA) because a) there are a bunch of trials to interpret, b) it’s directly relevant for cardiac prevention, and c) it’s not entirely clear how to translate the trial data.

The most recent data is now 5 years old. In 2018, which seems like yesterday, there were three primary prevention trials published:  ARRIVE (patient with elevated atherosclerotic cardiovascular disease risk), ASCEND (diabetic patients) and ASPREE  (older patients). The take-home from these 3 trials was that there was no net benefit to primary prevention ASA. The small reductions in non-fatal events were countered by an increase in major bleeding.

But. (Yes, there is always a but.) Most of these patients were not taking ASA at the time of trial enrollment. So, these were essentially studies of starting ASA.

But that’s not the main scenario I see in the office. In the office I mainly see patients who are taking ASA and the question of deprescribing comes up. To be honest, I’ve been recommending stopping the ASA, especially when patients are on an oral anticoagulant (OAC).

Now, a four-author group, two from Australia and two from Ireland, had the idea to meta-analyze those three trials looking specifically at patients who were taking ASA before the trial and were then randomized. The first author was Ruth Campbell.

That means part of this group were randomly assigned to stop the ASA. This is a good study idea because it comes up so often in the office.

• An observational study from Sweden found that patients in their registry who stopped ASA had a 28% higher risk of cardiovascular (CV) events. (That is an association, but if causal that is a big risk).

• The three 2018 primary prevention ASA trials had more than 47,000 patients. About 15% (about 7000 patients) were taking ASA before enrollment. (Of note, the ARRIVE Trial excluded patients on ASA).

• These 7222 participants provided a chance to evaluate outcomes based on randomization to continue or stop ASA among baseline users of ASA.

• The authors were able to get the source data from the authors of ASPREE and ASCEND.

• The main findings were that when they compared CV events in the 3500 ASA users randomized to placebo (stopping ASA) vs continuing ASA the hazard ratio (HR) was 21% greater (HR 1.21; 95% confidence interval [CI] ranged from 1.05-1.39). The absolute increase was a lot — 450 vs 374 in these groups of 3600 patients.

• The risk for major bleeding was not significantly different in the ASA stoppers vs ASA continuers (HR 0.86; CI 0.68-1.10).

Comments. This is provocative data, isn’t it? A 21% risk increase in CV events is a lot. This approaches the Swedish observational data, and there was not a huge harm reduction in terms of less bleeding.

Before we go further, we should say that all meta-analyses are basically observational studies. This is an exploratory analysis. Relative to new users of ASA, previous users comprised a much smaller group. To really answer the question, you’d need to do a properly powered randomized controlled trial (RCT) of ASA users and study deprescribing with randomization.

Interesting was that in each of the trials, there was not a significant subgroup interaction based on ASA use before the trial. I looked up ASPREE (primary prevention in the elderly).

• The risk reduction in ASPREE in previous ASA users was 0.78 vs 0.98 in non-users. The P for interaction did not reach significance, because of low numbers, but the 22% reduction is similar to the findings of this meta-analysis.

• The putative explanation of the finding that previous users do worse when the ASA is stopped is survivor bias. That is, people who have taken asa for a long time have shown that they can tolerate the drug.

• It reminded me of the EP clinic I used to do with Dr Zipes at Indiana University in the 1990s, years after CAST and the removal of encainide from the market. Patients would come to clinic and were able (somehow) to get encainide from us. I remember asking Dr Zipes, Hey what about CAST? His response was that these patients had been on the drug safely and effectively for years, even decades, and are tolerating it fine.

• It also reminds of the FRAIL AF study in the Northern Netherlands. Elderly frail patients doing well on warfarin were randomly assigned to staying on warfarin or switching to a direct acting OAC. The trial was stopped early because those who stayed on warfarin did much better.

I have to say, this analysis really makes me question the practice of stopping the ASA in patients who have been on it for a long time.

Let me know what you think, but looking harder at this meta-analysis makes me less certain about my deprescribing practice. Wait. Isn’t this almost always true? That is a deeper look at evidence almost always makes you less rather than more certain.

Cannabis Study

• Cannabis Use Tied to Increased Cardiovascular Risk

• It Sure Looks Like Cannabis Is Bad for the Heart, Doesn’t It?

The Journal of the American Heart Association has published a study correlating cannabis use and CV outcomes.

The topline results were that cannabis is associated with higher rates of CV events. More than a 100 news sites have covered the study. Pediatric News had the headline: “It Sure Looks Like Cannabis Is Bad for the Heart, Doesn’t It?” The paper even made the Today Show.

Veteran journalist Sue Hughes writing for theheart.org | Medscape Cardiology had excellent coverage. Her first sentence had the topline results. Her second sentence started with the phrase, “However, the study has many limitations….”

Indeed, this is true. Here is what the four authors did:

• They used data from the Behavioral Risk Factor Surveillance Survey from 27 American states. This is a telephone survey conducted by the Centers for Disease Control and Prevention (CDC) in which they call Americans and ask questions.

• Everything is self-reported. Cannabis use per the last 30 days, demographics like age and weight, and get this…even events. As in they asked people have you had a stroke or myocardial infarction (MI), or coronary heart disease (CHD)?

• They then formed three groups: Non-users, non-daily users, and daily users.

• Then they did correlations. With some minor variable adjustments, such as smoking (self-reported), age, weight, diabetes, etc. They did not know blood pressures or lipid levels or much else.

• They reported that cannabis use was associated with a higher rate of CV events. The odds ratio for a composite of CHD, MI, or stroke was 1.28, or 28% higher for daily users vs non-users. Elevated risks were also seen regardless of smoking. There also seemed to be a dose-response curve.

The authors list seven limitations, but they still make strong conclusions and the senior authors told journalists that “cannabis has CV risks” and their clinical perspective in the journal reads:

Patients should be screened for cannabis use and advised to avoid smoking cannabis to reduce their risk of premature cardiovascular disease and cardiac events.

Comments. This is an extremely weak observational study. It’s all from a phone survey.

For instance, when I look at their main table, it looks like cannabis is highly protective in those who drink alcohol (ETOH) daily (odds ratio 0.72;  CI  0.64 -0.81) Should we tell people who drink daily to take up cannabis to reduce their CV risk? Of course not.

• This study is fatally flawed because people who smoke cannabis are different from those who do not. Not only that, how about people who would talk with someone from the CDC vs those who would hang up. Do you really think data derived from a CDC phone survey in people who smoke weed every day is reliable?

• If I had to put a bet down on cannabis, I’d suspect that daily use is likely not a net positive for health. But studies this weak are not the way to sort out this knowledge.

• I saw someone online say, isn’t this the way we found out smoking was harmful. I would answer not really. Causation in smoking fulfills many of Bradford Hill’s criteria. Cohort studies show consistent relationship, there is a strong dose response curve, cessation reduces risk, and the plausible mechanisms, such as platelet aggregation and endothelial dysfunction.

The thing that bothers me about cannabis — scientifically that is — is that because it has been a banned substance, we have very little data on its potential benefits and harms.

With decriminalization, we will learn more, but not from these sorts of methodologically weak studies.

Finally, I will say it again, every time authors do such a study, and a journal publishes it, and health news mindlessly covers it, all for attention, trust in medical science diminishes. 

Low EF in Athletes

Circulation published an interesting paper in December of 2023 on sports cardiology. It is the matter of seeing a slightly low ejection fraction (EF) in highly trained athletes.

I see a fair number of athletes, so I have seen this and not known what to make of it. I’ve often had the athlete do a bunch of push-ups before the echocardiogram because that seems to improve it.

But still, why would an elite endurance athlete have a measured left ventricular (LV)EF of 46%?

Well, a group of Belgian and Australian investigators have published a neat study of 281 elite athletes. They did a battery of tests on these cyborgs and followed them over time.

Here is a summary of the seven main findings:

1. 44 of 281 or 16% of the athletes had a reduced EF on cardiac magnetic resonance imaging; 12 with low LVEF, 14 with isolated low right ventricular (RV) EF, and 18 with both reduced RVEF and LVEF. Age, sex, BMI, blood pressure, and sport type and training load were similar between the low and normal EF groups. So, no easy clues.

2. Athletes with lower EFs more frequently met criteria for reduced LV global longitudinal strain (GLS). Consistent with this, there was a moderate negative correlation such that lower EF was associated with reduced GLS  (r=–0.386; P < 0.00).

3. Late gadolinium enhancement (LGE), a marker of macroscopic fibrosis, was present in 16% of athletes with no difference between the reduced EF and normal EF groups. All 10 athletes with reduced EF had LGE in the interventricular septum at the hinge point of RV attachment. This was also the most common site of LGE in the normal EF, but four athletes in the normal EF had mid-epicardial lateral wall fibrosis.

4. During exercise, the reduced EF group picked up vigorously and had greater gains in EF during exercise (perhaps because they started from a lower point). HR and power output were similar at peak exercise between the normal and reduced EF groups.

5. As for rhythm monitoring, the reduced EF group had more premature ventricular contractions (PVCs), though the absolute number of PVCs per day was low.

6. Genetics wise, the authors used a polygenic risk score (PRS), which is a constellation of gene variants associated with dilated cardiomyopathy CM. The goal here was to determine of there was a genetic component to reduced EF in the athletes. The PRS was significantly higher in athletes with reduced EF compared with the those with normal EF, though when graphed out, the differences looked small. The caveat here is that I am over my skis here regarding evaluating PRS. In a multivariate analysis that included age and fitness, male sex and greater PRS were the only significant predictors of reduced EF.

7. During 4.5 years of follow-up, no athlete had heart failure (HF) or documented sustained arrhythmia. One athlete died from trauma. The other died of sudden death. This athlete had an abnormal RVEF and low normal LVEF, a PVC count of about 3000 per day, no LGE, and high genetic risk. A post-mortem exam revealed microscopic evidence of patchy fibrosis of both ventricles, suggesting the possibility of a cardiomyopathic process. But testing of cardiomyopathy genes was negative.  He had undergone regular testing for performance enhancing drugs that was negative.

Comments. This is a nice paper. The first author is Guido Claessen. Though it leaves a lot of unanswered questions. It’s weird for athletes to have a lowish EF. You don’t expect it. But it seems real, right? GLS was off in this group, and they had more PVCs.

The good news, at least in this small sample, is that it does not look to be a bad prognosis. Power output is similar on exercise testing, and none developed HF or arrhythmias.

The gene variant connection seems a bit weak to me. Yes, there was a significant relationship to a PRS, but a) I don’t know what that means, b) it seems like a small effect size.

Now, when I see these athletes with lowish EFs, I can recall this study, and feel relatively assured, though we definitely need more information.

One question or thought that came to my mind and the authors did not mention this is that I wonder if ETOH can be a factor.

For instance, they cited a study of Tour de France riders from years ago and a similar percentage had lowish EFs. Do these athletes drink ETOH? It’s probably not common in endurance athletes, but a recent story in cycling news implicated ETOH in a famous riders’ poor performance. And cyclists are people so….

Shared Decision Making and ICD Use

JAMA-Internal Medicine has published a research letter looking at ICD use for primary prevention after the SDM mandate that the Center for Medicare and Medicaid Services (CMS) imposed in 2018.

The mandate required doctors to perform SDM using a specific SDM tool before implanting primary prevention ICDs. Most use Dan Matlock’s University of Colorado SDM tool.

• This cohort study took a random sample of 20% of CMS data. They looked at the rate of new ICD implants before and after the mandate. They also compared primary prevention ICD implants, which required SDM, and secondary prevention implants, which did not require it.

• Before the mandate, primary-prevention ICD use was decreasing at a rate of 2.37 procedures per 100,000 beneficiaries per month (95% CI, −1.32 to −3.42; P < .001).

• They found no significant difference after the mandate (−0.77 procedures per 100,000 per month; 95% CI −2.13 to 0.58; P = .26).

• When comparing the primary-prevention group with the secondary-prevention group, there was a significant monthly reduction of 2.86 primary-prevention ICDs per 100,000 beneficiaries (95% CI, −5.17 to −0.57; P = .02).  However, they noted that this was likely related to reductions in the decline of secondary prevention ICDs after the mandate.

• The authors concluded that a broad mandate for SDM did not influence use of the ICD. They wrote that this goes against a paper published in 2013 which found an association between SDM and fewer preference-sensitive interventions.

They also commented that while they noted a statistically significant difference between primary and secondary prevention implants after the mandate, this was most likely not due to a reduction in primary prevention ICDs but rather a remarkable plateau (increase in the rate of secondary prevention ICDs). They speculated that this could have been due to doctors coding more ICD implants as secondary prevention.

Comments. I highlight this study because I want to say something about SDM. Shared decision making should never be used to decrease the use of any procedure. CMS may have mandated SDM for that reason, but it is morally and ethically bankrupt.

SDM is good medicine. Many of the decisions we make in cardiology (ASA, statins, treatment to different blood pressure goals, AF ablation, etc) are all preference sensitive. You could argue that all medical decisions are preference sensitive.

I define good medicine as the most evidence-based thoughtful therapy that aligns with a patient’s goals and preferences. I probably use the phrase “preference sensitive” 5 to 10 times every clinic day. I hope you do too.

I am glad the study found no significant change in ICD use after the mandate. And of course, mandating SDM was a bad idea, because it would be like mandating being a good doctor.

© 2024 WebMD, LLC

Any views expressed above are the author's own and do not necessarily reflect the views of WebMD or Medscape.

Cite this: John M. Mandrola. Mar 1, 2024 This Week in Cardiology Podcast - Medscape - Mar 01, 2024.