Mar 31, 2023 This Week in Cardiology Podcast

John M. Mandrola, MD


March 31, 2023

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 31, 2023, John Mandrola, MD comments on the following news and features stories.

First, I want to say thank you to the cardiovascular (CV) fellows at Mayo Clinic for choosing to have me come to speak about why I am a medical conservative. You see the phrase honored and humbled a lot, but it was a true honor to come to such a famous place and be around such thoughtful smart people.


A group of from Canada, including Gordan Guyatt, have published a meta-analysis of randomized controlled trials (RCTs) of patients at increased risk of CV disease that compared dietary programs.

  • Their outcomes were major adverse coronary events (MACE) and mortality.

  • They combined 40 trials looking at the Mediterranean diet, low-fat diets, etc.

  • The topline results were that they found moderate certainty evidence that programs promoting Mediterranean and low-fat diets reduced all-cause death and myocardial infarction (MI).

  • They also found that Mediterranean diets reduce stroke risk.

  • Of note, the effect sizes are small, in the range of 1% fewer events over many years of follow-up.

Comments. I do not understand these efforts. First, as the authors write, there are numerous limitations to this meta-analysis. Even in a trial, it is near impossible to isolate the effect of a type of diet. There are too many other variables. Then there is the matter of adherence. Control arms of these studies are heterogenous.

While I love the RCT, I am putting my cards down here and saying that outside of a prison setting, it seems pointless to attempt these sorts of trials with diets.

And even if you did such a study in a totally controlled setting, such as a prison, a) how would you translate it to regular people in the real world, and b) do you really think that after you control for smoking and exercise, the effect size between a Mediterranean diet and a regular low-fat diet would be clinically significant? I don’t.

  • At the risk of sounding overly simplistic, the best diets for patients are the ones that include the least processed food, ones that don’t cause weight gain, and ones that patients can maintain over the long run.

  • Whether Mediterranean or low-fat or Ornish is marginally better in the long- run will remain an unanswerable question and one that we should not waste time and money on.

  • So much more important in a country such as the United States is how to increase adherence to a diet that does not include food in packages or sugar-sweetened beverages.

A second diet study that also appeared this week bolsters my point. Publishing in JAMA Network Open, a group of European authors, led by Alysha Thompson, from Ireland, compared outcomes in people in the UK biobank who reported eating two types of plant-based diets. You didn’t know there were types? Well, yes, there are. There are healthy plant-based diets and not so healthy plant-based diets.

I will tell you a story. When I was bike racing, a cycling-colleague went to a vegan diet, but he never seemed to ride better or get any skinnier. I asked about it. Another friend said, yeah, he may not eat meat anymore, but Diet Sunkist and dry boxed cereal isn’t much better.

  • To no one’s surprise, the researchers, who used self-reported plant-based diet ingredients, noted that a diet characterized by high-quality plant-based foods and lower intakes of animal products may be beneficial for health.

  • The take-home is that just because patients report a plant-based diet does not mean it is a healthy diet.

CAC and Statins

Circulation Outcomes has published a paper, first author, Marc Pletcher, on the use of coronary artery calcium (CAC) scores to guide statin therapy. Two comments before we get to the core of this topic.

  • Clinicians overthink statins.

  • Re CAC, no single misunderstanding in modern medicine frustrates me more than what I see with CAC scans in my city. (Maybe it is different in your city.)

Where I am, it’s as if no clinician or patient in our community has read anything on CAC scans. I routinely see patients who had CAC scans who are already on statins, or elderly, or post-MI, post- CABG or post-stents. If you do a scan on someone who is post stent or on statins, you fail Medicine 101.

I have patients whose lives are torn up by paroxysmal atrial fibrillation (AF) but are more worried about their asymptomatic CAC score of 60. I will come back to this.

  • Most listeners of this podcast probably understand that proponents of CAC scans suggest the scores may help guide intermediate-risk people in the choice to take statins.

  • Most listeners probably get the idea that very low-risk people need not worry about taking a statin because a 25% reduction of a very low risk is still very low.

  • Conversely, very high-risk (> 20% 10-year risk) patients get benefits from statins regardless of a scan.

It’s those in the middle of risk who may be swayed by a zero CAC or high CAC. That’s where the sophisticated modeling study comes in.

They tested the cost-efficacy of three scenarios:

  1. Treat with statins without testing.

  2. Test and treat with statins if CAC is high.

  3. Neither test nor treat.

And they focused on the typical middle-aged adult about 55 years old who has an intermediate risk of 7.5%

Before I go on, please help me spread the word that before we advise patients on prevention techniques, we should always put their basic numbers in the PCE (pooled cohort equation) or American College of Cardiology (ACC) risk score equation and estimate a 10-year risk. It shocks me how many patients who ask me about their statin have no idea what their 10-year risk is. In fact, most patients who ask my advice on statins have no idea that there is a risk estimating equation. If you are in primary care at all, please show your patients how to know their 10-year risk estimate. They can look it up on their phones in seconds.

Back to the study:

  • The authors assume a cost of $225 for the CAC scan. They also consider low- vs higher-cost statins as well as long-term costs of CV events prevented.

  • Over a lifetime, they project that if statin costs are low, a hypothetical 55-year-old woman with high cholesterol would have essentially the same total healthcare costs in all three strategies, so the treat-all strategy is optimal because it prevents more coronary heart disease events at the same overall cost. 

  • The first take-home is that if statins are cheap and have no adverse effects, there is no reason to get an expensive test to decide whether to use them.

  • But that is not always the easy assumption, because statins can have a small negative effect on quality of life and not all statins are low-cost atorvastatin.

  • In this scenario, the authors found that the optimal economic strategy is to test with a coronary calcium scan and then identify and treat the higher risk, because then only those most likely to benefit would be exposed to the adverse effects and higher statin costs.

  • The take-home here is when the treatment is more costly and less well tolerated, testing to target therapy becomes the best strategy.

The model of Pletcher shows that testing for coronary calcium becomes the optimal strategy if statins reduce quality of life by as little as 1% to 2%, even when statins are inexpensive.

Keep in mind that one of the ways statins can reduce QOL is simply by requiring a patient to take a pill every day. The jargon here is disutility.

The authors concluded that CAC testing “in intermediate risk patients can be cost-effective but only if statins are costly or significantly affect quality of life.”

My Simple Thoughts on Statins and CAC. We overthink and misthink statins. In the community, people get put on primary-prevention statins as cholesterol-lowering drugs.

  • They reduce cholesterol, but that is almost a side effect rather than the effect. The effect is that statins reduce the risk of future CV events by 25%.

  • My idea is to show people their risk off statins and on statins and let them decide it is worth taking the drug.

  • In my opinion, CAC does not significantly or crucially reliably modify that risk enough to justify its cost or downstream effects (needless testing and treatment, like stents).

  • The one caveat I have come to in recent years, and I think patients need to know this, is that there is likely a cumulative effect of having exposure to statins and lower cholesterol.

This is where the number needed to treat (NNT) falls down. When you look at Kaplan Meier curves of statin trials, the separation grows over time. Treatment trials are short relative to the lifespan of a 55-year-old. The NNT at 5 years may be a lot larger than the NNT at 10 or 20 years. In other words, the benefit of primary prevention is, on average, likely to be a lot greater in younger vs older patients. In fact, the whole point of primary prevention is to become old.

  • My idea is simply to put statins over-the-counter.

  • Put a risk calculator in the aisle.

  • Have the American Heart Association and ACC use some of our professional fees to do a public health campaign to better understand CV risk and how to lower it with statins.

  • Let people decide if taking a pill every day is worth the 25% risk reduction – that likely grows greater over time.

Seeing Coronary Artery Disease (CAD)

The Annals of Internal Medicine has published a unique observational study from Copenhagen, Denmark. Andreas Fuchs and co-authors set out to define the features of subclinical coronary disease and its future risk.

They have something called the CGPS – Copenhagen General Population Study. This longitudinal study sounds similar to the National Health and Nutrition Examination Survey (NHANES) or the Multi-Ethnic Study of Atherosclerosis (MESA), in which persons are followed over time.

For this paper, individuals in the CGPS were offered Coronary computed tomography angiography (CCTA) scans. Nearly 10,000 asymptomatic persons over age 40 who did not have coronary artery disease (CAD) had the scans.

Crucially, the agreement was that these scans were only for research. Neither the person scanned nor the doctors were made aware of the results. This allowed for a natural history experiment.

  • CAD was labeled according to the luminal obstruction (obstructive vs non-obstructive) and extent (nonextensive vs extensive).

  • The primary outcome was MI. Secondary outcome was death or MI.

  • Follow-up was 3.5 years. These were 60-year-old people, more than half women. Only 10% were on statins, and even less on aspirin. Again, they were without symptoms or disease.

  • Subclinical coronary atherosclerosis was found in 46% of all participants and more frequently in men (n = 2484 [61%]) than in women (n = 1935 [36%]).

  • Most common: Nonobstructive subclinical coronary atherosclerosis (36%); single vessel obstructive disease was less frequent (8%); and multivessel disease or left main stenosis rare (2%).

  • Remember the two binary features, obstructive vs non-obstructive and extensive vs non-extensive.

  • For MI, using no disease as control, the risk was not elevated with non-obstructive, but it was much higher with obstructive (9 times higher).

  • The risk of non-extensive CAD was 2.3 times but for extensive it was 7.6 times.

When combining the two (obstructive and extent), non-obstructive and non-extensive – no significant hazard, but there was a graded increase in risk with positives of either, with the worse being both obstructive and extensive; here the risk was 12 times.

One small but important finding, in 14 participants with no subclinical coronary disease at baseline, an MI occurred during a median follow-up of 3.5 years.

The authors concluded that, in asymptomatic persons, subclinical, obstructive coronary atherosclerosis is associated with a more than 8-fold elevated risk for myocardial infarction.

Comments (medical and philosophical). This is a remarkable study because of its blinding. Patients with significant disease remained unaware of it and we get to see the natural history of disease.

One provocative feature was that the degree of obstruction was such a great predictor. So was extent of disease, but luminal stenosis was as well. Why is that interesting? The authors summarize this nicely.

  • Way back in the beginning, in the 1970s, it became known that high grade stenoses were high risk. That’s where fixing them came about. Then, over time, and with COURAGE, and BARI-2D and now ISCHEMIA, and even the 5-year results of STITCH, it became clear that atherosclerosis looked like more of a systemic disease. We all learned that MIs came from less obstructive plaques. Often unseen on angiography.

  • The interesting part of this study is that the degree of focal stenosis may be as important as we originally thought. (9 times the risk). And the old saw that obstructive plaques are stable may not be exactly true. That would be big. Imagine if yet another belief is discarded in the misinformation pile.

  • The authors cite a study I had forgotten about: PROSPECT, published in 2011 in NEJM, first author Gregg Stone. They studied patients with acute coronary syndrome with multivessel disease using ultrasound after percutaneous coronary intervention (PCI).

  • Recurrent events were evenly split between culprit vs non-culprit lesions. Nonculprit lesions associated with recurrent events were more likely than those not associated with recurrent events to be characterized by a plaque burden of 70% or greater or a minimal luminal area of 4.0 mm2 or less. In other words, more stenotic lesions.

Now we have two natural history studies that make us re-think the old saw about unstable lesions being non-obstructive. I’d make two conclusions here and one philosophical statement:

  • The first conclusion is akin to comments the authors made about screening. The great temptation here is to think we can use CCTA (even more than CAC) to screen. But that is a step that requires randomization. A trial. And there will be one. Called DANE-HEART. Good.

  • The second conclusion is that we eagerly await the authors analysis of plaque burden and severity quantifications. You know that I am skeptical of many things but perhaps, there will come a day when imaging can enhance prediction, or at least, teach us more about atherosclerosis.

  • Finally, to philosophy and doctoring tomorrow, my main points from the previous discussion of CAC and statins remain. Many patients with CAD did not have events. Some with no CAD had events. The imagers posit that imaging will better predict the future and tailor therapy.

  • But I see it as too imperfect. There will be tons of false positives, most will be scared for no reason, some will be relieved and still have an event. And everyone who is imaged falls prey to the risks of downstream testing and interventions.

Consider Table 5 in the Appendix. Here we get the absolute rates. The worse risk factor — obstructive — had an absolute risk of event of 1 per 100. Combined obstructive and extensive was 12 per 1000 patient years. So again, most people with bad disease don’t have events; here, even if not treated. Imagine if they were treated.

So, my idea for the prevention of heart disease today, is to live as heart healthy as possible, do not smoke, put numbers in the risk calculator, and if you like the risk reduction, take statins, if you don’t take statins, keep riding, or running, or playing golf, and by all means do not ignore symptoms.

Remember my friends, as we get better and better at treating a disease, early detection and screening become less important.

AF Ablation

I am going to be critical of my field, again. JAMA has published an RCT from researchers based in Ottawa. It’s called the AWARE RCT. The trial might seem specific to we who ablate AF, but I will bring in larger lessons. Keep listening.

You can close your eyes and imagine the standard AF ablation. Think of the left atrium as the US map, with four pulmonary veins (PVs) coming in at New England, Florida, San Diego, and Seattle. We use single dots – radiofrequency lesions — to ablate circles or electrical fences around the orifices of these PVs, the idea being that AF drivers come from the PVs and the electric fences keep out the drivers.

I know, you’d think we’d be more advanced, but that is what happens when a procedure is approved and well compensated. Learning slows.

Recurrences of AF occur after these circles are done (called PVI or pulmonary vein isolation) for two reasons — the fence springs a leak, ie, vein reconnection, or AF is being driven by a different geography. In years past it was almost always leaks in the line that we simply re-ablated. But technology has made us better at making these fences so now recurrent AF most often occurs because we weren’t burning in the right geography.

  • The AWARE trial tested the idea of making a double circle around the veins. This would a) reduce leaks or recurrent PV connection, and b) extend the area ablated in the left atrium (LA) and maybe, by chance, hit important geographies.

  • To be honest, most of us do a variant of this now. We can finish the original circle around the veins so quickly (thanks, technology) that while we are in the waiting period it makes sense to widen the circle and do a little more ablation.

  • The trial randomly assigned about 200 people to standard circle PVI or augmented double circle.

  • Investigators found no difference in freedom from AF. It didn’t matter. No subgroup did better.


  • First, thank you to lead author, Dr Girish Nair and colleagues. This is exactly how we should do things in medicine. This is evidence-based knowing, not eminence based knowing. Things that make sense, like extending lesions don’t always work.

  • It takes work, but randomly assigning patients is the way to learn what works.

  • Second, this study, can be added to the many that show that additional lesions beyond good old standard PVI do not improve our success. Complications were not different in this study, but every additional lesion that does not add to success rates increases the odds of a complication, like perforation or thermal damage to the esophagus.

  • Third, when I went to spend a week in Hamburg Germany in 2011, Feifan Ouyang and Karl Heinz Kuck made the argument that good quality PVI should be the standard. Extra lesions have not proved to work. Numerous studies have upheld this.

STAR AF 2 showed it in patients with persistent AF. The recent CAPLA study in persistent AF found that posterior wall isolation plus PVI was no better than PVI alone. Now AWARE – double circles no better than single circle. Less is more in AF ablation.

Finally, it needs to be said that the problem with AF ablation remains a knowledge problem. Since we don’t understand the basic causes of AF, we don’t know why PVI works or doesn’t work. We can’t answer the basic question that patients ask: How do you know where to burn?

The big breakthrough in AF will not come from a new energy source, ie, pulsed field ablation, but from the basic science labs.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.