Dec 10, 2021 This Week in Cardiology Podcast

John M. Mandrola, MD


December 10, 2021

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 on Apple Podcasts, Spotify, or your preferred podcast provider.

In This Week’s Podcast

For the week ending December 10, 2021, John Mandrola, MD comments on the following news and features stories.

AF Triggers

If you can find a pattern to paroxysmal atrial fibrillation (AF), the solution is easy: avoid the trigger. The UC San Francisco group has done, and is doing, great work looking at AF triggers — a topic that comes up nearly every day in a cardiology clinic. They recently presented and published the I-STOP-AF trial, first author Greg Marcus. The trial tested the hypothesis that “n-of-1” trials of self-selected AF triggers would improve quality of life in patients with AF.

An n-of-1 trial is when a patient acts as their own control. This differs from regular trials in which we get mean outcomes from two populations, the active arm and controls. An n-of-1 trial can demonstrate how a given individual will react to particular interventions. The best example of n-of-1 trials was the statin side effect trial called SAMSON, where patients took statin, placebo, or no pills, alternating by month.

In I-STOP-AF, one group of patients received instructions electronically to expose or avoid self-selected triggers in random 1-week blocks. The menu of triggers came from previous studies documenting common triggers, say alcohol or coffee or big meals. The patient chose which trigger they wanted to test. Most chose caffeine or alcohol.

Participants in this group were texted something like, “This is your week to expose yourself to the trigger,” and on other weeks, “This is your week to avoid your trigger.” The alternating weeks of avoidance/exposure went on for 6 weeks followed by a 4-week lifestyle-change period. In total, the trial blocks were 10 weeks. The control arm simply monitored their AF over the same period.

After the initial 10-week study period, all participants were given the option to end study participation or test up to four additional triggers—each of these were 6-week, alternating n-of-1 trials followed by a 4-week lifestyle change period.

Both groups were given Kardia smartphone electrocardiogram (ECG) recording devices. AF was assessed by self-report and the Kardia. The primary outcome comparing n-of-1 and control groups was the AF Effect on QualiTy of Life (AFEQT) score at 10 weeks. Nearly 500 patients participated; average age was 58 years.

  • The main result was no significant difference in AF quality of life between the two groups.

  • However, in the 4-week postintervention follow-up period, significantly fewer daily AF episodes were reported after trigger testing compared with controls over the same time period (P < .001).

  • It’s weird that those randomly assigned to alternate weeks of testing their triggers self-reported 40% fewer AF episodes but that did not lead to better quality of life.

  • The trial authors also included a meta-analysis of all the individualized n-of-1 trials and found only exposure to alcohol was associated with significantly increased risk of AF events.

The authors speculated on the tension between finding no difference in AF quality of life in those who repeatedly tested their triggers and had less AF: Maybe the AF questionnaire is not ideal; maybe the trial was underpowered. It’s hard to say. I think the main contribution of the study was that alcohol again came out as the only reliable trigger and although caffeine was the trigger that patients most often chose to test, it failed to exhibit any relationships with AF.

I-STOP-AF was a pragmatic trial. There was dropout and plenty of other limitations, which the authors described in a lengthy limitations section. One of the main limits is that Kardia mobile devices require active monitoring. You have to decide when to record your ECG. Newer devices such as smart watches may increase the sensitivity and specificity of AF detection.

Despite its limits I like this trial because it exposes the vast uncertainty in AF pathophysiology and treatment. Here we have a highly selected group of patients, a group you would expect to be sensitive to empowering stuff like supervised n-of-1 trials, who then have less AF in the intervention arm, yet this does not translate to better quality of life.

You all know I have been skeptical of digital health. I love technology as much as anyone, but there are many causal factors on the road to good health besides surrogate markers, such as your cardiac rhythm. You also have to consider the possibility of reverse causality: being connected and focused on triggers and AF could decrease one’s perception of health. Maybe the control arm just got busy having more fun living. Perhaps a contrarian lesson here is that less monitoring could be better. In the end, helping patients with AF is humbling. There is so much more than a rhythm monitoring. There are equal parts psychology and cheer-leading and friendship.

In an interview on | Medscape Cardiology, Professor Jag Singh, a Harvard electrophysiologist, talks with Greg Marcus from UCSF regarding their groups’ two late-breaking trial presentations at the 2021 American Heart Association (AHA) meeting. The 17-minute interview has an edited transcript and it’s worth a look.

They also discuss the CRAVE study from UCSF, which featured real-time ECG monitoring after exposure to caffeine. This was presented at AHA but not yet published. The short story was, again, that caffeine did not promote atrial ectopy or AF, but it did increase premature ventricular contractions and disrupt sleep.

Increases in BP During the Pandemic

A Cleveland Clinic group led by Luke Laffin published a research letter in Circulation chronicling a rise in blood pressure (BP) observed among adults during the pandemic. This was a longitudinal study of employees and their partners/spouses who participated in a wellness program operated by Quest diagnostics. The study included nearly a half a million individuals with data from 2018-2021.

  • BP from 2018 to 2019 showed no significant difference.

  • By contrast, annual BP was significantly higher in April to December 2020 than 2019 (P<0.0001).

  • The increases ranged from 1.1 to 2.5 mmHg and all age groups were affected.

  • Women and older adults had the largest increase.

The authors tell us in the discussion that in population-level studies, small 2 mmHg increases in systolic BP are associated with significant increases in mortality from stroke and ischemic heart disease. They propose the possibility of a forthcoming increase in major adverse cardiac events

I highlight this observation because it comports with my observations in the electrophysiology clinic. The vast majority of my patients have had worsening of their cardiometabolic risk factors. Rare is the patient who gained fitness, lost weight, and drank less alcohol during the pandemic.

I am no public health expert, but reports like this and our experiences in clinic, which describe the worsening metabolic health of people, illustrate one of the many negative externalities of COVID mitigation policies.

Direct Oral Anticoagulants After VTE

Both apixaban and rivaroxaban are used to treat venous thromboembolism (VTE), yet they’ve never been compared head-to-head. A University of Pennsylvania retrospective new-user cohort story published in the Annals of Internal Medicine attempts to sort out which drug works better. The data source is a US based commercial health insurance database. The primary endpoint was recurrent VTE. The primary safety endpoint was bleeding.

After propensity score matching, apixaban (vs. rivaroxaban) was associated with a lower rate for recurrent VTE (hazard ratio, 0.77 [95% CI, 0.69 to 0.87]) and bleeding (hazard ratio, 0.60 [CI, 0.53 to 0.69]).

The problem I hope everyone already knows is that these were non-randomized groups. Propensity matching is a technique used to try to match baseline characteristics, but the only real way to do that is with randomization.

Always remember, without randomization, the decision to use one therapy (apixaban or rivaroxaban) is made by a clinician. That clinician makes the decision for many reasons, some of which are on a spreadsheet, but others are not. Fortunately, there are ongoing randomized controlled trials (RCTs) enrolling now. Until then, I don’t think we should use this sort of data to believe apixaban is superior. I tend to think apixaban is the better anticoagulant, but I am often reminded that rivaroxaban is once daily and that is not nothing. Recall the drubbing that the twice-daily ticagrelor took against once daily prasugrel in the ISAR-REACT-5 of percutaneous coronary intervention (PCI) in acute coronary syndrome. Once daily drugs have a serious adherence advantage.


In the journal JACC-Heart Failure, a group of investigators published a report of 41 patients with post-acute sequelae of severe acute respiratory syndrome from SARS-CoV-2 (PASC) who had persistent dyspnea on exertion and underwent cardiopulmonary exercise testing (CPET).

CPET is different from standard stress testing in which you just look at exercise tolerance, BP, and ECG. In CPET, you get much more information on pulmonary function and oxygen delivery to the muscles. It’s what is used to test endurance athletes.

  • 18 men and 23 women with an average age of 45 years. Left ventricular ejection fraction (LVEF) normal at 59.

  • Peak oxygen consumption (VO2) was low at 20.3, which is 77% predicted.

  • 58% had peak VO2 of < 80% predicted.

  • 15 of 17 patients of with normal peak VO2 had ventilatory abnormalities including peak respiratory rate of more than 55/minute or dysfunctional breathing.

  • For the whole cohort, 88% of patients had ventilatory abnormalities with dysfunctional breathing,

  • Nearly half the patients met criteria for myalgic encephalomyelitis/chronic fatigue syndrome.

In sum, the majority of patients had both circulatory and ventilatory issues. The authors postulate that this could be caused by cardiac, lung, and muscle injury from acute infection.

While this is a detailed report and provides interesting observations, the problem I see in making conclusions and translating the data to help people are twofold. First, you don’t know what one’s cardiac and pulmonary function was before they had COVID infection. A sub-average maximum VO2 means nothing if you don’t know what that person could do before COVID. Second, you don’t know if these findings are specific for SARS-CoV-2.

In 2020, I had a viral pneumonia that turned into a bacterial pneumonia. I had bilateral infiltrates, fever, and O2 saturations in the 80s for a week. I wasn’t right for at least 3 months. I would have flunked a CPET. When I tried to exercise months later, I had utterly disordered breathing. Thus, we want and need to know whether these abnormalities in CPET follow other pulmonary infections.

Calcium and Cardiac Arrest

The community that studies resuscitation science does it right. This week, JAMA published another RCT, from a Danish group, studying the treatment of patients who have out of hospital cardiac arrest (OHCA).

The question was, does intravenous calcium improve outcomes. It’s plausible that calcium would help, given its role in contractility. Two small trials done in the 1980s found no significant differences with calcium administration, but both trials had point estimates that favored calcium.

Small trials that have numerically fewer events in the treatment arm, but not enough of a difference, could simply be underpowered and suffer from a false negative result. This is called type 2 error.

The best way to settle the question is not with a retrospective review of charts comparing survival in patients who received calcium vs those who did not. The best way is to randomly assign patients with OHCA to receive calcium or not. That’s what these investigators did. The primary endpoint is easy: return of spontaneous respiration (ROSC). About 200 patients were in each group. The results were clear: calcium did not work.

  • 19% in the calcium group had ROSC compared with 27% in the saline group (P = .09).

  • The trial was stopped because of potential harm.

This trial makes the podcast for two reasons: the small reason is that we now know not to give calcium for OHCA. But the main reason is that it is a shining example of how to answer important questions in Medicine. You have to have the courage to test things with prospective blinded randomized trials. Imagine if calcium was beneficial and all these years we have been omitting something helpful.

Social Media

Medscape Medical news covered a new survey from Press Ganey on consumer trends in 2021, the first chapter of which was titled: The Modern Patient Journey. The key finding was that 83% of respondents said they went online to read reviews of a physician after receiving a referral from another provider.

I realize what I am about to say seems like a banal thing to say on the podcast, but it relates to advice I have given in numerous lectures on social media—one I learned from Bryan Vartabedian. Doctor V says you can create your own story, or it will be created for you. You have 100% control over what you say about yourself and nearly 0% control of what others say about you.

His point, and I agree with it, is that it seems wise to have a web presence that you have created. A webpage, a blog, a Twitter feed, a place where, if a prospective patient Googles you, they will find out who you are. The problem with patient reviews is that the N is small and most people with good experiences are not motivated to write reviews. So, a couple of bad reviews really stand out. Imagine if before someone gets to your reviews they see your personal website, your blog, your Twitter feed—these are things that you control. For those publishing a ton of stuff academically, you will have an Internet presence, but most of us don’t publish a lot. It doesn’t take long to create a public presence. And then you get to define yourself.


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.