Oct 16, 2020 This Week in Cardiology Podcast

John M. Mandrola, MD


October 16, 2020

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.

In This Week’s Podcast

For the week ending October 16, 2020, John Mandrola, MD comments on the following news and features stories.

COVID Drug Trials News

Three big things happened in COVID science news. First, Eli Lilly's COVID-19 monoclonal antibody (mAB) is on hold for a potential (undisclosed) safety issue. I briefly mentioned mAB last week because despite the hype there are no phase 3 trials with clinical outcomes. As I discussed, the Bayesian prior on mABs is that these agents will not be game-changers for three reasons: Game-changers, or magic bullets are rare; upwards of 99% of people infected with SarsCoV2 will survive, so any therapeutic has a high-bar; and, with the help of scientist and science writer Dr. Hilda Bastian, I found a review article describing 21 mABs that failed in phase 3 trials in the last 5 years.

The other big COVID science news is that Johnson & Johnson said it was pausing all trials of its experimental COVID-19 vaccine after one participant developed an “adverse reaction.” No word on what the adverse reaction was. That is now the second US COVID vaccine trial on pause. The other is the Astra-Zeneca trial which was also paused for an adverse reaction. This Oxford-led trial has now resumed everywhere but in the United States.

Finally, the World Health Organization (WHO) consortium has announced results of the SOLIDARITY trial. Four drugs were compared with standard care in 11,266 patients with COVID in 405 hospitals in 30 countries: hydroxychloroquine, lopinavir, remdesevir, and interferon. The primary endpoint was in-hospital mortality. Just under 3000 patients received remdesivir. None of the four drugs reduced in-patient deaths.

Interim results of the trial were posted as a preprint and the authors’ final figure is a meta-analysis of the remdesivir trials. Now, given the large size of SOLIDARITY combined with all remdesivir trials shows no significant mortality benefit.

The WHO trial ought to seriously reduce the price paid for remdesivir, which is expensive. If the drug has any effect, it is likely quite modest. The dollars saved could be used in other more “scalable” interventions. One that comes immediately to mind is paying workers when they call in sick.

On the public health front, I discovered Dr. Stefan Baral, a public health specialist from Johns Hopkins whose area of focus is disparities. Dr. Baral grew up in Sweden and has family now living in Sweden. I highly recommend his Twitter feed. He is brave enough to talk candidly about how our interventions for this virus are increasing the already stark inequalities in health care outcomes in the United States.

BNP and COVID-19

A group from Stony Brook University in New York did a chart review of about 300 patients admitted with COVID-19 pneumonia in March and April. They excluded intubated patients or those who died within 24 hours of admission. Their research question was: In patients without a history of heart failure, does the N-terminal B-type natriuretic peptide (NT-BNP) provide any value in caring for patients with COVID? Thus, they studied the correlation of BNP with all-cause mortality. They found that baseline NT-proBNP was strongly associated with mortality.

Chart review studies that have no randomization and dichotomize a decidedly continuous marker don’t usually make the podcast, but there are teaching points. First is the issue of taking a continuous variable and picking a cutpoint. Google the words: the problem with dichotomization of variable and what pops up is a section of Vanderbilt professor Frank Harrel’s wiki page that lists 15 problems with this practice.

The second problem with this study, one that journalist Steve Stiles addresses in the news story with his interview of Dr. James Januzzi is what? How does knowing that a BNP is elevated *incrementally* help a clinician?

In the opening line of the news story, which almost always tells the reader the topline result, Steve writes: “Mortality goes up in tandem with admission natriuretic peptide levels in patients with severe COVID-19 pneumonia but no history of heart failure (HF)....” You could write that same sentence without the word COVID-19 or even without the word pneumonia. Mortality goes up in tandem with admission natriuretic peptide levels in patients with severe <insert any disease> but no history of heart failure (HF). This is because BNP is a marker for severe diseases. Diabetes, liver function, obesity, AF, tachycardia can also influence the BNP.

Sick people, especially those most severely affected by COVID-19, such as older patients with obesity, hypertension, or diabetes, might have an elevated BNP on presentation. When person comes in with dyspnea of unknown cause and the chest xray is inconclusive, a BNP may help. If normal it makes heart failure less likely. If it’s 30,000, heart failure is more likely. In a documented case of infiltrates with positive COVID test, a BNP probably adds little incremental supportive care. Dr. Januzzi’s quote in the news story is perfect: “If NT-proBNP is elevated, it's important for clinicians not to jump at a diagnosis of heart failure."  Why? Because doing so might cause you to diurese a patient who needs volume.

Do not ignore biomarkers, but do use them in the context of the whole clinical scenario.

Polypharmacy in Heart Failure

Last week I discussed the major discovery that SGLT2 inhibitor drugs benefit patients with heart failure with or without diabetes. Here is the problem: it is yet another pill a patient has to take, and people are only going to take so many meds. Guideline-directed medical therapy (GDMT) of heart failure already includes three drugs: RAS blockers, beta-blockers, and mineralocorticoid receptor antagonists (MRA).

But almost no patients with heart failure with reduced ejection fraction (HFrEF) have pure left ventricuar dysfunction. Many have ischemic heart disease or AFib or diabetes or hypertension or a slew of other conditions that are treated with.... pills.

A group from the Weill Cornell Medical College studied the issue of polypharmacy in 558 older patients who were admitted for HF. They used baseline data from the REGARDS study, which used a prospective, population-based US cohort. Some of the baseline characteristics are super important: The median age of patients admitted for heart failure was about 76 years. More than half were older than age 75. About 43% were women.

Let’s stop there.

In the original beta-blocker trials, the average age was 58 to 63 years. Women comprised 23% of the population in MERIT, 30% in the carvedilol trials. Thus, GDMT was established in young, mostly male outpatients. But the vast majority of patients we apply GDMT to are older, often female patients. In this new study, the vast majority of participants (84% at admission and 95% at discharge) took ≥5 medications; and 42% at admission and 55% at discharge took ≥10 medications.

  • The prevalence of taking ≥10 medications (polypharmacy) increased over the study period.

  • The authors separate the meds out into HF meds, non-HF but cardiac meds, and non-cardiovascular (CV) meds.

The most common non-HF med started was aspirin (ASA). This is appropriate in patients with a previous history of a cardiac event, like vascular stroke (not due to AFib), MI, or percutaneous coronary intervention (PCI). But there have been three large trials of ASA used in primary prevention and none showed a net benefit. The ASPREE trial specifically studied high risk elderly patients without prior CV disease, and found higher mortality on ASA.

The second most common non-HF drug was a statin. This may sound controversial but we have two very large statin RCTs in patients with HFrEF—GISSI and CORONA. These studies found no reduction in MACE with statin therapy. In GISSI, 40% of patients had an ischemic cause of HF. And in CORONA, 60% of patients had a history of MI.

The reason these studies did not show a benefit from statins is simple: competing causes of bad outcomes. Statins may reduce the rate of nonfatal CV events by a modest amount. But when you are old and have HF, that modest benefit is outweighed by other things. The bottom line is that coronary artery disease should not always require a statin. Statins, especially in the elderly and infirmed, offer an excellent opportunity for deprescribing.

A common non-HF med was potassium replacement. The authors highlight the low use of MRAs and suggest that adding an MRA may obviate the need for a potassium replacement. I love that idea. I call spironolactone my secret weapon. This drug, which has RCT level evidence, is underused.

Cardiac Arrest in In-Patients With COVID-19

JAMA-IM published a research letter from a group at the William Beaumont Hospital in Michigan on clinical outcomes in patients who have cardiac arrest in the hospital.

A recent review found that before the pandemic only one in four patients who have in-hospital cardiac arrest (IHCA) survive to discharge. Eight of 10 of these cases had nonshockable initial rhythms. When the initial rhythm is nonshockable, survival is only 10%. Keep in mind that survival to discharge is a low bar. Only a fraction of these survivors go on to live independently.

Now to COVID. In this single-center study, 1309 patients hospitalized with COVID-19 (4.6%) developed IHCA and underwent cardiopulmonary resuscitation (CPR). The research letter described these 54 patients. Only 2 had shockable rhythms. Pulseless electrical activity and asystole occurred in 52 of 54 patients. Return of spontaneous circulation was achieved in 29 patients (53.7%). Half of these had their code status switched to DNR while another half were recoded but still died. The survival to discharge was 0 of 54 (95% CI, 0-6.6).

The first thing to say about this study is that it is single-center and observational, but I am glad JAMA-IM published it. These are important observations for four reasons. From a therapy perspective, CPR should be considered a treatment—like a pill or a procedure. In this study, this treatment failed 100% of the time.

From a patient perspective, many patients when given information on what CPR actually looks like would choose not to have a nearly 100% ineffective procedure and highly aggressive procedure. From a healthcare cost standpoint, CPR and subsequent ICU admission is extremely costly. If we use dollars on futile care, that reduces the resources available for effective interventions.

From a safety perspective, CPR in a patient with COVID puts healthcare workers at significant risk. It would be one thing if the procedure had a chance of success. I am no ethicist, but given the dismal success of CPR in this disease, it would seem totally ethical and moral to not perform this procedure, especially in patients with nonshockable rhythms.

Some of you may disagree with this comment, I am happy to be critiqued. Let me know in the comments section or on Twitter.


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.
Post as: