COMMENTARY

Jul 10, 2020 This Week in Cardiology Podcast

John M. Mandrola, MD

Disclosures

July 10, 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 July 10, 2020, John Mandrola, MD comments on the following news and features stories.

Brief COVID Update: A New Surge

In the United States, the rate of growth of cases over the last 2 weeks is 1.15x, which is significantly greater than when I recorded my last podcast two weeks ago. Dr. Eric Topol Tweeted yesterday that every day sees a record new number of cases; Wednesday there were 62,000 new positives.

US deaths two weeks ago were 120,000 and this week they are 135,000, for a rate of rise of 1.06x, which is steady. But moments before I recorded, I looked at the COVID Tracking Project and it looks like the seven-day average for deaths is now moving up.

A lot has changed in two weeks. States that were quiet when New York, New Jersey, New Orleans, and Boston were surging are now seeing massive numbers of cases and hospital admissions. Some hospitals are close to being overwhelmed by COVID cases. Also, the percent of positive tests in some cities is striking and suggests widespread community spread.

COVID seems to be increasingly infecting younger people. While younger people do better, on average, this is still worrisome because increasing prevalence among the young will increase the odds of spreading the virus to older, more vulnerable people; and not every young person gets off easy. The good news: there numerous vaccine trials. Gosh we need a vaccine.

Sacubitril/Valsartan

Six years out from the landmark PARADIGM HF trial, I am gaining confidence with sacubitril/valsartan (SV). As an electrophysiologist, I see a lot of patients in evaluation for an implantable cardioverter-defibrillator (ICD), and a lot of patients with ICDs who have reached battery depletion. Right or wrong, the left ventricular ejection fraction (LVEF) is central in the ICD decisions. And over time I have seen a number of patients with improved EF after taking SV.

That is why I was drawn to a small but intriguing observational study from Italian investigators. Presented as an abstract at the virtual European Heart Rhythm Association meeting, the authors measured EF at baseline and then after 6 months of SV in patients with heart failure with reduced ejection fraction (HFrEF) and ICDs. The primary outcome was EF improvement and the clinical implications are obvious: if the drug can improve EF, we may be able to avoid ICD implantation and/or replacement.

At 6 months of follow-up, the mean absolute increase in LVEF was 3.9%, and 57 patients (25%) achieved an LVEF > 35%. I don’t mean to make too much out of a meeting abstract of an observational study but this is provocative and plausible, and consistent with my observations.

I went through the clinicaltrials.gov site looking for EF studies from PARADIGM and there are none. I did find a number of observational studies on the topic, the best of which is the PROVE-HF study published in JAMA last year. This was an uncontrolled study of about 800 patients on SV; it found that at 12 months, LVEF increased from 28.2% to 37.8% (difference, 9.4% [95% CI, 8.8% to 9.9%]; P < .001. The theory is that reverse and beneficial remodeling is the putative reason for the drug’s success in AF. Potential improvement in cardiac structure and function directly pertain to ICD net benefits. Recall that old trials of ICDs showed significant benefits, but the more contemporary DANISH trial, which was limited to nonischemic cardiomyopathy, found no ICD benefit. This is almost certainly related to the fact that as HF therapies improve the incremental benefit of ICDs decrease.

PRAGUE-17

Here is the thing about left atrial appendage occlusion (LAAO) with a device: it would be great if it worked. AF is a growing global problem. Taking a daily anticoagulant (AC) is a pain in the butt.

A brief comment on the background: we must remember that LAAO vs warfarin was hardly convincing. PROTECT AF did not pass FDA muster. The first co-primary endpoint of PREVAIL (a composite of stroke, systemic embolism, and cardiovascular death) missed its lax non-inferiority margin. Why? Because ischemic strokes with LAAO were more than twofold higher than with warfarin.

Crucially, patients who could not take AC were excluded from the regulatory trials. But these are the patients most commonly implanted with a device. That’s why the direct-acting oral anticoagulants (DOACs) vs LAAO comparison in PRAGUE-17 was so important. I wish that the trial results supported the authors’ conclusions. It did not.

PRAGUE -17 was a noninferiority (NI) trial. Which is a fine—indeed this is a good way to study LAAO. How would such a NI trial look? The DOAC trials show the way—RELY, ROCKET-AF, and ARISTOTLE all compared DOAC to warfarin and measured NI for efficacy, narrowly defined as stroke and systemic embolism. Then the authors tested safety in a superior design of DOAC vs warfarin.

But that is not what PRAGUE-17 authors did. Instead they chose everything bad that could happen to a patient with AF and put it into a composite: stroke or TIA, systemic embolism, bleeding, CV death, or a procedural-related complication.

In all, 47 primary outcome events occurred in the DOAC group vs 38 in the LAAO arm, resulting in a hazard ratio of 0.84 with confidence intervals of 0.53 to 1.31. The upper bound of 1.31 was less than the NI margin of 1.47 so the authors declared NI.

This is incorrect. You cannot combine endpoints of efficacy and safety in a NI trial. It’s circular logic in the case of stroke prevention. Dr William McIntyre said it well: If you include events that you expect to trend in opposite directions (thrombosis and bleeding), you bias to the null and make noninferiority easier to obtain." 

The likely reason PRAGUE-17 authors chose this design is not nefarious, and I mean no ill will. I am almost certain the composite endpoint was chosen for power reasons. But that does not change the fact that we cannot and we should not cite this study to suggest device NI to DOAC.

Statins

Observational studies are not useless, they can tell us about procedural complications, real-world uses of therapies, temporal trends, and sometimes raise good questions to be tested in randomized controlled trials (RCTs). The problem with observational comparisons is that the two groups are not randomized and thus one never knows if observed differences are due to the treatment or baseline differences.

Statins are most studied drug ever. But the authors asked an important question: do statins lower the risk of death in patients older than 75 who have not had a cardiac event? The problem is that no matter how many patients you include in a nonrandomized study, you don’t approach the value of an RCT.

This was a retrospective cohort study that used propensity score overlap weighting and included 326,981 participants from the VA health system. The study concluded: new statin use was significantly associated with a lower risk of all-cause and cardiovascular mortality. The problem was that in absolute terms the reduction in overall mortality with statins was far larger than the reduction in CV death or cardiac events.

For instance, overall mortality was about 1.9% less in those who took statins but CV death was only 0.3% lower and cardiac events was only 0.6% less. Why does the effect size for overall mortality dwarf reductions in CV death and cardiac events? Because healthy older adults are likely prescribed a statin.

If you want to know whether statins benefit older adults, you have to do trials. Citing an observational trial—no matter how many thousands of patients it includes—is the wrong answer for a patient who is deciding whether or not to take a statin. I mean no malice to the authors or JAMA but medical science would be fine without further observational trials of statin drugs.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
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:

processing....