At this moment in time the pre-pandemic cardiology research agenda needs to be completely reprioritized. There are two broad areas that now take precedence over all existing research concerns. On the one hand, researchers need to achieve a better understanding of the staggering incidence of deferred or delayed treatment of cardiovascular events and conditions as a result of the pandemic. Of course there will be important adverse consequences when serious events go untreated (but there also may be important insights that can be achieved about overdiagnosis and overtreatment).
On the other hand are urgent questions about the role of cardiovascular disease in patients with COVID-19. This includes both the increased risk from the virus in people with existing cardiovascular conditions or risk factors as well as the frightening possibility that COVID-19 itself may have serious short- and long-term adverse effects on the cardiovascular system. To date there have been no really good scientifically rigorous studies that provide useful and actionable information about this last truly urgent question. This represents a catastrophic failure of the cardiology research establishment.
The middle of a raging pandemic is no time for business-as-usual research studies, especially when the studies falsely purport to shed light on that pandemic. Such papers have all the appearance of high quality science. They have a ton of data and a lot of big words but they’re incapable of providing reliable or useful information that can aid in our understanding and control of the health emergency.
The problem is that academic medicine in general, and academic cardiology in particular, have become factories for the assembly-line production of low quality studies that either don’t add to our understanding of important questions in any significant way or, worse, influence the scientific discussion with hype, fear, or confusion. Even when such research is relatively harmless– for instance, if it doesn’t promote a drug or procedure that doesn’t work or is dangerous– it represents a wasted opportunity, effort that would better have been directed toward higher quality research. Low quality research is never desirable; now, in the time of the coronavirus, it is inexcusable.
A new study published in the Journal of the American College of Cardiology illustrates this problem. A large group of authors, including some of the most prominent names in cardiology, report the results of “an international, multicenter cohort study” in New York and Milan that retrospectively analyzed 305 hospitalized patients with COVID-19 who received a transthoracic echocardiogram (TTE) and an ECG in the hospital.
The 55 (!) authors of the paper report myocardial injury detected by biomarkers in 62.3% of their patients. The major finding of the study was that there was “a higher prevalence of ECG and echocardiographic abnormalities” in the patients with biomarker abnormalities, and that in-hospital mortality was higher in patients with biomarker abnormalities and, in that group, highest in those who also had echo abnormalities.
The major problem is that there’s no good context for this data; it’s purely observational and retrospective. There was no question that was addressed prospectively. In other words, the study is incapable of demonstrating cause and effect and can only be used to generate questions, not answers. This is a distinction the journal’s editors and reviewers should have insisted that the authors fully respect.
The authors acknowledge that “there was no systematic basis on which patients were selected to undergo echo evaluation. In fact, it is likely that only patients who were perceived to be at higher risk on clinical grounds underwent TTE.” Is it any surprise that there are many heart abnormalities in patients who undergo echo, and is it any surprise that patients with echo abnormalities have a bad prognosis?
James Stein, a cardiologist at the University of Wisconsin, pointed out a number of additional limitations in an email:
—“The study only describes people sick enough to be hospitalized with COVID-19 so it already only applies to ~5% of people who get it, maybe less.
—“The study only applies to those sick enough to have medical indication for an echo, so it applies to an even smaller number. The N = 305 is shockingly low since it comes from 7 medical centers in hotspots where likely thousands got hospitalized with COVID-19. I suspect some cases were missed.
—“Many patients had risk factors for structural heart disease such as hypertension, kidney disease, anemia, atrial fibrillation, or coronary artery disease. At any given time, if you randomly check troponins and echo in people like this you will find troponin elevations and echo abnormalities. The cardiologists in our hospital perform innumerable consultations for patients like this. A control group would have helped.
—“Critically a control group of people with acute illness or especially acute viral illness would tell how often these observations occur even without COVID-19.
—“Were the echo readers blinded to COVID-19 status? Unlikely. To cardiac risk factors? Unlikely. Troponin level? Unlikely. More bias can creep in.
—“The conclusion ‘myocardial injury with TTE abnormalities was associated with higher risk of death’ likely applies to everyone admitted to the hospital with any acute disease state.”
In other words, the study didn’t tell us– couldn’t tell us– anything we didn’t already know. People with cardiovascular disease who have COVID-19 have worse outcomes than COVID-19 patients without cardiovascular disease. Is that a surprise or helpful in any way?
It’s worth reemphasizing Stein’s point that we have no information at all about the larger COVID-19 population from whom the study was drawn. If we are going to understand the role of cardiovascular disease in the pandemic this information od desperately needed.
The study authors conclude that “TTE evaluation should be considered in patients with COVID-19 and biomarker evidence of myocardial injury to characterize the underlying cardiac substrate, for risk stratification, and to potentially guide management strategies.” This statement is unremarkable, but the exact same words could have been said in the complete absence of the study!
I don’t want to pick on this one study, though the presence of so many distinguished authors makes it a ripe target. The fact is that we’re seeing so much unhelpful research during the pandemic, and I’m afraid we’re going to see even more. It’s another sad example of our low times.
We’ve seen, for instance, specialists in cardiac magnetic resonance imaging publish a seriously flawed study that has been used to promote greatly expanded use of CMR in COVID-19 patients. Separately, participants with different positions in the nutrition wars have argued that the pandemic should be fought by adopting their own preferred nutritional strategy!
These are just a few examples. The most egregious case involved the two most prominent medical journals in the world, the New England Journal of Medicine and the Lancet. These journals published, and then retracted, two papers from a research group based on, apparently, entirely fraudulent data. The studies were highly impactful. It should come as no surprise that many of the co-authors of the two papers were highly prominent cardiovascular researchers.
The pandemic should not be the excuse for all sorts of shoddy research. Instead, the pandemic should be the compelling reason why only medical research of the absolutely highest standard should be encouraged and taken seriously.
This is a time to raise standards, not lower them.