6:55 a.m. - 2021-12-04
THE ABSTRACT PROBLEM
The latest social media Covid hot topic is an abstract appearing in CIRCULATION, the journal of the American Heart Association. Written by Steven R Gundry, a former cardiothoracic surgeon, who now promotes nutritious ways to avoid surgery, the abstract claims to have found a link between mRNA Covid vaccines, endothelial inflammation and T cell infiltration of the heart. Gundry used a test called PULS, which he calls "a clinically validated measurement of multiple protein biomarkers which generates a score predicting the 5 yr risk (percentage chance) of a new Acute Coronary Syndrome." Gundry claims that in "most" of his 566 patients the five-year risk of ACS went from 11% to 25% following vaccination with mRNA Covid vaccines. He concludes that these findings "may account for the observations of increased thrombosis, cardiomyopathy, and other vascular events following vaccination."

The abstract was accompanied by an Expression of Concern by the journal, pointing out several potential errors and concluding that "the abstract in its current version may not be reliable."

The Expression of Concern and subsequent discussions (on Twitter, naturally) include several criticisms:
1. There is at least one typographical error, misspelling the acronym for PULS as PLUS. A minor concern, and one that to this reader does not seem relevant to the validity of the abstract, though it does suggest carelessness on the part of the author.
2. There is no data in the abstract regarding T cell migration. A surrogate marker was used instead. Surrogate markers would normally be the starting point for a more thorough attempt to demonstrate T cell migration by other means.
3.There are no statistical analyses presented. One researcher on Twitter claims to have calculated the confidence limits using Gundry's data and found no real differences in the before and after numbers.
4. Anecdotal data for "vascular events following vaccination" are presented as conclusive. Misuse of the Vaccine Adverse Event Reporting System) has become common during Covid vaccine debates.
5. Myocarditis has been documented after vaccination, but it is less severe than the myocarditis that occurs with a Covid infection. In any event, the PULS system is meant to predict ACS, not myocarditis.
6. Another researcher on Twitter says there are no good data demonstrating the predictive value of PULS for ACS.
7. There is no mention of a control group for changes over time. In other words, what about Gundry's patients who did not get vaccinated, or those who received the Johnson and Johnson vaccine? His conclusions could be simply a post hoc error, i.e., A precedes B, therefore A must have caused B.
8. Gundry does not discuss how he determined "normal" levels.
9. Gundry says the changes became apparent in "most" of his patients, without giving a numerical value to "most."
10. Gundry's reasoning, therefore, goes something like this: "According to a test which uses SURROGATE MARKERS, MOST of my patients who received an mRNA vaccine are statistically at higher risk for ACS. Therefore, there is a link between vaccines and ANECDOTAL ACCOUNTS of other forms of cardiac damage."

Abstracts, even those published in prestigious journals, are not peer-reviewed. They include limited data, no graphs or tables, no references to earlier publications to back up their claims, no detailed statistical analysis. It is usually the policy of groups like the AHA to publish meeting abstracts en masse in CIRCULATION. All this is common practice. Not so common in an Expression of Concern such as the one that accompanied Gundry's article. Unfortunately, most of us are not trained scientists. We don't understand experiment design or statistics. But in our current political environment, anyone with a mistrust of vaccines can seize on "science," good or bad, to justify their prejudices. Scientists are partly to blame; publication of abstracts and preprints, both of which lack peer review, only leads to more confusion, and the Internet guarantees that scientists are no longer communicating only with each other.

previous - next