How Much Do We Worry About Silent Atherosclerosis?

A massive study in Denmark and Spain underscores our old knowledge that blood vessel disease begins early and is hard to avoid. Here is my take.

The title of the paper in the New England Journal of Medicine is Prevalence of Silent Atherosclerosis across Adult Life.

I was surprised that it was a featured study at the European Society of Cardiology congress early this month. I had thought we knew from pathology studies of young people who died of trauma that blood vessel disease (including in the coronary arteries) can begin early in life. Of course, pathology studies are limited because it’s a small sample of people who are unlucky enough to have died.

The Danish and Spanish team provide us a large systematic look at nearly 17,000 asymptomatic adults from young to older using modern imaging techniques. We actually learn some interesting and provocative findings. I write provocative because there is currently great interest among the tech community to inoculate themselves against heart disease.

The Study

The authors enrolled adults without disease and symptoms from five age strata. They then underwent 3D ultrasound of the carotid and femoral arteries as well as a coronary CT scan. The primary endpoint was the presence of atherosclerosis. It was not an outcome study. Nor was it a prospective study looking at progression in individuals.

There were 2000 adults in the age 18-29 and 60-70 group and 4000 each in the 30-39, 40-49, and 50-59 strata.

Findings

• More than half had silent atherosclerosis—63% of men vs 51% women.

• Even young people (18-29) had disease—8.7% of men vs 6.7% women. This rose to ≈ 90% by age 60-70.

• Plaque volume increased exponentially with age; in younger adults with coronary plaque, roughly 40-48% had a coronary calcium score of zero — meaning calcium scoring alone missed a substantial share of real disease.

• Clinical risk scores had terrible sensitivity. (This was a European study, so they used the SCORE2 classification vs the PREVENT or PCE that we use in the US.) The high-risk classification of SCORE2 had a great specificity of 99% (low false positive) but only 2% sensitivity. Meaning that if you used the clinical risk score criteria for high risk, you would miss almost everyone with actual silent disease. Even if you widen the net and include moderate risk on the clinical score, you still miss two-thirds of disease.

The authors gently suggested that these findings may have clinical value. They cite the low pick-up rate of risk scores and write that these scores and direct imaging capture different information, especially at younger ages.

[Our] results support the position that primary prevention strategies informed by imaging may improve risk assessment and may warrant starting early in life.

Comments

I laud the systematic nature of this study. It adds modern imaging of a contemporary population and documents what we have long suspected: blood vessel disease begins early and progresses throughout life, earlier in men but women too have progressive disease.

Some may be scared by these findings, especially in Danish and Spanish adults, as surely it would be worse in the Southeastern US. But I would not be scared.

First, the presence of atherosclerosis on an imaging test is still a surrogate marker. More and more, the outcome of interest in cardiology is not a shadow on an ultrasound or streak on a CT scan, but an MI or stroke. There is a correlation between extent of disease and risk of events, but it is only a correlation.

Second, I am also not put off by the fact that risk scores miss most cases of atherosclerosis. The job of the risk score is not to pick up a surrogate marker but predict risk of events so as to aid in primary prevention therapies (like cholesterol lowering meds, for instance). This study does not compare prediction from clinical risk scores vs anatomic imaging.

Third, perhaps the most important message from this study is limitation of coronary calcium screening. Andrew Foy and I have long argued that CAC scans have little value—in anyone. One of the many reasons for this is that they are much more likely to incorrectly misclassify a person. This study confirms that nearly half of people with actual disease had a CAC score of 0.

Fourth, I strongly disagree that these observations have any clinical use for improving risk assessment at any age. We know from the DANCAVAS I and DANCAVAS II trials, that a comprehensive cardiac screening program (including imaging) in Denmark (arguably a best-case-environment) did not improve mortality or major adverse cardiac outcomes. In fact, in DANCAVAS II, screening increased the risk of major bleeding.

If the authors wanted to use these findings to further study the risk prediction from anatomy vs risk scores, I would be supportive. This paper is interesting research, but I would be very reticent to unleash such screening into general populations—as it could lead to an increase in downstream testing and inappropriate stenting.

Finally, and this may sound unusual coming from a cardiologist, but one of the principles of early detection and screening for disease is that the better primary treatment becomes the less the impact of early detection through screening. My partners in interventional cardiology are truly amazing in their ability to stop heart attacks with urgent stenting. It is routine for a person to present with an MI due to an occluded vessel, be sent immediately to the cath lab, and have the artery opened with total preservation of heart muscle.

I make this point about the success of early intervention not to argue for total ignorance of heart health. On the contrary, I believe in the value of healthy eating, regular exercise and not ignoring common symptoms, such as new angina or difficulty breathing. But the quest to eliminate all risk of heart attack through intense imaging remains an unproven and likely unwise clinical strategy.

Atherosclerosis, like atrial and ventricular ectopy, and gray hair and wrinkles, and declining watts on a bike, surely remains part of the normal aging process.

Healthy living and exercise can slow the process, as can not smoking, and lipid-lowering therapy. But the idea that we must image our bodies to be healthier is unproven. Once again, I reiterate: be afraid of imaging when you feel well.