In Defense of the Coronary Artery Calcium Scan in Primary Prevention of Atherosclerotic Heart Disease

I had the pleasure of working alongside Dr. Anthony Pearson, known now as the skeptical cardiologist, in private practice for a decade. He is a thoughtful well-read cardiologist. We agree on many things. We don’t agree on the use of coronary artery calcium scoring. I have argued it is has no value. Dr. Pearson makes the pro-CAC case below. I am delighted to publish this view, even though I disagree with much of it. JMM

---

Although the skeptical cardiologist questions the value of many procedures and diagnostic tests in cardiology, he remains an enthusiastic advocate of the use of coronary artery calcium (CAC) testing to better personalize each individual's risk of heart attack and stroke.

You can read all about how the test is performed and why it is so valuable in my "Ultimate Guide to the coronary Artery Calcium Scan."

I am aware that controversy (at least on the social medium formerly known as Twitter) is raging over the utility of the coronary artery calcium scan in preventive cardiology.  Although I agree with the analyses of my esteemed former colleague John Mandrola on the vast majority of topics he reviews, I beg to differ when he criticizes CAC scoring.

To the majority of preventive cardiologists “in the trenches” like me who have decades of experience using the CAC to fine tune their patient’s risk of heart attack, stroke, and cardiovascular death, the value is obvious. 

The test provides direct information on the level of atherosclerosis in our patients’ coronary arteries! Atherosclerosis is the disease that causes atherosclerotic cardiovascular disease (ASCVD) which is responsible for the vast majority of heart attacks and cardiovascular morbidity and death.

Doesn’t it make sense to take a snapshot of the level of atherosclerosis actually in the arteries rather than relying on an indirect, inferior estimate of ASCVD risk?

The Evolution of ASCVD Risk Assessment

The Framingham Risk Study (FRS) which began assessing risk factors for Americans 70 years ago developed the traditional risk factor approach that serves as the foundation for determining which individuals will benefit from lipid-lowering therapy. In the last 30 years, however, multiple non-traditional risk markers have emerged that incorporate measures of subclinical atherosclerosis.

Since 2000 the landmark MESA study has been looking at the prevalence, correlates, and progression of these nontraditional risk markers of subclinical atherosclerosis in a multi-ethnic population without known ASCVD at baseline. A wealth of publications (69 at last count) have come out of the MESA database with many of them informing our understanding of the value of CAC.

MESA demonstrated that CAC is the single best predictor of ASCVD risk. CAC was much better than carotid intimal-media thickness (CIMT), brachial flow-mediated dilation (FMD), ankle brachial index (ABI), high sensitivity C-reactive protein (hs-CRP) and family history of coronary heart disease (FH). 

Using the MESA data and risk score with an app or online you can see how CAC modifies the risk estimate, sometimes revealing a greatly enhanced risk, sometimes revealing markedly lower risk, and sometimes making no change. 

The online MESA Risk Score Calculator is available here and the app can be downloaded for free (search for “MESA Risk Score”) on Apple and Android smartphones.

Limitations to the Standard Approach

Most physicians have been taught to follow the 2018 American Heart Association (AHA)/American College of Cardiology (ACC) cholesterol guidelines which recommend using pooled cohort equations (PCEs) to estimate 10-year atherosclerotic cardiovascular disease (ASCVD) risk and thereby decide who does or does not warrant lipid-lowering medications. 

These PCEs plug the standard ASCVD risk factors (blood pressure, smoking, hyperlipidemia, diabetes, age) into an algorithm that identifies high-risk (>20%), intermediate (10-20 %) and low-risk categories.

Why would one rely entirely on this method when we can personalize each individual's ASCVD risk?

Limitations of the PCE

• These scores do not apply to young adults (<40 years) or older adults (>75 years)

• In adults <55 years of age who experience an MI, the predicted 10 year risk is very low. Most would not have reached levels which triggered statin therapy. Given that sudden death may be the first symptom of MI, the standard 10 year ASCVD risk gives a false sense of security which means loss of opportunity to prevent death.

• Because they are based on population studies from 20 years ago PCE equations overestimate risk in the general population.

• Individuals with low to moderate ASCVD risk by the PCEs can account for 2/3 of individuals presenting with STEMI.

• They don’t take into account an individual’s family history of CAD. If your dad died of an MI at age 50 this does not put you into a higher risk category.

Due to the uncertainty of individual risk after this information is reviewed, good preventive cardiologists seek more information to further refine their patient’s risk.  

Four tests are generally recognized as useful for this. Three simple and inexpensive blood tests include the hs-CRP, apolipoprotein B, and lipoprotein (a) and the coronary artery calcium scan. 

Criticisms of CAC

The arguments frequently presented by the anti-CAC side of the this topic—typically from doctors who consider themselves medical conservatives—involve pointing at perceived downsides of CAC testing plus the absence of RCT level data that shows cardiac outcomes are improved.

I consider myself a staunch medical conservative and I’ve written extensively on the lack of value of various interventional procedures and misguided screening tests.

But CAC differs substantially from these low value interventions and tests in a number of critical areas.

First, CAC performs wonderfully as a screening test because a) it directly measures the process we are concerned about b) it does this with very high accuracy and reproducibility. False positives occur but they are so rare as to warrant a case report publication c) it is highly predictive.

The risks of CAC are minimal when the test is ordered and acted upon by an enlightened physician. 

Radiation and Other Risks 

The radiation exposure from CAC scans has decreased substantially in the last decade and is now very low, averaging 1 mSv. This is less radiation than the 1.2 mSv most Americans receive annually from natural exposure. 

Risks of radiation doses below 100 mSv have been described by US and international radiation protection organizations as meaningless because long-term effects are either too small to be observed or non-existent.

For comparison purposes, a typical mammogram uses 0.4 mSv, a chest X-ray uses 0.1 mSv. Two cardiac tests that are widely and often inappropriately used in the US, coronary angiography (5-10 mSv) and myocardial perfusion imaging stress tests (12-14 mSv), expose patients to 10 times the radiation that CAC scans do.

In addition, with CAC scans there is no risk associated with accessing the venous or arterial system and no risk of contrast reactions. There is little to no value in repeating a CAC scan so the maximum lifetime exposure will be 1 mSV.

Skill, equipment, technicians, and patient prep all matter with coronary CT angiography (unlike CAC). Unless great attention is paid to all of those there is a tendency to overestimate stenosis, especially if CAC score is high.

With mammography or PSA screening, the follow-up testing for the many false positives ranges from incredibly painful to dangerous.

If you see a good preventive cardiologist or internist in conjunction with your CAC and you are free of any cardiac related symptoms there should be NO painful, invasive, expensive or dangerous tests performed on you.

Downstream Testing and Cardiac Cripples

Some critics have emphasized the possibility of creating “cardiac cripples.” 

Identifying an abnormally high CAC score in a middle-aged individual should be immediately followed by an in-depth discussion of its meaning with an enlightened physician. 

For many individuals, this is life-transforming. If the discussion is done properly, the test result will radically improve their life and longevity. For the 25% who now find themselves with documented advanced subclinical atherosclerosis there is much greater motivation to make lifestyle changes: to lose weight, to exercise, to improve their diet and to stop smoking.

My patients are much more motivated to address diet and lifestyle factors increasing their risk after receiving direct confirmation of plaque in their coronary arteries.

When I spend the considerable amount of time necessary with such patients to explain that this is not a death sentence and that with aggressive management of risk factors they can live a totally normal life, they end up in a mental and emotional state more appropriate to their long term cardiac risk: aware of their personal risk, aware of their opportunity to basically eliminate heart attack and stroke, and highly motivated to modify lifestyle and dietary factors.

If you see a good preventive cardiologist or internist, in conjunction with your CAC, and you are free of any cardiac-related symptoms, there should be NO painful, invasive, expensive or dangerous tests performed on you.

Unfortunately, if you see an invasive cardiologist or even a general cardiologist who is not up to date on the appropriate treatment of stable ischemic heart disease you may end up getting a stress test (known for its false positives) or even worse invasive coronary angiography followed by a coronary stent you didn’t need.

Lack of RCT

I mention the lack of an RCT at the end of my piece. I don’t see it as a deal breaker. I see the CAC as a diagnostic test, like an echocardiogram that tells us if a murmur is from a valve problem or not. We have no RCTs on anything related to echocardiography yet we continue to order them and act on the results. 

We wouldn't think of managing a patient with a heart failure diagnosis without an echocardiogram which actually lets us know how the left ventricle is functioning. Yet, echocardiography is far more difficult to perform and interpret properly than CAC. It is very common for echo tests to be botched and to find marked disagreement in their interpretation which often leads to repeat and unnecessary, potentially dangerous down stream testing.

Guess what? There are no randomized trials of HF management with and without echocardiography.

Unlike many screening tests, the CAC is actually identifying and measuring the disease we are thinking about treating! 

Instead we often rely on the “risk estimates from pooled cohort equations” suggested by guidelines. Is there an RCT  showing the benefit of that approach? 

Preventively Yours,

-ACP