Subclinical atherosclerosis is associated with nearly 10-fold increase in heart attack risk

Coronary artery disease (CAD) is the most common form of heart disease in America. The primary cause is atherosclerosis, which is the buildup of fatty plaques in the arteries of the heart that can lead to heart attacks (myocardial infarction) if a major coronary artery becomes blocked. Despite the many treatments to reduce risk factors for atherosclerosis such as lowering levels of LDL (bad cholesterol), heart disease is still the number one cause of death in America with cancer a close second.

Symptoms of CAD include angina (chest pain), shortness of breath when exercising, and a fast heartbeat. But by the time these symptoms appear, the disease could be in an advanced state. Even worse, many have heart attacks without any symptoms, implying that earlier identification would help. In a previous post I described how an imaging method, cardiac CT, that can be used to diagnose CAD at an earlier stage. Indeed by some estimates, ~50% who die suddenly from CAD (i.e. heart attack) did not manifest symptoms.

Given the likely prevalent nature and medical significance of subclinical forms of CAD, i.e. before symptoms arise, one wonders how to characterize these subclinical cases and the extent of elevated risk. A new study addressed these questions taking advantage of a powerful heart imaging technique. The researchers used coronary computed tomography angiography to assess subclinical coronary atherosclerosis in a population of asymptomatic individuals.

What is coronary computed tomography angiography and how does it compare to cardiac CT?

Previously, I described cardiac CT as a procedure to screen for atherosclerosis at an earlier stage. More specifically, I detailed calcium scoring CT which is one type of cardiac CT. Coronary computed tomography angiography (CCTA) represents another type of cardiac CT.  While both tests involve the use of x-ray imaging CT technology, they provide different types of information.

CCTA is a diagnostic test that uses contrast dye and X-rays to produce detailed images of the coronary arteries and identify any blockages or narrowing in the vessels (Figure 1). In terms of procedure, CCTA requires the injection of contrast dye into a patient's bloodstream and takes longer to perform than calcium scoring CT. CCTA provides detailed images of the coronary arteries, which can be used to detect or rule out blockages and other abnormalities. 

Calcium scoring CT, on the other hand, is a screening test that is used to measure the amount of calcified plaque in the coronary arteries which correlates with atherosclerosis. It is a quicker and simpler test that does not require contrast dye injection, but cannot directly assess coronary artery blockage.

The ionizing radiation (i.e. x-ray) exposure is lower for calcium CT,  but is generally considered to be relatively harmless for both tests, and the benefits outweigh any risks.

In summary, CCTA and calcium scoring CT are both useful cardiac CT scans that provide different types of information. CCTA is a diagnostic test that provides detailed images of the coronary arteries, while calcium scoring CT is a screening test (i.e. faster without dye injection) that measures the amount of calcified plaque present in the coronary arteries. The choice of which test to use depends on the patient's specific circumstances and the reason for the test.

Returning to the paper, the trial design was a prospective observational cohort study. The cohort population consisted of 9533 asymptomatic persons aged 40 years or older without known heart disease. They were all imaged by CCTA, and categorized according to luminal (cavity) obstruction (non-obstructive or obstructive; ≥50% luminal narrowing) which measures the occlusion of the coronary arteries, and according to extent (nonextensive or extensive; ⅓ or more of the coronary tree). Thus, there were 4 categories: non-obstructive, nonextensive (control group); 2) non-obstructive, extensive; 3) obstructive, nonextensive; and 4) obstructive, extensive. The primary outcome was myocardial infarction (heart attack), and the secondary outcome was a combination of death from any cause or myocardial infarction. These outcomes were tabulated for each of the 4 groups over an observation period of approximately 3.5 years (range, 0.1 to 8.9 years).

Defining subclinical atherosclerosis as having the obstructive or extensive designations from the CCTA, the researchers found that 54% of the cohort subjects had no subclinical coronary atherosclerosis, 36% had nonobstructive (extensive) disease, and 10% had obstructive disease. During the trial period, 193 persons died and there were 71 myocardial infarctions. The number of events in each of the 4 groups were translated into adjusted relative risk scores using the non-obstructive-nonextensive group as the baseline (adjusted relative risk = 1). 

As expected the obstructive-extensive group had the highest adjusted relative risk of 12.48 (CI, 5.50 to 28.12), and the obstructive-nonextensive was next with a relative risk of 8.28. The combined relative risk of the two extensive groups was 7.65. For the secondary endpoint of death or myocardial infarction the adjusted relative risk for the obstructive-extensive group was smaller (most likely due to non-cardiac deaths) but still significantly above 1 with a value of 3.15 (CI, 2.05 to 4.83).

The two pieces of information that stand out are the high risk for those with subclinical atherosclerosis (almost 10-fold), and the significant portion of cohort population who had subclinical atherosclerosis (almost 50%). Remember these are people who did not exhibit any signs of CAD. Put simply, the vast majority of the heart attacks occurred in the subclinical group.

These results raise the question whether it is possible to perform large-scale screening for subclinical CAD. CCTA is not suited for high-throughput screening. However, it may be possible to develop a triage pipeline in which machine learning could be used to predict subclinical atherosclerosis based on atherosclerosis risk factors (e.g. obesity, diabetes, high LDL). Prediction positives would then be screened with calcium scoring cardiac CT, and those with high calcium scores could be diagnosed with CCTA.

Figure 1. Example of image from coronary computed tomography angiography (CCTA) showing atherosclerotic blockage of coronary artery (pointed to by yellow arrows) (link).