By Michael H. Crawford, MD, Editor

SYNOPSIS: Coronary artery CT angiography in asymptomatic, middle-aged subjects without known coronary artery disease showed coronary atherosclerosis is common but mostly mild and appears in women after a 10-year delay.

SOURCE: Bergström G, Persson M, Adiels M, et al. Prevalence of subclinical coronary artery atherosclerosis in the general population. Circulation 2021;144:916-929.

The prevalence of coronary artery atherosclerosis (CAA) in the general population without suggestive symptoms or apparent disease is unknown. Still, determining this status is important for prevention. Coronary artery calcium (CAC) scores often are used to detect CAA, but for symptomatic patients, coronary artery CT angiography (CCTA) could provide a more accurate assessment of CAA. The authors of the Swedish CArdioPulmonary BioImage Study (SCAPIS) sought to determine if this is true for an asymptomatic general population without known coronary artery disease (CAD).

Between 2013 and 2018, Bergström et al randomly recruited more than 30,000 participants ages 50-64 years. Subjects were excluded if there was known CAD or if CT exams were technically insufficient. Researchers obtained Agatston CAC scores and CCTAs. The main challenge to CCTA is defining the level of obstruction of calcified plaques. When investigators encountered so-called calcium blooming artifact, the lesion was graded as 1% to 49% stenosis, since calcium blooming tends to overestimate the severity of obstruction. The authors assessed each patient’s risk factor burden and calculated various risk scores. After applying the exclusion criteria, 25,182 subjects were included in the initial analysis: 49% men, mean age 57 years.

Researchers found detectable CAA in 42% of subjects, and 5% had a significant stenosis (> 50%). Severe CAA of the left main, proximal left anterior descending (LAD), or three-vessel, disease was less common (2%). The proximal LAD was the most commonly affected vessel. The delay in disease onset between men and women was about 10 years. The prevalence of CAA increased with higher risk factor scores. Subjects with only non-calcified plaques were unusual (2.4%). A mix of calcified and non-calcified plaques occurred in 8%. CAC-positive subjects followed a similar pattern, with 40% of the subjects testing positive and the severity paralleling the CAC score.

Among subjects with a CAC score > 400, 46% showed significant stenoses. In those with a CAC of 0, 5.5% exhibited CAA and 0.4% significant stenoses. In those with CAC scores of 0 and an intermediate risk profile, 9% showed CAA. A few patients with low CAC scores (1-10) did not exhibit CAA on CCTA. The authors concluded that in this random sample of the general population of Sweden, silent CAD by CCTA was common. Those with high CAC scores carried a high probability of significant stenoses. A CAC score of 0 did not exclude CAA, especially if subjects were at higher clinical risk.


This is the first article published about SCAPIS, a nationwide, randomized, population-based cohort where half the people asked to participate enrolled. Although one cannot exclude some selection bias, their characteristics were close to those of the general population and similar to other developed countries with largely white populations. Thus, the findings likely are widely applicable. SCAPIS consisted of an asymptomatic, middle-aged population, with the expected slight predominance of women and no history of CAD.

Overall, there were three major findings. First, 42% of subjects showed evidence of silent CAD. Significant stenoses (> 50%) were found in 5% of subjects (2% of cases were severe). Second, there was a 10-year delay in the onset and severity of CAA in women compared to men. Also, the distribution of CAA was identical between the sexes after accounting for the delay in women. Third, CAA detected by CCTA was associated with risk factor scores and the CAC score, but there were significant subgroups where CAC and risk scores were inaccurate. For example, among those with low risk-based scores, one-third of men and one-quarter of women had CAA. Likewise, in those with CAC scores of 0 to 10, 22% of women and 30% of men had CAA. Thus, using risk equations or CAC to determine whom among those with intermediate risk should start statin therapy is problematic, especially since the benefit of statins in such patients is uncertain.

There were some limitations to this investigation. This was a cross-sectional study, so there are no data on the progression of disease or outcomes. Also, those interpreting the CT scans saw both the CAC scores and the CCTA results, which could have biased their interpretations. Determining percent obstruction in heavily calcified lesions is problematic, too. Their assumption that heavy calcification overestimates the degree of obstruction may not be justified in all cases. The big issue with a study like this is whether CCTA is worth the expense for screening populations. The answer to this question may be forthcoming, as trials to assess the value of risk formulae vs. CCTA on outcomes are in progress.