Silent CAD present in 4 out of 10 Swedish individuals
AHA 2020 Prevalence of CAD in a Swedish population was 4 out of 10, using CCTA data from the SCAPIS cohort. A prediction model was developed to identify those with widespread atherosclerosis.
Coronary Artery Disease In The General Population- Results From The Swedish Cardiopulmonary Bioimage Study (SCAPIS)News - Nov. 14, 2020
Presented during the AHA Scientific Sessions 2020 by Göran Bergström (University of Gothenburg, Gothenburg, Sweden)
Introduction and methods
The aim of this study was to determine how many of middle-aged adults have silent coronary artery disease. In addition, a personalized screening strategy was developed to identify individuals at high risk of having wide-spread but still silent CAD. Data of SCAPIS (Swedish Cardiopulmonary Bioimage Study) was used, including data of individuals aged 50-64 years from the general population. 25,182 Individuals with no previous coronary heart disease were imaged using CCTA (coronary computed tomography angiography). Trained readers described the distribution, severity and the degree of CAD.
A prediction model for wide spread atherosclerosis was based on machine learning (AI). Data of participants with widely distributed atherosclerosis was used (those with ≥4 involved segments (SIS), ~13% of the study population. 120 Risk factors were tested. Two models were generated, one based on data collected at home (such as weight, circumference), and one based on clinical data.
The SCAPIS project is a collaborative effort of 6 universities in Sweden. It is a population study of coronary and pulmonary disease. 30.154 Individuals have been included and examined with advanced imaging techniques and by a comprehensive examination protocol.
Main results
- 4 out of 10 (42%) had any form of CAD.
- 1 out of 20 (5%) had severe silent CAD (clinically significant CAD – atherosclerosis that obstructs >50% of the blood flow in the vessel).
- Prevalence of CAD increased with age, sex and traditional risk factors (such as cholesterol, blood pressure).
- The home model was equally effective as the clinical model in identifying who had widespread CAD (around two thirds of participants with widespread CAD).
Conclusion
In a middle-aged population in Sweden, 4 out of 10 individuals had CAD and 1 in 20 had severe CAD. Using data collected at home, a prediction model could identify individuals with a high risk of having widespread CAD. The authors conclude that these findings may lead to a screening strategy for early detection of CAD.
Discussion
The discussant Pamela S. Douglas , MD (Durham, NC, USA) asked the question what clinically relevant coronary atherosclerosis is. The tiered testing proposal in SCAPIS relied on an anatomic risk model predicting ≥4 involved segments (SIS). Douglas said that the rationale for choosing the cut point of ≥4 SIS in unclear as clinical risk and/or mortality is higher in all forms of non-obstructive CAD, starting at 1 vessel involvement, compared to no CAD, as demonstrated in the CONFIRM study. In addition, other CTA findings, such as plaque characteristics, are equally prognostic.
Then she asked who are at highest risk for events. The clinical risk algorithm in SCAPIS failed to detect one third of patients with ≥4 SIS, and may not identify those with the greatest burden. Although those with stenosis (a tiny minority) have higher relative risk of MACE, most MACE events occur in large population with non-obstructive CAD. Douglas therefore stated: “Effective preventive strategies likely need to start with detection and treatment of patients with even minimal plaque”.
- Our reporting is based on the information provided during AHA Scientific Sessions 2020 -