Heterogeneity in the Association Between the Presence of Coronary Artery Calcium and Cardiovascular Events: A Machine-Learning Approach in the MESA Study

Abstract

Background: Coronary artery calcium (CAC) has been widely recognized as an important predictor of cardiovascular disease (CVD). Given the finite resources, it is important to identify individuals who would receive the most benefit from detecting positive CAC by screening. However, the evidence is limited as to whether the burden of positive CAC on CVD differs by multidimensional individual characteristics. We sought to investigate the heterogeneity in the association between positive CAC and incident CVD.

Methods: This cohort study included adults from MESA (Multi-Ethnic Study of Atherosclerosis) ages ≥45 years and free of cardiovascular disease. After propensity score matching in a 1:1 ratio, we applied a machine learning causal forest model to (1) evaluate the heterogeneity in the association between positive CAC and incident CVD, and (2) predict the increase in CVD risk at 10-years when CAC>0 (versus CAC=0) at the individual level. We then compared the estimated increase in CVD risk when CAC>0 to the absolute 10-year atherosclerotic CVD (ASCVD) risk calculated by the 2013 American College of Cardiology/American Heart Association pooled cohort equations.

Results: Across 3328 adults in our propensity score-matched analysis, our causal forest model showed the heterogeneity in the association between CAC>0 and incident CVD. We found a dose-response relationship of the estimated increase in CVD risk when CAC>0 with higher 10-year ASCVD risk. Almost all individuals (2293 of 2428 [94.4%]) with borderline risk of ASCVD or higher showed ≥2.5% increase in CVD risk when CAC>0. Even among 900 adults with low ASCVD risk, 689 (69.2%) showed ≥2.5% increase in CVD risk when CAC>0; these individuals were more likely to be male, Hispanic, and have unfavorable CVD risk factors than others.

Conclusions: The expected increases in CVD risk when CAC>0 were heterogeneous across individuals. Moreover, nearly 70% of people with low ASCVD risk showed a large increase in CVD risk when CAC>0, highlighting the need for CAC screening among such low-risk individuals. Future studies are needed to assess whether targeting individuals for CAC measurements based on not only the absolute ASCVD risk but also the expected increase in CVD risk when CAC>0 improves cardiovascular outcomes.

Keywords: atherosclerosis; cardiovascular disease; machine learning; precision medicine; propensity score.

Figure 1.

Flow of study sample selection. Sensitivity analysis was conducted using all 5594 samples before propensity score matching. CAC indicates coronary artery calcium; and MESA, Multi-Ethnic Study of Atherosclerosis.

Figure 2.

Association between the 10-year ASCVD risk and the estimated increase in the risk of cardiovascular events when CAC>0 compared to CAC=0. X-axis shows the 10-year ASCVD risk calculated by the 2013 ACC/AHA pooled cohort equations. Y-axis showed the estimated increase in the risk of cardiovascular events when CAC>0 (calculated by the causal forest model). Spearman correlation coefficient and Pearson correlation coefficient between the 10-year ASCVD risk and the estimated increase in the risk of cardiovascular events when CAC>0 were 0.60 (P value <0.001) and 0.61 (P value <0.001), respectively. ASCVD indicates atherosclerotic cardiovascular disease; CAC, calcium coronary artery calcium; and CVD, cardiovascular disease.

Figure 3.

Association between the traditional cardiovascular risk factors and the estimated increase in the risk of cardiovascular events when CAC>0 compared to CAC=0. X-axis shows 4 major traditional cardiovascular risk factors (age, BMI, systolic blood pressure, and LDL cholesterol levels). The vertical dashed line (blue) for each variable corresponds to age (65 years), BMI (30 kg/m²), systolic blood pressure (130 mmHg), and LDL cholesterol levels (100 mg/dL). Y-axis shows the estimated increase in the risk of cardiovascular events when CAC>0 compared to CAC=0 (calculated by the causal forest model). Spearman correlation coefficients (ρ) and Pearson correlation coefficient (r) between each traditional cardiovascular risk factor and the estimated increase in the risk of cardiovascular events when CAC>0 (vs CAC=0) and their P values are shown at the upper right of each plot. BMI indicates body mass index; CAC, coronary artery calcium; CVD, cardiovascular disease; and LDL, low-density lipoprotein.