Comprehensive Prediction of Subclinical Coronary Atherosclerosis in Subjects Without Traditional Cardiovascular Risk Factors

Abstract

It is not uncommon for asymptomatic individuals without identified cardiovascular risk factors to present with atherosclerosis-related adverse events. We aimed to evaluate the predictors of subclinical coronary atherosclerosis in individuals without traditional cardiovascular risk factors. We analyzed 2,061 individuals without identified cardiovascular risk factors who voluntarily underwent coronary computed tomography angiography as part of a general health examination. Subclinical atherosclerosis was defined as the presence of any coronary plaque. Of 2,061 individuals, subclinical atherosclerosis was observed in 337 individuals (16.4%). Clinical variables, such as age, gender, body mass index (BMI), systolic blood pressure (SBP), low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein cholesterol (HDL-C), were significantly associated with subclinical coronary atherosclerosis. The participants were randomly split into train and validation data sets. In the train set, a prediction model using 6 variables with optimal cutoffs (age >53 years for men and >55 years for women, gender, BMI >22 kg/m², SBP >120 mm Hg, HDL-C <55 mg/100 ml, and LDL-C >130 mg/100 ml) was derived (area under the curve 0.780, 95% confidence interval 0.751 to 0.809, goodness-of-fit p = 0.693). In the validation set, this model performed well (area under the curve 0.792, 95% confidence interval 0.726 to 0.858, goodness-of-fit p = 0.073). In conclusion, together with nonmodifiable risk factors, such as age and gender, modifiable variables, such as BMI, SBP, LDL-C, and HDL-C, were shown to be associated with subclinical coronary atherosclerosis, even at currently acceptable levels. These results suggest that stricter control of BMI, blood pressure, and cholesterol might be helpful in the primary prevention of future coronary events.