The cardiovascular risk factors associated with the plaque pattern on coronary computed tomographic angiography in subjects for health check-up

Background Although it is known that coronary computed tomographic angiography (CCTA) offers highly negative predictive value to exclude obstructive coronary lesions, the plaque pattern on CCTA has not been fully understood. The purpose of this study was to explore the difference of the plaque patterns on CCTA and to assess the cardiovascular risks in healthy subjects. Methods A total of 3914 subjects (mean age: 55 ± 10 years, M : F = 2649 : 1265) who underwent CCTA for health check-up between January 2009 and December 2012 were enrolled. According to coronary artery calcium score (CACS) and plaque pattern on CCTA, subjects were categorized into four groups (group 1: normal; group 2: “non-calcified” plaque; group 3: “calcified” plaque; group 4: mixed plaque). We analyzed cardiovascular risks and Framingham risk score (FRS) among the groups. Results The incidence of each group was group 1 in 55.0% (2152/3914), group 2 in 5.1% (200/3914), group 3 in 8.2% (319/3914), and group 4 in 7.2% (280/3914), respectively. There was no difference of FRS among the groups (6.4 ± 6.4%; 6.5 ± 4.6%; 8.2 ± 5.8%; 7.7 ± 5.7% p = 0.086). In multivariate analysis, HbA1c (OR = 2.285; 95%CI = 1.029 - 5.071; p = 0.042) in group 2; age (OR = 1.115; 95%CI = 1.034 - 1.202; p = 0.005) and smoking status (OR = 3.386; 95%CI = 1.124 - 10.202; p = 0.030) in group 3; and age (OR = 1.054; 95%CI = 1.011 - 1.099; p = 0.014) and hypertension (OR = 3.087; 95%CI = 1.536 - 6.202; p = 0.001) in group 4 were independent factors. Conclusions Our data suggest that more individualized therapy for reduction of cardiovascular risks associated with plaque pattern on CCTA could be considered in healthy subjects.


Background
Now it is well known that coronary artery disease (CAD) is the leading mortality cause due to sudden death or myocardial infarction in healthy subjects and a major public health problem in the world [1]. Therefore, it is potentially important that the endeavor to identify subclinical coronary atherosclerosis can lead to reduction in the rate of cardiovascular events. Although Framingham risk score (FRS), 10-year risk for coronary heart disease (CHD) represents a very useful diagnostic tool, there are some limitations for the estimation of the risk of cardiovascular morbidity and mortality such as overestimation in a low-risk population or underestimation the risk in a high-risk population [2].
Conventional invasive coronary angiography has been the standard method for diagnosing CAD [3,4]. In addition, the recent multidetector coronary computed tomographic angiography (CCTA) using dual-source computed tomography (DSCT) was introduced as a useful, non-invasive tool for the evaluation of coronary atherosclerosis and the prediction of cardiovascular morbidity [5,6]. However, the screening of coronary artery calcium score (CACS) using CCTA should not be recommended in asymptomatic individuals with low-risk (0 to 1 risk factor or a 10-year risk <10%) or high-risk (CHD risk equivalents or a 10-year risk >20%) according to Framingham criteria, but considered useful in patients with intermediate-risk (more than two risk factors or a 10-year risk 10-20%) [7,8]. Nevertheless, CACS has not only an excellent negative predictive value to exclude the presence of significant CAD [9,10], but provides also more important prognostic information for cardiovascular risk stratification than the biomarker, such as C-reactive protein [11,12].
Until now, the usefulness of CCTA for estimating and predicting of subclinical coronary atherosclerosis in healthy subjects has not been well established, although past studies mostly included symptomatic patients with significant or obstructive CAD including acute coronary syndrome [13,14]. The purpose of this study was to explore the difference of the plaque patterns on CCTA such as "non-calcified" or "calcified" plaque and to assess the cardiovascular risks in healthy subjects.

Study population
Between January 2009 and December 2012, a total of 3914 subjects (mean age: 55 ± 10 years, M : F = 2649 : 1265) who underwent CCTA for health check-up at the Heath Promotion Center of Seoul St. Mary's Hospital (The Catholic University, Seoul, Korea), were enrolled.
This study was approved by the Institutional Review Committee of St Mary's Hospital, the Catholic University of Korea and conducted in agreement with the Declaration of Helsinki. The participants were informed of the investigative nature of the study and written informed consent was obtained before enrollment (XC11RIMI0091S).

Biochemical parameters
Blood samples were taken at health check-up day. The lipid profile, including total cholesterol, triglyceride, high-density lipoprotein cholesterol (HDLcholesterol), and low-density lipoprotein cholesterol (LDL-cholesterol) levels were measured using enzymatic method by an automatic analyzer (7600-210; Hitachi Medical Corp., Tokyo, Japan). HbA1c was measured by G8 HbA1c analyzer (Tosoh Corporation, Tokyo, Japan). The biochemistry including fasting blood glucose and C-reactive protein from blood samples were measured by a biochemistry analyzer (7600-210; Hitachi Medical Corp., Tokyo, Japan).
All post-processing examinations were performed using retrospective electrocardiography (ECG)-gating. Scans were analyzed by consensus of two observers (YS Choi and JI Jung) with more than 3 years experience in cardiac CT imaging. CACS for vascular calcification were analyzed using a software (syngo.CT CaScoring; Siemens Healthcare; Forcheim, Germany).
For defining the quantity of coronary calcium, Agatston score, standard parameter, was used as the product of the area of calcification per coronary tomographic segment and a factor rated 1 through 4 dictated by the maximal calcium x-ray density in that segment, as described elsewhere [20]. The sum of all lesion scores for each major coronary artery including left main (LM), left anterior descending artery (LAD), left circumflex artery (LCX), and right coronary artery (RCA) was used to generate the total CACS.

Statistical analysis
All data are expressed as mean ± standard deviation (SD) for continuous variables and as a frequency percentage for categorical variables and statistical analysis was performed using the SAS statistical software version 9.2 (SAS Institute, Cary, NC, USA). Analysis among the groups for continuous variables was performed using ANOVA test and analysis of categorical data was performed using the Tukey's b-test as a post-hoc t-test. Analysis between the two groups was performed using unpaired t-test for continuous variables and chi-squared test for categorical data. The clinical variables related to FRS were assessed using Pearson correlation coefficient. To identify independent factors associated with the plaque pattern on CCTA, we used multiple logistic regression analysis and calculated odds ratios (OR) and 95% confidence intervals (95% CI). All statistical tests were 2-tailed and p <0.05 was considered statistically significant.

Clinical characteristics
The mean age of a total of 3914 participants was 55 ± 10 years; there were more male (n = 2649) than female (n = 1265) subjects. A total of 977 of 1265 (78.8%) female subjects were postmenopausal. The prevalence of hypertension, diabetes mellitus, dyslipidemia, familial history of cardiovascular risk and current smoking state were 32.5, 12.8, 16.0, 84.2 and 20.3%, respectively.

Discussion
The accumulation of atherosclerotic plaque without significant coronary stenosis happens over many years prior to acute cardiovascular events, including myocardial infarction or sudden cardiac death. Furthermore, CACS on CCTA, as a recent diagnostic tool, has been shown to be helpful in patients with low-and intermediate-risk who presents with atypical cardiac symptoms [7,8]. At the same time, effective strategies for earlier identification of subclinical coronary atherosclerosis requires in healthy subjects. In the present study, unlike previous research using CCTA in symptomatic patients with significant or obstructive CAD [13,14], we focused on the different cardiovascular risk factors associated with the plaque pattern on CCTA in healthy subjects. As a result, in this single-center, cross-sectional study of healthy subjects comparing CCTA with FRS, our data revealed that CCTA is reliable and effective for the estimation of the different cardiovascular risk factors associated with the plaque pattern on CCTA in healthy subjects. However, to predict the presence of subclinical coronary atherosclerosis, further investigations are required prospective study in larger populations via multicenter trials.
Association with FRS and the plaque pattern on CCTA FRS or CACS on CCTA for cardiovascular risk stratification is a useful tool. However, these tools alone may insufficient to identify subclinical coronary atherosclerosis in some part of the population. In addition, the combination of FRS and CACS may provide more accurate estimation of the risk of cardiovascular events [24]. In our study, although our data showed a weak correlation between FRS and traditional cardiovascular risk factors such as older age, obesity indicators, blood pressure and plasma lipid profile in subjects with "normal" coronary arteries on CCTA, there was no difference of FRS among the groups classified according to the plaque pattern on CCTA. As described in several previous published studies [7,8], these findings are particularly consistent with the facts that CACS on CCTA in clinical application can provide valuable prognostic evaluation and serve as an important tool for cardiovascular risk stratification of asymptomatic or healthy individuals, although CACS on CCTA should not be recommended as a tool to diagnose significant obstructive CAD in symptomatic patients.

Difference of the plaque pattern on CCTA
In analysis for the cardiovascular risk stratification of the development and progression of subclinical coronary atherosclerosis and the difference of the plaque pattern on CCTA such as "non-calcified" or "calcified" plaque, North et al. demonstrated the role of smoking status in the pathogenesis of "calcified" coronary plaque, similar  [25]. On the other hand, from the ROMI-CAT trial as a prospective, observational cohort study, Lehman et al. reported that smoking were independently associated with coronary atherosclerotic plaque burden progression on CCTA in patients with acute chest pain over 2 years, although rate of progression is dependent on plaque composition and may be higher for "non-calcified" when compared to "calcified" plaque [26]. In addition, in patients referred to the emergency department with chest pain, Yoon et al. reported that the patients ≥50% CAD of "non-calcified" plaque on CCTA were younger and had a higher prevalence of smoking [27].
Otherwise, the present study may have an important or interesting clinical implication in the association between HbA1c, as a key marker of diabetes control and "non-calcified" plaque from this observation. Interestingly, Hausleiter et al. demonstrated the role of "non-calcified" plaque, characterized by significantly higher total cholesterol, LDL-C, and C-reactive protein levels in patients with acute coronary syndrome [28]. Furthermore, more recent studies suggested that effective prevention has to be focused on the type of plaque composition [29][30][31]. Nicholls et al. demonstrated that "calcified" plaques are more resistant to undergoing changes in size in response to systemic interventions targeting atherosclerotic risk factors. On the contrary, "non-calcified" plaque might have a higher tendency to regress in response to established medical therapies [29]. Several studies also suggested that "mixed" plaque could convey a higher coronary risk including of acute coronary syndromes [29,30]. Thus, to overcome these various issues and problems for identifying subclinical coronary atherosclerosis, we need to conduct further research in a larger population including of ethnic differences.
There are several limitations that our study includes the relatively small sample size and possibility of referral bias from one center trial. First, the proportion of sample group was lower in group 2-4 than group 1 in present study. Second, there is a lack of knowledge about the analysis including statin therapy and differential hormonal effects based on gender. Last, our investigators suggest that prospective studies via large multiethnic populations and long-term follow up are required to determine the potential value of identifying the development and progression of subclinical coronary atherosclerosis and to predict the prognosis of CHD.

Conclusions
Although there was no difference of FRS among the groups classified according to the plaque pattern on CCTA, our data suggest that more individualized therapy for reduction of cardiovascular risks could be considered in healthy subjects.