The International Journal of Cardiovascular Imaging

, Volume 28, Supplement 2, pp 93–100

Subclinical coronary atherosclerosis in young adults: prevalence, characteristics, predictors with coronary computed tomography angiography

Authors

  • Kwang Nam Jin
    • Division of Cardiovascular Imaging, Department of Radiology, Bundang HospitalSeoul National University
    • Division of Cardiothoracic Imaging, Department of Radiology, Boramae Medical Center, Seoul MetropolitanSeoul National University
  • Eun Ju Chun
    • Division of Cardiovascular Imaging, Department of Radiology, Bundang HospitalSeoul National University
  • Chang-Hoon Lee
    • Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Lung Institute of Medical Research Center, College of Medicine, Seoul National University HospitalSeoul National University
  • Jeong A. Kim
    • Division of Cardiovascular Imaging, Department of Radiology, Bundang HospitalSeoul National University
  • Min Su Lee
    • Division of Cardiovascular Imaging, Department of Radiology, Bundang HospitalSeoul National University
    • Division of Cardiovascular Imaging, Department of Radiology, Bundang HospitalSeoul National University
Article

DOI: 10.1007/s10554-012-0143-0

Cite this article as:
Jin, K.N., Chun, E.J., Lee, C. et al. Int J Cardiovasc Imaging (2012) 28: 93. doi:10.1007/s10554-012-0143-0

Abstract

The purpose of this study was to assess the characteristics and predictors of subclinical coronary atherosclerosis in young adults. The study also evaluated predictors of cardiac events. We retrospectively enrolled 914 self-referred asymptomatic subjects under the age of 45 (552 men, 362 women) who had undergone both coronary CT angiography (CCTA) and coronary artery calcium scoring. Two radiologists analyzed plaque composition and degree of stenosis. For all subjects, we evaluated clinical risk factors and investigated cardiac events. Subclinical coronary atherosclerosis was found in 86 subjects (9.4 %). Among them, 64 subjects (74.4 %) had a single coronary artery plaque. On analysis of individual segments, the most common type of plaque was non-calcified plaque (NCP) (58 %), which was found in 63 subjects (6.9 %). Significant coronary artery stenosis was found in 9 subjects (0.1 %). Predictors of subclinical coronary atherosclerosis were age, male gender, hypertension, and diabetes mellitus. Male gender, diabetes mellitus, and amount of smoking were independent predictors of NCP. High low density lipoprotein cholesterol and calcium scores were also significant predictors of stenosis. Myocardial infarction developed in 1 subject, unstable angina in 2, stable angina in 1, and death in 1 (2.45 cardiac events per 1,000 person-years of follow-up). Multivariate analysis revealed hazard ratios of 2.2 for subclinical coronary atherosclerosis, 49.17 for NCP, and 105.58 for significant stenosis. The prevalence of subclinical coronary atherosclerosis in asymptomatic young adults is not negligible. CCTA has the potential to enhance risk stratification and prediction for coronary artery disease in asymptomatic young adults.

Keywords

Coronary arteryAtherosclerosisYoung adultsComputed tomography angiographyRisk factors

Introduction

Although the incidence of coronary artery disease (CAD) is low in individuals younger than 45 years old, early detection of CAD is important because afflicted individuals in these groups are more likely to experience catastrophic and unexpected events than older individuals [1, 2]. Symptomatic CAD at a young age is relatively uncommon, accounting for only about 3 % of all CAD cases [3]. However, the true prevalence of CAD, especially in younger, asymptomatic patients, has been grossly underestimated [4]. Indeed, intravascular ultrasound-based investigation in a cohort of transplanted hearts showed that atherosclerotic plaque was found in 28 % of asymptomatic donors younger than 30 years old [4]. In addition, the prevalence of risk factors for CAD, such as smoking, obesity, and lack of physical activity, has been increasing in young adults and children, and is predicted to increase the burden of CAD in the near future. Therefore, early detection of subclinical coronary atherosclerosis in asymptomatic young adults, which has the potential to cause significant cardiac events, is of immediate concern.

The most widely used method for risk stratification of CAD is the Framingham risk score (FRS). However it is known to underestimate the risk of CAD, especially in women and younger individuals [57]. Recently, coronary artery calcium scoring (CACS) as a marker of subclinical coronary atherosclerosis has been reported to provide incremental prognostic information, compared to the FRS for evaluation of CAD [8, 9]. However, it also has some limitations; it cannot evaluate the degree of stenosis or plaque characteristics, such as non-calcified plaques (NCP) [10].

With the recent advent of multidetector-row computed tomography (MDCT), coronary computed tomography angiography (CCTA) has arisen as a useful, non-invasive imaging tool that reveals either the degree of stenosis or the kind of plaque. CCTA has not been widely studied in subclinical coronary atherosclerosis in young adults any potential role of CCTA in screening these asymptomatic individuals for risk stratification has yet to be determined. Therefore, this study was designed to assess the plaque characteristics of subclinical coronary atherosclerosis in young adults and to further evaluate predictors of subclinical coronary atherosclerosis and cardiac events.

Materials and methods

Study population

We retrospectively selected asymptomatic adults under 45 years old without known cardiovascular disease who had undergone CCTA. The purpose of CCTA was as a general medical check-up. Among the registry of 15,355 subjects who underwent CCTA between January 2006 and December 2008 at our institution, we retrospectively enrolled 2,129 subjects under age 45 (1,347 men, 782 women; mean age 40 years). We excluded 1,159 subjects who had chest pain (n = 735), prior percutaneous coronary intervention (n = 7), previous coronary artery bypass graft (n = 12), a history of aortic valve or ascending aorta replacement due to aortic valvular disease or aortic aneurysm (n = 21), aortic dissection (n = 1), pericardial mass (n = 1), Kawasaki disease (n = 2), coronary artery fistula (n = 2), trauma (n = 1), Tetralogy of Fallot (n = 2), or any insufficient clinical data (n = 375). All individuals were asked whether they had chest pain or equivalent symptoms, according to the Rose angina questionnaire [11]. Asymptomatic individuals without known CAD were determined on the basis of patient history, which included the questionnaire and prior work-up that had been conducted by a physician.

We also excluded CCTA images with poor quality and a low degree of confidence for stenosis and plaque characterization (n = 56). Finally, a total of 914 asymptomatic subjects under the age of 45 who had undergone CCTA were enrolled. This study was approved by the Institutional Review Board of Seoul National University Bundang Hospital, with the requirement for patient consent waived.

Risk factor assessment and stratification

A medical history of myocardial infarction (MI), previous surgical operation, angina, hypertension, diabetes mellitus, current medication profile, and smoking was systematically acquired. Body weight, height, and blood pressure were also measured during their visit for a CT scan. Hypertension was defined as: a self-reported history of hypertension, the use of antihypertensive medications, or blood pressure ≥140/90 mmHg. Total cholesterol, triglyceride, high-density lipoprotein, low-density lipoprotein, fasting plasma glucose, blood urea nitrogen, and serum creatinine levels were all measured on the same day as the date of CT examination after a minimum 12-h fasting period. Levels were defined as elevated if total cholesterol was ≥240 mg/dl, triglycerides were ≥200 mg/dl, and LDL was ≥160 mg/dl. Low high density lipoprotein (HDL) cholesterol was defined as a level <40 mg/dl. Diabetes mellitus was defined as a self-reported history of diabetes, anti-diabetic treatment, or a fasting plasma glucose level of ≥126 mg/dl. We calculated the mean value for total smoking amount from the study population, including non-smokers. Daily smoking habits of current smokers were obtained as the average number of cigarettes smoked daily during the month before the interview, via a questionnaire. Past smokers were defined as those who had abstained from smoking for a period of >3 months at the time of the examination. The total number of pack-years was also calculated from the total number of years spent smoking, multiplied by the number of packs smoked per day. For the purpose of this analysis, total pack-years were divided into quartiles.

Framingham risk scores were calculated per the National Cholesterol Education Program (NCEP) guidelines [12]. All subjects were assigned to 3 different risk groups according to the revised NCEP-ATP III (National Cholesterol Education Program Adult Treatment Panel III) guidelines: high-risk (coronary heart disease [CHD], CHD risk equivalents, or 10-year risk >20 %), intermediate-risk (more than 2 risk factors and 10-year risk ≤20 %), and low-risk group (0–1 risk factor) [12].

Coronary CT angiography acquisition

All subjects underwent a CT scan with the use of a 64-slice MDCT scanner (Brilliance 64, Philips Medical Systems, Best, The Netherlands). Patients with a heart rate ≥70 beats/min received intravenous esmolol, 10–30 mg (Jeil Pharm. Co., Ltd., Seoul, Korea). The scan parameters were as follows: section collimation 64 × 0.625 mm, rotation time 420-ms, 120-kV tube voltage, and 800-mA tube current. For all subjects, both calcium scoring scans and retrospective ECG-gated helical CT scans were performed. A bolus of 80-ml of iomeprol (Iomeron 400, Bracco, Milan, Italy) was administered with an injection rate of 4–5 ml/sec, which was followed by a 50-ml saline chaser. Data, at the mid-diastolic phase (75 % of R–R interval) of the cardiac cycle, were used to reconstruct the initial image data. Additional reconstructions were performed if motion artifacts were present. The mean radiation exposure of CTA was 11.3 ± 0.82 mSv for men and 10.2 ± 0.72 mSv for women.

Image analysis

Two experienced radiologists (S.I.C. 8 years; E.J.C. 6 years for cardiac MDCT imaging), who were blinded to the clinical information, evaluated all scans with a 3-dimensional workstation (Brilliance, Philips Medical Systems, Best, The Netherlands). Consensus interpretation was performed to obtain a final MDCT diagnosis. Multiplanar reconstruction and maximum intensity projection images of short-axis, 2-chamber, and 4-chamber views, were used for the evaluation of the stenosis and plaque. The radiologists analyzed the characteristics of the stenosis and plaque per-segment, according to the modified American Heart Association classification [13]. For the evaluation of the degree of stenosis, the coronary lumen was semi-automatically traced at the maximal stenotic site and was compared with the mean value of a proximal and distal reference site. The severity of the stenosis diameter was graded as normal (0–24 %), mild (25–49 %), moderate (50–74 %), and severe (≥75 %) narrowing [14]. A stenosis of more than 50 % and 75 % were defined as significant and severe, respectively. Plaques were defined as structures that were 1 mm2 and within or adjacent to the vessel lumen, and which could be clearly distinguished from the lumen and the surrounding pericardial tissue. Plaques with more than 50 % of the plaque area occupied by calcified tissue (density ≥130 HU in native scans) were classified as calcified, plaques with 50 % calcium were classified as mixed, and plaques without any calcification were classified as non-calcified lesions [15]. CACS were measured with the scoring system previously described by Agatston et al. [16, 17]. Scores were categorized into four groups: 0; 1–100; 100.1–300; and ≥300.1 [9]. Subclinical coronary atherosclerosis was defined as the presence of any plaques on a coronary CT angiography in asymptomatic patients.

Mid-term follow-up

We assessed cardiovascular events using a series of questions regarding intervening hospital admissions and reviewed medical records for hospitalizations related to a complaint of chest discomfort, dyspnea, vascular or cardiac problems, or any major surgeries. A follow-up attempt was considered successful when surviving participants either returned to the clinic or completed a telephone interview. All relevant medical records were obtained. We calculated the percent of success and mean duration of follow-up. The study endpoint was cardiac events, which included cardiac death, non-fatal MI, unstable angina requiring hospitalization, or revascularization later than 90 days after CCTA. The time of coronary CT angiography and the time that cardiac events developed were recorded to estimate person-years of the follow-up period.

Statistical analysis

Continuous variables are expressed as means ± standard deviations and categorical variables as frequencies or percentages. A P value < 0.05 was considered statistical significance, except when testing for interactions, where P values up to 0.10 were accepted. Differences between continuous variables were analyzed by the unpaired Student t test. Categorical variables were analyzed by the Chi-square test or Fisher exact test, as appropriate. Multiple logistic regression analysis was employed to identify independent predictors of the presence of any plaque, significant stenosis, and any NCP on CCTA. Multivariable Cox proportional hazard regression analysis was performed to estimate the impact of plaque, significant stenosis, or clinical risk factors on the development of cardiac events. Proportional hazard assumption was tested with the STATA stphtest command. We also conducted Poisson regression (or exact Poisson regression if there were few events detected) to support the results from Cox regression. Statistical analyses were conducted using SPSS (Release 17; SPSS, Chicago, IL, USA) and STATA version 11.1 (STATA Corp, College Station, TX, USA).

Results

Clinical characteristics of the study population and CAC scores are summarized in Table 1. The study population consisted of 552 male (mean age, 40.4 years) and 362 female subjects (mean age, 40.7 years). There was no significant difference in mean age between men and women (P = 0.16). Diabetes mellitus, hypertension, obesity, and current smoking were significantly more frequent in men than in women. In the male group, current smoking was 61.4 % (n = 339), with a mean total pack-years of 13.2 ± 11.7 (range, 0–60), which was significantly higher than in women (P < 0.001). Female current smokers accounted for 12.7 % of the total females in the study, with mean total pack-years of 1.3 ± 4.2 (range, 0–34). Rates of high total cholesterol, high triglyceride, low HDL cholesterol, and high LDL cholesterol were significantly higher in men. On the basis of classification by the NCEP-ATP III guidelines, most subjects were classified into the low 10-year-risk group (men, 78.8 %; women, 97.5 %). Mean CAC scores were 1.35 ± 8.71 (range, 0–60) in men and 0.15 ± 2.01 (range, 0–34) in women.
Table 1

Summary of clinical characteristics and coronary artery calcium score according to sex

Parameter

Men (n = 552)a

Women (n = 362)a

P value

Age (year)b

 Median

41 (23–44)

41 (27–44)

 

 Mean ± SD

40.4 ± 3.4

40.7 ± 2.9

0.16

Diabetes mellitus

44 (8.0)

9 (2.5)

<0.001

Hypertension

124 (22.5)

22 (6.1)

<0.001

Obesity

49 (8.9)

9 (2.5)

<0.001

Smoking

 Non-smoker

68 (12.3)

284 (78.5)

 

 Ex-smoker

145 (26.3)

32 (8.8)

 

 Current smoker

339 (61.4)

46 (12.7)

<0.001

Smoking amount (pack-years)b

 Median

12.6 (0–60)

0 (0–34)

 

 Mean ± SD

13.2 ± 11.7

1.3 ± 4.2

<0.001

High total cholesterol

80 (14.5)

22 (6.1)

<0.001

High triglyceride

131 (23.7)

8 (2.2)

<0.001

Low HDL cholesterol

78 (14.1)

12 (3.3)

<0.001

High LDL cholesterol

69 (12.5)

10 (2.8)

<0.001

NCEP-ATP III risk stratification

 Low

435 (78.8)

353 (97.5)

 

 Intermediate

71 (12.9)

0

 

 High

46 (8.3)

9 (2.5)

 

CAC score

0

515 (93.3)

359 (99.2)

 

0.1–100

36 (6.5)

3 (0.8)

 

100.1–300

1 (0.2)

0

 

≥300.1

0

0

 

CAC coronary artery calcium, HDL high density lipoprotein, LDL low density lipoprotein, SD standard deviation

NCEP-ATP III The National Cholesterol Education Program Adult Treatment Panel III

aExcept where indicated, data are given as number of patients with percentages in parentheses

bNumbers in parentheses are ranges

Subclinical coronary atherosclerosis was found in 86 subjects (9.4 %) (Table 2). As described in Table 1, 788 patients were classified as the 10-year low risk group, 71 as the intermediate-risk group, and 55 as the high-risk group. And subclinical atherosclerosis was found in 7.2 % (n = 57) among the 10-year low risk group, 18.3 % (n = 13) among the intermediate-risk group, and 29.1 % (n = 16) among the high-risk group. NCP was the most common type of plaque in asymptomatic young adults. NCP, in any coronary artery segment, was found in 46 subjects (5.3 %) in the CACS = zero group and 17 subjects (42.5 %) in the CACS > 0 group. Nine subjects (0.1 %) had significant coronary artery stenosis. All subjects with significant stenosis had single-vessel disease in the left anterior descending artery. Per-segment analysis revealed that NCP was the most common type of plaque causing significant stenosis (n = 8, 88.9 %).
Table 2

Prevalence and characteristics of subclinical coronary atherosclerosis in young adults

Characteristics

Total (n = 914)a

NCEP-ATP III risk stratification

High

Intermediate

Low

Any plaqueb

86 (9.4)

  1 segment

64 (7.0)

8

9

47

  2 segments

19 (2.1)

5

4

10

  3 segments

2 (0.2)

2

0

0

  4 segments

1 (0.1)

1

0

0

No plaque

828 (90.6)

39

58

731

Non-calcified plaquec

63 (6.9)

13

8

42

Significant stenosis

9 (1.0)

5

1

3

 Number

  1 vessel

9

5

1

3

  2 vessels

0

0

0

0

  3 vessels

0

0

0

0

 Plaque composition

  Non-calcified

8

5

0

3

  Mixed

1

0

1

0

  Calcified

0

0

0

0

 Location

    

  Left main

0

0

0

0

  LAD

9

5

1

3

  LCx

0

0

0

0

  Right

0

0

0

0

 CAC score

  0

874

44

63

767

  0.1–100

39

11

8

20

  100.1–300

1

0

0

1

  ≥300.1

0

0

0

0

CAC coronary artery calcium, LAD left anterior descending, LCx left circumflex

NCEP-ATP III The National Cholesterol Education Program Adult Treatment Panel III

aExcept where indicated, data are given as number of patients with percentages given in parentheses

bNumber of subjects with any plaque

cNumber of subjects who have non-calcified plaque with or without any other type of plaque

Significant clinical predictors of subclinical coronary atherosclerosis were age, male gender, hypertension, and diabetes mellitus (Table 3). Age was not an independent predictor for NCP or significant stenosis. Male gender, diabetes mellitus, and amount of smoking were all independent predictors of NCP. Odds ratios for diabetes mellitus were 2.64 for subclinical coronary atherosclerosis (P < 0.001), 3.38 for NCP (P < 0.001), and 11.66 for significant coronary artery stenosis (P < 0.001). Amount of smoking was an independent predictor for NCP, but not for any plaque or significant coronary artery stenosis. Diabetes mellitus, high LDL cholesterol and CACS were considered to be independent predictors of significant coronary artery stenosis.
Table 3

Predictors of subclinical coronary atherosclerosis on coronary CT angiography in young adults

 

Any plaque (n = 86)a

Noncalcified plaque (n = 63)b

Significant stenosis (n = 9)

Univariate

Multivariatec

Univariate

Multivariate

Univariate

Multivariate

P value

OR (95 % CI)

P value

OR (95 % CI)

P value

OR (95 % CI)

P value

OR (95 % CI)

P value

OR (95 % CI)

P value

OR (95 % CI)

Age

0.01

1.11 (1.02–1.2)

0.03

1.10 (1.01–1.2)

0.12

1.07 (0.98–1.18)

0.21

1.06 (0.97–1.16)

0.73

1.04 (0.84–1.29)

Male gender

<0.001

7.28 (3.47–15.27)

<0.001

5.29 (2.39–11.71)

<0.001

1.04 (1.03–1.06)

0.04

1.02 (1–1.04)

0.99

0.6 (0.6–0.63)

Hypertension

<0.001

3.09 (1.9–5.04)

0.03

1.85 (1.08–3.15)

<0.001

2.68 (1.52–4.7)

0.14

1.59 (0.86–2.97)

0.01

6.77 (1.8–25.53)

0.28

2.42 (0.49–12.01)

Diabetes mellitus

<0.001

4.39 (2.3–8.37)

0.01

2.64 (1.3–5.35)

<0.001

5.27 (2.65–10.49)

<0.001

3.38 (1.59–7.16)

<0.001

22.32 (5.81–85.79)

<0.001

11.66 (2.3–59.03)

High total cholesterol

0.01

2.35 (1.33–4.13)

0.49

1.39 (0.55–3.49)

0.02

2.23 (1.16–4.26)

0.75

1.19 (0.41–3.5)

0.30

2.3 (0.47–11.22)

High triglyceride

0.03

1.81 (1.06–3.09)

0.34

0.74 (0.4–1.37)

0.11

1.66 (0.89–3.1)

0.20

0.62 (0.3–1.28)

0.15

2.83 (0.7–11.44)

0.61

0.65 (0.13–3.34)

Low HDL cholesterol

0.09

1.74 (0.92–3.28)

0.54

1.24 (0.62–2.48)

0.10

1.82 (0.89–3.71)

0.50

1.31 (0.6–2.84)

0.39

2.83 (0.7–11.44)

  

High LDL cholesterol

0.01

2.78 (1.52–5.06)

0.39

1.52 (0.58–3.97)

<0.001

2.75 (1.4–5.42)

0.33

1.73 (0.57–5.2)

<0.001

8.85 (2.33–33.67)

0.01

8.13 (1.81–36.44)

Smoking amount

<0.001

1.04 (1.02–1.05)

0.23

1.01 (0.99–1.03)

<0.001

8.38 (3.33–21.11)

<0.001

5.45 (2.04–14.57)

0.08

1.04 (1–1.09)

0.35

1.02 (0.97–1.08)

CAC score

<0.001

3.81 (1.94–7.45)

0.01

2.79 (1.3–6.03)

Obesity

0.30

1.35 (0.59–3.08)

0.37

1.30 (0.5–3.37)

0.55

0.94 (0.92–0.95)

CAC coronary artery calcium, CI confidence interval, HDL high density lipoprotein, LDL low density lipoprotein, OR odds ratio

aData are given as numbers of patients with any plaque in the coronary arteries

bData are given as number of subjects with non-calcified plaque in the coronary arteries with or without any other type of plaque

cMultiple logistic regression analysis

Successful follow-up occurred in 100 % of subjects; mean follow-up duration was 26.8 ± 14.6 months. Total person-years of follow-up were 2,039.2 across 914 subjects. The participating subjects had 5 cardiac events (2.45 per 1,000 person-years of follow-up) (Table 4). The mean interval between the date of CCTA and cardiac events was 36.4 ± 4.5 months. MI developed in 1 subject, unstable angina in 2, late revascularization in 1, and cardiac death in 1. Univariate cox regression analysis revealed that any plaque, NCP, significant stenosis, diabetes mellitus, CACS were significant risk factors for cardiac events. However, age, male gender, hypertension, obesity, smoking amount, high total cholesterol, high triglyceride, low HDL cholesterol, and high LDL cholesterol were not significant risk factors for cardiac events. In multivariate analysis, hazard ratios for cardiac events were 2.2 for subclinical coronary atherosclerosis (P = 0.36), 49.17 for NCP (P = 0.002), and 105.48 for significant stenosis (P = 0.001), respectively (Table 5). Diabetes mellitus and CACS were used as confounders for predicting cardiac events in multivariate analysis.
Table 4

Comparison of rate per 1,000 person-years of cardiac events and hazard ratios according to CT angiographic findings, diabetes mellitus, and CAC score

 

Number of subjects

Person-year of follow-up

Mean person-year of follow-up

Cardiac eventsa

Rate per 1,000 person year

Hazard ratiob (95 % CI)

P value

Any plaque (+)

86

187.38

2.18

4

21.35

3.39 (1.84–6.24)

<0.0001

Any plaque (−)

828

1,851.82

2.24

1

0.54

  

NCP (+)

63

144.80

2.3

4

27.62

67.54 (7.47–609.99)

<0.0001

NCP (−)

851

1,894.40

2.23

1

0.53

  

Significant stenosis (+)

9

20.56

2.28

3

145.92

179.91 (29.58–1,094.48)

<0.0001

Significant stenosis (−)

905

2,018.65

2.23

2

0.99

  

Diabetes mellitus (+)

53

125.65

2.37

2

15.92

8.64 (1.44–51.94)

0.018

Diabetes mellitus (−)

861

1,913.55

2.22

3

1.57

  

CAC score

     

4.35 (1.95–9.72)

<0.0001

0

874

1,949.24

2.23

2

1.03

  

0.1–100

39

86.80

2.23

3

34.56

  

100.1–300

1

3.16

3.16

0

0

  

CAC coronary artery calcium, CI confidence interval, NCP non-calcified plaque

aData are given as number of subjects

bUnivariate cox proportional hazard regression

Table 5

Evaluation of subclinical coronary atherosclerosis detected on CT angiography as predictors of cardiac events

 

Hazard ratioa (95 % CI)

P value

Adjusted relative riskb (95 % CI)

P value

Any plaque

2.20c (0.41–11.89)

0.36

32.42 (1.94–319.79)c

0.014

NCP

49.17c (4.13–585.81)

0.002

26.64 (1.31–1,728.82)c

0.003

Significant stenosis

105.48d (7.24–1,536.94)

0.001

103.87 (10.19–1,058.41)c

<0.0001

CI confidence interval, NCP noncalcified plaque

aMultivariable cox proportional hazard regression

bAdjusted with Poisson regression

cDiabetes mellitus and coronary artery calcium (CAC) group were analyzed as confounders

dCAC group was used as confounder

Discussion

The main findings of the present study are the following: (1) the prevalence of subclinical coronary atherosclerosis in young adults was not negligible and NCP was the most common type of subclinical coronary atherosclerosis; (2) 66.3 % (n = 57) and 53.5 % (n = 46) of patients with subclinical coronary atherosclerosis on CCTA (n = 86) were misclassified in the low-risk group by NCEP risk stratification and CACS, respectively; (3) individuals with a normal CCTA had excellent clinical outcomes during mid-term follow-up.

In this study, we found that 9.4 % of young adults have subclinical coronary atherosclerosis, which was quite similar to a recent CAC scan-based study. The CAC scan-based study showed that the prevalence of subclinical coronary atherosclerosis was 14.9 % in men and 5.1 % in women in ages 32–47 [18]. In our study, fewer subjects (6.2 % in men and 0.8 % in women) had CACS of more than zero and subclinical atherosclerosis compared to the CACS scan-based study. The reason for this difference is the presence of 46 subjects with NCP, accounting for 5.3 % of the subclinical coronary atherosclerosis. This suggests that CCTA can provide more accurate information on stenosis severity and plaque composition in the evaluation of subclinical coronary atherosclerosis. A recent study in asymptomatic young adults also showed that 27 % of the subjects had only NCP [19], which was higher than the result that showed 19 % of middle-aged asymptomatic subjects with NCP [20].

Previous studies characterizing subclinical coronary atherosclerosis in young adults were mainly based on angiographic findings in patients with MI or on histopathologic examinations on autopsies of young people. These studies reported that subclinical coronary atherosclerosis in young adults was less extensive and more commonly associated with single-vessel disease [2123]. Results of the present study correspond with these previous studies in terms of the predominant location in the left main coronary artery and single vessel involvement. Our results reflect more a accurate characterization of subclinical coronary atherosclerosis in young adults, however. If radiation exposure could be minimized sufficiently, the clinical application of CCTA in the detection of subclinical atherosclerosis may be possible in the generally asymptomatic population.

An high false-positive rate for CCTA, related mainly to heavily calcified plaques, decreases the diagnostic accuracy for significant stenosis [24]. However our study population showed that most subjects had a CACS score of zero (n = 874, 95.6 %) and non-calcified plaques were the predominant type of plaque. Therefore, the effect of false-positive observations may be minimal in our study.

Assessing cardiovascular risk in young adults can be used to the target preventive treatment of asymptomatic subjects at high risk for developing CAD. The presence of NCP or significant stenosis detected on the CCTA is an independent risk factor for predicting cardiovascular events, in addition to compensating for defects in known risk stratifying methods like NCEP-ATP III or CACS. We found that hazard ratios were 49.2 for NCP and 105.5 for significant stenosis, regardless of diabetes mellitus status or CACS score. Therefore, we suggest that CCTA could help identify young, asymptomatic subjects with subclinical coronary atherosclerosis who might benefit from intensified risk factor modification to prevent or retard the onset of clinical disease. Currently, CCTA is not justified as a screening tool for CAD in asymptomatic populations due to concern regarding radiation exposure. However, with the advent of MDCT technology, the radiation dose for CCTA has been minimized to <1 mSv. As such, the practical use of CCTA can be extended beyond symptomatic patients to asymptomatic younger adults.

Our study has some limitations. First, this study was a retrospective observational study with cases of the same ethnic background and geographical region. Results from this study should, therefore, be cautiously applied. Second, all participants in our study were self-referred, suggesting a possible selection bias. High-risk subjects might be included rather than the healthier general population. This selection bias could influence the relatively high prevalence of subclinical atherosclerosis. We also excluded 5 % of the lesions in the coronary segments with impaired image quality from plaque analysis, leading to a loss of potentially important information. Finally, the follow-up period for investigating the development of cardiac events was relatively short and the absolute number of cardiac events small. Regarding predictors of cardiac events, the significance of the statistical results in our study should be interpreted cautiously. Further study will be needed to reveal the future of CCTA in the prediction of progression to symptomatic coronary disease.

In conclusion, the prevalence of subclinical coronary atherosclerosis in young adults was not negligible. There were a considerable number of subjects with NCP, which could not be detected by CACS. CCTA has the potential to provide additional information about subclinical coronary atherosclerosis and to enhance the risk stratification, in addition to prediction of CAD.

Acknowledgments

This work was supported by a National Research Foundation of Korea (NRF) grant funded by the Korea government (MEST) (No. 2011-0023624) and grant 04-2009-004 from the SNUBH Research Fund.

Conflict of interest

None.

Copyright information

© Springer Science+Business Media Dordrecht 2012