This cross-sectional study analyzed 24,062 subjects screened for CAC in an annual health check-up at the Heath Screening Center of Kangbuk Samsung Hospital from 2010 to 2011. Subjects underwent cardiac multi-detector computed tomography (MDCT) estimation of the CAC score (CACS) using identical equipment and standardized protocols. Of the 24,062 subjects, two were excluded for missing waist circumference (WC) data, leaving 24,060 eligible subjects.

Informed consent was obtained from all participants. The study design and protocol were approved by the Institutional Review Board (IRB) of Kangbuk Samsung Hospital and are in accordance with the 1975 Declaration of Helsinki. After the IRB approved this study, specific datasets were released by the data management group of the Kangbuk Samsung Health Study after deletion of participants' personal information.

Anthropometric measurements included body weight, height, body mass index (BMI), WC, and systolic and diastolic blood pressure. BMI was calculated as weight in kilograms divided by height in meters squared (kg/m²). WC was measured in a standing position, at the midpoint between the anterior iliac crest and lower border of the last palpable rib, by a single examiner. Blood pressure was measured twice by a skilled nurse following at least 10 minutes of rest, and the mean value was calculated and recorded. Subjects who reported that they smoked were classified as 'current smokers.' Regular exercise was assessed using a physician-administered questionnaire in which subjects were asked if they participated in regular exercise 'of moderate intensity at least three times a week.' The presence of diabetes mellitus was determined according to self-reported data by the participants and the diagnostic criteria of the American Diabetes Association [10]. The presence of hypertension was defined by criteria recommended by the eighth report of the Joint National Committee on Prevention, Detection, Evaluation and Treatment of High BP (JNC8) [11] (i.e., a blood pressure ≥ 140/90 mmHg) or current use of antihypertensive medication or as self-reported by participants. Blood samples were collected from the antecubital vein following a minimum 12-hour overnight fast. The samples were analyzed for blood lipid profiles, fasting blood glucose levels, serum insulin levels, and glycated hemoglobin (HbA1c). Biochemical markers were measured using Bayer Reagent Packs on an automated chemistry analyzer (ADVIA 1650 Autoanalyzer, Bayer HealthCare, Tarrytown, NY, USA). Fasting blood glucose levels were measured using the hexokinase method. Lipid profiles, including values for total cholesterol (TC), triglycerides (TGs), low density lipoprotein cholesterol (LDL-C), and high density lipoprotein cholesterol (HDL-C), were measured by enzymatic colorimetric assay. Serum insulin concentrations were measured by immunoradiometric assay using an INS-IRMA kit (BioSource, Nivelles, Belgium). As a marker of IR, the HOMA-IR was calculated using the following formula [12]: HOMA-IR = [fasting insulin (IU/mL) × fasting glucose (mmol/L)]/22.5.

HbA1c was measured by immunoturbidimetric assay using a Cobra Integra 800 automatic analyzer (Roche Diagnostics, Basel, Switzerland) with a reference value of 4.4% to 6.4%. The methodology was aligned using the standards of the Diabetes Control and Complications Trial and National Glycohemoglobin Standardization Program (NGSP). The intra-assay coefficient of variation (CV) was 2.3%, and the interassay CV was 2.4%. Both these values were within the NGSP acceptable limits [13].

Participants underwent CAC quantification by MDCT (GE Healthcare, Tokyo, Japan). The total CACS was determined by summing the individual scores for the four major epicardial coronary arteries. Total CACS was reported in Agatston units [14]. CAC was categorized as present (Agatston score > 0) or absent.

MetS was defined according to the International Diabetes Federation (IDF), Modified ATP III, and European Group for the Study of Insulin Resistance (EGIR) criteria. To define high WC, ethnicity-specific values were used: high WC was defined as ≥ 90 cm in men or ≥ 85 cm in women in all three sets of MetS criteria [15]. The criteria were as follows:

IDF criteria [16]: WC ≥ 90 cm in men or ≥ 85 cm in women [15] plus two or more of the following criteria: (1) TG level ≥ 150 mg/dL or current treatment for this lipid abnormality; (2) HDL-C level < 40 mg/dL in men or < 50 mg/dL in women or current treatment for this lipid abnormality; (3) blood pressure ≥ 130/85 mmHg or use of antihypertensive medication; and (4) fasting plasma glucose level ≥ 100 mg/dL.

Modified ATP III criteria [17]: three or more of the following components: (1) WC ≥ 90 cm in men or ≥ 85 cm in women [15]; (2) TG level ≥ 150 mg/dL; (3) HDL-C level < 40 mg/dL in men or < 50 mg/dL in women; (4) blood pressure ≥ 130/85 mmHg; and (5) fasting glucose level ≥ 100 mg/dL. Note that the 2002 definition defined elevated fasting glucose level as ≥ 6.1 mmol/L (110 mg/dL); in 2004 this was changed to ≥ 5.6 mmol/L (100 mg/dL) in accordance with the American Diabetes Association's updated definition of impaired fasting glucose [18].

EGIR criteria [19]: fasting hyperinsulinemia (highest 25%) and at least two of the following criteria: (1) fasting glucose level ≥ 6.1 mmol/L (110 mg/dL); (2) blood pressure ≥ 140/90 mmHg or treatment for hypertension; (3) TG level > 150 mg/dL or HDL-C level < 40 mg/dL or treatment for dyslipidemia; and (4) WC ≥ 90 cm in men or ≥ 85 cm in women [15].

Statistical analyses were performed using SPSS version 18.0 (IBM Co., Armonk, NY, USA). For continuous variables, data are reported as the median (interquartile range) or mean ± SD and were analyzed by independent t test. For categorical variables, data are reported as proportions and were analyzed using the chi-square test. CACS, fasting insulin, and serum TGs were log-transformed for analysis to correct for skewed distributions. For ease of interpretation, these values are expressed as untransformed data in the tables in this paper. The associations between CAC and MetS components and cardiovascular risk factors were examined by Spearman correlation.

C statistics, which are analogous to the area under the receiver operating characteristic curve, were calculated to compare the ability of different logistic models to discriminate between the three sets of criteria with and without CAC. The DeLong algorithm was used to determine statistical significance [20]. Adjusted odds ratios (ORs) with 95% confidence intervals were derived according to the three sets of MetS criteria and by the number of MetS components in multiple logistic regression models for CAC, to determine which MetS criteria best predict CAC. The p values <0.05 were considered statistically significant.