IJCS | Volume 33, Nº1, January / February 2019

59 scanners were performed in helical acquisition mode, or in prospective axial and high-pitch spiral mode by the dual-source (two x-ray sources) scanner. 2. Image reconstruction For coronary artery calcium score calculation, images were reconstructed with a section thickness of 3 mm and 3 mm- interval. Coronary calcifications with attenuation ≥ 130 HU in an area ≥ 3 mm 2 were quantified, according to the algorithm proposed by Agatston et al. 12 CCTA images were reconstructed with a section thickness of 0.6 mm and increment of 0.3 mm in systole and/or automatically or manually determined (in case of spiral or prospective acquisition), to minimize cardiac motion artifacts. For better image quality, iterative reconstruction algorithms were performed. 3. Image interpretation All images (calcium score and CCTA) were analyzed on a dedicated workstation (Leonardo Definition, Siemens Healthcare, Erlanger, Germany). All CCTA images were analyzed by two observers; discrepancies were resolved by consensus. Coronary artery calciumwas quantitively determined by visual identification of coronary calcifications. Lesions in different coronary territories were automatically summed to determine the total calcium score. Per-segment analysis of CCTA images was performed following the Society of Cardiovascular Computed Tomography guidelines. 13 CADwas established at two levels: 1) calcium score > 0 (Agatston); 2) presence of atherosclerotic plaque (CCTA). Obstructive coronary diseasewas defined by the presence of any coronary stenosis ≥ 50%. Statistical analysis Binary data were described in absolute numbers and percentages. Continuous variables with normal distribution were presented as mean and standard deviation, whereas those without a normal distribution were presented as median and interquartile range. Data normality was tested by the Shapiro-Wilk test; coronary artery calcium score was the only variable that was not normally distributed. Categorical variables were compared using the chi-square test. Continuous variables were compared using the unpaired Student’s t-test. Calcium score between obese and non-obese patients was compared by the Mann-Whitney test. Amultiple linear regression model was used to assess the relationship between cardiovascular risk factors and the presence of obstructive CAD. For continuous variables of the model, β coefficients were used to indicate changes in the dependent variable (presence of obstructive CAD) for a unit change in each independent variable after controlling for confounding variables. For categorical variables (e.g. sex, smoking), β coefficient represents the difference in the dependent variable (presence of obstructive CAD) according to the status (e.g. male vs. female; smokers vs. non-smokers) after controlling of confounding variables of the model. Statistical analysis was performed using the STATA software (version 11, STATACorp, College Station, Texas, USA). The level of significance was set as 5%. Results A total of 1,814 consecutive patients with a medical indication for cardiac/coronary computed tomography angiography, were referred to a tertiary hospital in São Paulo between August 2010 and July 2012. We excluded from analysis patients whose indication for the test was not screening for CAD (e.g. patients with congenital heart disease, patients referred for evaluation of valve disease or pulmonary veins). In addition, we also excluded patients with history of CAD (myocardial infarction, angioplasty and /or surgical myocardial revascularization). A total of 1,383 patients were evaluated (Figure 1). Table 1 describes main epidemiological characteristics of the patients. Mean age was 58.5 +/- 11.5 years, and 66.3% (n = 917) of patients were men. In general, the prevalence of cardiovascular risk factors was not different between obese and non-obese subjects (Table 1), and the same was observed for the prevalence of obstructive CAD. Obstructive CAD was present in a similar percentage (18.4% in both groups) in obese patients (n = 58) and in those with BMI < 30 kg/m 2 (n = 197) (Figure 2). The presence of CAD, defined by the presence of coronary calcifications, was significantly different between the groups. Median calcium score was 1.4 and 14.7 Agatston units in the groups of non-obese and obese patients, respectively (Figure 2). In our sample, mean calcium score percentile, by age, sex and ethnicity was 61. In order to establish the role of each risk factor on the development of obstructive CAD, we used a multiple Pereira et al. Obesity and coronary artery disease Int J Cardiovasc Sci. 2020;33(1):57-64 Original Article