ABC | Volume 112, Nº3, March 2019

Original Article Eickemberg et al Abdominal adiposity and C-IMT in the ELSA-Brasil Arq Bras Cardiol. 2019; 112(3):220-227 healthy and non-obese Italians. 10 The only study found that evaluated the association between VAI and a subclinical measure of atherosclerosis – the CAC – coronary artery calcium score – was carried out with 33,468 Koreans with a mean age of 42 years. Similar to the present findings, but with a lower magnitude of association, the highest chance of having subclinical atherosclerosis (OR = 1.26, 95%CI: 1.14, 1.38) was shown in individuals with the highest tertile of VAI. 9 It was found in the current study that the chance of men and women with abdominal adiposity assessed by VAI of having high CIMT was 42% and 31%, respectively. This difference between the studies was possibly observed due to the characteristics of the investigated populations (healthy participants versus patients from a Korean university hospital). 9 Similar to VAI, the LAP showed an association between the presence of abdominal adiposity and CIMT. No previous evidence was found on the association between LAP and subclinical atherosclerosis. The LAP was developed to reflect combined metabolic and physical alterations, using WC and triglycerides. Therefore, it measures lipid overaccumulation and stands out as a cardiovascular risk factor in adults. This indicator has been investigated in the context of metabolic and cardiovascular diseases and mortality. An American cohort study with approximately 5,000 subjects treated at a cardiologic clinic between 1995-2006 showed an association between LAP and cardiovascular mortality (HR: 1.52 95%CI: 1.27, 1.82), adjusted for age, gender, smoking, diabetes, blood pressure, LDL-cholesterol and HDL-cholesterol. 36 However, more studies are needed, especially in Brazil, to broaden the knowledge of less popular indicators such as VAI and LAP. Evidence suggests that information not only on the fatty tissue accumulated in the abdominal region is provided through LAP and VAI, but also on fat deposition in areas such as the liver, muscle, heart and arteries. This lipid overaccumulation causes changes in intracellular metabolism and contributes to the occurrence of cardiovascular disease, including atherogenesis and death. 19 In the present study the associations between adiposity measures and CIMT were more significant for men than for women. Women have more total body fat (and subcutaneous), often in the legs and buttocks and, especially, before menopause. Men tend to accumulate fat in the abdominal region throughout life, so they are at higher risk for developing cardiovascular outcomes, 22 including atherosclerosis. Evidence shows differences in the progression of CIMT and adiposity due to the ethnicity/ skin color. 37 The cut-off points used in this study incorporated the differences between gender and ethnicity/skin color 20 and, perhaps because of that, no effect modification was detected. Through the coefficients of determination (R 2 ), the linear regression model variables, including each indicator alone, explained approximately 30% of the total CIMT variability. In our study, the models were adjusted for age, ethnicity/ skin color, level of schooling, smoking, HDL-cholesterol, LDL‑cholesterol and arterial hypertension. The study carried out by Santos et al., 25 using the ELSA-Brazil sample, found coefficients of determination (R 2 ) close to 40% when investigating the association of risk factors with CIMT through the variables: blood pressure, glucose metabolism, lipid profile and adiposity (body mass index, WC, hip circumference, WHR, waist-to-height ratio, neck circumference). It is noteworthy that, in addition to adiposity patterns, geographic, genetic, environmental and behavioral characteristics are also associated with the occurrence of atherosclerosis. The 75 th percentile of the distribution was used to categorize CIMT in the logistic regression analysis. Other values for this classification might have yielded more consistent results. However, studies show subjects with CIMT values above the 75 th percentile with a higher risk of developing cardiovascular disorders. 17,38 It is known that the atheroma plaques may be more representative of atherosclerosis than CIMT. 39 However, our population is relatively young, and when CIMT was dichotomized at 1.5 mm, a proposed classification for atheroma plaque according to the international consensus, 5 it showed a low frequency of participants with this condition (4% in men and 2% in women) (data not shown). The use of a stringent protocol for image acquisition and quality control provided reliable and accurate data of CIMT measurements in this study. To reduce the influence of the evaluator, the reading of all images was centralized, and the automated measurements were performed by software. Although we did not adjust the models by body mass index, we excluded subjects with class III obesity and those who underwent bariatric surgery from the analysis, aiming to filter the effect of abdominal adiposity without influence of excessive total body fat. This study has limitations. Data on menopause were not considered. When women reach menopause they lose the protection provided by the hormone estrogen and, as they get older, there is a greater accumulation of abdominal fat, as well as an increase in the occurrence of cardiovascular problems. 22 The literature is clear about the effect of age on atherosclerosis. 5 Although the analyzes were adjusted for age in this article, it did not allow the observation of the effect of adiposity on CIMT at different age groups. It is not possible to affirm causality due to the cross-sectional design of this study; however, it seems unlikely that arterial thickening occurs before the high accumulation of abdominal fat. ELSA-Brazil is an occupational cohort and generalizations for the Brazilian population are limited, despite similarities in the prevalence indicators observed in ELSA-Brasil and VIGITEL studies. 40 Conclusion The observed results reinforce the importance of abdominal adiposity for the condition of subclinical atherosclerosis. Abdominal adiposity, identified through WC, WHR, LAP and VAI, was associated with CIMT in both genders, with the traditional WC anthropometric indicator standing out. WC, when compared to the other indicators, and men, when compared to women, showed the most significant effects. Author contributions Conception and design of the research, analysis and interpretation of the data, statistical analysis and writing of the 225