ABC | Volume 111, Nº5, November 2018

Brief Communication Scorsatto et al Metabolically healthy obese women Arq Bras Cardiol. 2018; 111(5):733-737 with bioelectrical impedance (Biodynamics 450, Seattle, WA, USA), and blood pressure with an aneroid sphygmomanometer (Missouri, Curitiba, Brazil). Finally, in all participants we calculated the waist-to-height ratio (WHtR) in cm/cm. The visceral adiposity index (VAI) was calculated using the following sex-specific formula for women: VAI = WC 36.58 + (1.89 × BMI) TG 0.81 1.52 HDL × × Blood samples were collected after a 12-hour overnight fast. Serum was obtained by centrifugation of the samples at 4000 rpm for 15 minutes (Excelsa Baby I, Fanem, São Paulo, Brazil). Serum concentrations of glucose, triglycerides, high‑density lipoprotein (HDL)-cholesterol, and total cholesterol were determined by the enzymatic method in an automated biochemical analyzer (LabMax 240, Labtest Diagnostica SA, Brazil). Low-density lipoprotein (LDL)-cholesterol was calculated using the Friedewald formula. Serum insulin was measured by chemiluminescence, and insulin resistance was estimated using the HOMA index. 5 We distributed the HOMA indices in quartiles and classified the participants as metabolically healthy when their indices were within the three lowest quartiles (2.78), based on Pimentel et al. 4 The data are presented as mean and standard deviation (SD). The normality of the variables was tested using the Kolmogorov-Smirnov test. Intergroup comparisons were performed with the chi-square test for categorical variables and Student’s t-test for continuous variables. P values < 0.05 were considered statistically significant. We used receiver operating characteristic (ROC) curves to identify the cutoff points for WC and VAI values. The analyses were carried out with the statistical software SPSS 20.0 (SPSS, Chicago, IL, USA). Results We selected 292 women, 53 of whom were excluded after reporting a diagnosis of diabetes mellitus or use of hypoglycemic drugs. The final sample consisted of 239 individuals. A total of 181 participants (75.7%) were classified as MHO according to their HOMA index. The results showed that all anthropometric parameters and VAI were significantly greater in MUO, and that there were fewer hypertensive individuals and higher triglyceride values in the MHO group when compared with the MUO group (Table 1). Figure 1 shows the values of WC and VAI and their accuracy in identifying MHO women. Both groups presented similar ROC curves; the WC curve had a better negative likelihood ratio to discriminate MHO at a cutoff value of 108.2 cm. Discussion Regardless of the criteria used to define the MHO and MUO phenotypes, it is unclear whether MHO individuals have a lower risk of CVD or all-cause mortality when compared with MUO individuals. 6 A systematic review of the 14 studies that focused on the risk of CVD showed that most of the studies failed to demonstrate a significant association between MHO and increased risk of CVD and mortality, although MHO individuals may indeed have a slightly increased risk of CVD when compared with individuals with normal weight. 1,3 Berezina et al. 7 studied 503 patients with abdominal obesity and concluded that the MHO phenotype was associated with younger age, smaller WC, higher physical activity level, shorter duration of obesity, and presence of the G45G adiponectin genotype. 7 However, the greatest challenge is establishing a cutoff point for WC that can be applied to different obese populations. In our study, the prevalence of metabolic health was high; approximately 76% of obese individuals were MHO, and these results were influenced by which definition of metabolic health was used. According to those results, increased waist circumference, waist-to-height ratio, fat mass, blood glucose, insulin, triglycerides, VAI, and hypertension were associated with the MUO phenotype, suggesting that the criterion applied could identify individuals with higher CVD risk. This phenotype overlaps the so-called hypertriglyceridemic waist phenotype, associated with atherosclerosis, diabetes, and coronary artery disease. 1,3,6 Also, this higher prevalence of MHO suggests lack of evidence that BMI is a good marker of cardiometabolic risk and that there is a need for the development and validation of other markers that may help to guide diagnosis and treatment of obese individuals. 7 In a recent study, 8 including 296,535 participants of both sexes from the UK Biobank followed up for an average of 5 years, one standard deviation increase in waist circumference (12.6 cm for women and 11.4 cm for men) was associated with a hazard ratio (HR) of 1.16 (95% CI: 1.13–1.19) for women and 1.10 (95% CI 1.08–1.13) for men for CVD events. In our study, WC had greater measurement values and was an inexpensive and easy tool to apply in a clinical setting in order to discriminate Brazilian women with MHO from those with MUO. Also, WC and VAI identified MHO women with a similar area under the ROC curve. The VAI was a positive independent indicator of arterial stiffness, measured by brachial-ankle pulse wave velocity in 5,158 individuals over the age of 40 in a cross-sectional study conducted in Nanjing, China. 9 However, VAI is not so easily obtained in clinical practice. It is possible that WC and VAI could be markers of different aspects of MHO. The former is a tool that easily identifies MHO individuals, and the later assesses the effects of obesity on arterial stiffness and transition into an unhealthy state. Hamer et al. 10 followed up 2,422 men and women for over 8 years as part of the English Longitudinal Study of Ageing. These authors showed that the MHO phenotype is relatively unstable, since 44.5% of MHO individuals transitioned into an unhealthy state, and emphasized that the progress to an unhealthy state was linked with a significant increase in WC. 10 Visceral obesity is associated with pro-inflammatory activity and increased production of adiponectin linked to deterioration of insulin sensitivity, increased risk of diabetes, dyslipidemia, hypertension, atherosclerosis, and higher mortality. 10 The primary issue is that the number of obese individuals is continually increasing, and it would be unaffordable to treat all of them in the same fashion. When it comes to obese individuals, as a rule, they all exhibit higher WC measures than the values proposed as cutoff points by IDF andNCEP-ATP III. 3 In our study, 734