IJCS | Volume 31, Nº3, May/ June 2018

228 Faria et al. Bod composition and heart failure International Journal of Cardiovascular Sciences. 2018;31(3)226-234 Original Article The study was approved by the Research Ethics Committee of the institution (approval number 47828915.3.0000.5259). All patients were informed about the study’s purpose, and signed an informed consent form before being included, as volunteers, in the study. Statistical analysis Normality of the variables was tested by the Kolmogorov-Smirnov test. Descriptive statistics was used for characterization of the sample. Continuous variables were expressed asmean and standard deviation (±SD); the Student’s t-test and the Pearson correlation were used to analyze differences and correlations between independent samples, respectively. Categorical variableswere expressed as percentage, and associations between them were analyzed by the chi-square test or the Fisher’s exact test. Analyzes were performed using the STATA 14 softwae, and statistical significance was set at p < 0.05. Results In the present study, 41 volunteers of both sexes (n = 34, 82.9% were men) aged 61 ± 10.8 years were studied. The most common comorbidity was SAH (n = 33; 80.5%), followed by DM (n = 21, 51.2%), chronic kidney disease (n = 3; 7.3%) and chronic obstructive pulmonary disease (n = 3, 7.3%). With respect to HF classification, NYHA functional class I was the most prevalent (n = 18, 43.9%), and 34.1% (n = 14) of patients had ischemic HF. Eighteen (43.9%) patients had previous AMI, 14.6% (n = 6) had previous MRS, 9.8% (n = 4) had previous valve replacement, and 21.9 (n= 9) hadprevious stent implantation. No differences were found between men and women, except for the prevalence of DM, which was higher in women (n = 6, 85.7%) than men (n = 15, 44.1%) (Table 1). Regarding the anthropometric variables, BF% was significantly lower in men (mean of 27.2%) than women (mean of 35.8%). No differences were found in the other anthropometric parameters between men and women. PA (7.1° ± 1.4), estimated by BIA, and LVEF (37.4%) were higher inwomen thanmen, with no significant difference though. Clinical and anthropometric characteristics of the study population are described in Table 2. Mean BMI was 26.4 ± 3.6 Kg/m², with no difference between men (26.4 ± 3.4 Kg/m²) and women (26.5 ± 4.8 Kg/m²) (Table 2). Most participants were overweight (41.5%), followed by normal weight (39.0%) and obese subjects (19.5%). Anthropometric indicators of obesity (Table 3) showed that 61.8%of men and 57.1%of womenwere overweight/ obese, and 100% of women and 91.2% of men were at increased risk according to the C-index (totaling 92.7% of the study population). According toWC, 82.4% of men and 85.7% of women were at increased risk, and 76.5% of men had increasedWHtR. With respect to BF%, 67.7% of men and 71.4% of women were obese. No statistically significant difference in any of the indicators was found between men and women. Table 4 shows the correlation between obesity anthropometric indicators, PA and LVEF of the studied population. BMI showed a significant positive correlation with C-index, WC, WHtR, BF%, and PA; there was a positive significant correlation of C-index with WC, WHtR and BF%, a positive significant correlation of WC with BF% and WHtR, and between WHtR and BF%. The strongest correlations were observed of BMI with WC (r = 0.84) and WHtR (r = 0.83), of C index with WC (r = 0.80) and WHtR (r = 0.81), and between WC and WHtR (r = 0.85). PA showed a significant correlationwith BMI and amarginal correlationwithWHtR (r = 0.29, 0.06) and LVEF (r = 0.29, p = 0.07). Discussion Some studies have demonstrated the relationship of excess weight with left ventricular hypertrophy and concentric and eccentric remodeling, and with diastolic dysfunction followed by long-term systolic dysfunction, 20,21 indicating a direct effect of body composition on cardiovascular system. In this context, anthropometric assessment is crucial in the clinical practice, since an early diagnosis of obesity and an adequate intervention contribute to improve patients’ quality of life and prevent the worsening of health. 22 Borné et al. 23 investigated 26,653 individuals aged 45-73 years and showed that increased BMI, WC and BF% increased the risk of hospitalization for HF, and that this risk was even greater with combined exposure to both increased BMI and WC. In our study, mean BMI was 26.4 ± 3.4Kg/m 2 , andmost patients (41.5%) were overweight. Gastelurrutia et al. 24 evaluated HFREF and patients without reduced ejection fraction and identified that 42% of patients were overweight and 27%were obese. Although BMI has been used as an important indicator of body composition in epidemiologic studies, individual BMI values should be interpreted with caution. 10 Different from the general population, in HF patients, BMI is inversely correlated