IJCS | Volume 31, Nº3, May/ June 2018

231 Faria et al. Bod composition and heart failure International Journal of Cardiovascular Sciences. 2018;31(3)226-234 Original Article Table 4 – Correlation between obesity anthropometric indicators, phase angle and left ventricular ejection fraction BMI C-index WC WHtR BF% PA BMI C-index (0.46) p-value 0.002 WC (0.84) (0.80) p-value < 0.001 < 0.001 WHtR (0.83) (0.81) (0.85) p-value < 0.001 < 0.001 < 0.001 BF% (0.36) (0.38) (0.32) (0.53) p-value 0.02 0.01 0.04 < 0.001 PA (0.44) (-0.01) (0.22) (0.29) (0.06) p-value 0.004 0.95 0.17 0.06 0.7 LVEF (0.17) (0.12) (0.15) (0.17) (0.23) (0.29) p-value 0.29 0.47 0.34 0.29 0.15 0.07 BMI: body mass index; C-index: conicity index; WC: waist circumference; WHtR: waist-to-height-ratio; BF%: body fat percentage; PA: phase angle; LVEF: left ventricular ejection fraction reported in the study by Quirino et al. 27 showing that mean WC and WHtR values were higher than recommended in both men and women. Regarding the analysis of associations between anthropometric variables, Gomes et al. 28 found a positive significant correlation between BMI and WC. Colombo et al. 29 showed that BMI had a positive significant correlation with BF%, obtained by the sum of skinfold thickness measures, and both BMI and BF% had a significant correlation with WC. These correlations were found in our study also. Lobato et al., 30 found correlations between BMI and WC, and positive significant correlations of WC with WHtR and C-index, and between WHtR and C-index. In the study by Mendes et al., 31 involving patients with diabetes mellitus (DM), obesity and/or SAH, BMI was positively correlated with BF% (p < 0.001) and C-index (p = 0.009). Studies on C-index and WHtR as coronary risk predictors have been carried out in the Brazilian population and demonstrated the importance of these indicators in diagnostic assessment of patients. 15,17 We also obtained PA measures using BIA. These parameters have been increasingly used as a diagnostic tool in the clinical practice. In our study group, mean PA was 6.8 o  ± 1.1, with greater values in women (7.1 o  ± 1.4), but not significantly different than men. In healthy individuals, these values can vary from 4 to 10 degrees. 9 When increased, PA may be associated with greater amounts of intact cell membranes, indicating adequate health status, whereas low PA values suggest worsening of disease and cell death. 9 PA cutoff points vary between diseases – in HIV-infected patients, a PA lower than 5.3 o was associated with a unfavorable prognosis, 32 whereas lower survival rates were found in advanced cancer patients with PA lower than 4.4 o . 33 With respect toHF, Colín-Ramírez et al. 34 investigated a cohort of 389HF patients inMexico city anddemonstrated that PA is a good prognostic indicator. A PA lower than 4.2 o was more strongly associated with mortality (even after adjusting for age), serum hemoglobin and presence of DM. Another study reported a significant reduction in PA values in HF patients as compared with healthy controls (5.5 o vs . 6.4 o ). 35 Colín-Ramirez et al. 34 demonstrated the prognostic value of PA in HF patients, and showed that a lower PA was associatedwithmarkers ofmalnutrition, such as decreased BMI, worsening of functional class and kidney failure.