IJCS | Volume 33, Nº3, May / June 2020

248 no statistical significance was observed for individuals with or without central obesity (Table 2). A significant statistical association was observed between the HDL-C (p = 0.019), Castelli’s index I (p = 0.040) and central obesity among the elderly (Table 3). These and the non-HDL-C were included in the logistic regression model (Table 4), which showed a positive association between central obesity and low HDL-C (OR = 2.48; CI 95%: 1.15-5.37) and increased TC/HDL-C ratio (OR = 3.13; CI 95%: 1.01-9.76). Discussion The association between central obesity and biochemical changes has been widely discussed nowadays 19-21 , since visceral fat produces adipocytes which, in turn, are directly linked with inflammatory processes and cardiometabolic complications . 22 In this sense, this study contributes with the scientific literature because it shows a positive correlation between central obesity and the biochemical markers HDL-C and Castelli’s index I. A similar association was observed by several authors 22-24 who identified a positive correlation between WC and the lipid variables: TC, LDL-C, Non- HDL-C and TG. Roriz et al., 7 also observed an association between the visceral adipose tissue (VAT) area and Glucose, TG and uric acid. However, the biochemicals TC, LDL-C, HDL-C and VLDL-C did not present results with statistical significance. Castelli’s index I was adopted in this study because it is considered a major risk predictor for cardiovascular disease, 5,9 since HDL-C concentrations are inversely proportional to the incidence of atherosclerosis, due to its essential role in protecting the vascular bed, through cholesterol reverse transport. 25 In this process, the HDL-C removes the oxidized lipids of LDL, inhibits the fixation of adhesionmolecules andmonocytes to the endothelium and stimulates the liberation of nitric oxide. 26 In this context, the data presented are in line with the authors mentioned before, since in the logistic regression analysis we observed that the individuals with central obesity were 2.48 (p = 0.021) and 3.13 (p = 0.049) times more likely to develop changes inHDL-C and Castelli’s I index, respectively. Similarly, Cabral et al., 27 when assessing the association between the phenotype hypertriglyceridemic- waist (HTW) and the cardiometabolic risk in women, also observed a prevalence ratio (PR) of 3.41 (CI 95%: 2.42-4.81) for the HDL-C. These data corroborate the study conducted by Silva et al., 14 with older patients with type 2 diabetes, Table 2 - Descriptive analysis of the variables “age” and “biochemical markers of cardiometabolic risk”, stratified by central obesity, in elderly patients treated at a geriatric outpatient clinic in the city of Lagarto-SE Variables Sample Central obesity p No Yes (n = 34) (n = 125) Age 70.9 (7.5) 70.9 (8.4) 70.9 (7.2) 0.962 TC 192.4 (41.7) 190.7 (51.3) 192.9 (38.9) 0.791 LDL-C a 142.0 (119.0-173.0) 141.7 (114.7-174.3) 143.0 (118.0-173.0) 0.993 HDL-C a 41.0 (38.0-44.0) 41.0 (37.8-51.0) 40.0 (37.5-43.0) 0.309 Non-HDL-C 150.4 (42.4) 147.3 (51.2) 151.3 (39.8) 0.628 TG a 149.0 (137.0-167.0) 141.0 (130.5-162.5) 149.0 (139.0-167.5) 0.114 Glucose a 99.0 (91.0-121.0) 98.5 (87.7 -121.0) 101.0 (101.0-125.5) 0.334 Castelli’s index I 4.7 (1.3) 4.5 (1.5) 4.8 (1.2) 0.381 Castelli’s index II 3.5 (1.1) 3.3 (1.2) 3.5 (1.0) 0.575 Central obesity (WC > 80M; > 90H); High TC (> 200 mg/dL); High LDL-C > 160 mg/dL; Low HDL-C < 40 mg/dL in men and < 50 mg/dL in women, Non-HDL-C < 130 mg/dL, High TG (> 150 mg/dL); High Glucose (> 100 mg/dL); Castelli’s index I > 4.7; Castelli’s index II >3.1. a Data expressed as median (interquartile interval). Other results expressed as mean (standard deviation). Oliveira et al. Cardiometabolic risk in elderly Int J Cardiovasc Sci. 2020; 33(3):245-251 Original Article