Cut-off values of obesity indices to predict coronary heart disease incidence by time-dependent receiver operating characteristic curve analysis in 10-year follow-up in study of Yazd Healthy Heart Cohort, Iran

• Published in: ARYA atherosclerosis Vol. 18; pp. 1 - 10
• Main Authors: Hozhabrnia, Abdollah, Jambarsanq, Sara, Namayandeh, Seyedeh Mahdieh
• Format: Journal Article
• Language: English
• Published: Isfahan Isfahan Cardiovascular Research Center 15.05.2022
• Subjects: Abdomen; Angina pectoris; Angioplasty; Body mass index; Cardiovascular disease; Coronary vessels; Health care; Heart attacks; Heart surgery; Obesity; Overweight; Performance evaluation; Questionnaires; Risk factors; Vein & artery diseases; Womens health; Iran
• ISSN: 1735-3955; 2251-6638
• DOI: 10.22122/arya.v18i0.2219

Coronary artery bypass graft (CABG), percutaneous coronary intervention (PCI), myocardial infarction (MI), Rose Angina Questionnaire (RAQ) (chest pain) greater than 3, and electrocardiographic (ECG) changes in favour of the coronary artery disease (CAD) were considered as the CVD risks. Body mass index (BMI) is a usual index that is widely used to classify weight gain and obesity.4 Literature review showed that BMI could not accurately predict the presence of CVD risk factors or the incidence of heart disease in diverse populations.5 Notably, the CVD risk factor related to central obesity has been observed to be higher than general obesity as measured by BMI.6,7 The waist circumference (WC) has been shown to be closely linked to central obesity and modulate the limitations of BMI.8 Studies showed that the predictive power of WC for predicting risk of CVD was enhanced when it was corrected by height and hip circumference (HC).9 Many epidemiological studies proposed waist-to-height ratio (WHtR), as a more reliable predictor of CVD risks than WC.10,11 Although general and central obesity are associated with many risk factors of CVD, WC is highly correlated with BMI and differentiating these indices as epidemiological risk factors can be difficult.12 A Body Shape Index (ABSI) has been developed which has little correlation with height, weight, and BMI indices and a modest correlation with WC index.13 In comparison to BMI and WHtR, ABSI was found to be a poor predictor of CVD risk.14-16 Additional anthropometric indices that aim to quantify obesity have recently been proposed, including the abdominal volume index (AVI),9 body adiposity index (BAI),17 and body roundness index (BRI).15 In a cross-sectional study, the risk of coronary heart disease (CHD) was investigated by eight anthropometric indices of BMI, WC, waist-tohip ratio (WHpR), WHtR, AVI, BAI, BRI, and ABSI. Results showed that in rural areas of China, ABSI index in men and WHtR and BRI in women were the best indicators for estimating CHD risk.18 Accurate identification of overweight and obese people allows healthcare professionals to propose prevention and treatment programs to those at the highest risk of CVD.19 The receiver operating characteristic (ROC) is a well-established method for evaluating the predictive power of an anthropometric index in distinguishing between people experiencing CVD and those who do not.20 The area under the ROC curve (AUC) is used to measure the performance of an anthropometric index for CVD incidence, with a higher AUC value indicating superior anthropometric index performance.21 The classical ROC curve analysis supposes disease status for an individual as fixed over time; however, in practice, many disease outcomes are time-dependent. WC was metered at a point midway between the lower rib margin and the iliac crest without any pressure applied to the body surface and HC was taken at the level of maximal gluteal protrusion over light clothing.25 Anthropometric indices of BMI, WHpR, and WHtR for each subject were calculated using the following formulas:18 BMI = Weight (kg)/height2 (m) WHpR = WC (cm)/HC (cm) WHtR = WC (cm)/height (cm) The ABSI was calculated using the formula:13 ABSI = WC (m)/[BMI2/3 (kg/m2) height1/2 (m)] AVI was calculated as:26 AVI = [2 WC2 (cm) + 0.7 (WC - HC)2]/1000 BAI was calculated as:17 BAI = [HC (m)/height2/3 (m)] - 18 The formula was used to determine the BRI:15 BRI = 364.2 - 365.5 [1- n2 WC2 (m) height-2 (m)]1/2 Time to event for each subject was defined as the time interval between the inclusion in the study and incidence of CVD, death caused by CVD, the date of the last follow-up, and/or date of death due to other causes, whichever had occurred earlier.