Utility of anthropometric traits and indices in predicting the risk of coronary artery disease in the adult men of southern Andhra Pradesh

Abstract

Background/aim: Optimal cutoff values are influenced by ethnicity, geography, lifestyles, and physical activity, and hence, there is a need for establishing population- and disease-specific cutoff values to screen individuals/populations. Therefore, the present study was carried out to determine the optimal cutoff values of anthropometric variables for coronary artery disease (CAD) for the population of southern Andhra Pradesh.

Methods: One hundred sixty five patients with CAD and 87 controls were recruited, and 52 anthropometric variables were measured for them.

Results: Higher means in 22 anthropometric variables covering circumferences, skinfold thickness (sft), and indices were observed in patients than those in controls. Receiver operator curve analysis revealed that 18 variables including circumference, sft, and fat measures with an area under curve ranging from 0.61 to 0.72 were found to have the ability of predicting the risk of CAD. A stepwise discriminant analysis showed 9 variables to correctly classify 87.4% of subjects into CAD and controls. In logistic regression analysis, among these 9 variables, only circumferences of abdomen and foot; sft of supratellar, thigh and calf; and sum of subscapular/suprailiac, waist-hip ratio and lean body mass were associated with CAD and explained 73.4% of its variation.

Conclusions: Eighteen anthropometric variables were found to have the ability of predicting the risk of CAD. Longitudinal studies are needed to confirm the use of anthropometric variables in predicting the risk of CAD.

Keywords: Anthropometry; Coronary artery disease; Discriminant analysis; Logistic regression; Receiver operator characteristics.

Fig. 1

Receiver operator characteristic curves for circumference and skinfold thickness that showed AUC ≥ 0.6. Cir, circumference; WHR, waist-hip ratio; WAR, waist-arm ratio; RFPI, relative fat pattern index.

Fig. 2

Receiver operator characteristic curves central for obesity variables that showed AUC ≥ 0.6. Cir, circumference; sft, skinfold thickness; BFI, body fat index; CI, conicity index; WHtr, waist-height ratio; BF, percent of body fat; AVI, abdominal volume index; IAAT, intraabdominal adipose tissue.