Association between anthropometric indices and cardiometabolic risk factors in pre-school children

Abstract

Background: The world health organization (WHO) and the Identification and prevention of dietary- and lifestyle-induced health effects in children and infants- study (IDEFICS), released anthropometric reference values obtained from normal body weight children. This study examined the relationship between WHO [body mass index (BMI) and triceps- and subscapular-skinfolds], and IDEFICS (waist circumference, waist to height ratio and fat mass index) anthropometric indices with cardiometabolic risk factors in pre-school children ranging from normal body weight to obesity.

Methods: A cross-sectional study with 232 children (aged 4.1 ± 0.05 years) was performed. Anthropometric measurements were collected and BMI, waist circumference, waist to height ratio, triceps- and subscapular-skinfolds sum and fat mass index were calculated. Fasting glucose, fasting insulin, homeostasis model analysis insulin resistance (HOMA-IR), blood lipids and apolipoprotein (Apo) B-100 (Apo B) and Apo A-I were determined. Pearson's correlation coefficient, multiple regression analysis and the receiver-operating characteristic (ROC) curve analysis were run.

Results: 51% (n = 73) of the boys and 52% (n = 47) of the girls were of normal body weight, 49% (n = 69) of the boys and 48% (n = 43) of the girls were overweight or obese. Anthropometric indices correlated (p < 0.001) with insulin: [BMI (r = 0.514), waist circumference (r = 0.524), waist to height ratio (r = 0.304), triceps- and subscapular-skinfolds sum (r = 0.514) and fat mass index (r = 0.500)], and HOMA-IR: [BMI (r = 0.509), waist circumference (r = 0.521), waist to height ratio (r = 0.296), triceps- and subscapular-skinfolds sum (r = 0.483) and fat mass index (r = 0.492)]. Similar results were obtained after adjusting by age and sex. The areas under the curve (AUC) to identify children with insulin resistance were significant (p < 0.001) and similar among anthropometric indices (AUC > 0.68 to AUC < 0.76).

Conclusions: WHO and IDEFICS anthropometric indices correlated similarly with fasting insulin and HOMA-IR. The diagnostic accuracy of the anthropometric indices as a proxy to identify children with insulin resistance was similar. These data do not support the use of waist circumference, waist to height ratio, triceps- and subscapular- skinfolds sum or fat mass index, instead of the BMI as a proxy to identify pre-school children with insulin resistance, the most frequent alteration found in children ranging from normal body weight to obesity.

Fig. 1

Sensitivity of the anthropometric indices to identify children with insulin resistance. Receiver operating characteristic curves for anthropometric indices in the detection of insulin resistance, using a homeostatic model assessment insulin resistance. (n = 96)