Cardiometabolic risk stratification in pediatric obesity: evaluating the clinical utility of fasting insulin and BMI-SDS

Abstract

Background: Patients with obesity during childhood have an increased risk of fatal and non-fatal cardiovascular events during adulthood. The severity of obesity is commonly determined by BMI. However, children with relatively low BMI may have high cardiometabolic risk. Indeed, BMI-based obesity classifications might miss children at high cardiometabolic risk. Insulin has been suggested as a marker of cardiometabolic risk. In this study, we therefore estimated and compared cardiometabolic risk using either the BMI standard deviation score (BMI-SDS) or fasting insulin in an international cohort of children and adolescents with obesity and lean controls.

Methods: Study participants (712 with obesity and 99 lean controls), aged 3 to 18 years, were categorized according to their BMI-SDS as lean or obesity class I, II, or III, or by their fasting insulin quartiles as quartile 1, 2, 3, or 4 with the lean subjects in a separate control group. Prevalence of cardiometabolic risk factors was assessed in each group. Sensitivity and specificity analyses for cardiometabolic risk were conducted for both BMI-SDS and fasting insulin. Multiple regression, logistic regression, and receiver operating characteristic (ROC) analyses were performed between fasting insulin, BMI-SDS and cardiometabolic risk factors.

Results: An elevated prevalence of the cardiometabolic risk factors dyslipidemia, dysglycemia and hypertension was observed in both increasing BMI-SDS classes and increasing fasting insulin quartiles. Fasting insulin demonstrated higher areas under the curve (AUC) for detecting dyslipidemia, dysglycemia, and the combination of dyslipidemia, dysglycemia, and hypertension, compared to BMI-SDS. BMI-SDS demonstrated a higher AUC for detecting hypertension compared to fasting insulin. The same patterns were seen for the logistic regression. However, fasting insulin had an overall stronger association with the cardiometabolic risk factors studied compared to BMI-SDS.

Conclusions: In children and adolescents with obesity, fasting insulin provides complementary information to BMI-SDS in identifying those with elevated cardiometabolic risk factors. While neither marker alone offers strong predictive accuracy, incorporating fasting insulin into clinical assessment may help prioritize individuals who require more detailed evaluation. An elevated fasting insulin value may warrant further investigation among children and adolescents with obesity, independent of obesity class based on BMI-SDS.

Keywords: BMI-SDS; Cardiometabolic risk; Cardiovascular disease; Cardiovascular risk factors; Dysglycemia; Dyslipidemia; Hypertension; Insulin; Obesity; Pediatric obesity.

Fig. 1

Prevalence of cardiometabolic risk factors, stratified by obesity class based on BMI-SDS. Black bars = lean controls, yellow bars = obesity class I, blue bars = obesity class II, and green bars = obesity class III

Fig. 2

Prevalence of cardiometabolic risk factors, stratified by obesity class based on quartiles of fasting insulin. Black bars = lean controls, yellow bars = quartile 1, blue bars = quartile 2, green bars = quartile 3, and red bars = quartile 4

Fig. 3

Prevalence of hypertension in children and adolescents with obesity in different age groups. Within each group, the data is divided as follows: black bars = prevalence of systolic hypertension, yellow bars = prevalence of diastolic hypertension, blue bars = prevalence of combined systolic and diastolic hypertension, and green bars = either systolic hypertension, diastolic hypertension, or a combination of both. Numbers above the bars indicate the percentage of prevalence

Fig. 4

Receiver operating characteristic curves for cutoffs of fasting plasma insulin (black line) and BMI-SDS (blue line) as predictors of A dyslipidemia, B dysglycemia, C hypertension, and D a combination of the three in obese adolescents. Arrows indicate optimal cutoffs as defined by Youden’s index