Unpacking BMI Significance: Prevalence of Metabolically Healthy Obesity and Metabolic Profile Predictors Among Adolescents Enrolled in a Weight Management Program

ABSTRACT

Purpose: Adolescent obesity continues to rise, with body mass index (BMI) commonly used as an adiposity surrogate. While obesity correlates with metabolic syndrome risk, individuals with the same BMI do not have equivalent health risks. In 2018, the first pediatric consensus definition for metabolically healthy obesity (MHO) was proposed. Identifying MHO patients is clinically relevant for personalizing interventions by cardiometabolic phenotype. The objective of this study was to examine baseline MHO and metabolically unhealthy obesity (MUO) prevalence and identify metabolic and anthropomorphic predictors among adolescents enrolled in weight management. Methods: This study uses baseline data from 1,316 patients ≥ 11 years of age enrolled in a weight management program for obese adolescents in Baltimore, Maryland between 2005-2018. Anthropometric measures (including body fat by bioimpedance (%fat)), vital signs, and fasting labs were performed at intake. MHO definition was glucose <100, HDL > 40, triglycerides < 150, systolic blood pressure < 120, diastolic blood pressure < 80. MUO was defined as ≥ 1 abnormal value among MHO variables. Independent samples t-tests were used to compare mean %fat and BMI z-score of MHO and MUO groups. Bivariate logistic regression was performed to determine effects of age, sex, %fat, BMI, and BMI z-score on likelihood of MHO. Results: Mean age in the MHO group was 13.48 years (SD 1.88);mean age in the MUO group was 13.98 years (SD 2.03). 444 (33.7%) patients met criteria for MHO;872 patients had MUO. MHO teens had statistically significantly lower mean %fat (46.7% +/- 8.0% SD) vs. MUO (47.8% +/- 8.2% SD) (p = 0.034) and lower BMI z-score (2.37 +/- 0.33 SD vs 2.51 +/- 0.34 SD;p < 0.001) vs MUO. The MHO group was 66.9% female vs 54.5% females in MUO, with 38.9% lower odds of MHO for males vs. females (OR 0.611;CI 0.467 - 0.800). For every 1% increase in %fat, odds of MHO increased by 3.1%, (OR 1.031;CI 1.008 - 1.053). Each 1-year age increase led to 10.9% decrease in MHO odds (OR 0.891;CI 0.823 - 0.965). In addition, each 1 unit increase in BMI z-score was associated with a 64.5% decrease in odds of MHO (OR 0.355;CI 0.166 - 0.759). BMI change did not significantly change MHO odds. Conclusions: Among this cohort of obese adolescents enrolled in weight management, one-third had MHO. Factors associated with higher likelihood of MHO include female sex, younger age, and lower BMI z-score. Notably, BMI was not predictive of metabolic phenotype. These findings suggest potential for risk prediction for MUO profile to tailor interventions and resources accordingly. Next, we will evaluate metabolic profiles of patients enrolled during the COVID-19 pandemic. Sources of Support NICHD T32HD052459 (PI Trent), The Mount Washington Foundation.