Distinct Lipoprotein Curves in Normal Weight, Overweight, and Obese Children and Adolescents

Objectives: Pediatric lipoprotein curves are based on population-based samples. As obesity, may alter lipoprotein levels, cutoffs not adjusted for body mass index (BMI) are potentially inappropriate. We aimed to develop distinct serum lipid curves based on sex-and BMI-percentiles for children and adolescents. Methods: Cross-sectional analysis included all healthy children and adolescents (age range 2-17 years) with available serum lipid concentrations (n = 152,820 of approximately 1.2 million children and adolescents per study year). These children and adolescents were categorized according to sex-and age-stratified BMI-percentiles: 100,375 normal weight (5th-85th percentile), 26,028 overweight (85th-95th percentile) and 26,417 obese (>= 95th percentile) individuals. Excluded were individuals with hyperlipidemia, gastrointestinal disease, thyroid disease and lipid-lowering medications. Lambda-Mu-Sigma, smoothed percentile lipid curves were computed. Results: Obese children had a lipid profile pattern throughout childhood and adolescence similar to that of normal weight subjects but with a significant upward shift in total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), non-high-density lipoprotein cholesterol (non-HDLC), and triglycerides (TGs) and a downward shift in high-density lipoprotein-cholesterol (HDL-C). Obese boys had 13 mg/dL higher TC levels (P<0.001), 11 mg/dL higher LDL-C levels, 15 mg/dL higher non-HDL-C levels, and 5 mg/dL lower HDL-C levels (P<0.001). Obese girls had 6 mg/dL higher TC levels, 7 mg/dL higher LDL-C levels, 11 mg/dl higher non-HDL-C levels, and 6 mg/dL lower HDL-C levels (P<0.001). Conclusions: Across a large, nationally representative cohort of children and adolescents, lipoprotein levels were found to vary in relation to weight status. On the basis of these findings, it is suggested that when evaluating the lipid profile in the pediatric population, in addition to sex-based curves, clinical decision making may require consideration of BMI-stratified curves.