Risk for nonalcoholic fatty liver disease in Hispanic youth with BMI > or =95th percentile.

OBJECTIVES

To characterize children at risk for nonalcoholic fatty liver disease (NAFLD) and to explore possible mechanisms underlying the development of NAFLD in Hispanic youth with a body mass index > or =95th percentile.

PATIENTS AND

METHODS

Hispanic nonoverweight (n = 475) and overweight (n = 517) children, ages 4 to 19 y, were characterized in terms of body composition (dual-energy x-ray absorptiometry), maturation (Tanner stage), diet (24-h recall), physical activity (accelerometry), fitness (maximal oxygen uptake), and biochemical profile (fasting alanine aminotransferase [ALT], glucose, insulin, and lipids; inflammation markers such as adiponectin, leptin, C-reactive protein, and soluble intercellular adhesion molecule-1; and total antioxidants) using standard laboratory techniques. Risk for NAFLD was defined by fasting serum ALT values >97.5th percentile for age- and sex-specific reference values.

RESULTS

Fasting serum ALT was elevated in 24% of overweight children and in only 4% of nonoverweight children. Therefore, to identify risk factors associated with elevated ALT, the remaining statistical analysis was restricted to the overweight group. The percentage of overweight children with elevated ALT did not differ by sex, age, or Tanner stage. Weight, body mass index, z score, waist-to-hip ratio, fat-free mass, fat mass, and percent truncal fat mass were higher in the overweight children with elevated ALT. Fasting insulin, glucose, and homeostasis model-insulin resistance were higher in the overweight children with elevated ALT, as were triglycerides, total cholesterol, low-density lipoprotein, thyroid-stimulating hormone, and triiodothyronine. Fasting serum leptin, C-reactive protein, and soluble intercellular adhesion molecule-1 were significantly higher and adiponectin was lower in the overweight children with elevated ALT.

CONCLUSIONS

The risk for developing NAFLD was high in the overweight Hispanic children. The proportion of "at risk" children was not influenced by gender, age, or maturation. The risk for elevated ALT was predicted by the severity of obesity, central adiposity, hyperinsulinemia, hypertriglyceridemia, elevated thyroid-stimulating hormone, and systemic inflammation.