Liver dysfunction in paediatric obesity: a randomized, controlled trial of metformin

In a previous study we showed that metformin reduced BMI z-scores and fasting glucose and insulin concentrations, and increased whole body insulin sensitivity in obese adolescents with fasting hyperinsulinemia and a family history of type 2 diabetes. We analyzed the data from this study to determine (a) if metformin reduced serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) concentrations during the 6-month trial, and (b) if the response to pharmacotherapy varied along gender or ethnic lines.

The 6-month trial was randomized, double blinded and placebo controlled; a total of 14 metformin-treated (500 mg bid) and 15 placebo-treated subjects completed the study. There were no dietary restrictions.

In obese adolescents fed ad libitum, metformin (a) prevented the rise in ALT concentrations that were observed in placebo-treated subjects at the 3 to 5 month time-points (p < 0.05); (b) reduced (p < 0.01) the percentage of all ALT values exceeding 40 U/L; and (c) caused a modest (10%) but statistically significant (p < 0.05) reduction in serum ALT in Caucasian subjects. Metformin had no effect on ALT levels or the ALT to AST ratio in the five African American adolescents enrolled in the study but reduced their fasting insulin concentrations from 26.1 to 19.5 muU/mL (p < 0.05).

Our findings suggest that metformin might reduce the rates or severity of liver dysfunction in selected high-risk adolescents.