Metformin may help obese kids shed weight.

SAN FRANCISCO -- Metformin is associated with modest weight loss and some improvements in signs of metabolic syndrome in obese children with severe hyperinsulinemia, a government-sponsored randomized controlled trial has concluded.

But because this is the first such study in children, the drug can't yet be recommended for routine use in this population, Dr. Jack Yanovski said at the annual meeting of the Endocrine Society.

Dr. Yanovski, chief of the National Institutes of Child Health and Human Development's Unit on Growth and Obesity, presented the results of the first placebo-controlled trial of metformin in children aged 6-12 years. The 100 children who participated (mean age 10 years) were all obese (mean body mass index [BMI] 34.6 kg/[m.sup.2]) and severely hyperinsulinemic, with fasting insulin of at least 15 mcU/mL.

The cohort consisted mostly of female children (60%). Children randomized to metformin (53) were started at 1,000 mg/day; this was ramped up to a final dosage of 1,000 mg twice a day for the duration of the 6-month trial. All children but one were able to tolerate the dose. All children also took a daily multivitamin supplement.

By the study's end, mean BMI had decreased in the active group and increased in the placebo group (-1 kg/[m.sup.2] vs. + 0.23 kg/[m.sup.2]--a significant difference).

Children taking metformin also had a significantly decreased BMI z score, compared with those taking placebo (-0.11 vs. -0.04). Mean body fat mass also decreased significantly in the active group (-1.7 kg) and increased significantly in the placebo group (2 kg).

Some signs of metabolic syndrome improved in children taking metformin, although the changes were not significant. Serum glucose, homeostatic assessment model algorithm (HOMA) insulin resistance index, and total cholesterol all improved in the treated children.

Children in the metformin group experienced a significant decrease in serum vitamin [B.sub.12] concentrations, although all remained within normal range, and no child required additional supplementation.

The most commonly reported adverse events were liquid stool (60% metformin vs 2.5% placebo), nausea (24% vs 8%), and fatigue (14% vs 5%).

All were significantly more common in the metformin group; however, by the study's end, the incidence of liquid stool had decreased by 20% and the incidence of nausea had decreased by 8%.

Dr. Yanovski stated that he had no financial disclosures to make with regard to metformin.

BY MICHELE G. SULLIVAN

Mid-Atlantic Bureau 6-Month Change in Body Mass Index of Obese Children Metformin group (n = 53) -1.00 kg/[m.sup.2] Placebo group (n = 47) 0.23 kg/[m.sup.2] Note: Based on a study of children with a mean age of 10 years that were obese and severely hyperinsulinemic. Source: Dr. Yanovski Note: Table made from bar graph.