Diabetes Drug Perks Up Weight Loss in Obese Kids

However, the biguanide drug didn't affect levels of visceral fat or the prevalence of metabolic syndrome. While the changes were modest, the researchers still write that "metformin treatment may hold promise as a method to prevent or delay the appearance of impaired glucose homeostasis in children at high risk for the development of Type 2 diabetes." 

Metformin is known to stabilize or induce weight loss in adults and adolescents, but there have been no randomized controlled trials of such effects in severely obese children ages 6 to 12.

The research team at the National Institutes of Health in Bethesda, Maryland, studied 100 children ages 6 to 12 whose body mass index (BMI) averaged 34.6 kg/m2. The subjects and their parents met monthly with a dietitian.

In the randomized, double-blind trial, the kids were prescribed 1000 mg metformin (n = 53) or placebo (n = 47) twice daily for 6 months, followed by open-label metformin for another 6 months.

The first 6-month phase was completed by 85% of subjects (45 in the metformin group and 40 in the placebo group), while the second 6 months were completed by 67% (30 and 27 subjects, respectively). Analysis was by intent-to-treat using multiple imputation for missing outcome data.

After the first 6 months, the decrease in BMI Z score was significantly greater in the metformin group (-0.11 vs. -0.04, p = .02). BMI fell with metformin treatment while rising with placebo (-0.78 vs. 0.32, respectively, p = .006). Weight decreased less (1.47 vs. 4.85 kg, respectively, p < .001), as did total body fat mass (0.48 vs. 1.88 kg, p = .04).

The study also showed significant changes favoring metformin in abdominal and hip circumference and in triceps skinfold thickness. However, changes in intra-abdominal fat were similar in the two groups.

While non-Hispanic black children lost less weight than non-Hispanic or Hispanic whites, the effect of metformin didn't differ. There was also no significant impact of baseline insulin resistance or pubertal status on treatment success.

Dr. Yanovski's team also found that metformin lowered fasting serum insulin and plasma glucose significantly. While the homeostasis model assessment-insulin resistance (HOMA-IR) index -- a measure of hepatic sensitivity to insulin -- improved with active treatment, neither first-phase insulin secretion nor whole body insulin sensitivity estimated by hyperglycemic clamp study changed much.

Treatment group didn't affect cholesterol or triglyceride levels or high-sensitivity C-reactive protein. The report indicates that prevalence of metabolic syndrome didn't change.

Changes in BMI Z scores were maintained in metformin-treated individuals who continued throughout the 6-month open phase of the trial.

Side effects of metformin consisted primarily of gastrointestinal symptoms and fatigue, which tended to occur during the first month of the study. Clinicians reduced dosages in 17.0% of those in the metformin group and in 2.1% of those in the placebo group (p = .03). Only one child discontinued the trial because of medication intolerance.

The investigators note that the small study sample and limited length of treatment may have limited power for detecting more differences between groups.

In conclusion, they write, "Metformin had modest but favorable effects on body weight, body composition, and glucose homeostasis in obese insulin-resistant children participating in a low-intensity weight-reduction program."

Published online before print January 12, 2011, doi: 10.2337/db10-1185 Diabetes January 12, 2011 DB_101185