Metformin during pregnancy not associated with adiposity in children

Children of mothers who took metformin during pregnancy had similar adiposity at age 24 months to those whose mothers did not, a recent study found.

In a follow-up study of the Metformin in Women with Type 2 Diabetes in Pregnancy (MiTy) trial, researchers evaluated infants of participants who had received oral metformin, 1,000 mg twice daily, or placebo during pregnancy. Children's anthropometric measurements, including weight, height, and skinfold thicknesses, were taken at three, six, 12, 18, and 24 months. At 24 months, the researchers used linear regression to compare the Z score for body mass index (BMI) and the sum of skinfolds in the metformin versus placebo groups, after adjustment for confounders. Growth trajectories were assessed using fractional polynomials. The results of the study were published Feb. 3 by The Lancet Diabetes & Endocrinology.

Of 465 eligible children, 283 (61%) from 19 centers in Canada and Australia were included. At 24 months, no between-group difference was seen in mean BMI Z score (0.84 [SD, 1.52] with metformin vs. 0.91 [SD, 1.38] with placebo; mean difference, 0.07 [95% CI, –0.31 to 0.45]; P=0.72) or mean sum of skinfolds (23.0 mm [SD, 5.2 mm] vs. 23.8 mm [SD, 5.4 mm]; mean difference, 0.8 mm [95% CI, –0.7 to 2.3 mm]; P=0.31). The two groups also did not differ in weight; height; weight-for-length percentile; individual tricep, subscapular, and bicep skinfold thicknesses; Z scores for tricep and subscapular skinfolds; BMI Z scores for risk of overweight, overweight or obesity, and obesity; head circumference; waist circumference; or central-to-peripheral adiposity. Metformin use during pregnancy did not predict BMI Z score when children were 24 months of age (mean difference, –0.01 [95% CI, –0.42 to 0.37]; P=0.92). No overall difference was seen in BMI trajectory, but BMI trajectories among male children differed significantly by treatment (P=0.048), with the metformin group having a higher BMI between six and 24 months than the placebo group.

The researchers noted that they enrolled only 60% of the original cohort and that the significant findings could have been due to chance, among other limitations. They concluded that in this follow-up study of a randomized trial, children who were exposed to metformin in utero had similar anthropometrics up to 24 months of age compared to children who were not. “Given the increasing incidence of type 2 diabetes in pregnancy and the increasing use of metformin during pregnancy, we believe that these data are reassuring with regard to the use of metformin during pregnancy in women with type 2 diabetes and the long-term health of their children,” the authors wrote. “Future follow-up is needed to see if such findings persist.”

An accompanying comment agreed that the study data are reassuring from a clinical perspective and said that at the population level, benefits of metformin use in women with uncontrolled hyperglycemia in pregnancy “overwhelmingly” exceed risks. “The majority of babies exposed to metformin will have healthy birthweight and body composition, reduced perinatal complications, and normal postnatal growth. A small number of neonates might have small for gestational age; more data are needed about short-term and longer-term risks in this group,” the comment authors wrote. “In the future, a precision approach to disease management might allow improved identification of children at higher risk of restricted growth, who could be prioritised for alternative management. Until then, metformin remains an indispensable treatment option for uncontrolled hyperglycaemia in the antenatal diabetes clinic.”