Pregnancy can be a challenging time for women, especially those with diabetes. In order to reduce risk of complications to both mother and fetus, glycemic targets are stricter for both women with pre-existing diabetes and those who develop diabetes during pregnancy. Type 2 diabetes in pregnancy is associated with increased risk of preeclampsia in the mother, as well as the need for a caesarean section delivery; for the neonate there is an increased risk of congenital defects, macrosomia, preterm delivery, stillbirth and perinatal mortality1. Gestational diabetes, defined as the presence of any glucose intolerance with onset or first recognition in pregnancy, is also associated with a moderate risk of adverse perinatal outcomes2,3. Insulin is the preferred treatment option in patients with pre-existing and gestational diabetes, in addition to lifestyle and diet modification. However, guidelines for the use of metformin vary among sources4,5,6. As metformin crosses the placenta freely and reaches the same level as that of maternal circulation, there have been concerns about potential exposure to the fetus7.
The safety of metformin in early pregnancy can be derived from studies of the agent’s use in polycystic ovarian syndrome (PCOS). Insulin resistance is a common denominator in diabetes and PCOS, and metformin has been used to treat menstrual irregularity and as an adjunct therapy for treatment of infertility in PCOS. Studies have found that the continued use of metformin in pregnancy in women with PCOS was not associated with increased incidence of fetal anomalies or fetal birth weight8,9.
The use of metformin in patients with gestational diabetes was explored in the Metformin versus Insulin for the treatment of Gestational diabetes (MiG) trial. A total of 751 patients with gestational diabetes was randomized to receive treatment with either metformin (with supplemental insulin as needed) or insulin only. The primary outcome examined was a composite of neonatal complications such as neonatal hypoglycemia, respiratory distress, need for phototherapy, 5-minute Apgar score less than 7, and prematurity. In the metformin-treated group, 46.3% required supplemental insulin. In both the metformin (with supplemental insulin as needed) and insulin only-treated group, primary outcomes were similar (32.0% versus 32.2% respectively) and there were no serious adverse events in the metformin-treated group. Maternal weight gain was lower and patients in the metformin-treated group also expressed a preference to receive their assigned treatment again (76.6% versus 27.2%)10. A follow-up study on the offspring of the participants at two years found that in the metformin-exposed group, measures of subcutaneous fat (subscapular and biceps skinfolds) were higher despite having similar total body fat as that of the insulin-exposed group. This difference in fat distribution, more at the subcutaneous level than visceral level, may imply that in utero exposure to metformin led to a more favorable fat pattern distribution. Distribution of excess fat, whether visceral or subcutaneous, affects cardiometabolic risk with visceral fat conferring a higher risk11. However at 7 to 9 years, total and abdominal fat measures were similar and at year nine, children who had been exposed to metformin were larger by measurements of weight, arm and waist circumference, triceps skinfold and body mass index. The significance of these findings on the metabolic health of the offspring will require data from long-term follow-up12,13.
The Metformin in women with Type 2 diabetes in pregnancy (MiTy) trial findings were recently released. The study involved 502 women with type 2 diabetes who were on insulin between 6 and 22 weeks’ gestation. Patients were randomized to receive either metformin or placebo in addition to their usual insulin regime. Evaluation of primary outcomes showed no difference between the groups; these included pregnancy loss, preterm labor, birth injury, respiratory distress, neonatal hypoglycemia and neonatal intensive care unit admission. In the metformin-treated group, there was better glycemic control (glycated hemoglobin at 34 weeks was 5.90% versus 6.10%; p=0.015) and insulin requirement (1.1 units/kg/day versus 1.5 units/kg/day) was slightly reduced. Less maternal weight gain and need for caesarean section was also found. In the metformin-exposed neonates, birth weight was lower (3156 g versus 3375 g) as were mean adipose tissue measurements leading to lower incidences of macrosomia; however, the percentage of infants who were small for gestational age was higher (13% versus 7%). The follow-up study on the offspring from the MiTy study is underway14.
Because metformin action via mitochondrial modulation affects nutrition delivery and fetal programming, longer-term childhood metabolic data is required. Thus far, data on metformin’s effect on teratogenicity are reassuring and benefits are seen in terms of reduction in maternal weight gain and insulin requirement. Its use as a monotherapy may be limited to those with mild hyperglycemia as half of all patients may require insulin, as seen in the MiG trial. Ultimately, control of hyperglycemia is required during the gestation period to reduce complications affecting both mother and baby. Increasing our understanding of the role of available treatments will allow us to select suitable therapy for pregnant patients with either pre-existing or gestational diabetes.
Click here for Dr. Chan’s previous columns.
Any editorial comments about this article can be sent to: Jane.Savio@worldwidediabetes.org