November 20, 2020
Women with type 2 diabetes who take metformin during pregnancy to control their blood glucose levels experience a range of benefits, including reduced weight gain, reduced insulin doses, and fewer large for gestational age babies, suggest the results of a randomized controlled trial.
However, the drug was also associated with an increased risk of small for gestational age babies, which poses the question as to risk versus benefit of metformin on the health of offspring.
"Better understanding of the short- and long-term implications of these effects on infants will be important to properly advise patients with type 2 diabetes contemplating use of metformin during pregnancy," said lead author Denice S. Feig, MD, Mount Sinai Hospital, Toronto, Ontario, Canada.
Summing up, Feig said that, on balance, she would be inclined to give metformin to most pregnant women with type 2 diabetes, perhaps with the exception of those who may have risk factors for small for gestational age babies, for example, women who've had intrauterine growth restriction, are smokers, and those with significant renal disease, or have a lower body mass index (BMI).
Increased Prevalence of Type 2 Diabetes in Pregnancy
Feig said that across the developed world there have been huge increases in the prevalence of type 2 diabetes in pregnancy in recent years.
Insulin is the standard treatment for the management of type 2 diabetes in pregnancy, but these women have marked insulin resistance that worsens in pregnancy, which means their insulin requirements increase, leading to weight gain, painful injections, high cost, and noncompliance.
So despite treatment with insulin, these women continue to face increased rates of adverse maternal and fetal outcomes.
And although metformin is increasingly being used in women with type 2 diabetes during pregnancy, little data exist on the benefits and harms of metformin use on pregnancy outcomes in these women.
The MiTy trial was therefore undertaken to determine whether metformin could improve outcomes.
The team recruited 502 women from 29 sites in Canada and Australia who had type 2 diabetes prior to pregnancy or were diagnosed during pregnancy, before 20 weeks' gestation. The women were randomized to metformin 1 g twice daily or placebo, in addition to their usual insulin regimen, at between 6 and 28 weeks' gestation.
Type 2 diabetes was diagnosed prior to pregnancy in 83% of women in the metformin group and in 90% of those assigned to placebo. The mean A1c level at randomization was 47 mmol/mol (6.5%) in both groups.
The average maternal age at baseline was approximately 35 years and mean gestational age at randomization was 16 weeks. Mean pre-pregnancy BMI was approximately 34 kg/m2.
Of note, only 30% were of European ethnicity.
Less Weight Gain, Lower A1c, Less Insulin Needed With Metformin
Feig reported that there was no significant difference between the treatment groups in terms of the proportion of women with the composite primary outcome of pregnancy loss, preterm birth, birth injury, respiratory distress, neonatal hypoglycemia, or admission to neonatal intensive care lasting more than 24 hours (P = 0.86).
However, women in the metformin group had significantly less overall weight gain during pregnancy than those in the placebo group, at -1.8 kg (P < .0001).
They also had a significantly lower last A1c level in pregnancy, at 41 mmol/mol (5.9%) versus 43.2 mmol/mol (6.1%) in those given placebo (P = .015), and required fewer insulin doses, at 1.1 versus 1.5 units/kg/day (P < .0001), which translated to a reduction of almost 44 units/day.
Women given metformin were also less likely to require Cesarean section delivery, at 53.4% versus 62.7% in the placebo group (P = .03), although there was no difference between groups in terms of gestational hypertension or pre-eclampsia.
The most common adverse events were gastrointestinal complications, which occurred in 27.3% of women in the metformin group and 22.3% of those given placebo.
There were no significant differences between the metformin and placebo groups in rates of pregnancy loss (P = .81), preterm birth (P = .16), birth injury (P = .37), respiratory distress (P = .49), and congenital anomalies (P = .16).
Average Birth Weight Lower With Metformin
However, Feig showed that the average birth weight was lower for offspring of women given metformin than those assigned to placebo, at 3.2 kg (7.05 lb) versus 3.4 kg (7.4 lb) (P = .002).
Women given metformin were also less likely to have a baby with a birth weight of 4 kg (8.8 lb) or more, at 12.1% versus 19.2%, or a relative risk of 0.65 (P = .046), and a baby that was extremely large for gestational age, at 8.6% versus 14.8%, or a relative risk of 0.58 (P = .046).
But of concern, metformin was also associated with an increased risk of small for gestational age babies, at 12.9% versus 6.6% with placebo, or a relative risk of 1.96 (P = .03).
Feig suggested that this may be due to a direct effect of metformin "because as we know metformin inhibits the mTOR pathway," which is a "primary nutrient sensor in the placenta" and could "attenuate nutrient flux and fetal growth."
She said it is not clear whether the small for gestational age babies were "healthy or unhealthy."
To investigate further, the team has launched the MiTy Kids study, which will follow the offspring in the MiTy trial to determine whether metformin during pregnancy is associated with a reduction in adiposity and improvement in insulin resistance in the babies at 2 years of age.
Who Should Be Given Metformin?
During the discussion, Helen R. Murphy, MD, PhD, Norwich Medical School, University of East Anglia, UK, asked whether Feig would recommend continuing metformin in pregnancy if it was started preconception for fertility issues rather than diabetes.
She replied: "If they don't have diabetes and it's simply for PCOS [polycystic ovary syndrome], then I have either stopped it as soon as they got pregnant or sometimes continued it through the first trimester, and then stopped."
"If the person has diabetes, however, I think given this work, for most people I would continue it," she said.
The study was funded by the Canadian Institutes of Health Research, Lunenfeld-Tanenbaum Research Institute, and the University of Toronto. The authors have reported no relevant financial relationships.