Metformin during pregnancy not linked to adverse birth outcomes

Credit: Freestocks / Unsplash

Women who continue metformin therapy through pregnancy are at little increased risk to experience a nonlive birth versus women who switched from metformin to insulin monotherapy, according to new research.¹

Data from a registry cohort analysis conducted by a team of Harvard-based investigators found that gestational metformin use—the first-line treatment for patients with type 2 diabetes (T2D)—may not be associated with significant risk of neither nonlive birth nor congenital malformations among newborns. The findings are encouraging—not only relative to the increased prevalence of T2D and gestational diabetes among childbearing-age women, but with regard to the utility of metformin in the management of polycstic ovary syndrome (PCOS) effects as well.²

Investigators led by Yu-Han Chiu, MD, ScD, assistant professor in the department of public health sciences at Penn State College of Medicine and former postdoctoral fellow at the Harvard T.H. Chan School of Public Health’s CAUSALab, sought to evaluate the teratogenicity of metformin use among women in their first trimester of pregnancy. Despite being a primary pharmacotherapy for T2D, there is “limited evidence” about metformin’s safety in early pregnancy, the team wrote.¹

Indeed, prior research has shown an existing “grey area” in the prescription of metformin for pregnancy, wrote Verma et al.³

“Even though its benefits have been more or less established through findings from various controlled trials, one has to weigh the benefits against the risks posed by the drug, as not much research has been conducted regarding the mechanisms of its metabolism in the fetus,” Verma et al wrote. “All the information on metformin as a drug for pregnancy comes from studies that are often underfunded, do not have enough study participants, and do not conduct long-term follow-ups. As a result of this, regarding the drug's safety, there are no readily available definite facts.”

To address this void in clinical evidence, Chiu and colleagues conducted their observational cohort assessment using pregnant women with pregestational T2D who received metformin monotherapy prior to their last menstrual period.¹ They emulated a target trial with 2 treatment strategies:

• Continued metformin and initiated insulin therapy within 90 days of last menstrual period.
• Discontinued metformin treatment and initiated monotherapy insulin within 90 days of last menstrual period (Control arm).

The team sought outcomes including the overall risk and ratio of nonlive births and congenital malformations among live births among each treatment arm, using covariate-adjusted estimations derived from standardization.

Using registrants from the US Medicaid health care administration database from 2000 – 2018, investigators identified 12,489 pregnant women who met the study’s eligibility criteria. Of them, 2407 were included in the assessment—850 (35.3%) were in the control arm, and 1557 (64.7%) were in the continued metformin arm.

The final analysis showed that the estimated risk for nonlive births was 32.7% among women who discontinued metformin and received insulin monotherapy, versus 34.3% among those who continued metformin plus insulin (risk ratio [RR], 1.02; 95% CI, 1.01 – 1.04).

Estimated risk for congenital malformations among live-birth newborns was 8.0% (95% CI, 5.7 – 10.2) among women receiving insulin monotherapy, versus only 5.7% (95% CI, 4.5 – 7.3) among those receiving continued metformin plus insulin—indicating a 28% reduced risk of such outcomes among pregnant women receiving metformin, though the finding was not statistically significant (RR, 0.72; 95% CI, 0.51 – 1.09).

Though the assessment was potentially limited by the residual confounding of outcomes based on patients’ glycemic control and body mass index (BMI), among other risk factors, the team concluded their findings provide a greater insight into the impact of gestational metformin use on newborn outcomes.

“Compared with switching to insulin monotherapy, continuing metformin and adding insulin in early pregnancy resulted in little to no increased risk for nonlive birth among women receiving metformin before pregnancy,” Chiu and colleagues concluded. “Under conventional statistical criteria, anything between a 49% decrease and a 9% increase in risk for congenital malformations was highly compatible with our data.”

References

1. ^{Chiu YH, Huybrechts KF, Patorno E, et al. Metformin Use in the First Trimester of Pregnancy and Risk for Nonlive Birth and Congenital Malformations: Emulating a Target Trial Using Real-World Data. Ann Intern Med. Published online June 18, 2024. doi:10.7326/M23-2038}
2. ^{Shmerling RH. Is metformin a wonder drug? Harvard Health Publishing. Published April 8, 2024. https://www.health.harvard.edu/blog/is-metformin-a-wonder-drug-202109222605#:~:text=Menstrual%20irregularities%20and%20fertility%20problems,Weight%20gain%20from%20antipsychotic%20medicines.}
3. ^{Verma V, Mehendale AM. A Review on the Use of Metformin in Pregnancy and Its Associated Fetal Outcomes. Cureus. 2022;14(10):e30039. Published 2022 Oct 7. doi:10.7759/cureus.30039}