By Sarah Dzubay and Maria I. Rodriguez, MD, MPH
Sarah Dzubay is an MD candidate, Oregon Health & Science University, Portland. Dr. Rodriguez is Professor, Obstetrics & Gynecology, Division of Complex Family Planning, Department of Obstetrics & Gynecology, Oregon Health & Science University, Portland.
A multinational cohort study of more than 50,000 pregnant people found that those with periconceptual use of non-insulin antidiabetic medications, such as sulfonylureas, dipeptidyl peptidase 4 inhibitors, glucagon-like peptide 1 agonists, or sodium-glucose cotransporter 2 inhibitors, did not have increased risk of congenital malformations compared to those who used insulin periconceptually.
Cesta CE, Rotem R, Bateman BT, et al. Safety of GLP-1 receptor agonists and other second-line antidiabetics in early pregnancy. JAMA Intern Med 2024;184:144-152.
Diabetes is a significant cause of morbidity and mortality in the United States, affecting an estimated 38.1 million people.1 Between 2016 and 2021, rates of pregestational diabetes increased by 27% up to 10.9 cases per 1,000 live births and rates of gestational diabetes increased from 6% to 8.3%.2,3
The standard of care for diabetes management in pregnancy is comprised of lifestyle modifications, such as diet and exercise, and medical management.4 Insulin is the mainstay of medical management for diabetes in pregnancy for both pregestational and gestational diabetes since insulin does not cross the placenta and, thus, has been demonstrated to be relatively safe in pregnancy.4
Given the prevalence of diabetes, particularly type 2 diabetes, there has been an increased interest in alternate methods to treat this condition. These are called non-insulin antidiabetic medicines (ADMs), and they have limited safety evidence in pregnancy relative to insulin.
Understanding the safety of these new ADMs in pregnancy is important; 42% of pregnancies in the United States are unintended, and exposures to these non-insulin ADMs are probable.5 Additionally, if any prove safe in pregnancy, it is possible they could prove to be a more convenient option than insulin.
This cohort study was conducted within the International Pregnancy Safety Study Consortium and included prospectively collected data from six large healthcare databases from the United States, Finland, Iceland, Norway, Sweden, and Israel.6 The U.S. cohort was comprised of pregnant women and their liveborn infants in the MarketScan Research Database, a national database of individuals with employer-sponsored private insurance, between 2012 and 2021.
The cohort from the Nordic countries was derived from nationwide population health registries and included all pregnancies that resulted in a liveborn singleton infant between 2009 and 2020. The Israeli cohort included women who were enrolled in Israel’s second largest health system’s database for at least one year prior to conception and who had a singleton liveborn infant between 2010 and 2020. Pregnancies with known fetal chromosomal abnormalities or exposure to a known teratogenic agent were excluded from analysis.
The study population was people with pregestational type 2 diabetes mellitus (T2DM) who became pregnant and had a liveborn infant. For study purposes, T2DM was identified using algorithms based on chart diagnoses and prescription refills. Periconceptual was defined as 90 days prior to the last known menstrual period through the end of the first trimester.
Exposures were divided into seven groups based on prescription refills during the periconceptual period: no use of an ADM; metformin only; insulin with or without metformin; or one or more second-line ADMs (sulfonylureas, dipeptidyl peptidase-4 [DPP-4] inhibitors, glucagon-like peptide 1 [GLP-1] receptor agonists, or sodium glucose cotransporter 2 [SGLT2] inhibitors).
Study outcomes were any major congenital malformations identified between birth and 1 year of life in the Israeli and Nordic cohorts, and within 90 days of birth in the U.S. cohort. Additional analyses were performed for the subcategory of major cardiac malformations. Covariates included maternal age, medical comorbidities, and use of other prescription medications.
Analyses were conducted separately for each cohort. The investigators calculated both the proportion of pregnancies exposed to second-line non-insulin ADMs by birth year and the weighted prevalence of malformations for all exposure groups. For each cohort and each second-line non-insulin ADM exposure group, crude and adjusted relative risks of congenital malformations were estimated using a log-binomial model with 95% confidence intervals (CIs). Crude and adjusted estimates from each cohort then were combined using fixed-effect meta-analysis techniques.
The investigators found that use of all non-insulin, second-line ADMs except for sulfonylureas increased during the study period. Compared to women using insulin, women using non-insulin, second-line ADMs were slightly younger in the United States and Israel and older in Nordic countries. Those using GLP-1 agonists had the highest prevalence of polycystic ovary syndrome (PCOS) and obesity, and those using SGLT2 inhibitors had the highest prevalence of hypertension and cardiovascular disease. Those using second-line, non-insulin ADMs had a high prevalence of co-refills for insulin (37% to 82%) and metformin (25% to 90%). The prevalence of major congenital malformations was higher in the T2DM study population compared to the full pregnancy cohort (5.28% vs. 3.76%, respectively).
Within the cohort with T2DM, the prevalence of congenital malformations was lowest among those without exposure to any ADM (4.77%) or with exposure only to metformin (5.32%), with higher prevalence among all groups with some exposure to insulin (7.83%) or a second-line, non-insulin ADM (6.14% to 9.71%). Similar trends were found for cardiac congenital malformations. However, meta-analyzed relative risks for congenital malformations in those exposed to a second-line, non-insulin ADM compared to those exposed to insulin, adjusted for maternal age and comorbidities, did not demonstrate increased risk of major congenital malformations.
Periconceptual exposure to sulfonylureas showed a non-statistically significant increased risk at 1.18 (95% CI, 0.94-1.48), DPP-4 inhibitors demonstrated a non-statistically significant reduction in risk at 0.83 (95% CI, 0.64-1.06), GLP-1 receptor agonists showed a non-statistically significant reduction in risk at 0.95 (95% CI, 0.72-1.26), and SGLT2 inhibitors had a non-statistically significant reduction in risk at 0.95 (95% CI, 0.65-1.46). Similar results were found when analyzing specifically for cardiac malformations.
COMMENTARY
This study provides new data regarding the relative safety of non-insulin, second-line ADMs. In line with prior research, the study found that there is an elevated risk of congenital malformations in all patients with T2DM, relative to the general population.
The findings that exposure to non-insulin, second-line ADMs do not demonstrate an increased risk of malformations relative to insulin is reassuring, given that more pregnancies are likely to be exposed periconceptually to these medications as their use continues to increase. However, animal models have demonstrated that use of GLP-1 agonists may lead to delayed fetal growth and development, and that SGLT2 inhibitor use may increase the risk of renal abnormalities.7 These safety concerns have led to recommendations to stop these treatments preconceptually.
There are limitations to the Cesta et al study, including imprecise risk estimates because of limited exposure to second-line ADMs in the study population, as well as confounding by comorbidity, type of ADM used, and glycemic control.
Additionally, this study does not provide insight into whether second-line ADMs increase the risk of infertility, miscarriage, or other non-malformation adverse effects, such as neonatal hypoglycemia or macrosomia. Therefore, although this study adds to our knowledge about second-line ADM use in the periconceptual period and can provide reassurance to patients whose pregnancies may have been unintentionally exposed to these medications, further research is needed before clinical guidelines are adjusted to include treatment options outside of insulin and metformin in pregnancy.
REFERENCES
- Centers for Disease Control and Prevention. National Diabetes Statistics Report. https://www.cdc.gov/diabetes/php/data-research/?CDC_AAref_Val=https://www.cdc.gov/diabetes/data/statistics-report/index.html
- Gregory ECW, Ely DM. Trends and characteristics in prepregnancy diabetes: United States, 2016-2021. Natl Vital Stat Rep 2023;72:1-13.
- [No authors listed]. QuickStats: Percentage of mothers with gestational diabetes,* by maternal age – National Vital Statistics System, United States, 2016 and 2021. MMWR Morb Mortal Wkly Rep 2023;72:16.
- American College of Obstetricians and Gynecologists’ Committee on Practice Bulletins – Obstetrics. ACOG Practice Bulletin No. 201: Pregestational diabetes mellitus. Obstet Gynecol 2018;132:e228-e248.
- Rossen LM, Hamilton BE, Abma JC, et al. Updated methodology to estimate overall and unintended pregnancy rates in the United States. NCHS Vital and Health Statistics 2023;2:1-29.
- Cesta CE, Rotem R, Bateman BT, et al. Safety of GLP-1 receptor agonists and other second-line antidiabetics in early pregnancy. JAMA Intern Med 2024;184:144-152.
- Muller DRP, Stenvers DJ, Malekzadeh A, et al. Effects of GLP-1 agonists and SGLT2 inhibitors during pregnancy and lactation on offspring outcomes: A systematic review of the evidence. Front Endocrinol (Lausanne) 2023;14:1215356.
