A 28-year-old woman in her second pregnancy has a history of adult-onset diabetes mellitus.
Q. A 28-year-old woman (gravida 2) presents at 6 weeks' gestation in her second pregnancy. She has a history of adult-onset diabetes mellitus since age 21 and has been taking the oral hypoglycemic agent glyburide since her last delivery under your care. Her hemoglobin A1c value is 8%.
Is oral hypoglycemic treatment of type 2 diabetes mellitus in pregnancy safe for the mother and fetus? If so, which agent should be used?
A. Oral hypoglycemic agents have been shown to be safe and effective in the treatment of type 2 diabetes, which is typically associated with significant insulin resistance, when used in nonpregnant adults.
In the first trimester, a hemoglobin A1c value less than 7% represents an acceptable level of glycemic control and minimizes the probability of adverse outcomes, including fetal malformations and miscarriage.1
Hemoglobin A1c levels between 7.2% and 9.1% are associated with a fetal malformation rate of 14%. Therefore, the patient described above does not have adequate glucose control. Because her current glycemic control is suboptimal and because there is insufficient data on use of oral hypoglycemic agents in women with pregestational diabetes, it is recommended that she switch to insulin therapy in addition to following a strict diabetic diet and exercise plan.2
Glyburide is a second-generation sulfonylurea that lowers blood glucose levels by directly stimulating the insulin release from pancreatic islet beta cells. Until recently, it was believed that the transplacental passage of glyburide was negligible.3 However, recent studies have suggested that there is low-level placental transfer of glyburide and/or its active metabolites to the fetus.4
Metformin, a biguanide, enhances the sensitivity of both hepatic and peripheral tissues to insulin, reducing hepatic glucose output and increasing glucose disposal in muscle and adipose tissue, but it has been shown to readily cross the placenta.5
Recent meta-analyses have not demonstrated an increase in fetal malformations or neonatal deaths with first trimester in utero exposure to oral hypoglycemic agents.6,7
In spite of these encouraging findings, long-term follow-up studies are lacking. Because lengthy follow-up data are unavailable, and because only limited data are available regarding the safety of first-trimester treatment with glyburide, this patient would be advised to switch to insulin therapy even had she presented with a normal hemoglobin A1c.
Insulin treatment can be difficult for patients to adhere to because it is costly, inconvenient, uncomfortable, and can require multiple injections throughout the day. Had the above-described patient or one with adequate glucose control in early pregnancy refused insulin treatment, glyburide could be continued at an increased dose (if the maximum dosage of 20 mg daily had not been achieved) because of the clear benefits of achieving euglycemia. Metformin could be considered as an alternative. However, the patient should be advised that available data regarding the safety and long-term effects of in utero exposure to oral hypoglycemic agents are limited.
Q. Can oral hypoglycemic agents be used for the treatment of gestational diabetes?
A. Although insulin remains the preferred initial treatment for women with gestational diabetes requiring medication for many, some experts have adopted glyburide for select women.8,9 Since gestational diabetes typically reflects an inadequate maternal response to increased insulin resistance in pregnancy, oral hypoglycemic agents may be an appropriate alternative to insulin therapy when gestational diabetes cannot be controlled by diet and activity alone. Because gestational diabetes is typically diagnosed after 20 weeks' gestation, the risk for fetal malformation related to this treatment is low.
Several trials of oral hypoglycemic-agent therapy have been conducted in women with gestational diabetes. In a randomized study, Langer et al found that glyburide was an effective alternative to insulin therapy and that its use was not associated with an increase in neonatal complications in women with gestational diabetes.10 Only 4% of mothers required conversion to insulin for failed oral hypoglycemic therapy. Outcomes including macrosomia, large-for-gestational-age birth weight, perinatal death, and neonatal hypoglycemia were similar between groups, but the power of the study to evaluate these newborn outcomes was low.
A recent meta-analysis including 745 glyburide-exposed and 637 insulin-exposed pregnancies from 9 mostly observational studies also suggested that glyburide therapy was not associated with an increased risk for macrosomia, large-for-gestational-age birth weight, neonatal intensive care unit admission, or neonatal hypoglycemia.11
In a study of 137 women with polycystic ovary syndrome (PCOS) who became pregnant, continuation of metformin treatment significantly reduced the rates of miscarriage, gestational diabetes requiring insulin treatment, and fetal growth restriction.12 Bolton et al compared 66 women with PCOS treated with metformin in the first trimester with 66 control participants matched for age and parity and found more frequent diagnoses of gestational diabetes with metformin (9% vs 0%; P=.012) but similar frequencies of small- (13.6% vs 19.7%) and large- (9.1% vs 15.2%) for-gestational-age babies and of congenital anomalies between the groups.13 Although these investigators claimed less neonatal hypoglycemia (18.5% vs 24.5%) and less need for newborn intravenous glucose therapy (6.3% vs 12%) with metformin, statistical comparisons were not provided.
In a recent randomized, controlled trial comparing the efficacy and safety of metformin versus insulin in patients with gestational diabetes, no increase in perinatal complications for those initially assigned to metformin was found, but 46% required supplemental insulin therapy.14
Because it is prudent to avoid fetal exposure to medications when there is an alternative with less transport to the fetus, it is reasonable to consider initiating treatment with glyburide rather than with metformin. However, the comparative effectiveness of these medicines has not been well studied, and individual circumstances may further influence which medication is initially used in a given patient.