Glyburide Should Not Be First-Line for Gestational Diabetes

Miriam E. Tucker

May 01, 2018

The oral sulfonylurea glyburide should not be considered as a first-line treatment for gestational diabetes, new research suggests.

Results from the first randomized trial to compare neonatal outcomes of women with gestational diabetes treated with oral glyburide versus subcutaneous insulin were published online May 1 in JAMA by Marie-Victoire Sénat, MD, PhD, Department of Obstetrics and Gynecology, Hôpital Bicêtre, Le Kremlin Bicêtre, France, and colleagues. 

The study, designed to demonstrate noninferiority, "failed to show that use of glyburide compared with subcutaneous insulin does not result in a greater frequency of perinatal complications. These findings do not justify the use of glyburide as a first-line treatment," Sénat and colleagues write.

However, the authors also point out that the results suggest the increase in perinatal complications with glyburide — macrosomia, neonatal hypoglycemia, and hyperbilirubinemia — may be no more than 10.5% compared with insulin, and that the risk should be balanced with ease of use and patient satisfaction with glyburide.

"In clinical situations in which an oral agent may be necessary, mothers, informed by their physicians, would be appropriate decision makers based on their own weighing of benefits and risks."

But in an accompanying editorial, obstetrician-gynecologist Donald R. Coustan, MD, of Brown University, Jamestown, Rhode Island, and endocrinologist Linda Barbour, MD, University of Colorado Hospital, Aurora, note another concern, the fact that few data are available on the long-term effects of glyburide on the fetus.

"Whether glyburide levels are high enough to stimulate fetal insulin or whether exposing the fetal pancreas to an insulin secretagogue during critical stages of development would be harmful or helpful is not known," they write.

Given that, and the possible 10.5% increased perinatal complication risk, Coustan and Barbour conclude: "Use of glyburide may be most appropriate when insulin injections are not acceptable or practical. When glyburide or other oral agents that cross the placenta are prescribed for pregnant women, concerns about the equivalence in perinatal outcomes and unanswered questions regarding long-term effects on offspring should be discussed frankly."

Both the American Diabetes Association and American College of Obstetricians recommend only insulin as first-line gestational diabetes treatment, and it's the only gestational diabetes medication approved by the US Food and Drug Administration. Nonetheless, glyburide use in gestational diabetes has become common in the United States (but not in Europe).

Noninferiority Not Proved

The study included 914 women with gestational diabetes who required pharmacologic treatment after 10 days of dietary management randomized to glyburide or insulin.

Glyburide was started at 2.5 mg/day and could be increased by 2.5 mg, if necessary, 4 days later, and by 5 mg every 4 days thereafter, up to a maximum of 20 mg/day. The starting dosage for insulin was 4 to 20 IU of short- and intermediate- or long-acting insulin given 1 to 4 times daily, with adjustments according to self-measured blood glucose levels.

The primary outcome was a composite of macrosomia, neonatal hypoglycemia, and hyperbilirubinemia, with the noninferiority margin set at 7% based on a one-sided 97.5% confidence interval (CI).

That outcome occurred in 27.6% of the glyburide group and 23.4% of the insulin group. The 4.2% difference gave a one-sided 97.5% CI of −∞ to 10.5% (P = .19), with the upper confidence limit exceeding the noninferiority margin of 7%. Results were similar after adjustment for multiparity and gestational age at treatment (4.4% difference; one-sided 97.5% CI, −∞ to 10.5%; P = .20).

Hypoglycemia accounted for most of the difference, occurring in 45% of newborns in the glyburide group versus 32% with insulin (P = .02).

There were no perinatal deaths in the glyburide group and two with insulin (unexplained uterine death and medical termination because of severe brain abnormalities).

Rates of admission to neonatal intensive care didn't differ, nor were there significant between-group differences in birth injury, ponderal index, pH level of less than 7, lactate levels, or respiratory distress syndrome.

Maternal blood glucose control during pregnancy was significantly better in the glyburide group, with 71.7% versus 63.2% with insulin achieving fewer than 20% of fasting values of 95 mg/dL or greater (P = .003) and 57.8% versus 49.3%, respectively, achieving fewer than 20% of postprandial values greater than 120 mg/dL (P = .051).

However, maternal hypoglycemic episodes (< 60 mg/dL) were 25.3% more frequent in the glyburide group (P < .001).

There were no significant differences in mode of delivery, preterm delivery, or perineal trauma.

Of the 68.5% who completed a satisfaction questionnaire during their maternity stay, 78.7% of the glyburide group said they would choose the treatment again for a subsequent pregnancy, versus just 19.9% of those treated with insulin (P < .001).

Was the Dosing of Glyburide Correct?

One issue Coustan and Barbour question is whether glyburide was dosed correctly in the study, citing pharmacokinetic data suggesting that the drug may be underdosed in pregnancy given more rapid clearance.

They point out that glyburide should be given 1 hour before meals because its peak effect is 3 to 4 hours post-ingestion, but that the study authors give no information on the timing of glyburide administration with respect to meals or how high the average patient's glyburide dose was titrated.

The editorialists conclude, "Failure to demonstrate noninferiority does not mean that glyburide is inferior to insulin. However, given that this collaborative group set out to determine if adverse outcomes attributed to glyburide in other smaller or less rigorous studies and systematic reviews could be invalidated by a larger randomized clinical trial, the previously proposed disadvantages of glyburide have not been disproven."

Sénat, Coustan, and Barbour have reported no relevant financial relationships.

JAMA. Published online May 1, 2018. Abstract, Editorial

For more diabetes and endocrinology news, follow us on Twitter and on Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.