From the Journals

Glyburide failed to show noninferiority in gestational diabetes

View on the News

Consider dosing, patient selection

The researchers were “reasonable” to conclude that insulin should remain the first-line pharmacotherapy for gestational diabetes, according to Donald R. Coustan, MD, and Linda Barbour, MD, MSPH, whose editorial accompanied the study in JAMA.

“Use of glyburide may be most appropriate when insulin injections are not acceptable or practical,” they wrote. They suggested “frankly” counseling pregnant women about glyburide crossing the placenta and about “unanswered questions regarding long-term effects on offspring.”

Ideally, pregnant women should receive glyburide 1 hour before meals so that its effect peaks 3-4 hours later, according to the experts. But the study authors did not describe treatment timing with respect to meals, did not adjust initial dosing based on fasting or postprandial hyperglycemia, and only increased the dose every 4 days, they noted.

Although insulin was dosed much more flexibly, the glyburide group had better fasting glucose than did controls (72% vs. 63%; P = .003), the editorialists noted. Glyburide is most likely to succeed in younger women without fasting hyperglycemia and whose gestational diabetes begins later in pregnancy. Better dosing and patient selection might make glyburide more effective while also helping prevent maternal hypoglycemia and adverse perinatal outcomes, they contended.

Dr. Coustan is with Brown University, Providence, R.I. Dr. Barbour is with University of Colorado at Denver, Aurora. They reported having no conflicts of interest. These comments paraphrase their editorial ( JAMA. 2018;319[17]:1769-70 ).



A randomized, multicenter trial failed to find glyburide noninferior to insulin for treatment of gestational diabetes, investigators reported.

The composite rate of macrosomia, neonatal hypoglycemia, and hyperbilirubinemia was 27.6% with oral glyburide and 23.4% with subcutaneous insulin (P = .19) therapy, said Marie-Victoire Sénat, MD, PhD, of Hôpital Bicêtre in Paris, and her associates. The upper limit of the 97.5% confidence interval for the difference between groups was 10.5%, exceeding the prespecified noninferiority margin of 7%. “These findings do not justify the use of glyburide as first-line treatment,” the researchers wrote. The report was published online May 1 in JAMA.

Glyburide is a common add-on therapy for gestational diabetes in the United States but is not used regularly in Europe. The treatments exert similar glycemic control, but meta-analyses and recent studies have linked glyburide to increased rates of neonatal macrosomia and hypoglycemia. However, trials comparing glyburide with insulin focused on maternal glycemic control and thus “were not optimally designed to investigate neonatal complications,” the researchers wrote.

For the study, they randomly assigned 914 women whose gestational diabetes persisted despite dietary intervention to receive either 2.5 mg glyburide once daily or 4 IU to 20 IU insulin one to four times daily. Patients up-titrated treatment as needed based on self-measured blood glucose levels. Glyburide first was increased by 2.5 mg on day 4 and thereafter by 5 mg every 4 days in morning and evening doses to a daily maximum of 20 mg. Prandial insulin was increased by 2 IU every 2 days, while basal or intermediate insulin was dosed at 4 IU to 8 IU at bedtime and increased by 2 IU every 2 days.

The difference in the composite endpoint still exceeded 4% between groups even after the researchers controlled for multiparity and gestational age at treatment. Rates of each individual complication were higher with glyburide than with insulin, although only hypoglycemia reached statistical significance (12.2% for glyburide versus 7.2% for insulin; P = .02).

Maternal hypoglycemia affected 3.8% of the glyburide arm and 1% of the insulin arm (P = .02), and 72% of glyburide patients maintained good fasting glycemic control versus 63% of insulin recipients (P = .003). Also, 58% of glyburide recipients had good postprandial glucose control versus 49% of insulin recipients (P = .051).

Questionnaires indicated that patients were more likely to find glyburide tolerable and to report that they would use it again, if needed, during a future pregnancy (P less than .001 for between-group comparisons). “Although the data do not allow a conclusion that glyburide is not inferior to insulin in the prevention of perinatal complications, the results suggest that the increase in complications may be no more than 10.5% compared with insulin,” the investigators wrote. “This result should be balanced with the ease of use and better satisfaction with glyburide.”

Dr. Sénat reported having no conflicts of interest. One coinvestigator disclosed ties to Ferring Laboratories.

SOURCE: Sénat M-V et al. JAMA. 319(17):1773-80.

Next Article: