Approximately 7 percent of pregnancies are complicated by diabetes, 90 percent of which are classified as gestational diabetes mellitus, and the incidence is rising. Kelly and colleagues conducted a review summarizing the issues related to gestational diabetes and examining the possibilities of using oral hypoglycemia medications during pregnancy.

A search of electronic databases and a study of expert guidelines revealed mainly retrospective chart reviews, case reports, and consensus development statements. Better quality evidence is anticipated, especially when the Hyperglycemia and Adverse Pregnancy Outcome (HAPO) study results become available.

Controversy exists regarding the definition of gestational diabetes and screening guidelines. Some investigators do not recognize gestational diabetes as a distinct entity but rather as a manifestation of insulin resistance and a risk factor for later overt type 2 diabetes. Expert groups differ in recommendations about screening for gestational diabetes. Most recommend screening for high-risk mothers (e.g., 35 years and older, body mass index greater than 30 kg per m², corticosteroid use, personal or family history of diabetes, high-risk ethnic group), but routine screening, as is common in the United States, is not universally recommended or practiced. The authors of a 2004 Cochrane review found no evidence that screening for gestational diabetes made any difference in perinatal outcomes.

Controversy also exists about the adverse effects of gestational diabetes and the effectiveness of current screening and treatment protocols. Hyperglycemia in the first 10 weeks of pregnancy while organogenesis is taking place (A1C level greater than 7.5 percent) is teratogenic and associated with a two- to threefold increase in congenital malformation and miscarriage. Unfortunately, by the time most women know they are pregnant, much organogenesis has already occurred and women with type 1 or 2 diabetes may not have adjusted their glycemic control accordingly. Women with undiagnosed type 2 diabetes also are at significant risk of hyperglycemic complications. Whatever the cause, mothers who have hyperglycemia in early pregnancy are not identified by current screening at 24 to 28 weeks of gestation.

After 12 weeks of gestation, maternal glucose crosses the placenta, and fetal beta cells can produce insulin. If the maternal glucose level is elevated after 12 weeks of gestation, fetal insulin production increases. The growth hormone effects of this insulin lead to fetal macrosomia. About 20 percent of deliveries in women with gestational diabetes are macrosomic (i.e., birth weight greater than 4,000 g [8 lb, 13 oz]). Treatment of gestational diabetes is associated with a non-significant reduction in macrosomic births and cesarean deliveries, but diagnosis of gestational diabetes increases the likelihood of cesarean delivery (to about 30 percent of mothers with gestational diabetes). In regions of Canada where routine screening was discontinued, macrosomia rates did not change. In contrast to mothers who have type 1 or 2 diabetes, mothers with gestational diabetes do not have elevated rates of congenital malformation compared with the general population. Estimates of perinatal mortality in women with gestational diabetes vary greatly. Current estimates are that perinatal mortality rates in women with gestational diabetes are close to or lower than rates in comparable control patients.

In the majority (60 to 95 percent) of mothers with gestational diabetes, the disease can be managed appropriately with diet alone. For other patients, intensive insulin therapies have been associated with lower rates of macrosomia and cesarean delivery—but increased rates of maternal hypoglycemia (around 20 percent) and risk of neonatal hypoglycemia. Compliance is poor with many regimens; some regimens require blood glucose monitoring seven times per day plus four insulin shots per day. Little evidence is available comparing different insulin regimens.

More recently, glyburide (Micronase), an oral therapy, was approved for use in trials. Researchers found that it did not cross the placenta and achieved results comparable to those of insulin therapy in a large study beginning at 11 weeks of gestation. Most mothers prefer oral therapy, and glyburide has been used successfully to treat mothers whose gestational diabetes was not controlled by diet. Pregnancy outcomes have been excellent in studies of glyburide therapy for gestational diabetes in the second and third trimesters. Older studies suggest that metformin (Glucophage) could be safe and effective in early pregnancy, and new research on this agent is being conducted.

The authors conclude that screening artificially increases the prevalence of gestational diabetes but does not conclusively improve outcomes. The authors hope that the HAPO study of 25,000 pregnancies will answer some of the questions about gestational diabetes. In the meantime, making a diagnosis at 24 to 28 weeks of gestation appears to only modestly lower rates of macrosomia but has little clear additional advantage.