Maternal glucose levels that were high but below the diagnostic threshold for gestational diabetes were strongly associated with high fetal insulin levels and birth weights in an international study of 23,316 pregnant women.
There were also weaker—but still significant—associations between maternal hyperglycemia that fell short of overt gestational diabetes and a host of neonatal problems: hypoglycemia in the neonate, the need for cesarean delivery, premature delivery, shoulder dystocia or birth injury, the need for intensive neonatal care, hyperbilirubinemia, and preeclampsia, investigators reported in the New England Journal of Medicine.
These findings “indicate the need to reconsider current criteria for diagnosing and treating hyperglycemia during pregnancy,” reported Dr. Boyd E. Metzger of Northwestern University, Chicago, and his associates in the Hyperglycemia and Adverse Pregnancy Outcome study (N. Engl. J. Med. 2008;358:1991–2002).
The researchers assessed the 23,316 gravid women “to clarify the risk of adverse outcomes associated with degrees of maternal glucose intolerance less severe than overt diabetes mellitus.” The study subjects underwent standard oral glucose tolerance testing at 24–32 weeks' gestation at 15 medical centers in nine countries.
Cord blood specimens were obtained at delivery to assess serum C-peptide levels, an indicator of fetal β-cell function.
High levels of fasting, 1-hour, and 2-hour plasma glucose were strongly correlated with birth weight above the 90th percentile and C-peptide levels above the 90th percentile, and the rates of these problems increased as plasma glucose levels increased.
There were weaker but significant correlations between maternal hyperglycemia and two other primary outcomes of this study (cesarean delivery and clinical neonatal hyperglycemia), as well as five secondary outcomes. A similar dose-response relationship was seen between increasing maternal glucose level and rising rates of these problems, Dr. Metzger and his associates said.
In a separate study of gestational diabetes published in the same issue, Dr. Janet A. Rowan of Auckland City (New Zealand) Hospital and her associates in the Metformin in Gestational Diabetes trial found that metformin was “noninferior” to insulin in safety and efficacy, and was preferred by patients with overt disease.
In that open-label study, Dr. Rowan and her associates compared oral metformin with insulin therapy in 733 women who had overt gestational diabetes and were followed at 10 New Zealand and Australian obstetric hospitals.
The composite outcome of numerous neonatal complications, including hypoglycemia in the infant, was no different between the metformin group and the insulin group, at 32% in both. There also were no differences between the two groups in neonatal anthropometric measures or in umbilical cord serum insulin concentrations.
The women preferred metformin to insulin. However, 46% of those who took metformin eventually required supplemental insulin as well, Dr. Rowan and her associates said (N. Engl. J. Med. 2008;358:2003–15).
Further follow-up data on the offspring are needed to determine the long-term safety of metformin use in pregnancy, they noted.