Gestational diabetes diagnosis: Does lower glycaemic threshold affect maternal, birth outcomes?

Using a lower glycaemic threshold led to more diagnoses of gestational diabetes, though its impact on maternal and birth outcomes remains uncertain, results of the GEMS* trial showed.

“[T]hose who had been assigned to the lower glycaemic criteria group were more than 2.5 times as likely to receive a diagnosis of and treatment for gestational diabetes as those assigned to the higher glycaemic criteria group when tested in mid-pregnancy,” said the investigators.

“[However,] we found no significant difference in the incidence of birth of a large for gestational age (LGA) infant between the two trial groups.”

Participants were 4,061 women (mean age 31.5 years, median BMI 26.6 kg/m²) at 24–32 weeks (median 27.3 weeks) gestation with singleton pregnancies. They were randomized 1:1 to be assessed for gestational diabetes using either higher or lower glycaemic criteria for diagnosis. The higher glycaemic criterion was fasting plasma glucose (FPG) level ≥99 mg/dL or a 2-hour level of ≥162 mg/dL, while the lower glycaemic criterion was FPG level ≥92 mg/dL, a 1-hour level of ≥180 mg/dL, or a 2-hour level of ≥153 mg/dL.

Diabetes mellitus or a history of gestational diabetes were exclusion criteria. About 34–36 percent had a family history of diabetes.

Among the participants, 15.3 and 6.1 percent were diagnosed with gestational diabetes according to the lower and higher glycaemic criteria, respectively. Women diagnosed with gestational diabetes received usual management for the condition including nutritional therapy, blood glucose monitoring, and pharmacologic treatment if required.

A total of 2,019 and 2,031 infants were born to women in the lower and higher glycaemic criteria groups, respectively. Women in the lower and higher criteria groups had a comparable likelihood of delivering an LGA** infant (8.8 percent vs 8.9 percent; adjusted relative risk [adjRR], 0.98, 95 percent confidence interval, 0.80–1.19; p=0.82). [N Engl J Med 2022;387:587-598]

There was no significant between-group difference for infant anthropometric measures at birth such as weight, length, head circumference, small-for-gestational-age (SGA) status, or macrosomia, or for gestational age at birth or preterm birth. Hypoglycaemia at birth was diagnosed and treated in more infants in the lower than higher criteria group (10.7 percent vs 8.4 percent; adjRR, 1.27).

This last finding may have been due to infants being born to mothers with gestational diabetes being identified as needing screening for hypoglycaemia, the investigators noted.

Women in the lower glycaemic criteria group were more likely to undergo induced labour than those in the higher criteria group (33.7 percent vs 30.2 percent; adjRR, 1.12), as well as require pharmacologic therapy for gestational diabetes (10.9 percent vs 4.6 percent; adjRR, 2.40). They also had more exposure to health services, though this did not appear to lead to maternal benefits, said the investigators.

Subgroup analysis in women with “in between” criteria

A prespecified subgroup analysis was conducted in women in both groups whose glucose levels were between the lower and higher glycaemic criteria (195 and 178 women, respectively, did and did not receive treatment for gestational diabetes). The birth of LGA infants was less likely among women in the lower vs higher criteria groups (6.2 percent vs 18.0 percent; adjRR, 0.33), with four women needing to be diagnosed and treated for gestational diabetes to prevent one LGA birth.

Incidence of macrosomia and shoulder dystocia and mean gestational age at birth were lower in the lower vs higher criteria group, while neonatal hypoglycaemia incidence was higher. Serious health outcomes were documented in 0.5 and 3.9 percent of infants in the lower and higher criteria groups, respectively.

Gestational weight gain was lower in the lower vs higher criteria group (10.0 vs 11.9 kg), as was pre-eclampsia incidence (0.5 percent vs 5.6 percent). Conversely, induced labour was more common in the lower vs higher criteria group (56.9 percent vs 30.3 percent), as was pharmacologic treatment for gestational diabetes (63.6 percent vs 2.3 percent).

Moving forward

Gestational diabetes can lead to problems both for the pregnant woman (eg, induced labour, Caesarean delivery, pre-eclampsia) and the infant (eg, LGA, operative birth, shoulder dystocia, birth injuries), said the investigators.

“Treatment for gestational diabetes improves mothers’ and babies’ health but it has been unclear what level of blood sugar should be used to make the diagnosis,” remarked lead investigator Professor Caroline Crowther from the Liggins Institute, University of Auckland, Auckland, New Zealand.

The results showed that the greater number of gestational diabetes diagnoses and treatment with the lower glycaemic threshold did not lead to improved health benefits.

“[However,] our subgroup analysis suggests clinically important, short-term maternal and infant health benefits for the women who received a diagnosis of a milder degree of gestational diabetes and also received treatment,” the investigators said. Follow up is warranted to assess if these short-term outcomes have long-term implications.