Real-Time Glucose Monitors Cut Oversized Baby Births in Gestational Diabetes
A major RCT finds rt-CGM reduces large-for-gestational-age births by 68% vs finger-stick monitoring in gestational diabetes.
Summary
The GRACE trial randomized 375 pregnant women with gestational diabetes to real-time continuous glucose monitoring (rt-CGM) or standard self-monitoring of blood glucose (SMBG). Women using rt-CGM had significantly fewer large-for-gestational-age (LGA) newborns — just 4% versus 10% in the SMBG group, a 68% relative reduction. Serious adverse events were similar between groups. However, a higher-than-expected rate of small-for-gestational-age (SGA) infants appeared in the rt-CGM group (19% vs 13%), raising questions about whether tighter glycemic control may overcorrect in some cases. The trial, spanning four university hospitals across Austria, Germany, and Switzerland, provides the strongest evidence yet that rt-CGM meaningfully improves neonatal outcomes in gestational diabetes.
Detailed Summary
Gestational diabetes affects millions of pregnancies worldwide and significantly raises the risk of complications including excessively large babies, cesarean delivery, and long-term metabolic disease in both mother and child. Continuous glucose monitoring has transformed diabetes management in non-pregnant populations, but its benefit in gestational diabetes has remained debated — making the GRACE trial a critical piece of evidence.
Researchers conducted an open-label, multicentre, multinational randomised controlled trial across four university hospitals in Austria, Germany, and Switzerland. Women aged 18–55 with singleton pregnancies and gestational diabetes were randomised 1:1 to wear real-time CGM (Dexcom) through delivery or to perform standard finger-stick blood glucose self-monitoring. The primary outcome was the proportion of large-for-gestational-age (LGA) newborns, assessed using customised GROW birthweight percentiles.
Of 345 participants with evaluable data, only 4% of rt-CGM users delivered LGA infants compared to 10% in the SMBG group — an odds ratio of 0.32 (95% CI 0.10–0.87, p=0.014). This represents a clinically meaningful two-thirds reduction in LGA births. Serious adverse events were not statistically different between arms, suggesting the technology is safe in this context.
A notable and unexpected finding was a higher rate of small-for-gestational-age (SGA) infants in the rt-CGM group (19% vs 13%), though this difference did not reach statistical significance. The authors suggest that aggressive glycemic control may inadvertently restrict fetal growth in some women, a hypothesis requiring dedicated follow-up research.
For clinicians and pregnant women managing gestational diabetes, these findings support the integration of rt-CGM into standard care protocols. The trial also highlights the need to define optimal glucose targets that reduce LGA risk without increasing SGA incidence — a nuanced balance with direct implications for fetal health.
Key Findings
- rt-CGM reduced large-for-gestational-age births by 68% versus finger-stick monitoring (4% vs 10%, p=0.014).
- Serious adverse events were similar between rt-CGM and SMBG groups (12% vs 15%, p=0.39).
- SGA infants were more common with rt-CGM (19% vs 13%), though not statistically significant.
- 375 women with gestational diabetes were enrolled across Austria, Germany, and Switzerland.
- Tight glycemic control from rt-CGM may risk overcorrection, potentially restricting fetal growth.
Methodology
Open-label, parallel-group, multicentre RCT enrolling 375 women with gestational diabetes at four university hospitals across three countries. Randomisation was 1:1 with minimisation to balance gestational age, prior GDM, and preconceptional BMI. Primary endpoint assessed in the intention-to-treat population using GROW customised birthweight percentiles.
Study Limitations
The trial was open-label, introducing potential behavioural bias in glucose management between groups. The higher-than-expected SGA rate in the rt-CGM arm warrants caution and further investigation into optimal glycemic targets. The study was funded by Dexcom, which may influence generalisability to other CGM devices.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.
Enter your email to subscribe: