- βCGM mean glucose level independently predicted maternal complications with strong accuracy (AUC 0.818)
- βA simple average CGM threshold of ~87 mg/dL helped identify high-risk pregnancies
- βSkin autofluorescence and oxidative stress tests were shown to be non-predictive, helping focus resources on more effective tools
| Journal | Journal of clinical medicine |
| Year | 2025 |
| Authors | Kakuto, Ohara, Koide |
| PMID | 41464700 |
What Was Studied and Why It Matters
Researchers in Japan wanted to find better tools for identifying which pregnant women are most at risk for complications from gestational diabetes (GDM) β a type of diabetes that develops during pregnancy. They compared two approaches: a painless skin scan called skin autofluorescence (SAF) and continuous glucose monitoring (CGM), a wearable sensor that tracks blood sugar around the clock.
This matters because gestational diabetes affects roughly 1 in 7 pregnancies worldwide, and catching high-risk cases early can lead to better care for both mother and baby.
How the Study Worked
The research team followed 115 pregnant Japanese women who had already shown elevated blood sugar on an initial screening test (called a glucose challenge test). At around 29 weeks of pregnancy, each woman had her skin scanned for a measurement called SAF β which reflects long-term sugar buildup in body tissues β and had blood drawn to check for signs of oxidative stress (cellular damage linked to high blood sugar). A smaller group of women also wore CGM devices for 14 days so researchers could track detailed blood sugar patterns.
What the Findings Showed
The headline result was clear: the skin scan (SAF) did not predict who would develop complications. Neither did the oxidative stress blood test. However, the story was very different for CGM.
- Women who wore CGM sensors and experienced maternal complications had significantly higher average blood sugar levels.
- A CGM metric called mean glucose level (MGL) was the only independent predictor of maternal adverse events.
- Women with an average CGM glucose above approximately 86.8 mg/dL had more than 10 times the odds of experiencing a maternal complication.
- The CGM-based prediction model was strong, with an accuracy score (AUC) of 0.818 β considered very good in medical research.
Importantly, no single test β including CGM β reliably predicted complications for newborns, which suggests neonatal risk may depend on other factors not yet fully understood.
What This Means for Real Patients
If you are pregnant and have been flagged for elevated blood sugar, this research suggests that wearing a CGM device could give your care team much richer information than a one-time skin scan or blood test. Rather than just knowing your fasting glucose or A1C, your doctor could see how your blood sugar behaves throughout the day and night β after meals, during sleep, and during activity.
For women managing gestational diabetes, reliable CGM supplies are essential. MDS Diabetes offers a range of CGM sensors and diabetes management supplies that can support continuous monitoring throughout pregnancy, making it easier to stay within healthy glucose ranges.
Talk to your obstetrician or endocrinologist about whether CGM is appropriate for your pregnancy, especially if your initial glucose screening came back elevated.
Bottom Line
A painless skin scan designed to measure long-term sugar buildup did not predict pregnancy complications in this study β but continuous glucose monitoring did. An average CGM blood sugar above about 87 mg/dL was strongly linked to maternal complications. For pregnant women with abnormal glucose screening results, CGM appears to be the most useful risk-assessment tool currently available, and daily monitoring with the right supplies may make a meaningful difference in outcomes.