Early continuous glucose monitoring-derived glycemic patterns are associated with subsequent insulin resistance and gestational diabetes mellitus development during pregnancy.
Level 3 - non-randomized controlled study
Prospective observational cohort study
PubMed 39538266 · doi:10.1186/s13098-024-01508-4
What was done
A prospective cohort study in Singapore evaluated 167 pregnant women (mean age 31.7 years, mean BMI 22.9 kg/m², mean gestation 20.3 weeks). Interstitial glucose was measured using continuous glucose monitoring (CGM) for 3–10 days at 18–24 weeks' gestation. Follow-up testing at 24–28 weeks' gestation included fasting serum insulin and oral glucose tolerance testing. Insulin resistance (IR) was defined as updated HOMA-IR ≥ 1.22, and gestational diabetes mellitus (GDM) was diagnosed using 2013 WHO criteria. Associations were evaluated with modified Poisson models adjusted for confounders, and 24-hour interstitial glucose centiles were estimated with Generalized Additive Models for Location, Scale and Shape.
What was found
In this cohort, 25% of women developed IR and 18% developed GDM. After adjustment for confounders, higher early-gestation CGM metrics were significantly associated with subsequent IR risk: - Mean daily glucose: RR 1.42 (95% CI, 1.16 to 1.73) - Glucose management indicator: RR 1.08 (95% CI, 1.03 to 1.12) - J-index: RR 1.04 (95% CI, 1.02 to 1.06) - Standard deviation: RR 1.69 (95% CI, 1.37 to 2.09) - Coefficient of variation: RR 1.03 (95% CI, 1.00 to 1.06) - Mean amplitude of glycemic excursions: RR 1.4 (95% CI, 1.14 to 1.35) Similar associations were observed for GDM risk. Centile curves showed that women with suboptimal glycemic status (IR and/or GDM) had higher 24-hour glucose profiles and wider glycemic fluctuations than women with normal status.
Why it matters
This study shows that continuous glucose monitoring can detect elevated interstitial glucose levels and increased glycemic variability weeks prior to standard 24–28 week GDM screening. It suggests early-gestation CGM could potentially identify women at risk for maternal dysglycemia.
Limits
The sample size was modest (n = 167) and conducted at a single site in a predominantly normal-weight Singaporean population, limiting generalizability. CGM wear duration varied widely (3 to 10 days). The study did not measure maternal or neonatal clinical outcomes beyond biochemical diagnosis.
Cited by
- supports Frequent spiking and crashing of blood glucose can contribute to the development of metabolic dysfunction and eventually pre-diabetes.