Continuous glucose monitoring metrics and pregnancy outcomes in women with gestational diabetes: a secondary analysis of the DiGest trial

OBJECTIVE: Continuous glucose monitoring (CGM) is increasingly used in gestational diabetes mellitus (GDM) but optimal metrics, ranges and targets in this population are undefined. We assessed associations between CGM metrics and pregnancy outcomes in gestational diabetes. RESEARCH DESIGN AND METHODS: During the DiGest study, 425 women with GDM (diagnosed at median (IQR) 25.1 (18.3- 27.7) weeks) and BMI ≥25kg/m2 received a dietary intervention, with masked Dexcom G6 CGM at 29 (n=361), 32 (n=215) and 36 weeks (n=227) gestation. For this secondary analysis, we used logistic regression, receiver-operator-curves and Youden index to assess associations and predictive ability of CGM metrics including pregnancy-specific time-in-range (TIRp; 63-140mg/dL; 3.5-7.8mmol/L) and pregnancy outcomes. RESULTS: CGM metrics at 29 weeks were significantly associated with LGA and SGA. Participants achieving mean glucose <110mg/dL (6.1mmol/L), TIRp ≥90% or pregnancy-specific time-above-range (TARp) <10% at 29 weeks had a significantly lower risk of LGA (OR 0.41 (95%CI 0.22-0.77); OR 0.38 (0.20-0.70); OR 0.39 (0.20-0.73) and SGA (OR 0.26 (0.08-0.79); OR 0.30 (0.10-0.91); OR 0.19 (0.06-0.62)). TARp<10% and mean nocturnal glucose <110mg/dL were associated with reduced odds of preterm birth (OR 0.40 (0.17-0.94); OR 0.42 (0.19-0.97)). A stricter range (63-120mg/dL; 3.5-6.7mmol/L) had similar performance overall, but had no single statistically-robust TIR/TAR target across all outcomes. CONCLUSIONS: In women with GDM, CGM mean glucose <110mg/dL (6.1mmol/L), ≥90% TIRp or <10% TARp using a range of 63-140mg/dL (3.5-7.8mmol/L) at 29 weeks of gestation was associated with a low risk of suboptimal offspring outcomes.