Introduction: Pregnancy-specific thresholds for TIR, TAR and TBR have been validated. However, recommendations on other CGM-derived metrics (GMI, CV, mean glucose) are lacking. We aimed to better assess the association between CGM metrics and neonatal outcomes (NeO).
Methods: In 37 pregnant women with insulin-treated diabetes (21 T1D, 11 T2D, 5 GDM), clinical, metabolic and CGM data at each trimester were prospectively evaluated (June 2018 - September 2023). Mann-Whitney test was used to compare CGM metrics in women with and without adverse NeO. Logistic regression and ROC-curve were used to identify predictive variables.
Results: Adverse NeO were reported in 75% women (90% T1D vs 40% GDM, p=0.038). At 3rd trimester, women with adverse NeO had higher mean glucose, TAR, GMI, CV and lower TIR, with no difference in HbA1c (Table 1). At logistic regression, TAR was the strongest risk factor (OR: 1.142 [1.012-1.289], p=0.032), regardless of diabetes type and HbA1c. The ROC-curve identified an increased risk of adverse NeO for TAR ≥18% (AUC: 0.845 [0.707-0.982], p=0.03, sens 71%, spec 89%).
Conclusion: At 3rd trimester, women with adverse NeO showed higher mean glucose, TAR, GMI, CV and lower TIR, suggesting that all CGM metrics should be evaluated in pregnancy, pending proper identification and validation of pregnancy-specific thresholds. To minimize the risk of adverse NeO, TAR<18% should be targeted.
F. Citro: None. C. Bianchi: Consultant; Novo Nordisk, Merck Sharp & Dohme Corp., Lilly Diabetes. M. Aragona: None. P. Marchetti: Speaker's Bureau; Eli Lilly and Company, Novo Nordisk. A. Bertolotto: None.
EC and MUR under PNRR – M4C2-I1.3. Project PE_00000019 “HEAL ITALIA”
