Login / Signup

Performance of the Dexcom G7 Continuous Glucose Monitoring System in Pregnant Women with Diabetes.

Sarit PolskyAmy M ValentElvira IsganaitisKristin N CastorinoGrenye O'MalleyStayce E BeckPeggy GaoLori M B LaffelFlorence M BrownCarol J Levy
Published in: Diabetes technology & therapeutics (2024)
Background: We evaluated accuracy and safety of a seventh-generation real-time continuous glucose monitoring (CGM) system during pregnancy. Materials and Methods: Evaluable data for accuracy analysis were obtained from 96 G7 sensors (Dexcom, Inc.) worn by 96 of 105 enrolled pregnant women with type 1 ( n  = 59), type 2 ( n  = 21), or gestational diabetes ( n  = 25). CGM values were compared with arterialized venous glucose values from the YSI comparator instrument during 6-h clinic sessions at different time points throughout the sensors' 10-day wear period. The primary endpoint was the proportion of CGM values in the 70-180 mg/dL range within 15% of comparator glucose values. Secondary endpoints included the proportion of CGM values within 20% or 20 mg/dL of comparator values ≥ or <100 mg/dL, respectively (the %20/20 agreement rate). Results: Of the 1739 pairs with CGM in the 70-180 mg/dL range, 83.2% were within 15% of comparator values. The lower bound of the 95% confidence interval was 79.8%. Of the 2102 pairs with CGM values in the 40-400 mg/dL range, the %20/20 agreement rate was 92.5%. Of the 1659 pairs with comparator values in the 63-140 mg/dL range, the %20/20 agreement rate was 92.3%. The %20/20 agreement rates on days 1, 4 and 7, and 10 were 78.6%, 96.3%, and 97.3%, respectively. Consensus error grid analysis showed 99.8% of pairs in the clinically acceptable A and B zones. There were no serious adverse events. The sensors' 10-day survival rate was 90.3%. Conclusion: The G7 system is accurate and safe during pregnancies complicated by diabetes and does not require confirmatory fingerstick testing. Clinical Trial Registration: clinicaltrials.gov NCT04905628.
Keyphrases
  • clinical trial
  • type diabetes
  • pregnant women
  • randomized controlled trial
  • blood pressure
  • machine learning
  • low cost
  • study protocol
  • phase ii