Continuous Glucose Monitoring in Adults With Type 1 Diabetes With 35 Years Duration From the DCCT/EDIC Study
Research Design and Methods: CGM-derived metrics were compared for daytime and nighttime periods using blinded CGM for a minimum of 6.5 days (average 11.9 days) and correlated with HbA1c levels, routine use of diabetes devices, and other characteristics in 765 participants.
Results: Participants were 58.9±6.5 years of age with diabetes duration 36.8±4.9 years and HbA1c 7.8±1.2%; 58% used insulin pumps and 27% used personal, unblinded CGM. Compared to daytime, nighttime mean sensor glucose was lower, % time in range 70-180 mg/dL (TIR) was similar and hypoglycemia more common. Over the entire recording period, only 9% of the 765 participants achieved >70% TIR and only 28% achieved <1% of observations <54 mg/dL. Indeed, participants with the highest percentage of hypoglycemia had the lowest HbA1c levels. However, use of insulin pumps and CGM decreased the % time <54 mg/dL.
Conclusions: In adults with long-standing type 1 diabetes, short-term blinded CGM profiles revealed frequent clinically-significant hypoglycemia (<54 mg/dL) during the night and more time in hyperglycemia during the day. The small subset of participants using routine CGM and insulin pumps had fewer hypoglycemic and hyperglycemic excursions and lower HbA1c levels. Thus, strategies to lower meal-stimulated hyperglycemia during the day and prevent hypoglycemia at night are relevant clinical goals in older type 1 diabetes patients.