The HypoDE study, a randomized multi-center trial, showed that rtCGM use reduces the number of low glucose events (<55 mg/dl for at least 20 minutes) per 28 days from 10.4 to 3.4 events compared to SMBG (13.5 to 13.2 events) in MDI-treated type 1 diabetic patients with hypoglycemia problems. In this post-hoc analysis, we analyzed which baseline variables predict successful reduction of low glucose events. The criterion was a reduction to 0 events or by 50% from baseline. In a block-wise, multivariate logistic regression, demographic and medical variables, patient-reported-outcomes, biochemical hypoglycemia at baseline (based on 4-week masked rtCGM use), hours per day in time in range and hyperglycemia, and use of rtCGM were used as predictors. Model fit of the block-wise regression suggests that none of the baseline characteristics did predict hypoglycemia avoidance. However, entering use of rtCGM led to a rise of Nagelkerke´s R² from 0.061 to 0.331. This indicates that rtCGM use was the most important predictor for hypoglycemia avoidance. In the CGM Group, 76.1% of the patients compared to 26.3% in the Control group could avoid hypoglycemic events (Odds ratio 9.8, 95% CI 4.2 to 23.0). This suggests that rtCGM use is effective in patients with type 1 diabetes and hypoglycemia problems regardless of diabetes duration, glycemic control, severity of hypoglycemia problems, and exposure to biochemical hypoglycemia at baseline.

Predictors of hypoglycemia avoidance

 Characteristics Nagelkerke´s R² Delta - Nagelkerke´s R² 
Block 1 Demographic variables (age, diabetes Duration) 0.005  
Block 2 Block 1 + medical variables (HbA1c, severe hypoglycemia in the past 12 month, hypoglycemia unawareness 0.022 0.017 
Block 3 Block 2 + Patient reported outcomes (fear of hypoglycemia, diabetes distress, satisfaction with glucose monitoring, self-reported health status) 0.029 0.027 
Block 4 Block 3 + biochemical hypoglycemia at baseline (time spent ≤70 and ≤55 mg/dl per day, baseline low glucose Events) 0.058 0.009 
Block51 Block 4 + time in range and time in hyperglycemia (>180 mg/dl) 0.061 0.003 
Block 6 Block 5 + Treatment (rtCGM vs. SMBG) 0.331 0.270 
 Characteristics Nagelkerke´s R² Delta - Nagelkerke´s R² 
Block 1 Demographic variables (age, diabetes Duration) 0.005  
Block 2 Block 1 + medical variables (HbA1c, severe hypoglycemia in the past 12 month, hypoglycemia unawareness 0.022 0.017 
Block 3 Block 2 + Patient reported outcomes (fear of hypoglycemia, diabetes distress, satisfaction with glucose monitoring, self-reported health status) 0.029 0.027 
Block 4 Block 3 + biochemical hypoglycemia at baseline (time spent ≤70 and ≤55 mg/dl per day, baseline low glucose Events) 0.058 0.009 
Block51 Block 4 + time in range and time in hyperglycemia (>180 mg/dl) 0.061 0.003 
Block 6 Block 5 + Treatment (rtCGM vs. SMBG) 0.331 0.270 

Disclosure

N. Hermanns: Speaker's Bureau; Self; Berlin-Chemie AG. Advisory Panel; Self; Abbott. Research Support; Self; Abbott. Speaker's Bureau; Self; Abbott. Research Support; Self; Berlin-Chemie AG. Advisory Panel; Self; Eli Lilly and Company. Speaker's Bureau; Self; Eli Lilly and Company. Advisory Panel; Self; Roche Diabetes Care Health and Digital Solutions. Research Support; Self; Ypsomed AG, Dexcom, Inc.. Speaker's Bureau; Self; Novo Nordisk Inc. L. Heinemann: Stock/Shareholder; Self; Profil Institute for Metabolic Research, ProSciento. Consultant; Self; Roche Diabetes Care Health and Digital Solutions. G. Freckmann: Speaker's Bureau; Self; Ascensia Diabetes Care. Research Support; Self; Ascensia Diabetes Care. Speaker's Bureau; Self; Roche Diabetes Care Health and Digital Solutions. Advisory Panel; Self; Roche Diabetes Care Health and Digital Solutions. Research Support; Self; Roche Diabetes Care Health and Digital Solutions. Advisory Panel; Self; Abbott, Novo Nordisk Inc.. Consultant; Self; Sensile Medical AG. Speaker's Bureau; Self; Ypsomed AG. D. Waldenmaier: None. D. Ehrmann: Speaker's Bureau; Self; Berlin-Chemie AG.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.