Abstract

Objective: To compare the numerical and clinical accuracy of four continuous glucose monitors (CGM): Guardian^RT, DexCom^TM, Navigator^TM, and Glucoday®.

Research Design and Methods: Accuracy data for four CGM were collected in two studies: Study 1 enrolled 14 adults with Type 1 Diabetes (T1DM) at the University of Virginia (UVA); Study 2 enrolled 20 adults with T1DM at the PROFIL Institute for Metabolic Research, Neuss, Germany. All participants underwent hyperisulinemic clamps including 1.5-2 hours of maintained euglycemia at 5.6 mmol/l followed by descent into hypoglycemia, sustained hypoglycemia at 2.5 mmol/l for 30 minutes, and recovery. Reference blood glucose was sampled every 5 minutes. The UVA study tested simultaneously Guardian^RT, DexCom^TM, and Navigator^TM; the PROFIL study tested Glucoday®.

Results - Numerical accuracy: During euglycemia the mean absolute relative differences (MARD) of Guardian^RT, DexCom^TM, Navigator^TM, and Glucoday® were 15.2%, 21.2%, 15.3%, and 15.6% respectively. During hypoglycemia, the MARD were 16.1%, 21.5%, 10.3%, and 17.5% respectively.

Clinical accuracy: Continuous Glucose Error-Grid Analysis (CG-EGA) revealed 98.9%, 98.3%, 98.6%, and 98.7% A+B-zone hits in euglycemia. During hypoglycemia, the A+B-zone hits were 84.4%, 97.0%, and 96.2% for Guardian^RT, Navigator^TM, and Glucoday® respectively. Due to frequent loss of sensitivity, there were insufficient hypoglycemic DexCom^TM data to perform CG-EGA.

Conclusions: The numerical accuracy of Guardian^RT, Navigator^TM, and Glucoday® was comparable, with advantage to the Navigator^TM in hypoglycemia; the numerical errors of the DexCom^TM were approximately 30% larger. The clinical accuracy of the four sensors was similar in euglycemia and was higher for the Navigator^TM and Glucoday® in hypoglycemia.