In addition to point accuracy, continuous glucose monitoring (CGM) devices should be evaluated on trend accuracy to determine whether they provide accurate and timely tracking of glucose fluctuations. We evaluated trend accuracy in a study of the investigational G6 Factory-Calibrated System ("G6 FC", Dexcom), which provides real-time glucose measurements and trend indicators. A total of 76 subjects enrolled (49 children ages 6-17 years and 27 adults) at four U.S. clinical sites. 73 had type 1 diabetes. Participants attended 1 or 2 clinic sessions during which CGM readings were compared to reference venous YSI glucose obtained every 15 minutes. Trend accuracy was assessed using R-deviation, which computes the difference between “instantaneous” rates of glucose change for the CGM and the reference rate of glucose change. Approximately 94% of data had less than ±1 mg/dL/min difference from the reference rate of change (95% for adults vs. 93% for children). When YSI glucose values were falling rapidly (<-2 mg/dL/min), the G6 FC showed that glucose values were falling moderately to rapidly 83% of the time; when the glucose values were rising rapidly (>2 mg/dL/min), the G6 FC showed that glucose values were rising moderately to rapidly 91% of the time (Table). Accurate tracking of glucose trends with the novel G6 FC CGM system will provide additional assurance to the patients relying on these data for diabetes management decisions.
G6 FC System's Concurrence on Rate of Change
| CGM Rate of Change (mg/dL/min) | YSI Rate of Change (mg/dL/min) | Matched Pairs (N) | |||||
| < -2 | [-2, -1) | [-1, 0) | [0, 1) | [1, 2] | >2 | ||
| < -2 | 45.0% | 37.5% | 17.5% | 0.0% | 0.0% | 0.0% | 40 |
| [-2, -1) | 3.6% | 59.5% | 33.6% | 2.9% | 0.4% | 0.0% | 274 |
| [-1, 1) | 0.2% | 3.9% | 57.2% | 34.8% | 3.6% | 0.3% | 2707 |
| [1, 2] | 0.0% | 0.8% | 1.6% | 36.2% | 50.2% | 11.3% | 257 |
| > 2 | 0.0% | 0.0% | 2.4% | 6.6% | 27.7% | 63.3% | 166 |
| Matched Pairs (N) | 33 | 285 | 1656 | 1054 | 274 | 142 | 3444 |
| CGM Rate of Change (mg/dL/min) | YSI Rate of Change (mg/dL/min) | Matched Pairs (N) | |||||
| < -2 | [-2, -1) | [-1, 0) | [0, 1) | [1, 2] | >2 | ||
| < -2 | 45.0% | 37.5% | 17.5% | 0.0% | 0.0% | 0.0% | 40 |
| [-2, -1) | 3.6% | 59.5% | 33.6% | 2.9% | 0.4% | 0.0% | 274 |
| [-1, 1) | 0.2% | 3.9% | 57.2% | 34.8% | 3.6% | 0.3% | 2707 |
| [1, 2] | 0.0% | 0.8% | 1.6% | 36.2% | 50.2% | 11.3% | 257 |
| > 2 | 0.0% | 0.0% | 2.4% | 6.6% | 27.7% | 63.3% | 166 |
| Matched Pairs (N) | 33 | 285 | 1656 | 1054 | 274 | 142 | 3444 |
T. Johnson: Employee; Self; Dexcom, Inc. X. Zhang: Employee; Self; Dexcom, Inc. A. Balo: Employee; Self; Dexcom, Inc..
