Reduction in hemoglobin A1c with real-time continuous glucose monitoring: Results from a 12-week observational study

Background: Real-time continuous glucose monitoring (CGM) was studied in 140 adults with diabetes over a 12-week period of home use. Hemoglobin A(1c) (HbA1c) was measured on day I (baseline) and at weeks 6 and 12. Methods: On day 1, participants received the CGM device (STS (R) System, DexCom, Inc., San Diego, CA) and underwent training on proper use. Insertion of the first sensor was performed under staff supervision. Subjects inserted subsequent sensors on their own. After calibration, the device (a 3-day sensor, receiver, and transmitter) provided users with real-time glucose values updated at 5-min intervals, glucose trend graphs, configurable high/low alerts, and a hypoglycemia alarm (<= 55 mg/dL). Study participants were given supplies sufficient for 3 weeks of device use. Follow-up visits were performed at 3-week intervals for resupply and to download CGM data, with a final visit at the end of week 12. Results: Overall, a reduction in HbA1c of 0.4 +/- 0.05% (least squares mean +/- SE) was observed, P < 0.0001. Significant HbA1c reductions were observed across subgroups of subjects with both type I and 2 diabetes, and those delivering insulin by multiple daily injections and pumps. The largest HbA1c reduction (1.4 +/- 0.4%) was observed in subjects with baseline HbA1c > 9.0%. Increased CGM use was associated with greater reductions in HbAlc. Conclusions: This observational study showed that home use of real-time GCM was safe and well tolerated and associated with a clinically and statistically significant reduction in HbAlc. Large-scale randomized, controlled outcome studies of CGM are indicated.