Background: Continuous glucose monitoring (CGM) has become an essential tool in diabetes management. In order to use CGM for treatment decisions and to evaluate diabetes therapy, CGM systems have to be reliable over a wide range of glycaemia as well as in situations with rapidly changing glucose levels. 
Aims: The aim of this study was to evaluate the performance of the Enlite 2 sensor calibrated either retrospectively (blinded iPro 2 system) or prospectively (real-time 640G system, both Medtronic) and applied under real-life conditions in a head-to-head comparison.
Method: Ten subjects with type 1 diabetes (age 30.5 ± 10.7 years, 60% female, BMI 24.4 ± 4.6 kg/m2, diabetes duration 11.3 ± 10.7 years, HbA1c 7.0 ± 0.8%) were included in this monocentric study. The study consisted of a 24-hour inpatient phase at the research centre followed by a 5-day home phase. At the research centre routine clinical conditions were mimicked by meal and exercise tests followed by an overnight rest. During the home phase patients were allowed to follow their regular life-style. The sensors were inserted 24 hours prior to the test in parallel and were calibrated according to manufacturers’ instructions. Reference plasma glucose samples were taken every 5 minutes at the research centre (Super GL analyser) and at least 7 times per day during the home phase (Bayer Contour Next glucose meter).
Results: A total of 606 and 926 data pairs were available (iPro and 640G, respectively). Sensor accuracy was determined for each CGM system according to ISO 15197: 2013 guideline (±15%; ± 0.83 mmol/L for glucose <5.55 mmol/L). The systems fulfilled these criteria as follows (overall, day, night): 58.4%, 46.9%, 71.8% (iPro2) and 57.8%, 49.5%, 68.7% (640G).
Discussion: 640G showed superior performance in hyperglycaemia, for the other glycaemic ranges both systems showed similar performance. Sensor accuracy was worst during hypoglycaemia for both systems. As compared to day-time better sensor performance was observed during nighttime. Future CGM generations need to be improved regarding accuracy especially in the low glucose range and during dynamic phases.



Clinical Trial Registration Number: DRKS00009604. Supported by: European Commission FP7-305343.

Standardised evaluation of retrospectively vs prospectively calibrated CGM systems at the research centre and at home