We retrospectively assessed CGM profiles of T1DM patients. Data were collected from four clinical centers (Medical Department M, Aarhus University Hospital, Denmark; Profil Institute for Metabolic Research, Neuss, Germany; German Diabetes Research Institute at the Heinrich–Heine University of Duesseldorf, Germany; and Department of Pharmaceutical Technology and Biopharmacy, University Center of Pharmacy, University of Groningen, The Netherlands). All centers participated in the clinical development of the SCGM 1 system (Roche Diagnostics, Mannheim, Germany).
A total of 22 subjects were included in the study. Subjects were recruited from their respective outpatient clinics. The data sets of these subjects were collected from a larger population of more than 200 series in two phases. First, other researchers unfamiliar with the present study’s aim and methods selected data for further evaluation based on the following criteria: (1) sufficient technical quality of the measurements, (2) elimination of artifacts, studying all recordings manually, (3) T1DM, and (4) data collection in inpatient setting, resulting in a pool of 91 subjects with relevant data quality and characteristics regarding CGM data. Next, we inspected records of insulin and carbohydrate intake to discriminate between subjects eligible for analysis and subjects with insufficient data amounts. During datacollection, all subjects were encouraged to live their normal everyday lives, maintaining and controlling their normal therapy (primarily insulin) as usual, using their own glucose meters. They were further encouraged to maintain their normal level of physical activity on all study days, walking or indoor cycling in the ward. They were not given access to CGM or other experimental data during data collection.
All subjects received written and oral information according to the Declaration of Helsinki II and signed consent forms. The study was approved by the local ethics committees of the four centers participating in the study and was performed according to Good Clinical Practice guidelines.
SCGM 1 System
The SCGM 1 system is based on the glucose oxidase principle and consists of a sensor unit device and a belt-held sensor holding the microdialysis system. The systemallows up to 120 h of minutely dialysate glucose measure-ments. Data are stored by custom-designed software, and online display of dialysate glucose is transferred wirelessly from the sensor unit to the portable data manager. Additional information (insulin administration, meals, exercise, etc.) can be entered as separate events in the data managing device. The sensor unit uses a roller pump that provides a push–pull flow, resulting in a perfusion of the microdialysis membrane with 0.3 ml/min.The perfusion fluid (Ringer chloride, sodium ion, 147 mmol/liter; potassium, 1.4 mmol/liter; serum calcium, 2.3 mmol/liter; chloride, 156 mmol/liter, pH 6; osmolality, 290 mosmol/kg) passes through the catheter, achieving approximately 95% equilibration with the interstitial fluid. Glucose oxidase is mixed with the dialysate and passes the ex vivo sensor, creating a current in the nano-ampere range. The current is averaged over 60 s, and data are stored.
The microdialysis probe was inserted into the subcutaneous abdominal adipose tissue after skin puncture with a 16 G needle. At the end of each experiment, the last half hour of in vivo measurement was discarded to avoid inclusion of data derived after the explantation of the catheter. Subsequently, the membrane was placed in glucose of a known concentration, and repeated calibration procedures were performed to assess the individual lag time of each catheter.
In order to calibrate the dialysate glucose values to capillary blood glucose, spot measurements were performed up to 20 times per day as described later. On the basis of the spot measurements performed throughout the experiment and the lag time-corrected (inherent physical microdialysis lag time of 31 min) corresponding interstitial values, a linear regression approach was used to calibrate the system.
Hemoglobin A1c was measured by high-performance liquid chromatography at all sites (normal range 4.8–6.2). Spot measurements of capillary blood glucose were performed by the glucose oxidase method on a Glucotrend™ blood glucose monitoring device (Roche Diagnostics, Mannheim, Germany).
Data Analysis and Statistics
The sensor glucose profiles were calibrated by fitting the paired meter data and sensor data to a line and adjusting the sensor data to the gain and offset identified by the fitting.
Meals were detected from the recordings of carbohydrate and insulin intake. The highest carbohydrate intake close (± 2 h) to the times 8 a.m., 12 a.m., and 6 p.m. was classified as a main meal. Two events of carbohydrate intake separated by less than 15 min, of which one was a main meal, were summed to one main meal, whereas main meals were excluded if subsequent meals followed within 16–120 min. Subjects with less than three includable main meals in the recording period were excluded.
For each main meal, peak time was identified as the time elapsed between meal time and the time of the highest CGM value in the interval 40–150 min after meal start.
Interindividual variation in mean peak time was calculated using one-way Anova (including only the first three meals of each subject). Intraindividual variations in peak time were calculated as the intraclass correlation coefficient (ICC), one-way random (including only the first three meals of each subject). All statistical analyses were done using SPSS (SPSS 19, IBM, Chicago).