Collection of continuous glucose monitoring and reference glucose values to develop and test algorithms for a smart continuous glucose monitor.

In recent years, Continuous Glucose Monitoring (CGM) systems have become
available for patients with diabetes. Most CGM sensors measure glucose
subcutaneously in the interstitial fluid. This provides the least discomfort
for patients and is a suitable method for CGM at home. CGM is an essential part
in the development of artificial pancreas systems (AP-systems), in which the
CGM data can be used by an algorithm to automatically determine and administer
the appropriate amount of insulin needed to establish euglycemia by an insulin
pump. Several initiatives are underway to develop these AP-systems.
Inaccuracy of CGM hampers the development of these AP-systems however, in part
because there is a lag between blood glucose and subcutaneously measured
glucose by means of CGM. This lag can cause suboptimal control by the
AP-system. Therefore this lag needs to be addressed. One method of addressing
this lag is by adjusting the algorithm within CGM*s to take this lag into
consideration. To improve this algorithm data is needed comparing CGM data to
blood glucose data. During this trial we will collect this data using one of
the leading CGM brands of sensors: The Seven Plus sensor. The DexCom Seven Plus
sensor exploits the glucose-oxidase principle, and measures an electrical
current which is related to the interstitial glucose (IG) concentration. The
current is transformed into a glucose concentration using a calibration
procedure that exploits one or more blood glucose (BG) references. The current
and the BG references are collected in different sites and the existence of a
dynamic system between them is proved. If the dynamic system is not taken into
account, the output of the sensor could lack in accuracy.

The aim of this study is the collection of clinical data to develop and test
calibration, filtering and prediction algorithms to be embedded in the
so-called *smart sensor* for the enhancement of continuous glucose monitoring
outcomes. In particular, the collected data will be used for the identification
of the parameters, i.e. the gradient and time constant, which characterize the
BG-to-IG dynamic model.

The study is an open-label study. The study will include a total of 12 patients
(3 in each of the four participating clinical centers of the consortium).
Drop-outs will be replaced. The study will take 7 days. On 6 days the patient
will wear the CGM in his/her own surroundings (e.g. at home), on 1 day the
patient will be admitted to the clinical research unit.

Patients will visit the CRC on day 1 and 7 for insertion and removal of the CGM
sensor, estimated duration of these visits will be 2 hours and 30 minutes
respectively. Patients will be admitted to the clinical research ward on day 3,
4 or 5 for a duration of 26 hours. Blood samples will be drawn during the night
and after dinner. Risk to the patient includes haematoma or infection around
the blood collection catheter sites or the CGM insertion site.