Hormone Responses During Hypoglycemia and the Accuracy of Continuous Glucose Monitors

The present study is being undertaken to compare counterregulatory hormone responses to a mild and gradual reduction in plasma glucose in young children with T1DM versus responses in adolescents. The studies will be performed under the close supervision of the professional staff of each DirecNet center and frequent bedside monitoring of plasma glucose concentrations will ensure that clinically significant hypoglycemia is prevented from developing. All subjects will be admitted to the CRC and have an IV line for blood sampling inserted on the evening prior to study to reduce stress on the morning of the study. The study procedure will be simplified and made less invasive in comparison to a clamp or standard insulin infusion study (i.e. only the one IV for blood sampling will be needed) by limiting enrollment to insulin pump-treated subjects who will have their basal rates modestly increased to produce the hypoglycemic stimulus. Monsod and colleagues used the same procedure of increasing the basal insulin infusion dose to induce a gradual fall in plasma glucose in youth with type 1 diabetes in a study that compared the ability of injections of glucagon and epinephrine to treat mild hypoglycemia. It is particularly important to note that once the blood glucose level falls below 60 mg/dl, a blood sample will be obtained and hypoglycemia will then be immediately corrected by intravenous administration of exogenous glucose. In our recent DirecNet study, ~25% of children and adolescents had plasma glucose levels below 60 mg/dl during a typical night and this rose to ~50% of subjects when there was antecedent exercise in the late afternoon. Moreover, the frequency of mild as well as severe hypoglycemia is substantially higher in pre-school children than in children and adolescents. This safe and rigorously designed study will provide important new information regarding the role of inadequate counter-regulation on the increased risk of hypoglycemia in very young children with T1DM.

Real-time continuous glucose sensing systems offer the potential to markedly lower the risk of hypoglycemia in youth with T1DM. However, the DirecNet inpatient accuracy study demonstrated that the first generation of these devices was inaccurate when blood glucose was lowered to less than 70 mg/dl. In that study, children with T1DM between 3-17 years of age were admitted to the CRC for approximately 26 hours during which they wore 1-2 Medtronic MiniMed CGMS and 1-2 Cygnus GlucoWatch G2 Biographer continuous glucose monitors. In every subject in that study, blood samples were obtained every 30-60 minutes from an indwelling intravenous catheter for measurement of reference plasma glucose levels in the DirecNet Central Laboratory. The Guardian-RT continuous glucose monitoring systems is a real-time continuous glucose monitor that has considerable promise for use in children with diabetes. Therefore, a secondary aim of this study is to obtain very important data regarding the accuracy of this system during hypoglycemia in young children, as well as adolescents.

The Guardian-RT has been approved by the FDA for detecting trends and tracking patterns in adults (18 and older) and are indicated for adjunctive rather than replacement of standard home glucose monitoring devices. The sensor has been approved by the FDA for use for up to 72 hours but can function for a longer period of time.

The primary objective of this study will be to compare the glucose level at which counter-regulatory hormone responses occur during hypoglycemia in young children with diabetes, with the glucose level counter regulatory hormone responses that occur in older children with diabetes. We hypothesize that the children in the younger age group will not have a counterregulatory response until a lower glucose level is reached compared with children in the older age group.

Secondary objectives will be:

1. To assess signs and symptoms at different glucose levels of hypoglycemia in younger children and compare those with the older children. Furthermore, symptoms will be compared to counter regulatory hormone levels in the two age groups. Symptoms will be assessed using physiologic data and using an age-appropriate questionnaire that would be completed during the test by the subject (where appropriate) and by a parent.
2. To assess whether there is a difference in counter-regulatory hormone response in each age group, among those who had at least 2-3 episodes of hypoglycemia per day or night prior to the study compared with those who had an occasional or no episodes of hypoglycemia prior to the test. Subjects will wear a Guardian RT for 6 days (+1 day) prior to the test to assess episodes of hypoglycemia.
3. To examine the accuracy of the Guardian-RT during hypoglycemia in children with type 1 diabetes.

Beginning the Study

When a child enters the study, the following will be done:

1. Informed consent is obtained from eligible subjects (age 3 to <7 or 12 to <18 years, T1D for >1 year, insulin pump being used).
2. On the day of enrollment a hemoglobin A1c is obtained and instructions are given for use of the Guardian RT. The study personnel will supervise the subject or parent inserting the sensor in the clinic. The subject will be instructed to complete at least four glucose measurements a day using the study HGM. Instructions will also be given for response to Guardian RT alarms prior to the CRC admission.
3. The subject will return for an 18-hour overnight CRC admission approximately 6 days (+ 1 day) after the enrollment visit.

• Subjects will continue using the Guardian RT sensor inserted prior to the admission.
• For subjects of sufficient size to accommodate additional devices, a second Guardian-RT sensor will be inserted. An intravenous catheter will be inserted for reference measurements (glucose, epinephrine, norepinephrine, cortisol, glucagon and GH), which will be drawn during the subcutaneous insulin infusion test the following morning to send to a central laboratory.
• For subjects of sufficient weight (subjects >14.9kg at reinfusion centers and >26.3kg at discard centers) to accommodate the volume of blood required, blood glucose measurements will be made every 30 minutes during the admission to allow for assessment of the accuracy of the Guardian-RT.
• For subjects of sufficient weight to accommodate the volume of blood required, blood glucose measurements will be made every 15 minutes for two hours after dinner. This will allow for assessment of the accuracy of the Guardian-RT in detecting change during a period of rising blood glucose.
• At approximately 8:00 a.m. the subcutaneous insulin infusion test will start.
• Prior to starting the test, a HypoMon® may be placed around the subject's chest using an adjustable strap.
• The glucose concentration as measured by the study HGM must be >110 mg/dL to start the test.
• At the start of the test, the basal insulin rate will be increased by approximately 25-50% to provide a gradual decline in blood glucose. A small priming bolus dose of insulin equal to approximately one hour of the subject's usual basal dose may also be given at the discretion of the investigator in addition to the 25-50% increase in the basal insulin.
• The basal insulin rate may be increased an additional amount and additional bolus insulin doses may be given at the discretion of the investigator in order to get a gradual decline in the glucose concentration.
• Blood samples will be collected for the laboratory and glucose will be checked with the study HGM with venous blood every 15 minutes until the glucose level reaches 100 mg/dL. Thereafter, the study HGM will be used to check the glucose levels with venous blood every 5-10 minutes depending on the rate of fall of the glucose level until the end of the study.
• Blood samples will be collected for laboratory determination of hormone concentrations at baseline (before increasing the insulin infusion) and when the glucose levels are <90, <80, <70, and <60 mg/dL.
• Subjects (or parent if appropriate) will be asked questions regarding symptoms of hypoglycemia each time the glucose level is checked with the study HGM.
• Once the endpoint is reached (the first time the glucose is <60 mg/dL using the study HGM), the basal rate will be returned to normal, the subjects will be treated with intravenous glucose and breakfast will be provided.

1. Prior to discharge, the sensors will be removed and downloaded and the subject's insulin pump will be downloaded if possible

About 6 days (+1 day) following the enrollment visit, subjects will have an inpatient CRC admission of approximately 18 hours. Subjects will be admitted at approximately 3:00 PM to allow sufficient time to calibrate the sensors before dinner is provided.

• Areas where a Guardian RT sensor was worn during the first week will be assessed by study personnel for any skin irritation.
• The Guardian RT, HGM, and pump data from the previous week will be reviewed and changes will be made to diabetes management as needed.
• Subjects will continue using the Guardian RT sensor last inserted at home. If the sensor is not functioning properly, a new sensor will be inserted.
• For subjects of sufficient size, an additional Guardian-RT sensor will be inserted and calibrated approximately two hours later. An intravenous catheter will be inserted in an arm vein for collection of blood samples during the admission. The area where the catheter will be inserted may be numbed with Elamax or EMLA cream prior to catheter insertion.

Once the study endpoint is reached (the first time the glucose is <60 mg/dL using the study HGM), the basal rate will be returned to normal and the subjects will be treated with intravenous glucose. An additional blood sample will be collected for laboratory determination of glucose and hormone concentrations 15 minutes following the treatment with intravenous glucose. Subjects will then be given breakfast and discharged. Prior to discharge, the sensor(s) will be removed and downloaded. The subject's insulin pump will be downloaded if possible.