Sensor-Augmented Insulin Delivery: Insulin Plus Glucagon Versus Insulin Alone

This study aims to test an insulin and glucagon delivery algorithm designed to be used in conjunction with a continuous glucose monitoring system. This combined glucose sensing/hormone delivery approach is a step on the way to eventual development of an artificial (or automated) pancreas. The insulin and glucagon delivery algorithm is based on the difference between the current blood glucose and the target glucose (proportional error) and the rate of change in blood glucose (derivative error), both adjusted for the recent glucose history. This algorithm is called the Fading Memory Proportional-Derivative (FMPD) Algorithm. The principal investigator of this study has published previous research regarding the use of this algorithm and found it to be well-suited to control blood glucose in type 1 diabetic animals. The addition of glucagon was helpful; better glycemic control with fewer glucose excursions were observed when small intermittent infusions of subcutaneous glucagon were given during times of impending low blood sugar (Ward et al. 2008).

The objective of the current human study is to compare glycemic control in persons with Type 1 Diabetes using the FMPD Insulin plus Glucagon Delivery Algorithm vs. the FMPD Insulin-Alone Algorithm. Subjects will undergo two 28-hour sensor-augmented glycemic control studies. Each subject will be fitted with two short term continuous glucose monitoring systems and two subcutaneous (SC) infusion catheters. These catheters will allow for SC delivery of insulin and glucagon (or insulin plus a glucagon placebo). The accuracy of the wire sensors will be verified every 10 minutes with a venous blood glucose test. For the first 4 hours, the insulin and glucagon delivery will be controlled by venous blood in order to assess and compare the accuracy of the two sensors, after which the more accurate sensor (if it remains accurate) will control the FMPD algorithm. The main outcomes of our study are time spent in the target range (70 - 180 mg/dl) and the percentage of studies requiring intervention due to hypoglycemia (glucose < 70 mg/dl). The accuracy of the sensors over the life of the study will also be evaluated.

The specific system used in this study of frequent blood testing and the use of two separate infusion pumps is not feasible for every day use for individuals with diabetes. However, if the glucose control algorithm (with or without the use of glucagon) provides effective blood glucose management over long time periods the calculation program may be integrated into a continuous blood glucose monitoring system with an insulin and glucagon pump.

• Drug: Insulin, Asp(B28)-
  Insulin dosing and frequency calculated by Fading Memory Proportional Derivative algorithm
  Other Names:

  • Aspart insulin
  • Novolog
• Drug: Glucagon
  During incipient hypoglycemia, glucagon was given in an attempt to prevent overt hypoglycemia. Dosing and frequency was calculated by the Fading Memory Proportional Derivative algorithm
  Other Name: Glucagen
• Drug: Placebo
  Saline solution 0.9%
  Other Name: Normal saline, NaCl 0.9%

• Placebo Comparator: Insulin + Placebo
  Glycemic control of subject participants was managed by the closed-loop system which delivered insulin and normal saline (instead of glucagon) as a placebo, based upon algorithm calculations.
  Interventions:

  • Drug: Insulin, Asp(B28)-
  • Drug: Placebo
• Active Comparator: Insulin + Glucagon
  Glycemic control of subject participants was managed by the system which delivered insulin and glucagon based upon algorithm calculations.
  Interventions:

  • Drug: Insulin, Asp(B28)-
  • Drug: Glucagon
• Experimental: Pilot Study
  Pilot studies designed to assess safety of the system. Includes 6 participants undergoing 7 studies.
  Interventions:

  • Drug: Insulin, Asp(B28)-
  • Drug: Glucagon
  • Drug: Placebo

Inclusion Criteria:

• Age 21-65, history of Type 1 Diabetes Mellitus for > 3 months.

• Women:

  • For women of childbearing potential, a negative urine pregnancy test is required on the first day of the study prior to sensor insertion AND the subject must agree to use contraception prior to and during the study.
  • For menopausal women or those who have undergone surgical sterilization, no pregnancy test or contraception will be required.
• Willingness to attend all clinic visits and participate in two 28-hour studies or one 9-hour study.
• Hemoglobin A1C of 5.0-10%. (Values below 5.0 suggest a severe tendency towards hypoglycemia, and values above 10% suggest severely uncontrolled diabetes with risk for ketoacidosis.)
• Body mass index of 19-35.

Exclusion Criteria:

• Pregnancy, lactation or refusal to use contraception.
• Use of any investigational drug during the 30 days prior to screening.
• Enrollment or participation in any other research studies 30 days prior to and during the entirety of sensor insertion.
• Current alcohol abuse, substance abuse, or severe mental illness (as judged by the Principal Investigator (PI)).
• Any prior cerebrovascular accident or major permanent neurological damage such as aphasia, hemiparesis, or dementia.
• A history of cerebrovascular disease or cardiovascular disease regardless of the time since occurrence.
• Coronary artery disease (symptomatic or asymptomatic) as manifested by unstable angina, acute coronary syndrome, myocardial infarction or therapeutic coronary procedure (e.g., Percutaneous Transluminal Coronary Angioplasty (PTCA), stent placement, Coronary Artery By-pass Grafting (CABG)) within the prior 6 months.
• Any degree of heart failure (as defined by New York Heart Association)..
• Renal insufficiency (serum creatinine of > 2.5).
• Current foot or leg ulceration.
• Peripheral arterial disease with uncontrolled claudication.
• Active uncontrolled malignancy except basal cell or squamous cell skin cancers.
• Concurrent illness, other than diabetic mellitus, that is not controlled by a stable therapeutic regimen.
• Hemoglobin A1C of less than 5.0 or greater than 10%.
• A total bilirubin level above 1.5 mg/dl.
• Medications: Oral or parenteral corticosteroids (glucocorticoid therapy) are exclusions; topical corticosteroids are not.
• Any chronic immunosuppressive medication (such as cyclosporine, azathioprine, sirolimus, or tacrolimus).
• Visual impairment that would prevent reading the display of the Medtronic Guardian® Receiver.
• Physical impairment that would prevent using the buttons of the Medtronic Guardian® Receiver.
• Serum Alanine Transaminase (ALT) or Aspartate Transaminase (AST) ≥3x the upper limit of normal.
• Serum albumin level of < 3.2 g/dl.
• Severe anemia as defined by a hematocrit of < 28%.
• Severe serum electrolyte abnormality (sodium, potassium, carbon dioxide, chloride).
• Cardiac rhythm disturbance characterized by: 2nd or 3rd degree heart block, bradycardia of less than 50 bpm (exception of bradycardia in an aerobic athlete), tachycardia of greater than 100 bpm, or any arrhythmia judged by the investigator to be exclusionary.
• A history of Human Immunodeficiency Virus (HIV) infection.
• An active hepatitis infection.
• Known allergy to any type of insulin
• Insulin resistance requiring more than 200 units of insulin per day
• Known bleeding disorders or chronic usage of warfarin.
• Any known seizure disorder.
• Past history of pheochromocytoma or a family history of Multiple Endocrine Neoplasia (MEN) 2A, MEN 2B, neurofibromatosis, or von Hippel-Lindau disease.
• Hypoglycemic unawareness or chronic hypoglycemia.
• A severe hypoglycemic event which required hospitalization within the past two years.
• Adrenal insufficiency.
• Insulinoma.
• Use of both acetaminophen and ascorbic acid.
• Impaired mentation or psychiatric diagnoses
• Uncontrolled candidiasis.
• Any known allergy to glucagon.

• Juvenile Diabetes Research Foundation
• Oregon Health and Science University