The glucose lowering effects of GLP-1 agonists are well established in subjects with type 2 diabetes, however, these have not been studied prospectively in subjects with type 1 diabetes. The investigators have, therefore, designed this study to investigate the central hypothesis that in patients with type 1 diabetes, Liraglutide has a glucose lowering effect. A major secondary objective of this study is to elucidate the mechanisms responsible for its glucose lowering effects and those involved in reducing the insulin dose. The specific aims of this proposal are:

Hypothesis 1: Treatment with Liraglutide in patients with type 1 diabetes decreases HbA1c, fasting, postprandial and the overall mean glucose concentrations while decreasing the dose of insulin required.

Hypothesis 2: Treatment with Liraglutide in patients with type 1 diabetes decreases basal and postprandial glucagon concentrations and increases basal and postprandial C-peptide concentrations.

Hypothesis 3: Treatment with Liraglutide in patients with type 1 diabetes delays gastric emptying.

The control of glucose homeostasis in subjects with type 1 diabetes is fragile since exogenous insulin cannot compensate for changing requirements and is not precise either in terms of the dose or the bio-availability of the insulin injected. Furthermore, in the near total absence of insulin secretion, the physiological post prandial inhibition of glucagon secretion by the α-cell is also probably deficient in all type 1 diabetics. Thus, there is a need for therapies beyond insulin that can further improve glycemic control and reduce fluctuations in glucose in these subjects. The investigators have recently shown that Liraglutide, a glucagon like peptide (GLP)-1 analogue with duration of action of 24 hours, when added to insulin in subjects with well controlled type 1 diabetes reduces mean and standard deviation of blood glucose, HbA1c and insulin requirements. Since C-peptide concentrations did not alter following Liraglutide, it is likely that the suppression of glucagon may have contributed to this effect. The glucose lowering effects of GLP-1 agonists are well established in subjects with type 2 diabetes, however, these have not been studied prospectively in subjects with type 1 diabetes. The investigators have, therefore, designed this study to investigate the central hypothesis that in patients with type 1 diabetes, Liraglutide has a glucose lowering effect. A major secondary objective of this study is to elucidate the mechanisms responsible for its glucose lowering effects and those involved in reducing the insulin dose. The specific aims of this proposal are:

Hypothesis 1: Treatment with Liraglutide in patients with type 1 diabetes decreases HbA1c, fasting, postprandial and the overall mean glucose concentrations while decreasing the dose of insulin required.

Aim 1.1: To compare the HbA1c, mean fasting, glucose, mean weekly glucose, standard deviation of weekly blood glucose concentrations as recorded by continuous glucose monitoring and the dose of insulin required prior to and following 52 weeks of treatment with 1.8 mg of liraglutide daily.

Aim 1.2: To compare the postprandial glucose concentrations following a test meal before and after 52 weeks of treatment with 1.8 mg of liraglutide daily.

Hypothesis 2: Treatment with Liraglutide in patients with type 1 diabetes decreases basal and postprandial glucagon concentrations and increases basal and postprandial C-peptide concentrations.

Aim 2.1: To compare the basal and postprandial glucagon and C-peptide concentrations following a test meal before and after 52 weeks of treatment with 1.8 mg of liraglutide daily.

Hypothesis 3: Treatment with Liraglutide in patients with type 1 diabetes delays gastric emptying.

Aim 3.1: To compare the gastric emptying as measured by acetaminophen absorption before and after treatment with 1.8 mg of daily subcutaneous liraglutide.

Inclusion Criteria:

1. Type 1 Diabetes on continuous subcutaneous insulin infusion (CSII; also known as insulin pump) or multiple (four or more) injections of insulin per day.
2. Regularly measuring blood sugars four times daily.
3. HbA1c of less than 8.5%.
4. Well versed with carbohydrate counting.
5. Age 18-75 years.
6. BMI 20-40 kg/m2

Exclusion Criteria:

1. Type 1 diabetes for less than 6 months;
2. Coronary event or procedure (myocardial infarction, unstable angina, coronary artery bypass, surgery or coronary angioplasty) in the previous four weeks;
3. Hepatic disease (transaminase > 3 times normal) or cirrhosis;
4. Renal impairment (serum eGFR < 30ml/min/1.73m2);
5. HIV or Hepatitis B or C positive status;
6. Participation in any other concurrent clinical trial;
7. Any other life-threatening, non-cardiac disease;
8. Use of an investigational agent or therapeutic regimen within 30 days of study.
9. history of pancreatitis
10. pregnancy
11. inability to give informed consent
12. history of gastroparesis
13. history of medullary thyroid carcinoma or MEN 2 syndrome.

Out of 69 patients enrolled in the study, 13 patients dropped out from the study before baseline/randomization visit (during the 30 days insulin dose optimization).

Therefore, baseline data is from 29 and 27 patients only.

56 weeks

Safety data is collected for 56 weeks (4 weeks before randomization and 52 after randomization)