Optimising mealtime insulin bolusing algorithms for dietary fat to improve glycaemic control in adults with type 1 diabetes.

Part 1: Relationship Between Amount and Type of Dietary Fat and Postprandial Glycaemia A randomised, within-subject trial comparing capillary postprandial glycaemia for meals of varying amounts and types of fat over 5h with the same insulin dose for all meals (based on the insulin: CHO ratio) in 20 adults with T1D using insulin pump therapy. Dietary Fat: * 4 amounts of fat (0g, 20g, 40g, 60g) added to 45g CHO * 3 different types of fat (saturated, monounsaturated and polyunsaturated fat) added to 45g CHO Part 2: Optimal Prandial Insulin Bolusing Strategy for Dietary Fat A randomised, within-subject trial in the same participants to identify the optimal insulin bolus dose and delivery pattern, using the Model Predicted Bolus, for meals of varying fat content and amount in 20 adults with T1D using insulin pump therapy. * 1 type of fat (Monounsaturated fat) x 3 amounts of fat (20g, 40g, 60g) added to 45g CHO Design Rationale: The types and amounts of fat, for which the ICR did not provide adequate insulin coverage in Part 1 (assessed as blood glucose levels outside target range of 4-10mmol/L), will then receive the same test meals with a MPB in the same participants (Predicted to be meals B, C and D). Using a previously validated MPB to determine the optimised subcutaneous insulin dose allows us to achieve tight postprandial glycemic control that is directly applicable to the clinical setting. The bolus delivery pattern for the MPB will be derived from the data from Part 1. Model Predicted Bolus (MPB): Our model uses ‘Model Predictive Control’ to identify the acute changes in physiological parameters affected by meal composition and insulin dosing and predicts the optimal insulin bolus amount and pattern to achieve the minimal area above 10 mmol/L and below 4 mmol/L. In simple terms, the known variables (the macronutrient content of the meal, the insulin dose given and resulting blood glucose results) are used to determine individual physiological parameters ( the unknown variable) in Part 1. The model then 'rearranges the formula' to predict the optimal insulin dose (i.e the MPB, unknown variable) for the known individual physiological parameters, meal macronutrient composition and desired blood glucose result. This MPB can then be validated in people with T1D to confirm the efficacy of the prediction. The model has been previously been validated for predicting T1D mealtime insulin in the following paper: Bell KJ, Toschi E, Steil GM, Wolpert HA. Optimized Mealtime Insulin Dosing for Fat and Protein in Type 1 Diabetes: Application of a Model-Based Approach to Derive Insulin Doses for Open-Loop Diabetes Management. Diabetes Care, 2016; 39(9):1631-4. Test Meals & Insulin doses Part 1: Insulin doses calculated using participants' insulin: CHO ratio (standard clinical practice). Test Meal A: 0g monounsaturated fat with 45g of CHO (avocado on toast) Test Meal B: 20g monounsaturated fat with 45g of CHO (avocado on toast) Test Meal C: 40g monounsaturated fat with 45g of CHO (avocado on toast) Test Meal D: 60g monounsaturated fat with 45g of CHO (avocado on toast) Test Meal E: 20g polyunsaturated fat with 45g of CHO (margarine on toast) Test Meal F: 20g saturated fat with 45g of CHO (butter on toast) Part 2: Insulin doses calculated using the MPB. Repeat Test Meal B: 20g monounsaturated fat with 45g of CHO (avocado on toast) Repeat Test Meal C: 40g monounsaturated fat with 45g of CHO (avocado on toast) Repeat: Test Meal D: 60g monounsaturated fat with 45g of CHO (avocado on toast) Wash out period: All test sessions must be completed on separate days but may be consecutive days. There is no washout period between testing sessions in Part 1 and Part 2 (i.e the final session of part 1 and the first session of part 2 may be conducted on consecutive days). Study Procedure: Participants will be instructed to avoid alcohol and exercise for 24h prior to the session and not make any manual insulin adjustments (correction bolus or temporary basal rate) after midnight. In the case of hypoglycaemia, participants will be instructed to treat their hypoglycaemia according to their usual clinical care and their test session will be rescheduled. On the day of the test session, participants will arrive at the metabolic kitchen at the Charles Perkins Centre, University of Sydney between 6:30-8:30am after an overnight fast. Two baseline capillary blood glucose tests will be performed on arrival to confirm eligibility to commence the testing session. The fasting BGL must be between 4-10 mmol/L for the session to be commenced. If eligible, capillary blood samples will taken 30, 15 and 0 minutes prior to the consumption of the test meal. The allocated insulin dose will be administered subcutaneously via an insulin pump by the CDE 15 minutes immediately prior to the consumption of the test meal. The test food will be served with 250mL of plain water and participants will be given 12 minutes to consume both the food and water and then no other food or drink will be provided for the remainder of the 5h test session (except water). Capillary blood samples will be collected at 15 min, 30 min, 45 min, 1h, 1.5h, 2h, 2.5h, 3h, 3.5h, 4h, 4.5h and 5h. If hypoglycaemia occurs (> 3.5 mmol/L), the test session will be terminated, the event recorded and the participant treated appropriately.