People with Type 1 diabetes (T1DM) are usually provided guidance on how best to control glycaemia around meal times through adjusting their rapid-acting insulin dose according to the carbohydrate content of the meal (e.g. 1 IU per 10/15 g of carbohydrate; Schmidt et al., 2014). However, a potential issue around this method is the role of dietary fat in the calculation of insulin requirements (Wolpert et al., 2013). The fat component of the meal has the potential to influence the insulin dose requirement to normalise postprandial glycaemia (Wolpert et al., 2013). Although normalising postprandial glycaemia is vital, postprandial lipaemia is also an important consideration for long-term health, and at present there is scant data in this area in T1DM. In addition, changing the macronutrient composition of foods and altering insulin doses may carry important implications for vascular function and prospective appetite regulation.

This research will examine the glycaemic and lipaemic responses after consuming a mixed meal similar in carbohydrate content, but differing in fat content. Moreover, this research will assess whether acute postprandial reductions in insulin sensitivity can be offset through increasing the dose of rapid-acting insulin for such meals. Venous blood samples will be collected before and for 6 hours after meals, for the determination of glycaemic and lipaemic responses, as well as metabolite and hormonal parameters. In addition this study will assess the impact of mixed meals and adjusting insulin dose on vascular function and subjective ratings of appetite.

The findings from this study will benefit patients with type 1 diabetes by the provision of more refined self-management strategies for insulin dosage around meal-times.

Study design and Methodology General Design: 18 male or female type 1 diabetes (T1DM) individuals aged between 18 and 50 years old will be invited to attend four laboratory sessions, each separated by 7 days. Participants will be recruited in clinic and through advertising in local media. Participants will complete four main trials in a randomised and counter balanced fashion. Main trials will involve the manipulation of both the meal composition (fat content) and rapid-acting insulin dose. During each visit blood samples will be measured over a 6 hour postprandial period to determine glycaemic, lipaemic, hormone, and metabolite parameters.

Sample size calculation:

Sample size requirement was estimated using Eq. 1 Hopkins (2000):

n = 8s2/d2 [Eq. 1] Where n is the sample size, s is the typical error in measurement and d is the meaningful effect size. The magnitude of d was derived from 0.8 of the between subject variation, which was calculated using existing data. The postprandial incremental area under the curve (AUC) for triglyceride concentrations following a mixed-macronutrient meal in people with T1DM is typically 0.74 mmol.l-1.hour-1, with a between-subject standard deviation of 0.99 mmol.l-1.hour-1 (Levetan et al. 2003). Therefore d was calculated as 0.79 mmol.l-1.hour-1. The CV of this measure derived from repeated trials is 23% (Weiss et al. 2008), giving a typical error in the region of 0.18 mmol.l-1.hour-1. These data indicate a sample size of 18 participants provides greater than an 80% chance of detecting a statistically significant effect with a P value of < 0.05.

Inclusion/exclusion criteria Participants: For inclusion in the study, volunteers will be either male or female and aged 18-50 years old, free from any diabetes complications apart from background diabetic retinopathy, not taking any prescribed medication other than insulin, and be treated with a stable insulin regimen composed of a combination of slow/long acting insulin glargine/determir and a fast acting insulin analogue (lispro or aspart, glulisine), and have a HbA1c of <9.5% (80 mmol/mol). Participants will be currently using the carbohydrate counting method for administering meal time rapid-acting insulin.

Prior to main trials: Participants will be instructed to follow a recommended diet plan for 48 hours prior to testing. Participants will be required to read and sign an informed consent form and medical questionnaire which will be reviewed during each visit to the laboratory. 24 hours prior to each trial, participants will be fitted with a real-time continuous glucose monitoring system (CGM). The CGM consists of a sensor which is inserted into subcutaneous tissue which is secured in place with a waterproof adhesive dressing. Sensor readings are obtained and stored in the memory of the monitor which is attached to the indwelling sensor. Glucose readings are provided on the screen of the monitor in real-time. Alarms will be set on the monitor to notify participants if glucose levels are low (<3.9 mmol/L) or high (>14 mmol/L) during the pre-trial period. For calibration purposes, the participant will be required to input blood capillary blood glucose values a minimum of 4 times per day (at meal times and before bed) using their blood glucose finger-prick meter. The CGM will be removed on the day after the trial at a time convenient for the participant, so that late-evening and nocturnal glycaemia can be captured. The CGM monitor will also act as a safety measure to alert participants to glycaemic excursions following the intervention. Participants will be instructed not to use real-time CGM glucose values to determine corrections to blood glucose (additional carbohydrate / insulin administration).

Main trials: All trials will be conducted in a randomised and counter-balanced fashion. For each visit, the participants will be required to arrive at the NIHR Clinical Research Facility laboratory on a morning (~07:00 - 09:00 AM). On the first visit, anthropometric variables will be collected (mass, stature, BMI) prior to experimentation. On each visit, the participant will assume a seated and rested position whilst an 18 gauge cannula is inserted into the antecubital vein of their non-dominant arm for collection of venous blood. This will be kept patent with periodic infusion of saline. After cannulation is complete, a 12 ml resting blood sample will be collected. Participants will then perform one of four trials:

Trial 1) Carbohydrate only meal: Participants will consume a standardised carbohydrate meal (80 g of carbohydrates, 25 g protein, 0 g fat: meal composition, white rice, chicken, curry sauce; 420 kcal) and will self-administer (into the subcutaneous tissue of the abdomen, as per their regular routine) a rapid-acting insulin dose calculated as per the carbohydrate-counting ratio (e.g. 1 IU of insulin per 10 g of carbohydrates).

Trial 2) Participants will replicate Trial 1, but on this occasion the meal consumed will have an additional 50 g of fat (via addition of Ghee). This fat will be added to the sauce within the meal (80 g of carbohydrates, 25 g of protein, 50 g of fat; 735 Kcal). Participants will administer their rapid-acting insulin as per the carbohydrate counting method (i.e. the same IU of insulin as per Trial 1).

Trial 3) Participants will replicate Trial 2, but will administer a rapid-acting insulin dose that has been increased by 15%.

Trial 4) Participants will replicate Trial 2, but will administer a rapid-acting insulin dose that has been increased by 30%.

Once participants have consumed the meal, they will then remain at rest for 6 hours, with 12ml blood samples per hour to capture the time-course changes in glycaemia, lipaemia, hormone, and metabolite concentrations. During the first 2 hours, an additional 1ml sample will be taken at 15 and 45 minute time periods to determine blood glucose and lactate only. Thus, total amount of whole blood collected within each trial is 160ml, which is less than half a typical soft drinks can.

Blood sampling process For each blood sample, 1 ml of whole blood will be collected and used for the immediate quantification of glucose and lactate (Biosen C, EKF Diagnostics, UK). On samples where the additional 11 ml of blood is taken, the remaining blood will be transferred into tubes designed to collect blood plasma and serum. Both containers will then be centrifuged at 3000 x g for 15 minutes to obtain plasma and serum, which will be stored at -80 C° for later analysis of insulin (Invitron, UK), triglycerides (Randox Labs, UK), Apoliopoprotein-B48 (Oxford Biosystems, UK), Apolipoprotein-B-100 (Oxford Biosystems, UK), non-esterified fatty acids (Randox Labs, UK), TNF-α (Sigma Aldrich), GLP-1 (Oxford Biosystems, UK), PYY (Sigma Aldrich, UK), Total Antioxidant capacity (Eagle Biosciences, UK), and Human Myeloperoxidase (Eagle Biosciences, UK).

During each blood draw, participants will be asked to complete a 10 point Likert Visual Analogue Scale (VAS) to determine subjective feelings of hunger and fullness. Following the blood sample and completion of the VAS at 6 hours post-meal, participants will be discharged from the laboratory.

Inclusion Criteria:

For inclusion in the study, volunteers will be -

• either male or female and aged 18-50 years old
• free from any diabetes complications apart from background diabetic retinopathy
• not taking any prescribed medication other than insulin
• treated with a stable insulin regimen composed of a combination of slow/long acting insulin glargine/determir and a fast acting insulin analogue (lispro or aspart, glulisine)
• have a HbA1c of <9.5% (80 mmol/mol)
• using the carbohydrate counting method for administering meal time rapid-acting insulin.