Individualised Postprandial Glucose Responses in Type 1 Diabetes

Assessing Interpersonal Variability in Postprandial Glucose Responses to Food in People With Type 1 Diabetes

Type 1 diabetes (T1D) is a lifelong disease which stops the body from producing insulin - an important hormone that controls blood sugar (glucose) levels. People with T1D use insulin replacement therapy, usually in the form of injections, to help control blood glucose levels, however keeping glucose levels within normal ranges is usually a challenge. Mealtime glucose control is fundamental to good diabetes management and are an important contributor to long-term diabetes complications. However, many individuals experience variability in glucose levels around mealtimes. The objective of this study is to establish whether and which parameters are important predictors of mealtime glucose levels in people with T1D. The investigators will monitor glucose levels using the latest glucose monitoring technology and collect blood samples to: Characterise changes in glucose levels in individuals with T1D in response to different meals Determine whether and what food characteristics and personal factors are linked to individual glucose responses to different meals. The investigators will recruit 150 individuals with type 1 diabetes. Firstly participants will attend a preliminary visit, where a blood sample will be donated to study laboratory blood markers of vascular and metabolic health accompanied by a full medical examination in which body composition will be established. During this visit participants will also complete questionnaires about their lifestyle, and be fitted with two wearable devices to monitor glucose levels and physical activity levels under free-living conditions. After 4-weeks of wearing the devices, participants will attend two experimental laboratory visits where breakfast and lunch will be served and blood samples taken. This will enable us to observe glucose and metabolic responses to feeding under controlled conditions.

Recruitment: Participants will be recruited using advertisements through university channels. The investigators will also approach local and national diabetes charities (Diabetes UK, JDRF) who the investigators work with closely to advertise through written publications (patient magazines), websites, social media, and patient support groups. In addition, the investigators will contact patients who have previously participated in our research who have consented to being contacted about future research. Contact information for the research team will be provided in advertisements; an initial telephone call will be available to interested participants to discuss the study, answer any questions, and obtain contact information to send a study information pack including a participant information sheet. Participants will be given 6-weeks to decide whether to participate in the study. Preliminary visit: Patients will attend a preliminary visit to our dedicated laboratories at the University of Sunderland to assess eligibility and obtain written informed consent by a GCP trained member of the research team. At this visit, consented participants will be fitted with a blinded real-time CGM and an accelerometer and provided with a food diary. During this visit, patients will complete a questionnaire to assess their medical history and physical activity. Both CGM and accelerometer devices will be worn by participants for the duration of their study involvement; both devices complete with data capture will be returned by participants to the research team either in person or via post (using a prepaid envelope). CGM measurement: A small CGM device (CGM; Freestyle Libre Pro; Abbott Diabetes Care) will be inserted during the preliminary visit to the laboratory. It is a small discreet, water-resistant sensor, with a thin filament (<0.4m thick) that is inserted (painlessly) 5mm beneath the skin surface on the upper arm. The sensor will be inserted into the subcutaneous tissue on the anterior of the upper arm. The insertion site will be taken as equidistant between the most medial portion of the upper arm and marked with indelible ink so that initial placement can be replicated on subsequent insertions in the unlikely event that a sensor should fail. The sensor is factory calibrated making it far more convenient and removing the risk for sensor inaccuracies from calibration user error (capillary meter inaccuracy, not washing hands etc) seen with other CGM devices. Glucose data generated by the CGM is masked (i.e., participants cannot see their glucose data and thus cannot change their treatment regimen based on this information). A single reader device is used to activate and retrieve the data. The accuracy and safety of this device has been established in T1D across the full range of glucose levels likely to be observed in this study, including those in the normal range; the latest generation of the CGM sensor has a MARD of <9% as of 2018. Moreover, as an un-blinded version of this device is now made available to people with T1D with treatment indications through the NHS, the use of this technology is deemed acceptable to people with T1D. Study participants who are using Freestyle Libre to monitor their glucose patterns can continue to use the device as usual but will be asked to wear an additional blinded sensor. In total, patients will be requested to wear the CGM device for 6 continuous weeks, consisting of 4-weeks observation followed by 2-weeks observation during which laboratory experimental testing will occur. Accelerometer: Participants will be issued and fitted with an accelerometer (GeneActiv, worn on the arm) which will provide a comprehensive assessment of sitting/standing patterns (i.e. posture allocated) and physical activity levels. The data collected from this device will be used to estimate daily energy expenditure (using a priori cut points) and calculate physical activity levels including time spent in different postures and activities during discrete parts of the day both before and after each laboratory visit. This will allow for quantification, standardisation and replication of physical activity patterns during free-living episodes. In total, patients will be requested to wear the accelerometer device for 2 continuous weeks. Medical history: To include: age, gender, duration of diabetes, insulin therapy (average total daily dose and basal insulin dose), associated medical conditions, family history of T2D and cardiovascular disease, hypoglycaemia unawareness (assessed through a combination of the Clarke and Gold methods). Physical activity assessment: Patients will complete the IPAQ (International Physical Activity Questionnaire), the Exercise Benefits/Barriers Scale Questionnaire, and The Barriers to Physical Activity in Type 1 Diabetes Scale. These will assess current levels of physical activity participation, as well as general and diabetes-specific enablers and barriers to physical activity. Anthropometrics: Anthropometric data will include: weight, height, waist circumference and waist-hip-ratio. Body composition will be determined using the bioelectrical impedance method [Seca mBCA 525] as detailed above. Blood pressure will be taken alongside ankle brachial index measurements. Estimated glucose disposal rate (eGDR; a measure of insulin resistance) will be calculated for all patients. Dietary assessment: Patients will complete a food frequency questionnaire (Dietary Instrument for Nutrition Education) and will be provided with a food scale and a food diary to perform a weighed food log. In total, participants will complete the weighed food diary for 1-week. Blood donation: A single 50mL blood donation will be made which will be retrospectively analysed for parameters of vascular and metabolic health. Blood will be obtained using venepuncture technique by a trained phlebotomist. Experimental laboratory visits Pre-laboratory phase: Prior to the first experimental visit, the investigators will assess glycaemic control (CGM) and physical activity patterns (accelerometer) for 4- and 1-weeks, respectively. Patients will be required to use their dietary recording sheets to replicate their dietary patterns in the 48-hours prior to each experimental laboratory visit. For standardisation of glycaemic control, as well as biochemical parameters, participants will be provided with a standardised meal to be consumed on the evening before each experimental laboratory visit based upon their weight (~635kcal; vegetarian lasagne; Tesco). Since an acute exercise session may enhance insulin action for up to 48-hours, participants will be asked to reframe from moderate-to-vigorous physical activity/exercise during the 48-hours preceding each laboratory visit to avoid the potential impact of this on glycaemia and hypoglycaemia risk. Additionally, during the 24-hour prior to each laboratory visit, participants will be asked to reframe from caffeine and alcohol which are known to influence glucose levels. To avoid any confounding influence of hyper/hypoglycaemia or illness, a researcher will call the morning of each testing day - if blood glucose is outside the range of 4-12 mmol/L or if the participant is unwell with blood ketones >0.6 mmol/L, the experimental visit will be postponed. The dosage and timing of insulin administrations as well as any supplements/vitamins being taken will be standardised for the duration of each trial period. To control for the potential confounding effect of insulin regimen (CSII vs MDI) and medication, insulin regimen and medication will be included as a covariate in the statistical model. Participants using CSII will be advised to avoid making changes to their insulin pump delivery settings and insulin regimen between study visits. Participants treated with MDI will be advised not to change their insulin regimen between study visits. Due to the likelihood of changes in glycaemic control and/or insulin requirements over longer periods of time, pre-menopausal women will complete each condition with a 1-week washout, as for all other participants. Menstrual phase will be recorded for each condition and controlled for as a covariate in statistical models. Main experimental visits: In a randomised crossover design, patients will attend two separate morning-time (~08:00am) laboratory-based visits (lasting ~4.5-hours in duration) within the dedicated clinical testing facility at the Univeristy of Sunderland, each interspersed by one week. Each experimental condition will be conducted in a controlled laboratory setting to maximise internal validity and minimise the inherent variability of less-controlled 'real-world' settings. Except for toilet breaks, participants will be seated in a comfortable lounge-chair and instructed to minimise excessive movement. On each occasion and upon arrival to the laboratory following an overnight fast, participants will assume a seated and rested position while a 20-gauge cannula (Vasofix, B. Braun, Melsungen AG, Melsungen, Germany) is inserted into the antecubital vein of their non-dominant arm; resting, fasted venous blood samples will be collected prior to experimental testing. Following the initial blood sample, patients will be given a standardised mixed-macronutrient breakfast-based meal on each occasion. At 4-hours post-breakfast, participants will be provided with a standardised mixed-macronutrient lunch meal before being discharged home. Each meal will be matched for carbohydrate (equating 1g.carbohydrate.kg.BM-1 ), fat, protein, and fibre content but differing in glycaemic index through manipulation of carbohydrate composition to represent either a low or high glycaemic index meal. Insulin administration will be standardised across visits based on the carbohydrate counting method. Meals will be preceded by an insulin bolus, calculated from each participant's usual insulin-to-carbohydrate ratio (with no correction dose), administered 10-minutes before eating, with those treated on MDI standardising insulin injection site. The timing and amount of water consumed with each meal, and ad libitum throughout the duration of the first visit will be recorded so that this can be replicated on subsequent visits. Blood sampling: Periodic blood samples will be drawn at 30-minute intervals across the 4-hour observation window and commencing immediately prior to the consumption of the breakfast meal). At each time point, 10mL of venous whole blood will be taken and dispensed into Vacutainers before centrifugation at 3000 rev/min for 15-minutes at 4ºC. Plasma will be separated and stored at -80 degrees centigrade for retrospective analysis, or immediate analysis for a panel of vascular inflammatory and thrombotic parameters, including plasma levels of C-reactive protein (CRP), complement C3, fibrinogen, plasminogen activator inhibitors (PAI)-1, interleukin 1b, will be assessed using standard ELISA techniques. An additional 5mL of venous blood will be taken at rest on one arm of the study for C-peptide and routine blood tests (including HbA1c, U&Es, LFTs, full lipid profile, urinary albumin creatinine ratio (unless these have been done within three months of enrolment). A total of 8 blood samples will be taken with the total volume of blood equalling 85mL, which is approximately 1/5th of a full blood donation. The total volume of blood to be taken is safe and is similar to other studies that the investigators have performed previously in people with T1D. Further, the investigators have worked hard with our PPI representatives to minimise the participant burden of all study procedures; the volume and frequency of blood sampling has been deemed acceptable by our PPI representatives. Post-laboratory observation period: Following the last blood sample at 7-hours post-breakfast, participants will be discharged from the laboratory. Glycaemia will continue to be monitored under free-living conditions for a total of 48-hours post-intervention using CGM. As the investigators are interested in the 48-hour glucose responses following our intervention (including the dawn phenomenon), participants will be given two standardised meals to consume as their evening meal and breakfast and be required to replicate their dietary intake across trials for the duration of the 48-hour post-intervention period, in order to control for the potentially confounding effects of diet on glycaemic control.

Consumption of standardised low-GI mixed-macronutrient breakfast-based meal followed by a standardised low-GI mixed-macronutrient lunch meal 4 hours later.

Consumption of a standardised high-GI mixed-macronutrient breakfast-based meal followed by a standardised high-GI mixed-macronutrient lunch meal 4 hours later.

Gender/sex, race/ethnicity, sexual orientation, and religion are not exclusion criteria for this study Inclusion Criteria: A diagnosis of T1D for a minimum of 5 years; Currently treated on a stable insulin regimen for a minimum of 6 months consisting of continuous subcutaneous insulin infusion (CSII) therapy or multiple daily injections (MDI) of a combination of rapid-acting and long-acting insulin; Familiar and currently using the carbohydrate-counting method for determining mealtime insulin dose; Not currently pregnant; No overt diabetes complications including end stage renal failure requiring dialysis; Free from hypoglycaemia unawareness assessed through a combination of the Clarke64 and Gold65 methods66; No recent (<6-months) history of diabetic ketoacidosis (DKA); Free from medical conditions relating to a haematological disorder, gut mobility or digestion; No history of anorexia, bulimia, or any other disordered eating; No history of deep vein thrombosis; No history of heart attack or stroke within 6 months prior to recruitment; No history of malignancy; No existing medical or psychiatric conditions likely to interfere with the study; No dietary allergies or intolerances likely to interfere with the study; Able to understand written English and provide written informed consent. Exclusion Criteria: A diagnosis of T1D for less than 5 years; Not currently treated on a stable insulin regimen for more than 6 months consisting of continuous subcutaneous insulin infusion (CSII) therapy or multiple daily injections (MDI) of a combination of rapid-acting and long-acting insulin; Unfamiliar and not currently using the carbohydrate-counting method for determining mealtime insulin dose; Currently pregnant or planning on becoming pregnant during study involvement; Presence of overt diabetes complications including end stage renal failure requiring dialysis; Presenting with hypoglycaemia unawareness assessed through a combination of the Clarke and Gold methods; Recent (<6-months) history of diabetic ketoacidosis (DKA); Presenting with medical conditions relating to a haematological disorder, gut mobility or digestion; Presenting with a history of anorexia, bulimia, or any other disordered eating; Presenting with a history of deep vein thrombosis; Presenting with a history of heart attack or stroke within 6 months prior to recruitment; History of malignancy; Presence of existing medical or psychiatric conditions likely to interfere with the study; Presenting with dietary allergies or intolerances likely to interfere with the study; Unable to understand written English and provide written informed consent.

Submission of a research proposal to the lead contact. The submission will be reviewed prior to approval for IPD sharing.