The Impact of a Predictive Hypoglycaemia Alert Function in Physical Activity for People With T1DM

Overview

Observational, within-subject, crossover study To assess the impact of Dexcom G6 RT-CGM with a predictive hypoglycaemia alert function on the frequency, duration and severity of hypoglycaemia occurring before, during and after regular physical activity in people with type 1 diabetes At Imperial College Healthcare NHS Trust have established the multi-disciplinary Imperial Physical Activity and Diabetes (IPAD) clinic to empower, educate and enable people with diabetes to manage their blood glucose when they undertake physical activity. The investigator utilise the skills and expertise of a consultant diabetologist, a diabetes dietitian, a consultant in sports \& exercise medicine, and a diabetes specialist nurse with expertise in diabetes technology. The investigator have access to diagnostic \& therapeutic radiology, physiotherapy and psychology services.

CLINIC VISIT 1: Screening Following informed consent study participants will give a full medical and medication history. Venous blood tests will be taken to assess HbA1c and renal function. C-peptide will be measured in participants who have not had previous tests to confirm T1DM diagnosis.. Women of childbearing age will have a urine pregnancy test. An ECG will be performed in participants with longstanding Type 1 Diabetes Mellitus (\>10 years) or participants who are \>40 years old. CLINIC VISIT 2: Run-in If participants meet the inclusion criteria they will be enrolled on to the study, within a month of screening and at a time that is convenient to them. For the 10 days of the run-in period, participants CGM will be blinded, meaning that they will be wearing the CGM sensor but will not see any of the glucose data. They will wear a sensor which will be attached to a transmitter which will send data to a Dexcom receiver. The receiver must be kept within 6 meters of the transmitter most of the time as the transmitter can only store three hours of data. After the run in they will be able to use the Smartphone app instead of carrying the receiver if they wish. They will be taught to insert the sensor at the run-in visit, and provided with spares in case of sensor failure. Participants will be instructed to test their capillary blood glucose levels and self-manage their type 1 diabetes as normal during the run-in period. The run in phase will last 10 days. Data on food intake, and insulin usage will be collected for the full duration of the study using a smart phone app (mySugr) in those treated with insulin injections, or using the built-in system in the pumps of those using insulin pump therapy. Both the app and the insulin pumps allow participants to input their own insulin doses, or use the bolus calculator which uses the participant's individual insulin:carbohydrate ratio to help calculate insulin doses based on the carbohydrate intake. There will be no change in participants' insulin dosing, and participants will make all their insulin dosing decisions themselves, based on their pre-study insulin:carbohydrate ratios. The advantage of using the app or bolus advisor is that the tracking of the insulin action (which lasts for 3-5 hours) allows participants to see how much insulin is still active. This is important when the CGM is unblinded as it can prevent participants giving additional insulin unnecessarily when they see their blood glucose levels on the CGM. Those using insulin injections will be provided with a smart insulin pen, a device similar to the participants' usual insulin delivery device use, but with a memory capable of storing dosing data until the end of the study (Inpen, produced by Companion Medical). A structured education refresher focusing on carbohydrate counting, hypoglycaemia avoidance, recognition, and management, insulin action and the theory and practicalities of CGM will be provided. All education will be delivered by the research dietitian with a checklist of topics that are provided as part of routine care. Participants will wear a GPS-enabled sports watch with built-in heart rate monitor (Garmin Forerunner 935) for the entire duration of the study. Participants will be asked to download the Garmin Connect App for data from the Garmin watch to be transmitted to the app via Bluetooth. They will be provided with anonymised log in details for the app and a charger for the watch. Volunteers will be asked to record their activity type and perceived physical exertion after every session of exercise. Quality of life data will be collected by mixed quantitative and qualitative methodologies. Validated questionnaires (HFS-II, Gold, RPAQ, DDS17) will be used to assess fear of hypoglycaemia, hypoglycaemia awareness, recent physical activity and diabetes distress. The investigator will undertake a short interview about the participants perceptions and experience of CGM. CLINIC VISIT 3: Randomisation Ten days after visit 2, participants will attend the research clinic for the randomisation appointment. The CGM, Garmin data and insulin bolus data (from mysugr or insulin pump) will be downloaded. Participants will have their weight and body composition measured using electrical impedance scales. Participants will start using their CGM unblinded, so that they are able to view their glucose levels either on the receiver or on the Dexcom app on their smartphones, depending on choice. Participants will be randomised to real-time continuous glucose monitoring with Predictive Hypoglycaemia Alert Algorithm enabled (ON group) or real-time continuous glucose monitoring with Predictive Hypoglycaemia Alert Algorithm disabled (OFF group). Randomisation will be done using sealedenvelope.com and stratified by insulin delivery modality and previous CGM experience. Participants will be provided with sensors for the rest of the study. Participants will be instructed to test their capillary blood glucose if symptoms of hypo- or hyperglycaemia occur, in case of sensor failure or if the sensor glucose is out of the desired range, in line with the device license. Participants will change sensor every ten days (or sooner in the event of sensor failure). Low glucose alert settings will be standardised at 4.4 mmol/L (80mg/dL) for all participants at the start of the study and can be reduced to 4 mmol/L (70mg/dL) at week 2 during the telephone visit depending on participant preference. High glucose alerts may be personalised. Participants will be provided with a contact number for technical support but insulin titration decisions will be made by the participant. Two weeks after visit 3 and 4, all participants will receive a telephone call from the research dietitian to discuss any queries they have regarding the CGM, sports watch or any other element of the study. Those in the ON group will also be able to adjust the predictive hypoglycaemia alert level from 4.4 mmol/L (80mg/dL) to 4 mmol/L (70mg/dL) if preferred. Each study period will last 40 days. CLINIC VISIT 4: Cross-over Forty days after visit 3, participants will attend the research clinic and be switched onto the opposite intervention (ON/OFF) for the second phase of the study, which will last 40 days. Participants will continue to use the mySugr app or bolus advisor on their pumps, the Garmin watch and inpen or pump. The CGM, Garmin data and bolus data from mysugr will all be downloaded. The questionnaires assessing fear of hypoglycaemia, hypoglycaemia awareness, recent physical activity and diabetes distress will be repeated, as well as a short CGM interview on perceptions and experience of CGM. VISIT 5: End of study Forty days after visit 4, and 90 days after the start of the study, the study will end. All participants will be reviewed by the research dietitian at the research clinic for data to be collected and uploaded and to address any issues arising from the study. Smart insulin pens, G6 CGM and garmin watches will be returned and downloaded. Insulin pumps, mysugr apps and blood glucose meters will also be downloaded. CSV files will be downloaded from mySugr. There is no patient identifiable information in the CSV file. Garmin data is sent from phone to the Garmin connect app via Bluetooth, without any identifiable information, and CSV files can be downloaded from Garmin connect with a secure login. There is no patient data on the inpen, which will be downloaded onto Imperial College London computers, and will be associated with the anonymous email containing the study ID number for each participant. Insulin pumps and blood glucose meters will be downloaded onto NHS computers. HbA1c will be rechecked at this point via venous blood tests. Weight and body composition measurements will also be repeated. Validated questionnaires to assess fear of hypoglycaemia, hypoglycaemia awareness, recent physical activity and diabetes distress, will be repeated. A short interview about the participants perceptions and experience of CGM will be undertaken.