This project will aim to investigate the clinical efficacy and metabolic effects of a pre-exercise dose of caffeine with a low (10g) dose of carbohydrate (CAF+lowCHO) without modification of insulin degludec on exercise metabolism in people with T1D.

Treatment of type 1 diabetes (T1D) involves lifelong use of exogenous insulin to manage blood glucose concentration. As with the rest of the population, people living with T1D are recommended to engage in regular exercise for a variety of health and fitness reasons . However, glycaemic control during exercise remains a particular challenge for this population due to rapid changes in insulin sensitivity and the impact of additional hormones which increase the risk of exercise-related hypoglycaemia. Current guidelines to prevent exercise-induced hypoglycaemia suggest insulin dose reduction and/or ingestion of carbohydrates in the context of the exercise bout. However, these adaptations are often difficult to apply, as insulin dose adjustments require knowledge of insulin pharmacokinetics and advanced planning which is not always possible. None of these strategies provide complete assurance that hypoglycaemia will not occur and high carbohydrate intake can be counterproductive if weight management is the target. Furthermore, modern very long-acting insulin analogues, which are favoured by many people with T1D, do not offer the option to rapidly or transiently reduce insulin before exercise. When using such insulins, dose reductions may take two to three days to achieve an adapted steady state, increasing the risk of inadequate insulin following exercise. Collectively, these factors increase the risk of further deterring patients from exercise. Simple, alternative strategies to reduce the risk of hypoglycaemia, both during and after exercise are needed.

Caffeine (1,3,7-trimethylxanthine) is the most commonly consumed chemical stimulant in the world that is naturally found in many foods and is frequently added to sports supplements due to its ergogenic effects in a range of sporting events. Caffeine has numerous physiological effects throughout the body including increased lipolysis in adipose tissues and hepatic glucose production in the liver alongside a decrease in glucose uptake in skeletal muscle. These responses have led to the suggestion that acute caffeine intake may attenuate exercise-associated hypoglycaemia in people with T1D. Ingestion of modest amounts of caffeine (200-250 mg, equivalent to 3-4 cups of coffee each day) has been shown to augment the symptomatic and hormonal responses to hypoglycaemia in participants with and without T1D. Caffeine has also been shown to reduce the frequency of moderate episodes of hypoglycaemia occurring overnight. The paucity of data on caffeine and exercise in individuals with T1D, in conjunction with caffeine's popularity both socially and as a sports supplement, suggests this deserves further attention.

A clear example whereby caffeine supplementation may be of use is in patients using an ultra-long acting basal insulin analogue such as insulin degludec. The administration is via subcutaneous injection once daily, and it has a duration of action that lasts up to 42 hours (compared to 18 to 26 hours provided by other marketed long-acting insulins such as insulin glargine and insulin detemir). On average, the half-life at steady state is approximately 25 hours independent of dose. Compared to the other basal insulin analogues, the risk of hypoglycaemia appears to be lower with insulin degludec, however, hypoglycaemia still occurs. In the case of physical exercise, the inability of the patient using such long-acting insulins to make rapid adjustments can translate to the occurrence of exercise-related hypoglycaemia due to an inability to reduce insulin already onboard, hence the need for new strategies to prevent this undesired phenomenon. When using such insulins, dose reductions may take two to three days to achieve an adapted steady state, increasing the risk of inadequate insulin following exercise. Applying a novel in-house developed lipid chromatography-mass spectrometry (LC-MS) assay, members of our research group observed that a single bout of aerobic exercise increases systemic insulin degludec concentrations in adults on stable basal insulin degludec regimens. Therefore, if these individuals wish to engage in regular exercise, as recommended in international guidelines, current treatment strategies may not be sufficient. For patients treated with modern basal insulin analogues, it seems more adequate not to modify the ultra-long acting insulin doses, as this can often result in more confusion than improvement, but to apply alternative strategies for recreational exercise. Caffeine ingestion pre or post exercise may offer a simple means to better manage glycaemia in the context of exercise in patients using these insulins and have the added benefit of reducing carbohydrate requirements in the context of exercise.

Inclusion Criteria:

• Type 1 diabetes for ≥1 year and negative C-peptide (<100pmol/l)
• Male and female aged 18-45 years old
• HbA1c <8.5% (69 mmol/mol) based on analysis from the central laboratory unit of Bern University Hospital
• Using multiple daily injections
• Using insulin degludec (Tresiba; Novo Nordisk A/S, Bagsværd, Denmark) as basal insulin for a minimum of 3 months
• Written informed consent
• Able and willing to adhere to safe contraception during the study and for 2 weeks after completion of the study. Safe contraception comprises double barrier methods (hormonal contraception [like: oral contraceptive pills or intrauterine contraceptive devices] together with a mechanical barrier [like: condom, diaphragm]).

Exclusion Criteria:

• Physical or psychological disease likely to interfere with the normal conduct of the study as judged by the investigator
• Continuous subcutaneous insulin infusion (using an insulin pump)
• Hypoglycaemic unawareness (Gold likert score ≥4) or having experienced any episode of a severe hypoglycaemic event within the last 6 months (i.e. need of third-party assistance).
• Current treatment with drugs known to interfere with metabolism e.g. systemic corticosteroids, SGLT2 inhibitors, Glucagon like 1 peptide- receptor agonists, or metformin.
• Relevant diabetic complications as judged by the investigator and based on medical record check (no cardiovascular disease and no significant microvascular disease)
• Microalbuminuria (as defined by area under the curve >30 mg/g)
• Body mass index more than or equal to 30 kg/m2
• Uncontrolled hypertension (>180/100 mmHg)
• Pregnant or planning to become pregnant during the study period (females only)
• Breastfeeding

In each of the conditions, participants will receive a taste-matched drink containing either • caffeine and 10 g rapid-acting carbohydrate (CAF+lowCHO); 20 g rapid acting carbohydrate (20g CHO); or placebo. The drink will be made by an independent member of the research team to ensure that neither the investigators or participants know the order of the drinks (double-blind)