The impact of structured exercise in type 1 diabetes

Monitoring exercise using wearable technology can help people with type 1 diabetes enjoy more benefit from being active, by clarifying the many factors affecting glycaemic response. Professor Mike Riddell of the Muscle Health Research Centre at York University, Toronto, presented some promising preliminary results from the Type 1 Diabetes Exercise Initiative (T1DEXI project) at the Advanced Technologies and Treatments for Diabetes (ATTD) meeting in Barcelona recently. Dr Susan Aldridge reports.

Everyone should be doing at least 150 minutes per week of accumulated activity, which could be in the form of activities of daily living, like household chores, or structured exercise. The latter, by definition, differs from other activity because it’s structured and done with purpose. “Structured exercise is something I think we should do on top of the recommended 150 minutes, because it comes with additional benefits, particularly for those of us living with type 1 diabetes,” Professor Riddell says.

Positron emission tomography (PET) scans show how glucose disposal into muscles increases rapidly on exercise. Vigorous exercise, like cross-country skiing, increases glucose in all the muscles in the body, and the brain, up to 50-fold, and it enhances insulin sensitivity in skeletal muscle and in liver.

“Exercise also improves aerobic capacity, which is a great predictor of longevity and overall cardiovascular health,” Professor Riddell explains. “It improves the risk of microvascular disease and may even have the potential to improve our HbA1c – by around 0.4% – and time in range.”

However, different forms of exercise produce different glycaemic trends in type 1 diabetes. “Aerobic exercise at a steady state for long period of time, where the workload remains constant, is typically associated with a drop in glucose, but it’s a variable drop depending on who you are and how your body responds to exercise,” Professor Riddell continues. “We think we know some of the variables – some of these are easy to measure, some not. Intermittent exercise drops glucose variably, because bursts of exercise like power lifting and sprinting can lead to a rise in glucose that might need insulin correction.”

The T1DEXI study

“We know that there’s variability in response to exercise in type 1 diabetes, which makes blood glucose control challenging, but we really don’t know the factors that are predicting that variability all that well,” Professor Riddell explains. “Some people avoid exercise altogether if they’re living with type 1 diabetes, for fear of hypoglycaemia. The artificial pancreas systems help with this, but they don’t eliminate it.”

The Type 1 Diabetes Exercise Initiative (TIDEXI) Study, which is being undertaken by a consortium of researchers, including Professor Riddell, is a large observational study of ‘at home’ exercise in type 1 diabetes. By generating large datasets, TIDEXI aims to better understand how different forms of exercise influence glycaemia and glucose control in type 1 diabetes. The findings will shed light on the key individual and/or event level variables that drive glycaemic changes during exercise. Eventually, this will help improve current exercise guidelines for type 1 diabetes and potentially inform newer insulin delivery systems and decision support networks, making exercise safer and more effective in improving overall diabetes self-management. “The project is a large one, and we couldn’t do it without the generous support of the The Helmsley Charitable Trust and management from the Jaeb Centre for Health Research,” Professor Riddell notes.  

TIDEXI recruited people over 18, who’d had type 1 diabetes for at least two years, who were relatively active and using a pump, multiple daily injections or hybrid closed-loop. They were randomised to one of three exercise videos – aerobic, high-intensity interval training or resistance training – each of 25 to 30 minutes duration to be followed three times a week for four weeks.

They wore a number of sensors to track their glucose and activity. New for this study was the TDEXI smart phone app, developed by Peter Jacobs and colleagues at Oregon Health and Science University, to track activity levels, carb intake, insulin use and many other variables. Food capture photography was also used on selected days to better understand how macronutrients, and maybe micronutrients, impact glycaemia during exercise and sport. 

T1DEXI findings

“We ended up with a very robust dataset of nearly 500 individuals,” says Professor Riddell. “We captured 15,105 discrete physical activity events.  Walking was the most common form of exercise, and there were also 3,000 study video exercise sessions and lots of daily activities like house cleaning.”

The main outcome data was changing glucose over time. Aerobic exercise caused the biggest fall in glucose levels, followed by interval training, followed by resistance training. This was not a novel finding, but it was good to have it confirmed by home-based exercise. “What is novel is that we found a number of factors that can predict the variability in response of glucose to exercise,” says Professor Riddell. “For example, baseline heart rate is a predictor of fall in glucose – the lower the baseline heart rate, the greater the change in glucose during exercise. Is this due to fitness, or stress? We don’t know but it’s a novel finding.” There was also large variability in individual glucose responses, with some participants having a rise in glucose over 30 minutes, while others had a fall of 60 mg/dl. Factors affecting the response include exercise type, insulin on board, sex of the participant, time in range and time below range in the previous 24 hours and insulin on board.  

On days when people were doing the exercise videos, all three forms of structured exercise were associated with an acute improvement in time in range over the next 24 hours. “This confirms to me that structured exercise sessions of as little as 30 minutes a day will improve your time in range, compared with a non-structured exercise day,” Professor Riddell says. “The change in time in range is clinically relevant. We’re getting over 10% increase in range, mostly by knocking down time below range.”

And there is much more to come from TIDEXI. Future analysis of the TDEXI data set will look at other forms of physical activity and exercise and determine the impact of nutritional intake and other potential factors on glycaemia during exercise, to help design new guidelines and algorithms and automated insulin delivery systems for exercise management. And you can do your own analysis if this is your interest area, for the data set is being made publicly available, and will be online in CDISC format very soon.  

For more from Professor Riddell on exercise and diabetes management, see the following contributions he has made to Horizons:

• Exercise and insulin: a potent synergy (Insulin@100)
• This sporting life (The long and the short of it)

Any opinions expressed in this article are the responsibility of the EASD e-Learning Programme Director, Dr Eleanor D Kennedy.