Diabetes treatment trialled

“An 'artificial pancreas' can be used to regulate blood sugar in children with type 1 diabetes,” BBC News reported.

This story is based on research on how to optimise devices that can sense glucose levels and adjust the amount of insulin a child with type 1 diabetes receives overnight. This is important in maintaining glucose levels while a child sleeps where standard glucose monitoring is not practical.

This was a small but well-conducted study that showed promising results for this technology in keeping glucose levels within an acceptable range overnight. However, further research would be needed before a commercially available system were to be available.

Where did the story come from?

This research was carried out in the UK by Dr Roman Hovorka and colleagues from the University of Cambridge. The study was published in the peer-reviewed medical journal The Lancet .

The research was funded by the Juvenile Diabetes Research Foundation, the European Foundation for the Study of Diabetes, the Medical Research Council and the National Institute for Health Research - Cambridge Biomedical Research Centre.

Generally, the press did not examine the science behind this story too deeply and the small size of this study should be highlighted as it means that further research may be needed to check how well the technology might work outside of a research institution. The media also said that the research was carried out over 54 days, suggesting that the child patients received this treatment for almost two months. However, the children received overnight treatment on one to four occasions only.

What kind of research was this?

This randomised crossover study investigated whether a new system for delivering insulin could prevent nocturnal hypoglycaemia (low blood sugar at night) in both children and adolescents.

Type 1 diabetes occurs because your body cannot produce any insulin, a hormone that is needed to control the amount of glucose (sugar) in your blood. When you eat, your digestive system breaks down food and passes its nutrients into your bloodstream. Normally, insulin is produced by your pancreas to take any glucose out of your blood and move it into your cells where it is broken down to produce energy. However, if you have type 1 diabetes, there is no insulin to move glucose out of your bloodstream and into your cells.

If you have type 1 diabetes, you will need to take insulin injections for life. You must also make sure that your blood glucose levels stay balanced by eating a healthy diet and carrying out regular blood tests.

Monitoring glucose levels and administering insulin is problematic while diabetic patients are asleep. One treatment is to continually infuse insulin overnight, but the infusion is given at a constant rate and it does not respond to changing glucose levels during sleep.

Continuous glucose monitoring devices and insulin pumps have been developed and combined to form systems where the insulin is delivered as required based on measured glucose levels.

As yet, these ‘closed loop systems’ have not managed to deliver optimal accuracy and reliability. These researchers wanted to assess whether the drawbacks of existing closed loop systems prototypes could be overcome by adjusting the control algorithms.

In this type of crossover study, participants are given either a new treatment or the standard treatment in their first session, followed by the alternative in a second session. This allows the researchers to compare two treatments tested at different times in the same patient.

What did the research involve?

Children aged between 5 and 18 years with type 1 diabetes were enrolled between April 2007 and September 2008. Sometimes it is difficult to assess children from how they are physically feeling on whether they are hypoglycaemic. The researchers excluded children who had, on several occasions in the past, severe hypoglycaemia but had not been aware of it. They also excluded children who had any type of nerve damage.

The study had three parts, the first of which was a crossover study comparing the closed loop delivery system with continuous insulin delivery overnight. The second part looked at the closed loop system overnight when the participants had received a slowly or rapidly absorbed meal (high or low glycaemic index). The third looked at the closed loop system versus the continuous insulin infusion when the participants had exercised prior to sleeping.

For the first part, 13 patients were treated with an overnight treatment or standard treatment on two occasions one to three weeks apart. The insulin pump delivery was optimised for each patient by analysing their glucose levels periodically over a 72-hour period two weeks before the first treatment.

The second part involved seven patients from the first part of the study, who were aged fom 12 to 18 years and were studied on two further occasions. These patients were asked to eat meals that had either a high or low glycaemic load. This relates to the amount of carbohydrate a food has and how quickly the food affects blood sugar levels. The patients were then put on the closed loop system overnight. On the second occasion, they received the alternative meal.

The third part of the study involved 10 patients aged between 12 and 18 years. These patients did an exercise test so they could determine an appropriate level of exercise for the children who were of different ages and may have had different levels of fitness.

The patients’ oxygen intake was measured both at rest and when they were exercising on a treadmill at 50% of their peak level for 15 minutes.

The exercise test was performed before the patients were assigned to either the overnight treatment with continuous insulin infusion or the closed loop system.

The glucose levels of the sleeping patients were continuously monitored during all treatments to check whether they were in the appropriate range. The researchers also compared different algorithms for working out how much insulin to give based on the glucose levels in the closed loop system.

What were the basic results?

In the first part of the study, the patients’ glucose levels were in the target range for longer in the closed loop group compared with the continuous infusion, but this was not statistically significant. There was no difference between continuous infusion and the closed loop system in preventing hypoglycaemia. On average, the closed loop system and the continuous infusion system administered the same insulin dose.

The patients’ blood glucose levels were the same using the closed loop system following high or low glycaemic load meals.

Patients who had been given the closed loop delivery system after evening exercise spent more time within the optimum glucose range, but this was not statistically significant.

No significant differences between the two systems were found from the individual parts of the study. However, when the data from all three parts was pooled, it showed that patients who had closed loop treatment spent longer within the target glucose range than those who had the continuous infusion. They also spent less time with levels of glucose that were lower than the target range. This was also the case when patients had initial high or low glucose levels.

How did the researchers interpret the results?

The researchers conclude that overnight manual closed loop insulin delivery can improve glucose control and reduce the risk of hypoglycaemia in young patients with type 1 diabetes. They say that sensing errors are perceived to be the main obstacle for safe and effective closed loop glucose control. However, in their study the glucose sensors on the closed loop system and the blood glucose sensors were the same. The researchers suggest that advances in glucose sensing can further improve the performance of closed loop systems.


This study found some evidence that closed loop systems are better at maintaining appropriate glucose levels overnight than continuous insulin infusion in children and adolescents.

Further larger studies would be useful in assessing and optimising the technology. If the same study were to be carried out in a larger sample of patients, important differences between the two systems might become more apparent.

This small but well-conducted study is a step forward in managing blood glucose levels overnight, with the potential for improving the quality of life for young patients with type 1 diabetes. Further research would be needed before a commercially available system were made available.

NHS Attribution