Eating behaviour and activity levels are difficult to predict in toddlers and preschoolers with type 1 diabetes, so insulin requirements can fluctuate considerably. In addition, hypoglycaemia can occur very quickly in this age group due to the high insulin sensitivity and the children are not yet capable of correctly interpreting the corresponding warning symptoms themselves. The fear of unnoticed hypoglycaemia is extremely stressful for parents – besides the constant blood glucose checks and insulin injections. Small children with type 1 diabetes are therefore particularly often equipped with a sensor for continuous glucose monitoring (CGM) and an insulin pump.
This therapy is only a relatively small step away from closed-loop systems, in which both devices are linked together through an algorithm that operates independently and in which the amount of insulin injected adjusts to the measured blood glucose levels. The system is referred to as hybrid since a scheduled meal still has to be announced to the system.
The KidsAP02 study, led by Julia Ware from the University of Cambridge, tested a hybrid closed-loop system developed at her university in 74 children between 1 and 7 years (mean age 5.6 years) with type 1 diabetes in a cross-over trial at 8 centres in England, Luxembourg, Austria and Germany (Leipzig). An app designed for tablets was used for 16 weeks, then the groups were switched. The primary endpoint of the trial was the proportion of time in which the measured glucose level was within a range of 70 to 180 mg/dl (time in range).
With the device being turned on an average of 95% of the time, the parents' acceptance of the system was obviously high. This coincided with a significant improvement in glycaemic control. The children were in the target range 71.6% of the time with the app, and only 62.9% of the time without the algorithm. The difference of 8.7 percentage points was highly significant.
It was mainly the prevention of hyperglycaemia that led to better glucose control – the difference was 8.5 percentage points. However, the more stringent control was not achieved at the expense of increased hypoglycaemia, as was often feared. Though hypoglycaemia duration was not shortened, it was not prolonged either (2.6% vs. 2.4% of the time < 63 mg/dl, 1.0 vs. 0.9 < 54 mg/dl). As expected, the overall lower blood glucose levels also had a favourable effect on the HbA1c value, which decreased from an average baseline value of 7.0% to 6.6%. This difference was also significant.
One of the children with type 1 diabetes experienced a seizure due to severe nocturnal hypoglycaemia while on closed-loop treatment. In this case, it turned out the parents had set the nightly target value to 80 mg/dl against medical advice and then failed to react to the audio alarm of the glucose sensor for several hours.
The authors point out that especially in younger children with type 1 diabetes, maintaining good blood glucose control is essential, as the developing brain is particularly susceptible to damage. It is known that hyperglycaemia in particular has an adverse effect at this age and is associated with lower IQ scores over time. In this light, even small improvements in blood glucose control can have a significant impact.
Source:
Julia Ware et al; Randomized Trial of Closed-Loop Control in Very Young Children with Type 1 Diabetes; N Engl J Med (2022); 386:209-219; DOI: 10.1056/NEJMoa2111673