The Renal Study

The impact of automated insulin delivery on glucose management in people with diabetes and advanced chronic kidney disease

This study tested whether an automated insulin delivery (AID) system (sometimes called an “artificial pancreas”) could help people with diabetes who also have advanced chronic kidney disease (CKD). Researchers compared using the AID system for 8 weeks against standard insulin treatment with continuous glucose monitoring (CGM) in 40 adults across hospitals in Australia and Denmark.

People with both diabetes and advanced kidney disease cannot filter insulin properly through their kidneys, making blood sugar levels unpredictable and hard to control. This instability increases the risk of serious health complications. Additionally, dialysis treatment (used when kidneys fail completely) can cause sudden swings between dangerously high and low blood sugar levels. Standard insulin injections struggle to keep up with these rapid changes.

The study included a medically complex group, with a median age 60 years, 60% with type 1 diabetes and 40% with type 2 diabetes. Most participants (83%) had advanced kidney disease but were not yet on dialysis; 10% used peritoneal dialysis and 8% used haemodialysis. Many participants had serious health complications including heart disease and vision problems. Notably, 25% of participants required hospital admission during the short 16-week study period for unrelated medical issues.

Participants received training on carbohydrate counting and diabetes management, then used a continuous glucose monitor for 3 weeks to establish a baseline. They were then randomly assigned to use either the MiniMed 780G automated insulin pump system or continue their usual insulin treatment (injections or open-loop pump) for 8 weeks each. The automated system gradually increased its capabilities over the first 8 weeks, starting with low-glucose alerts and eventually providing fully automated insulin adjustments. The study revealed significant implementation barriers. Of 66 people screened, 15 withdrew during the training phase because they found the frequent appointments and complex education overwhelming. However, all those who persisted were able to successfully use the system

The results of this study indicate that the use on the AID system significantly improved blood sugar management. Participants spent 73% of their time within the target glucose range (3.9–10.0 mmol/l) using AID, compared to 60% with usual care, an improvement of approximately 3.3 additional hours per day in the healthy range. This improvement was driven primarily by reductions in high blood sugar episodes. Hypoglycaemia (dangerously low blood sugar) remained rare in both groups and did not increase with AID use. No serious adverse events were attributed to the device itself. The benefits were more pronounced in people with type 1 diabetes (19.4% improvement in time in range) than type 2 diabetes (5.3% improvement), likely because type 1 diabetes involves more severe insulin deficiency and greater blood sugar variability. Participants reported stable mental health, quality of life, and sleep quality throughout the study. Treatment satisfaction improved with both approaches, possibly due to the comprehensive education and glucose monitoring provided to all participants.

The study concluded that automated insulin delivery is feasible, safe and effective for people with diabetes complicated by advanced kidney disease. It provides superior glucose control compared to standard care, particularly in type 1 diabetes. However, implementation in this vulnerable population requires intensive education, close clinical support and careful device training. The study suggests that optimising the user interface to reduce physical and cognitive demands would help broader adoption in this medically complex group.

To read the publication, click here