Closed-Loop "Artificial Pancreas" May Be Helpful in Type 1 Diabetes

Laurie Barclay, MD

September 18, 2010

February 4, 2010 — A closed-loop "artificial pancreas" system may improve blood glucose control in children and adolescents with type 1 diabetes, according to the results of a phase 2, randomized crossover trial published online in the February 5 issue of The Lancet.

"Closed-loop systems link continuous glucose measurements to insulin delivery," write Roman Hovorka, PhD, from the University of Cambridge, United Kingdom, and colleagues. "We aimed to establish whether closed-loop insulin delivery could control overnight blood glucose in young people."

The investigators performed 3 randomized crossover studies in 19 patients with type 1 diabetes, using allocation by computer-generated random code. The age range of the participants was 5 to 18 years, and mean duration of type 1 diabetes was 6.4 ± 4.0 years. The first study compared standard continuous subcutaneous insulin infusion with closed-loop delivery (n = 13; Artificial Pancreas Project at Cambridge [APCam] 01), the second compared closed-loop delivery after rapidly and slowly absorbed meals (n = 7; APCam02), and the third compared closed-loop delivery and standard treatment after exercise (n = 10; APCam03).

Participants were blinded to plasma and sensor glucose levels, and investigators in APCam01 were also blinded to plasma glucose levels. Patients' standard pump settings were used during control nights, whereas during closed-loop nights, a control algorithm calculated the rate of insulin infusion based on glucose measurements every 15 minutes, and a nurse adjusted the insulin pump. Analysis was per protocol, with primary endpoints of time for which plasma glucose concentration was in a target range of 3.91 to 8.00 mmol/L or below target range: 3.90 mmol/L or lower.

Seventeen participants were evaluated during 33 closed-loop and 21 continuous-infusion nights. The main study endpoints were not significantly different between treatment groups in APCam01. A total of 12 closed-loop participants were analyzed: 52% (interquartile range [IQR], 43 - 83) achieved the target range for plasma glucose concentration vs 39% [IQR, 15 - 51] of those who received standard treatment; P = .06. The proportion of those who achieved plasma glucose concentrations less than 3.90 mmol/L was 1% [IQR, 0 - 7] vs 2% [IQR, 0 - 41] for those who received standard treatment; P = .13)

The other 2 studies also showed no significant differences in primary outcomes between treatment groups. For 6 patients analyzed in APCam02, target range occurred in 53% (IQR, 48 - 57) with the rapidly absorbed meal vs 55% (IQR, 37 - 64) with the slowly absorbed meal (P = .97), and concentration less than 3.90 mmol/L occurred in 0% (IQR, 0 - 4) vs 0% (IQR, 0 - 0; P = .16).

For 9 patients analyzed in APCam03, target range occurred in 78% (IQR, 60 - 92) with closed-loop vs 43% (IQR, 25 - 65) with control (P = .0245, not significant at corrected level; concentration less than 3.90 mmol/L occurred in 10% [IQR, 2 - 15] vs 6% [IQR, 0 - 44]; P = .27).

Closed-loop delivery was associated with increased time in the target range (60% [IQR, 51 - 88] vs 40% [IQR, 18 - 61]; P = .0022) and decreased time for which glucose concentrations were 3.90 mmol/L or lower (2.1% [IQR, 0.0 - 10.0] vs 4.1% [IQR, 0.0 - 42.0]; P = .0304), based on a secondary analysis of pooled data.

During closed-loop delivery, there were no events with plasma glucose concentration lower than 3.0 mmol/L, whereas there were 9 events during standard treatment.

"Closed-loop systems could reduce risk of nocturnal hypoglycaemia in children and adolescents with type 1 diabetes," the study authors write.

"Advancements in glucose-sensing technologies could further improve performance of closed-loop systems. Fully automated closed-loop delivery will need wireless data transmission to replace manual control of the pump by nurses," they add.

In an accompanying comment, Eric Renard, MD, PhD, from Centre Hospitalier Universitaire de Montpellier in France, noted various technical concerns associated with closed-loop insulin delivery.

"Notwithstanding the nurses' intervention — which could be easily replaced by a wireless direct connection — to programme the pump rate in agreement with the controller proposal, the absence of any sensor or pump failure during the trials allows realistic extrapolation towards a home environment," Dr. Renard writes. "Starting with overnight control before addressing control at meal times and during various activities is the most rational way forward. This seems to be the track to follow towards progressive implementation of automated insulin delivery at home."

The Juvenile Diabetes Research Foundation, the European Foundation for Study of Diabetes, the Medical Research Council Centre for Obesity and Related Metabolic Diseases, and the National Institute for Health Research Cambridge Biomedical Research Centre supported this study. Smiths Medical supplied study pumps. Abbott Diabetes Care supplied Freestyle Navigator devices and sensors. Medtronic supplied Guardian RT devices and discounted sensors. Some of the study authors report patent applications and/or various financial relationships with Minimed Medtronic, Abbott Diabetes Care, Lifescan, Novo Nordisk, BBraun, and/or Medtronic International Trading Sàrl. Dr. Renard has disclosed no relevant financial relationships.

Lancet. Published online February 5, 2010.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.