Bionic and artificial pancreas systems found effective in real-life settings

Recent tests of artificial pancreas systems have found that the technology could provide better glucose control for patients with type 1 diabetes than current care does.

A crossover study tested a wearable, automatic bionic pancreas that delivered both insulin and glucagon subcutaneously based on data from a continuous glucose monitor, user-provided information about meal consumption, and an algorithm that determined dosing every 5 minutes. In a study of 20 adults and another of 32 adolescents, glucose control while using the bionic pancreas for 5 days was compared with an insulin pump. Results were published by the New England Journal of Medicine on June 15.

The adults, who wore the bionic pancreas at home and work, had better mean glucose control (138 mg/dL [7.4 mmol/L] vs. 159 mg/dL [8.8 mmol/L]; P<0.001) and less time with a glucose reading below 70 mg/dL (3.9 mmol/L) while using the device (4.1% of the time vs. 7.3%; P=0.01). The adolescents, who were at a diabetes camp, had similar glucose control improvement but no significant difference in time with low glucose. They had less frequent interventions for hypoglycemia on the bionic pancreas. The researchers noted that they asked adults to limit alcohol intake and there were occasional wireless connectivity problems. But overall, despite technological challenges, the device improved control over the current standard of care, the researchers concluded.

Another crossover study in adults compared 4 weeks on an artificial pancreas (insulin delivered according to an algorithm and continuous glucose monitoring) to usual insulin pump therapy. The 24 participants spent significantly more overnight time with their glucose in the target range of 70 to 144 mg/dL (3.9 to 8.0 mmol/L) when they had the artificial pancreas (mean difference between groups, 13.5%; P=0.0002). There were 2 severe hypoglycemic episodes during the artificial pancreas period and none during the control period, but they were not found to be related to the algorithm. The researchers also noted that patients had better glucose control during the day and similar total insulin use with the artificial pancreas as with usual care. This study also had issues with disruptions in wireless connectivity, but the researchers concluded that extended home use of the technology is feasible.

An accompanying comment noted that such devices need to be tested in long-term clinical trials, which should assess usability, safety, effectiveness for everyday use, and the effects on patients who currently have difficulty controlling their glucose. Smartphones, and their potential to control an artificial pancreas, could be key to the initiation of widespread use, the editorialists said. The study and comment were published by The Lancet Diabetes and Endocrinology on June 16.