Artificial Pancreas Improves Glycemic Control After Total Pancreatectomy

Use of a portable, fully automated bi-hormonal artificial pancreas (BIHAP) with closed-loop glucose control led to better glycemic control for patients who had a total pancreatectomy, a Dutch randomized crossover trial found.

Among 10 such outpatients who developed type 1 diabetes after surgery, receiving the system for both glucagon and insulin delivery was associated with significantly more time spent in the euglycemic range of 70-180 mg/dL, as assessed by continuous glucose monitoring, compared to usual care with insulin pump or pen therapy for 7 days (median 78.3% vs 57.4%, P=0.03).

And BIHAP treatment led to less time spent in hypoglycemia (<70 mg/dL; median 0% vs 1.6%, P=0.004), reported Marc Besselink, MD, PhD, of the University of Amsterdam, and the Dutch Pancreatic Cancer Group in JAMA Surgery.

During BIHAP, patients also spent less time with a glucose level over 250 mg/dL (1.2% vs 8.4%, P=0.049), although the difference wasn't significant for levels over 180 mg/dL (21.7% vs 38.9%, P=0.19).

Glycemic variability was also significantly lower during BIHAP treatment by all measures, except for the high blood glucose index.

"The development of artificial pancreas may, thus, have an important role in mitigating the morbidity caused by [total pancreatectomy] and may even accelerate the expansion of indications," Pauli Puolakkainen, MD, PhD, and Ville Sallinen, MD, PhD, both of the University of Helsinki in Finland, stated in an accompanying editorial.

Total pancreatectomy leads to type 1 diabetes and life-long insulin replacement therapy, since alpha and beta cells are no longer present, Besselink's group noted. With such complete loss of endogenous production of both insulin and glucagon, total pancreatectomy patients remain at risk for severe hypoglycemia.

A novel BIHAP treatment device that automatically maintains blood glucose levels within a target range via a continuous closed-loop system improved glucose control for type 1 diabetes patients in the APPEL5 study.

"This portable device contains both insulin and glucagon and has a reactive glucose control algorithm that is self-learning to cope with day-to-day variations in insulin sensitivity," they noted. "With all alternative systems of insulin treatment, manually entering the carbohydrate content of each meal is still required."

Such similar devices have also proven superior to usual care for glucose control.

For this study, Besselink and colleagues enrolled 10 adult outpatients who had total pancreatectomy more than 3 months prior, typically due to cancer. They were randomized 1:1 to receive 7-day treatment with a fully closed-loop BIHAP (Inreda Diabetic BIHAP) then their usual diabetes care (insulin pen or pump) or the two strategies in reverse order. Current diabetes care for most was insulin pen therapy (80%).

Patients were seen at the University Medical Centers in Amsterdam between Aug. 21 and Nov. 16, 2020 and underwent a 5-day training period prior to BIHAP treatment. Excluded were those with impaired awareness of hypoglycemia.

Median age was 62.5, and 70% of participants were men. Average BMI was 22.6, and diabetes disease duration averaged 4.5 years.

Notably, 70% were satisfied with BIHAP and wished to continue treatment under the caveat that the number of alarms and device size would be reduced.

Common adverse events during BIHAP treatment included skin irritation from the subcutaneous infusion set/sensor (40%), headache (20%), and nausea (10%). No serious adverse events were reported.

"While encouraging, the results are somewhat preliminary in their clinical applicability," the editorialists said. "Limitations and adverse effects of BIHAP include the need for 4 sensors inserted subcutaneously, carrying a 345-g [0.76-lb] device, complaints of skin irritation, and minor, but frequent, problems in the function of the device."

The authors acknowledged limitations to the data including the small sample size and short treatment period. Quality of life was not assessed and no cost-effectiveness analysis was performed. Glucose outcomes were based on the use of different glucose monitoring devices.

Comment