Artificial Pancreas Effective in Hospitalized Diabetic Patients

Closed-loop insulin delivery systems, also known as the artificial pancreas, proved effective devices for treating hospitalized seriously ill patients with complications of diabetes, according to a study published in The Lancet: Diabetes & Endocrinology.

Hood Thabit, PhD, and colleagues at Addenbrooke’s Hospital, Cambridge, United Kingdom assessed whether fully closed-loop insulin delivery without mealtime boluses was as safe and effective as standard subcutaneous insulin therapy.

They enrolled 40 patients and assigned 20 to the closed-loop intervention. They found the patients using the artificial pancreas spend 59.8% of their time in the target glucose range compared to 38.1% of patients getting subcutaneous injections.

No episodes of severe hypoglycemia or ketoanemia were reported in either group and the only adverse event was unrelated to the insulin therapy (a gastrointestinal bleed emergency).

“Taken together, our findings suggest that an automated closed-loop insulin delivery system can potentially provide health-care professionals with a valuable clinical tool to manage inpatient hyperglycemia safely and effectively while possibly reducing workload,” they wrote.

Patients in the closed-loop group “stated that they were happy with the glucose concentrations in hospital during the study,” and most said they would recommend the system to a friend or family member if that person were to be hospitalized.

The system used was comprised of an abdominal cannula, a glucose monitor, and a closed loop system.

According to the team, a subcutaneous cannula was inserted by the investigator in the abdomen for delivery of insulin lispro (Humalog, Eli Lilly, IN, by an insulin pump (Dana R Diabecare, Sooil, Seoul, South Korea). A subcutaneous, real-time, continuous glucose monitor (Freestyle Navigator II, Abbott Diabetes Care, Alameda, CA, was inserted in the abdomen or upper arm by the investigator and calibrated according to the manufacturer’s instructions. After successful sensor calibration when sensor glucose concentrations became available, automated closed-loop glucose control was started by the investigator and continued for 72 h. The low glucose alarm on the continuous glucose monitoring receiver was initially set at a threshold of 3·5 mmol/L.

The FlorenceD2W-T2 automated closed-loop system comprised a model predictive control algorithm (version 0.3.65) residing on a control algorithm device (Dell Latitude 10 Tablet, Dell, TX, USA) linked by a USB cable to the continuous glucose monitoring receiver (FreeStyle Navigator II). The tablet communicated with the study pump (Dana R Diabecare) via the Bluetooth wireless communication protocol.

The researchers said they did not seek formal feedback from staff but noted that “there was no interruption to the ward workflow and patients’ usual ward-related activities from the closed-loop intervention.”

The article is entitled, “Closed-loop insulin delivery in inpatients with type 2 diabetes: a randomized, parallel-group trial.”