Artificial pancreas uses adaptive algorithm to manage type-2 diabetes

In 2016, the US Food and Drug Administration (FDA) issued its first approval of an "artificial pancreas" for type-1 diabetes that automatically monitors blood sugar levels and doses the patient with the appropriate amount of insulin. In the latest evolution of this technology, the same type of system is now being tested in people who have type-2 diabetes.

The University of Cambridge and its University Hospitals NHS Foundation Trust first created an artificial pancreas intended to help manage type-1 diabetes. In their latest update on this effort, the researchers, as well as colleagues from the University of Bern and its university hospital, demonstrated that this same technology can be used with type-2 diabetes patients who have kidney failure.

Kidney failure is a common outcome for people who have type-2 diabetes, the latter of which occurs when the pancreas can no longer meet the body's insulin demands. Diabetes remains the most common cause of kidney failure, complicating treatment by increasing one's odds of experiencing unusually high and low blood sugar levels.

Because of this complication, treating type-2 diabetes patients who also have kidney failure is difficult; oral medications aren't an option in most cases, meaning the patient must manually monitor their blood sugar levels and inject insulin. The researchers note that it is hard to develop an optimal dosing regimen for these patients, however, and extreme blood sugar changes can have severe health outcomes for these individuals.

A different version of the artificial pancreas is being used to treat type-2 diabetes patients. Compared to the one used for type-1 patients, this version features a fully closed loop system, according to the researchers. With this, the faux pancreas is fully automated.

The medical device utilizes a glucose sensor, insulin pump, and small computer to determine the amount of insulin administered. A smartphone is also used in the mix to "make further adjustments" to the insulin dose. Compared to controls, participants who used the artificial pancreas were able to remain in their target blood range 53-percent of the time, on average, versus the control's 38-percent.

As the algorithm used to control insulin dosing adapted to the patients, the amount of time spent in the target range increased. A trial of the technology with outpatients who have type-2 diabetes without the need for dialysis is currently taking place.