Artificial Pancreas Keeps Tight Rein on Blood Sugar

June 28, 2010— -- ORLANDO -- A closed-loop insulin delivery system -- the so-called "artificial pancreas" -- appears to improve glucose control in patients with type 1 diabetes even after a large dinner accompanied by wine, compared with insulin pump therapy, researchers said here.

In a small study, patients who had a high-carbohydrate dinner with a glass of wine spent more time in the target blood glucose range when they used the closed-loop system versus patients who relied on an insulin pump, according to Dr. Roman Hovorka of the University of Cambridge, and colleagues.

They reported their findings at a symposium at the American Diabetes Association meeting here.

The artificial pancreas is a device that can automatically regulate blood sugar levels in patients with type 1 diabetes by releasing insulin when alerted to high levels of sugar, and withholding it when levels are low. Right now, patients with type 1 diabetes -- a group that accounts for between 5 and 10 percent of all diabetics -- either self-inject insulin or wear an insulin pump that releases the hormone into their body throughout the day.

The artificial pancreas is essentially taking insulin pumps a step further, by adding a continuous glucose monitor as well as an algorithm. The monitor constantly reads glucose levels, which the algorithm interprets and subsequently tells the insulin pump whether to release or withhold insulin.

Hovorka said at a press briefing that "social drinking in the evening causes hypoglycemia early in morning."

His group reported findings earlier this year in The Lancet that showed improved overnight glucose control and reduced risk of hypoglycemia with the closed-loop system compared with standard therapy.

In order to test how a large meal and liquor would affect that control, the researchers enrolled 12 adults with type 1 diabetes in a cross-over study.

Those hospitalized were treated on two separate nights with either closed-loop insulin delivery or conventional insulin pump therapy.

On both occasions, patients consumed a mixed meal that included 100 grams of carbohydrates along with 0.75 mg/kg ethanol as 13 percent white wine.

They also took insulin with their meal, which was eaten between 8:30 pm and 10 pm. The closed loop or "artificial pancreas" was then commenced from 10 pm until noon the following day.

Continuous monitoring was used to tabulate glucose values at 15 minute intervals.

The researchers found that the closed-loop system significantly improved time in the targeted glucose range compared with insulin pump therapy.

Moreover, there was reduced duration and severity of hyperglycemia with the closed loop, as measured by the Blood Glucose Index.

Finally, there was one episode of severe hypoglycemia with insulin pump therapy, but and none with the "artificial pancreas."

"We plan to move into home studies, hopefully later on this year," Hovorka said, noting that thus far his studies have only been done at night.

Dr. Aaron Kowalski, research director for the artificial pancreas project at the Juvenile Diabetes Research Foundation, which helped support the research, said during the press briefing that closed-loop systems should be available "in the near term."

But, he cautioned, it's "important to [note] that it's not going to … let us walk away from diabetes tomorrow. There are a number of important clinical steps that will have a tremendous impact."

Researchers are reluctant to speculate as to when such a device would become available. Currently, they are working to optimize the best combination of continuous sensors with insulin pumps.

Medtronic has developed a open-loop system that transmits sensor data to an insulin pump but does not include the algorithms to instruct the pump to deploy or withhold insulin infusions.

Other challenges with the device include the accuracy of continuous glucose monitoring due to the difference between blood levels of glucose and interstitial levels. Kowalski says these issues are mitigated by good calibration of devices.

Another hindrance to effective use of an artificial pancreas is compliance, particularly among teenagers.

Marilyn Ritholz of the Joslin Diabetes Center in Boston has been evaluating which psychosocial factors are associated with better outcomes with use of the artificial pancreas.

Patients who are more devoted to interpreting the information garnered from their devices had better outcomes, as do those who have better support from their spouses, she said at the briefing.

In addition, many patients report concerns about body image when wearing the devices, as it makes them "more aware of their diabetes," Ritholz said. "It makes them feel different. Some report feeling 'somewhat robotic.'"

She added that the success or failure of the device "is as dependent on the human experience as it is on the development of the technology."