Skip to Content

Artificial Pancreas Tracks Two Hormones

A new system controls diabetics’ blood sugar by monitoring both insulin and glucagon.

Managing type 1 diabetes is a feat of organization and control. The better a diabetic can keep blood sugar in check, the less likely are long-term health complications. But even with devices like automated insulin pumps, which release a continuous dose of insulin, diabetics still need to remember to add an extra dose of insulin at meals, and many spend a significant amount of time each day with blood-sugar levels either higher or lower than normal.

Scientists have been working to develop an “artificial pancreas” system that would monitor blood-sugar levels and normalize them automatically, without any human input. A preliminary clinical trial detailed this month in Science Translational Medicine marks a significant advance in creating a fully automated system. The system was able to control blood sugar in a small group of diabetic patients, even when patients ate high-carbohydrate meals, which is one of the major challenges for artificial-pancreas systems.

Most artificial-pancreas systems under development pair a blood-sugar monitor with an insulin pump. The device in this study adds another component that monitors the hormone glucagon, which counteracts insulin. Glucagon helps prevent blood-sugar levels from dropping too low if too much insulin is given. Although diabetics still produce glucagon, it doesn’t always function properly.

“Our feeling is that glucagon is an important extra measure of safety,” says Steven Russell, an endocrinologist at Massachusetts General Hospital, who co-led the research. Russell explains that hypoglycemia can be a major problem for diabetics–paradoxically, the better diabetics control their blood sugar, the more they are at risk. Hypoglycemia, which occurs when blood sugar drops too low, can lead to sweating, trembling, dizziness, and confusion, and in some cases it can be life-threatening.

The new study was primarily designed to test an algorithm the team developed to predict the amount of insulin and glucagon needed to keep blood-sugar levels normal. Edward Damiano, a bioengineer at Boston University and co-leader of the study, says that because insulin is absorbed and cleared from the body slowly, the algorithm can’t simply respond to the current blood-sugar level but must also anticipate where it is headed. “At each dose, it keeps track of the rising insulin it’s given, as well as the decay of previous doses,” he says.

The system was tested in 11 adults with type 1 diabetes for a period of 27 hours, during which the subjects ate three high-carbohydrate meals. Rather than relying on a glucose monitor under the skin, the researchers took direct blood-sugar readings from the blood every five minutes. Software then calculated the amount of insulin and glucagon needed. Doses were administered by nurses.

In the first trial, the system kept blood-sugar levels normal for six of the subjects, but the other five experienced hypoglycemia that needed to be rescued by drinking fruit juice. The researchers found that these five patients took much longer than anticipated to absorb and clear the insulin they received. So they adjusted the parameters of the system to match a slower insulin absorption rate, and retested the same subjects. The system was then able to keep the blood-sugar levels of all participants under control, although levels were slightly higher in those who had absorbed insulin quickly in the first trial.

Bruce Buckingham, a pediatric endocrinologist at Stanford University who was not involved in the work, says that demonstrating that the system works even after subjects have eaten large meals is a key achievement. “Dealing with meals is the real obstacle” in developing any artificial pancreas, he says. A recent study from a group at Cambridge University, U.K., tested a similar system, but only overnight, when patients were not eating. Buckingham says that another challenge for the devices will be handling periods of exercise, which also causes blood sugar to fluctuate.

The team behind the new device is planning a further trial using an FDA-approved continuous glucose monitor and an automated system for delivering the two hormones. This trial will compare the dual-hormone system with an insulin-only one, and it will cover two days and include a period of exercise.

Aaron Kowalski, director of the Artificial Pancreas Project at the Juvenile Diabetes Research Foundation, which partially funded the study, says that the insulin-glucagon system represents the future of artificial-pancreas technologies. “Our ultimate goal is to try to come as close to a human physiology without diabetes as possible,” he says. But bringing such a system to market will be challenging. Kowalski says that there are no FDA-approved pumps that deliver two different substances, and glucagon is currently used only in emergencies when diabetic patients become dangerously hypoglycemic–it comes in the form of a powder that must be reconstituted.

In the more immediate term, Kowalski believes that insulin-only artificial-pancreas devices will become available much sooner. An insulin pump that can shut off automatically when blood sugar drops too low has already been approved in Europe. And the Artificial Pancreas Project is working with Animas Corporation, which makes glucose monitors, on a system that monitors blood sugar but only automatically delivers insulin when blood-sugar levels fall above or below a certain range. Kowalski believes that such a device, while not entirely automating the delivery of insulin, can offer diabetics crucial peace of mind.

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

How scientists traced a mysterious covid case back to six toilets

When wastewater surveillance turns into a hunt for a single infected individual, the ethics get tricky.

Google DeepMind’s new generative model makes Super Mario–like games from scratch

Genie learns how to control games by watching hours and hours of video. It could help train next-gen robots too.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.