Further to Aaron Kowalski's recent JDRF update here on developing a closed-loop system for treating diabetes -- and of course the new OmniPod/DexCom deal -- I ought to share some official updates from the recent Global Diabetes Summit in Ohio, courtesy of Close Concerns:

• "In a session on continuous glucose monitoring (CGM), Dr. Alan Marcus, Vice President and Global Director of Medical Affairs for Medtronic Diabetes, revealed that Medtronic Diabetes will release an enhanced senor-augmented pump that shuts off insulin delivery in response to a low alarm if the patient fails to act." The editors say this is the first of a series of important "interim steps" toward achieving a closed-loop system. Medtronic definitely believes that "the future of diabetes management is an integrated insulin pump and CGM system," they say, and they are hard at work on the "incremental addition of automatic features" including improvements in the way these devices gather data, talk to each other, respond, and ultimately automate. A sensor-augmented pump will be released in 2008 in Europe (only). No introduction yet in the US, presumably because it takes so much longer to get through the US regulatory system. Thank you, FDA?

Dr. Barry Ginsberg estimates that closed-loop systems would become available at the earliest in 2013-2015, based on testimony of numerous independent experts. Close Concerns reports:

"Great progress has been made toward a closed-loop system, but numerous challenges remain." The current generation of continuous monitors has a mean inaccuracy of about 12%. The 95% confidence limits are ± ~30%, meaning that a blood glucose reading of 200 indicates that blood glucose has a 95% chance of being between 140 and 260. Frequent samples partially overcome this problem, as do slow corrections (more readings during correction period). For a closed-loop system, accuracy will need to increase to at least 99%. It will also be difficult to detect when a meal is eaten, and it will be impossible to tell how much or what is eaten. Patients will need to enter some information into the device. A final challenge is that insulin bio-availability {aka absorption} varies (for regular insulin, action varies by ±35% for 95% confidence interval) and duration of insulin action (i.e. insulin on board) changes. It will be necessary to know how much insulin is present to avoid overdosing." Them's lots of challenges, I'd say.


Oooh, no thanks... if it looks like that, I'm willing to wait a bit.

By the way, Close Concerns also reported results from it survey of 525 medical school students this month. Conclusion? "This is probably the most dispiriting story we've reported on in 75 issues of Diabetes Close Up -- our main finding was that virtually none of these students plans to pursue diabetes as a specialty. There are several reasons why -- the money is one thing, the hours are another, the lack of success in working with patients is yet another. We were struck not only by the declining interest in treating diabetes but also by declining interest in general medicine. Where will patients go in 20 years?" I don't know, either. Time for something to change.

Disclaimer: Content created by the Diabetes Mine team. For more details click here.


This content is created for Diabetes Mine, a consumer health blog focused on the diabetes community. The content is not medically reviewed and doesn't adhere to Healthline's editorial guidelines. For more information about Healthline's partnership with Diabetes Mine, please click here.