One of the terms most-oft overheard at the recent ADA Conference was the "closed-loop system" for diabetes: when will we have it? is it really viable?
For those unfamiliar with the term (as I was, novice that I am!), "closed-loop" means a device that combines continuous blood glucose sensing and insulin delivery. Mimicking the function of the human pancreas, the "closed loop" system would monitor glucose levels and, in response, deliver an appropriate amount of insulin. Essentially, this system would perform as an artificial pancreas.
Researchers -- at least those I spoke with -- now openly admit that they were overly ambitious in predicting the arrival of a functioning and marketable closed-loop system, fabled to be ready in a few years (that was five years ago!) But hopes still run high. The JDRF has it on their short list of aggressive goals. And Medtronic MiniMed's Long Term Sensor System (LTSS), which links an implantable long-term glucose sensor with an implantable insulin pump, is looking promising in clinical tests. They hope to bring an artificial pancreas to market by 2008.
The technology is tricky, indeed, but the biggest stumbling block seems to be the concept itself.
A system that monitors BG levels, and then provides doses automatically based on that data would be the first to put people's lives in the hands of fully automated technology. The system would go ahead and administer insulin based on its own readings and calculations. In a truly closed-loop system, this means you wouldn't even have to enter basal rates, as the unit would react to high BG readings to account for the carbs you ate. If any portion of the system breaks down, the consequences could be grave -- because there is no "human intervention" to use common sense, in the case that a BG reading seems off, for example.
"The FDA's going to have a really difficult time approving a system like that," a researcher from Abbott noted.