This proposal seeks to continue the development of a reliable implantable sensor for glucose. Such a device could eventually be used as part of an artificial pancreas consisting of a sensor, a microprocessor-driven data acquisition and control system and a pump to be used for insulin infusion. More immediately, however, the goal is to develop a monitoring system which will continuously record blood glucose levels. Continuous monitoring is expected to be useful for two purposes: 1. To provide an alarm system which will warn the diabetic of the impending onset of a hypoglycemic episode and 2. To provide a glycemia-time profile which should be useful to the clinician in devising an insulin administration regimen. The needle-type subcutaneously implanted sensor for the above purpose has been extensively tested in rats and dogs and its reliability has been established. It remains therefore to test the sensor in humans to determine: 1. Does the sensor function reliably in humans? and 2. Do the data produced yield clinically useful results? To answer these questions a series of clinical studies are proposed. The first series involves testing sensor and control unit performance on healthy volunteers. This will be accomplished at Hotel- Dieu in Paris. Starting in the second year (Year 8) clinical trials will be carried out simultaneously in Paris and KU Medical Center. Studies will be made on Type I insulin dependent diabetics to provide additional support for the reliability and ease of use of the system and to study particularly nocturnal hypoglycemia. Long term (1 week intervals) implantation will also be investigated as well as the evaluation of effectiveness of diabetes management using the monitoring unit. There remain some important fundamental and practical questions relating to sensor and control unit function which must be resolved. The sensor will be modified to render it unresponsive to acetaminophen. Several aspects of in-vivo performance will also be studied. For the monitoring unit it will be necessary to improve signal processing algorithms and to develop user-friendly data analysis techniques.
Showing the most recent 10 out of 33 publications
