Nanoparticle, Raman-based Fiber-optic Glucose Sensor

Walsh, Joseph

Abstract

NIH estimates approximately 17 million Americans have diabetes mellitus, with approximately 1/10 of these being type 1 and the remaining being classified as type 2, gestational, and other. The disease can be characterized by long-term complications involving eyes, kidneys, nerves, and blood vessels are common but are limited with tight glycemic control characterized by near-normal glycosylated hemoglobin. The goal of the proposed research is to prove that a sensor based on Surface Enhanced Raman Spectroscopy (SERS) can be used to measure glucose levels in vivo with sufficient accuracy for feedback control of insulin delivery. We propose to develop the fiber-optic-based sensor, test the sensor in vitro and in vivo, and acquire sufficient data in vivo to specify the system requirements for use in human trials. Much of the previous work in the field of glucose sensing has focused on minimally invasive means of measuring blood glucose levels; such systems will allow patients to avoid repeated painful finger-pricks but glucose control would still be based on patient-dependant feedback control. Implanted electrochemical sensors have been FDA-cleared since mid-1999, but they must be physician implanted and their use in real-time glucose control has been limited. Our preliminary SERS work, based on noble metal nanoparticles has shown the potential accuracy and precision of a SERS sensor for glucose. We propose that an advanced version of this SERS sensor could be passed through the skin into the subcutaneous space, in a manner similar to the placement of insulin-pump-based catheters, and replaced every 3 days, which is the current norm for insulin-pump catheters. The nanoparticle-based SERS system provides a direct measure of glucose concentration, can monitor glucose levels continuously, is not based upon consumables (e.g. enzymes or substrates), and like all Raman-based analysis is specific to glucose and relatively insensitive to confounding analytes. We envision using these sensors to close the feedback loop by using the continuous measurement of glucose to control insulin levels, thus freeing the patient of repeated measurements, carbohydrate-counting, and insulin-dosing problems.