The diabetes challenge

17 Jun 2002

Optical sensor technologies could provide a solution to the age-old problem of continuous blood-glucose measurement for diabetics. But as Nadya Anscombe discovers, obstacles must be overcome before one of the many devices in development reaches the market.

From Opto & Laser Europe April 2002

How the implantable sensor works

Diabetes affects millions of people worldwide. At present, diabetics must take blood samples up to six times per day to determine the amount of insulin to inject - an inconvenient and unreliable process. However, optical sensor technologies are being developed that could change this.

Many companies hope to tap into this potentially lucrative market by developing a continuous monitoring system that would allow diabetics to monitor their glucose levels painlessly and reliably. But after 10 years of work, no manufacturer has yet commercialized such a system.

The ultimate aim would be to combine such a sensor with an implanted glucose pump to create an artificial pancreas. Most firms are now turning to optical rather than electrochemical sensor technologies.In the US, MiniMed, BioTex, Animas, SpectRx, and Sensors for Medicine and Science Inc are all developing implantable optical sensors. Some companies are also trying other approaches. MicroSense International, for example, has developed a handheld probe that measures glucose levels without drawing blood, and Cell Robotics has developed a finger perforator that uses a laser to make a small hole in a patient's finger for blood collection.

Several optical methods could be used for the continuous monitoring of glucose, but the two that show the most promise are: fully implantable devices, which measure glucose in whole blood and involve implanting a light source and detector and sending the result out of the body via a radio-frequency signal; and optochemical devices, which measure glucose via the skin's interstitial fluid and involve implanting a chemical sensor just under the skin and stimulating it with external light.

Both of these technologies have some disadvantages, however - fully-implantable devices require surgery and last for about a year, while optochemical sensors require light to travel through skin, which is a scattering medium.

But both types of optical sensors have an important advantage over the chemical or electro-chemical sensors that are predicted to come to market soon. When a foreign object enters the body, it is immediately coated with a protein layer and later a collagen-like layer known as encapsulation tissue to protect the body. This layer can interfere with the accurate measurement of glucose using chemical or electro-chemical sensors, and means the sensor needs to be calibrated several times. But light of the correct wavelength passes freely through this tissue and reduces this problem.

Animas, which has developed a fully implantable sensor, says that the solution is to implant into a part of the body in which this tissue growth is minimal. Yizhong Yu, chief scientist at Animas, told OLE: "We've found the right place for the sensor - now we need to find a semiconductor light source that works at 2 µm - a wavelength where glucose has some interesting features." He says that finding a source is difficult because there has been no demand in the past for sensors of this wavelength.Although he acknowledges that fully implantable sensors involve the disadvantage of the patient having to undergo surgery, Yu believes that these sensors are further developed than non-invasive sensors. "There is no technology pathway leading to a non-invasive sensor," he said.

A less invasive approach, however, has been developed by researchers from the Lawrence Livermore National Laboratory (LLNL) and MiniMed, a maker of insulin pumps. They have developed an optical sensor that can be implanted under the skin without surgery and will last for a year before replacement. It measures glucose in interstitial fluid, rather than whole blood.

The device is a small disc with a polymer-based fluorescent chemical sensor. In the absence of glucose, the sensor's molecules have a low level of fluorescence. The presence of glucose alters the molecules' electron configuration so that they become much more fluorescent and emit light of a specific colour. If developmental work goes to plan, a small handheld instrument will shine light onto the skin and a small detector will measure the resulting fluorescence.

LLNL researcher Tom Peyser told OLE: "This approach measures the intensity of the fluorescence. We are also developing an approach that measures the lifetime of the fluorescence." This second method would be more tolerant to instrument errors - many factors, such as whether a patient wears a watch, can change the detector's readings using the first method.

This is just one of the many problems faced by developers of optical glucose sensors. Groups around the world have been developing these technologies for more than 10 years, but a commercial optical sensor is not predicted to come onto the market for another five years.

Tom Peyser concluded: "Fluorescent glucose sensors are clearly a promising approach to an old, important and unsolved problem in biomedical engineering. This is becoming a crowded field of research."