As a diabetes researcher, exercise physiologist and individual with type 1 diabetes, I am always curious about how the latest diabetes technology fits into an exercise program. Exercise is, after all, one of the three cornerstones of diabetes management, along with diet and medication.

Excited to try noninvasively measuring my interstitial glucose with the new GlucoWatch Biographer from Cygnus, I obtained the required doctor's prescription and ordered one this past May, a month after the new product became available.

After carefully reading through the manufacturer's patient information sheet, I noticed that the sheet did not say that you can't exercise with the GlucoWatch Biographer. It did warn, however, that "skipped readings and unexpected shut offs may occur due to excessive perspiration, jarring or dislodging of the device from the skin." The sheet also advised that the Biographer will usually begin monitoring again after you stop perspiring.

We all know that exercise can make us sweat, but I began to wonder how hard I could exercise or how much I could sweat and still have the GlucoWatch work correctly.

Being a researcher, I devised a study involving eight to 10 people, half with type 1, and the other half of the same age and gender but without diabetes. I convinced these people to wear a GlucoWatch while exercising both outdoors and indoors, doing 45 minutes of easy activity, 30 minutes of moderate exercise and 15 minutes of hard exercise on separate occasions. On another day, they simply rested outdoors or indoors for 45 minutes.

All of the participants live in eastern Virginia. Granted, it can be quite hot and humid here during the summer, but we did have air conditioning for our indoor activities.

When resting outdoors for 45 minutes in conditions anywhere from 75 to 91 degrees F and 30 to 100 percent humidity, slightly more than one-third of the participants got all their readings. However, while simply resting, the same number were unable to get any readings, whether they were perspiring lightly or not at all. Apparently, even less than "excessive" sweating adversely affects the Biographer's function. Temperature and humidity differences did not effectively explain the varying results. Indoor resting results were only slightly better, with 43 percent obtaining all readings, and 29 percent obtaining none.

The exercise results were even more disheartening. During outdoor exercise, the GlucoWatch skipped 75 to 100 percent of its readings during and immediately following all intensities of exercise. In addition, it completely shut off following an activity 37 to 50 percent of the time. Once this happens, you must start over with a new Autosensor (cost: approximately $5) and wait through the three-hour warm-up period again if you still want readings for the rest of the day.

Exercising indoors in an air-conditioned area helped slightly, but only for light activities. For moderate and heavy workouts, the GlucoWatch still failed to give any readings 85 to 100 percent of the time. It shut off completely more than half the time following moderate activity, despite being in a cooler environment.

Two of the participants were determined to get readings. One discovered how to place a Band-Aid on his arm to cover the Biographer's perspiration sensor. This helped slightly, but it's likely to have caused erroneous readings as a result of sample dilution with his sweat. Another exerciser blew a fan directly on his arm during all his indoor activities and obtained more readings than most. Neither method seems necessarily accurate or feasible, however.

Cygnus apparently has yet to address the sweating issue, which certainly needs to be resolved before the GlucoWatch will be useful with exercise. Admittedly, it is a difficult problem to solve, given this new method for noninvasive sample extraction. Until it's resolved, however, I will be using my blood-glucose meter, not the GlucoWatch Biographer.