Measuring Tear Glucose Levels   (January 28, 2014)

On January 16, the Google Official Blog had an entry about measuring glucose levels in tears, that has garnered considerable attention. It's vaguely titled ( Introducing our smart contact lens project). Among other claims, the blog stated that "We’ve completed multiple clinical research studies which are helping to refine our prototype." More on this in a moment.

The idea that a surrogate for blood glucose could be measured by the glucose level in tears seemed to excite both the media and people with diabetes and physicians (see Dr. Fran Cogen's comments). It's not new, and it's only one of a long list of bodily substances that have been suggested for use as surrogates for BG. The list includes saliva (which I've written about previously), ear wax, and skin, as well as tears.

A review article in 2011, Noninvasive Diagnostic Devices for Diabetes through Measuring Tear Glucose, said that contact lenses "are being considered by at least three research groups who are working with such sensors as an alternative tool to continuously and noninvasively monitor the level of glucose in tears."

Google's Project Co-Founders (Dr. Brian Otis and Dr. Babak Parviz) both have doctorates in electrical engineering and both are faculty members at the University of Washington.  Dr. Parviz previously worked on a similar project while employed by Microsoft. The Microsoft version of the contact lens claimed to be able to detect glucose levels in tears, but there's apparently no data about how accurate the device might have been, nor whether the tear-glucose (TG) levels correlate with BG levels.

(Another article, Measurement of tear glucose levels with amperometric glucose biosensor/capillary tube configuration, reported that "A strong correlation between tear and blood glucose levels was found, suggesting that measurement of tear glucose is a potential noninvasive substitute for blood glucose measurements.")

One publication from Dr. Parviz is A contact lens with embedded sensor for monitoring tear glucose level. There's nothing in this article that indicates that this study was done in humans (or animals); it seems a lab experiment (in-vitro) pure and simple. I'm a physician, not an electrical engineer, but the description of what they did is rather hazy to me. It seems they developed a sensor, mixed some chemicals together (including a glucose solution -- the researchers apparently did not use real tears) and measured electrical current.

The astute reader will notice that I have not mentioned any human trials of any these TG devices. I was therefore surprised at the claim in the Google blog that "multiple clinical research studies" have been performed. In the medical research field, "clinical research studies" means studies in humans -- as compared to bench studies (which are called in-vitro studies) or animal studies (which are called pre-clinical studies).

I therefore went to ClinicalTrials.gov, where almost every recent clinical research study is listed, but couldn't find anything. So my suspicion level went up. Have there been any human studies, or are the authors so compartmentalized in the field of electrical engineering that they don't use the medical research lingo correctly? I wrote to both co-founders (their e-mail addresses are at the UW website along with their credentials and photos) and asked them.

Eventually, I got a reply -- although not from them: it came from Google's Manager of Global Communications & Public Affairs! She refused to discuss any results ("The data will be shared with the FDA at the appropriate time") but confirmed that "We’ve completed a number of clinical research studies, including ones on people to test comfort and functionality." So, there's at least some human studies done. Hopefully more will be underway soon.

We shall see...