Shed a Tear for Diabetes: The Development of a Tear Glucose Sensor


Has diabetes ever made you cry?  If Dr. Jeffrey La Belle and his team from the School of Biological and Health Systems Engineering at Arizona State University are successful in their current endeavor, diabetics may be able to put their tears to good use.  In collaboration with the Mayo Clinic Arizona, Dr. La Belle and his team have been developing home glucose monitors which measure glucose using tear fluid from an eye rather than a drop of blood. The team has working prototypes- small meters with strips that are touched lightly to the eye, and the necessary algorithms to convert measurements of glucose concentration in tears to the blood glucose equivalents that we are familiar with. They have partnered with BioAccel, an Arizona nonprofit, to move the new technology into clinical trials with the hope that making glucose testing less invasive will ease the burden of glucose control for the millions of people living with diabetes.

We spoke to Dr. La Belle to learn more about the tear glucose sensor, what it can do, and when we might see it.

How did you come up with the idea of testing tear glucose?

The idea isn’t new.  The work began in the 1930s, but there have been many issues over the past 80 years about capturing the tear fluid, not to mention detecting such low levels of glucose. There are many groups trying to come up with the next generation of devices, but we decided to go for what could we deliver today (well, as rapidly as one can).

How does glucose in tear fluid correlate with glucose in blood?

This can best be summarized by Baca, et al, 2007 who worked with Dr. Asher at University of Pittsburgh. In Dr. Sanford Asher’s group they summarized all the literature from the 1930s to present day and found (many) reasons why it does and does not correlate.  It mainly comes down to rapid sample collection, small volumes, not allowing the sample to evaporate, and doing so without stimulating the eye/conjunctiva.  If you do that all correctly, the levels are closely related in time.  A chief problem is that the tear glucose levels are 10-100 times lower than in blood.  That means in blood you would find 100 mg/dL, in tear, that’s about 1 mg/dL.

In your paper, you mention the importance of gaining a better understanding of the correlation and kinetics of tear glucose/blood glucose– has any work been started toward this end?

Yes, we have recently begun this work with support from a local non-profit called BioAccel.  With their kind assistance and our collaboration with the Mayo Clinic Arizona, we are doing this now.

How does your device work?  Does it require touching the eye?  (That sounds just as unpleasant as a finger prick!)

Actually, we touch the eye, but only the white part, the conjunctiva, for around 5 seconds.  This is not different than placing a contact lens on your eye (but our device does not sit on the eye for very long).

In your paper you also mention the need for further research into the effects of chronic tear sampling on the eyes. Do you anticipate that repeated tear sampling will cause irritation or damage to the eye?

We hope not.  People wear devices in their eyes all day, months at a time (contact lenses), but we want to do good due diligence and risk management.

Does the known relationship between diabetes and ocular complications make any frequent eye testing a severe risk?

There are two eye related issues we are concerned with, one being chronically dry eyes, the other being watery eyes.  It is always impossible to build a device or therapy that works for every person, but these issues will be addressed in subsequent testing.  We are really still at the front end of the tests.

How would contacts, lens solution, Visine, and the like affect the accuracy of the tear glucose measurements?

This is not too difficult to solve.  You know how some medications state take with food or do not take with food…. they come with doctor’s orders.  We could simply state to obtain tear glucose readings prior to placing contact or solutions in eyes.  Of course, we would have to perform studies to verify those orders, times, etc.  As far as contacts go in general, since we are only measuring off of the conjunctiva, they should pose no barrier to our device.

Could someone with dry eyes (who doesn’t produce a lot of tears) use a product like this?

Again, something to study in the future.

Given the concerns about accuracy and long-term usability, have any pharmaceutical or medical device companies expressed interest in taking a tear glucose product to market?

Well, that’s the big question.  We’ll see.  We obviously hope so, as this would remove the need for a diabetic to carry lancets and alcohol wipes, so some of the paraphernalia typically required would be removed.  And many people said they look forward to a device which would not require finger pricking.

Would the cost of using a tear fluid sensor be less than the cost of using blood glucose test strips?

This goes hand-in-hand with the above statement.  Right now we know we are adding on a little material for each device, hence prices would be higher.  But by removing the need for lancets and alcohol wipes, we hope the additional costs would be no different than factoring in those associated costs.

Jeff La Belle and Scout


Notify of
Inline Feedbacks
View all comments
Copyright © 2009-2021 Diabetes Media Foundation, All Rights Reserved.
ASweetLife™ is a trademark of the Diabetes Media Foundation, All Rights Reserved.
Would love your thoughts, please comment.x