Dogs for Hypoglycemia?

A few years ago, when we first discussed the idea of dogs sniffing out hypoglycemia (low blood glucose) here at Diabetes Flashpoints, the idea was widely met with skepticism and even ridicule — The Colbert Report even devoted part of a segment to mocking the idea. But with a new study confirming the ability of dogs to smell an indicator of dangerously low blood glucose, the idea may be gaining a new level of respectability.

Published in the July 2016 edition of the journal Diabetes Care, the study confirms that there is indeed a chemical released in the breath during hypoglycemic episodes that dogs can smell. The researchers came to this conclusion by using eight women with Type 1 diabetes as test subjects. These participants’ blood glucose was raised and then lowered in a controlled setting, and levels of volatile organic compounds (VOCs) were measured in their breath.


As noted in a UPI article on the study, the researchers discovered that levels of one particular chemical, isoprene, rose significantly at the onset of hypoglycemia. Aside from hypoglycemia, though, there was no relationship between blood glucose levels and breath isoprene levels, and there was also no significant relationship between blood glucose and any of the other VOCs that the researchers measured (acetone, methyl nitrate, ethanol, ethyl benzene, and propane). Other research has confirmed that dogs can smell isoprene, which has no smell to humans at the levels measured in the study.

But when it comes to detecting hypoglycemia and alerting their owner to the problem, dogs have some serious competition: continuous glucose monitoring (CGM). As noted in a MD+DI article published last month, the CGM company Dexcom has data showing that users of its monitors use traditional (fingerstick) blood glucose monitors less frequently that non-CGM users do, yet still experience better blood glucose control. Dexcom is petitioning the U.S. Food and Drug Administration (FDA) to approve a claim by the company that its CGM systems can safely be used to make treatment decisions, such as injecting insulin or taking glucose tablets. Currently, CGM users are instructed to always confirm a CGM reading with a fingerstick test before taking any steps to raise or lower their blood glucose level.

What’s your take on dogs versus CGM — are you interested in potentially using either one to alert you to episodes of hypoglycemia? Do you think dogs might offer any advantages that CGM doesn’t, or vice versa? If you don’t currently use a continuous glucose monitor, are you hesitant to add yet another device to your life — one that is meant to be worn at all times? If you already use CGM, have you ever “broken the rules” and used a reading to make a treatment decision? If so, how has this turned out for you? Leave a comment below!

  • S. Kerr

    how low must a blood sugar be for isoprene level to rise and be detected by a trained dog? In this research study the level was 2.8 mmol/L (50.45 mg/dl.) The ADA defines hypoglycemia as glucose below 70 mg/dl.