Sniffing Out Danger

This article reprinted from R&D Magazine.

Editorial

March 2005

While at the 56th Annual Pittsburgh Conference on Analytical Chemistry and Applied Spectroscopy (PITTCON) in Orlando, I was deluged with new product introductions on all types of sophisticated, high-end analytical instrumentation and software. A visit to one exhibit booth near the end of the week, however, put much of these product introductions in perspective.

In this booth, I noticed a golden Labrador dog lying on the floor, an abnormal scene at most trade shows where animals are generally not allowed—except now for the occasional police K-9 units. The other exception, of course, is for service animals, the most readily apparent example being that of the guide dog for visually impaired people. In this case, though, the dog (named Armstrong) was a medical-alert service dog trained to alert his diabetic handler to hypoglycemia (low blood sugar).

Guide dogs have a long history of helping visually impaired people, with formal training programs developed after World War I. Medical alert dogs, however, have only been in place for just a little over ten years and are relatively few in number. Most breeds of dogs can be trained for medical alert duties, mostly depending upon their disposition. They can be trained in about three months to provide alerts to heart attacks, strokes, diabetes, epilepsy, panic attacks, anxiety attacks, and post-traumatic stress disorders. The dogs have been shown to give their owners a warning of between 20 to 45 minutes before a seizure occurs with a warning that is specific, but always consistent, with each dog and owner.

Very little is known about how the dogs know when their owner is about to have a seizure. It's believed that the dogs are picking up on unique signs of seizures like physiological or behavioral changes like dilating pupils or changes in facial expressions or color. Other changes that might be picked up are changes in human odors. Armstrong's owner, Mark, claimed that his dog could pick up on a potential hypoglycemic seizure in him from 20 feet away.

When comparing a dog's ability to sense a potential seizure through its sense of smell, the level of detection by the dog is on the order of parts per trillion (ppt). The practical sensitivity of portable electronic noses like those that you might find at PITTCON or other similar conferences is currently only on the order of parts per million—a million times less than what a dog may be capable of detecting. To create an electronic nose with a ppt sensitivity and apply it in a medical application like this would be cost prohibitive—service animals are generally provided free to their owners through charitable organizations.

Private research in actually what causes the dogs to pick up on their owner's medical seizures is just now beginning, but due to the relatively small size of the service animal community, the research will likely not be very substantive, especially when compared to the intensity of most other life science research efforts. It would be interesting to know just how much funding has been provided by either the NIH or the NSF to investigate these very real life saving animals and situations. It would be interesting, indeed, to know why there's not more government-supported research being done in this area.

To learn more about Armstrong, go to www.Dogs4Diabetics.com.

This article reprinted from R&D Magazine.