Seizure Warning Dogs

[wittgenst3in]

Just found a article on the web about dogs who can supposedly detect when their owner is about to have a seizure:

http://www.freep.com/news/health/dogs22_20040622.htm

Seems quite interesting, and would be very useful for sufferers of epilepsy if true. However it seems to me like they could quite easily be fooling themselves.

My mother had a story from when she was little. She had a small dog, and when her parents went out she would get a little concerned. The dog would pick up on this and start to look worried and scared too, my mother would then look at the dog and think it must have heard an intruder or something. This causes her to worriedly look around as if something is outside, and the dog picks up ..... etc etc.

Her parents would come home to find both her and the dog a nervous wreck, for no good reason.

I'm wondering is something similar might be the case here.

Do you think it's possible to psychologically force or 'condition' someone to have a seizure?

 

[Goshawk]

I don't see "psychological conditioning" listed as a cause of epilepsy.
http://yourmedicalsource.com/library/epilepsy/EPI_causes.html

And from the article:

In all, 40 percent of the dogs in the study displayed specific reactions to a seizure, and about 15 percent of them appeared able to anticipate a seizure, the researchers report.

The gist of the article is an excited, "Dogs can predict when a child is going to have a seizure!", but then it says that only 15% of them were reportedly able to do so. I don't think that 15% constitutes a statistically significant percentage, especially if you're allowing for reporting errors, exaggeration, misidentifying the dog's behavior, and selective memory when talking to this enthusiastic pre-sold doctor person.

It's also only 15% of 122 dogs, which I make it as 18.3 dogs. Eighteen dog owners who think their dogs can predict seizures is not a big statistical bulge.

 

[wittgenst3in]

I don't see "psychological conditioning" listed as a cause of

It's also only 15% of 122 dogs, which I make it as 18.3 dogs. Eighteen dog owners who think their dogs can predict seizures is not a big statistical bulge.

Good point. I can see the possibility of only 'talented dogs' being able to do it though.

They say they want to actually do trials which looks like the next logical step.

 

[Deetee]

Hmmmm.....
The full article is fairly brief on detail and possible biases.
I do believe there is something about dog's perceptive behaviour that can detect behavioural changes in humans, and there have been reports of dogs anticipating the onset of diabetic comas (smell of glucose and ketones?- even humans can detect ketoacidosis by smell)

However, taking a sceptical view of these totally uncorroborated and subjective reports........
The median anticipation time was 2.5 (0.9, 15.0) minutes with a range from 10 seconds to 5 hours.
So anything a dog does up to 5 hours prior to a seizure is ascribed to its ability to predict the seizure.....?

Anticipatory behaviors were never demonstrated without a subsequent seizure
This in itself is highly suspicious of reporting bias. No "test" is that accurate. If dogs are that sensitive, some false positives are anticipated.
Also, you will note that the most common form of doggy alert was "face licking" (60%) . Now show me any family with kids and a dog where this does not occur routinely and frequently as part of normal interaction? .....And then the families claim "NO FALSE POSITIVES"????

A Sheltie-Spitz cross would forcibly sit on her toddler and not allow her to stand prior to a drop attack
Did anyone consider the possibility that the dog might have precipitated an attack? I can't imagine that having a dog forcibly squashing a toddler would have no physiological consequenses...hyperventilation-lowering of fit threshold is one I can think of. Similarly, having a bloody great Pyrenees mountain dog stalking a toddler around for hours might disturb the child somewhat.

I would have been happier with a publication highlighting the possibility of this type of behaviour if it were of better quality.

How might this be scientifically tested? Ideally you would induce fits under experimental conditions and observe the dogs behaviour prior to this. Zero chance of ethical approval tho.

Edited to add - I now see the face-licking was only a respose to the fit, and not anticipatory. Is it really very unusual for a dog to lick the face of its "master" when the master has just collapsed unconscious onto the floor?

 

[wittgenst3in]

I can't imagine that having a dog forcibly squashing a toddler would have no physiological consequenses...hyperventilation-lowering of fit threshold is one I can think of. Similarly, having a bloody great Pyrenees mountain dog stalking a toddler around for hours might disturb the child somewhat.

That's what I thought might be the case. "Gee honey, juniour's having more seizures nowadays, better make sure that dog stays with him all the time".

Don't get me wrong; I love dogs especially noble ones like German Shepherds, but I'm not sure I'd feel the same if it was twice the size of me.

How might this be scientifically tested? Ideally you would induce fits under experimental conditions and observe the dogs behaviour prior to this. Zero chance of ethical approval tho.

I think you'd just have to wait long enough and you'd get a seizure. No need to induce anything. Alternatively does anyone know if portable EEG packs are available. I knwo they did this with cardio monitors a while ago, but I'd imagine there's a hell of a lot more data in a brain wave.

I did a lab session last semester with a automatically diagnosing ECG machine. We ran tests on perfectly healthy 21yr olds, and got printouts of conditions as long as your arm! (Acute myocardial infarctions, sinus arrythmia etc.) The whole thing is heavily dependent on how you move around, who is touching the bed, etc. Not to say the machine dosen't work, but if the leads are'nt just right, and the bed is knocked I can see this scaring patients unnecessarily.

 

[bug_girl]

There is a big difference between untrained pets and trained seizure dogs. Several people i know have *trained* dogs, and they are endorsed by the epilepsy foundation:
http://www.epilepsyfoundation.org/epilepsyusa/aboutseizuredogs.cfm

Additional links:
http://www.4pawsforability.org/seizuredogs.htm
This one has a nice discussion of the difference between seizure support dogs, and seizure response dogs. Two very different things.

In case anyone forgot, i have epilepsy

 

[Suezoled]

Bug Girl, you also have antenna...

 

[bug_girl]

Alas, the antennae have been remarkably uneffective at detecting oncoming "spells".
I even dented one very badly once in a fall.

 

[Suezoled]

Alas, the antennae have been remarkably uneffective at detecting oncoming "spells".
I even dented one very badly once in a fall.

So your sense of smell is reduced??? Oh No!

 

[wittgenst3in]

Alas, the antennae have been remarkably uneffective at detecting oncoming "spells".
I even dented one very badly once in a fall.

Sorry to hear about that.

So if you don't mind me asking, what sort of warning, if any, do you get? I gather from the antenna thing it's not enough to sit down somewhere?

If you are upset do you think you are more likely to have a seizure?

 

[Deetee]

Alas, the antennae have been remarkably uneffective at detecting oncoming "spells".
I even dented one very badly once in a fall.

And just look what happenned to your eyes! (Mine usually turn black after trauma)

 

[bug_girl]

So your sense of smell is reduced??? Oh No!

Yes. Now I smell terrible! (ba dump bump )

I have temporal motor epilepsy.
i am lucky enough to have a clear warning aura (flashing lights/loss of sight in one eye).
I also have been really fortunate that drugs work fairly well on me, and that as the time since my initial head injury passes, i seem to have fewer and fewer problems. I just had my 13yr anniversary this year.
All of this is much preferable to being dead. or a veggie.

definately, when i am tired or angry, it is much, much worse. PMS has a whole new meaning for me

 

[Rolfe]

Bug Girl obviously has this one taped, but I thought I'd just add my tuppenceworth.

I watched a very interesting programme on local BBC TV about a woman with severe non-drug-responsive epilepsy who had a trained seizure dog, and claimed he completely revolutionised her life. The dog was trained by means of treats and rewards to respond to the virtually subliminal signs that the woman was about to have a seizure. The time-scale was pretty short, I thought, but it was still just enough for the woman to get somewhere safe and lie down. The dog stayed with her and wouldn't let her rise until the signs showed that the fit was over.

She explained that the dog had been trained so as to regard her having a fit as the most important experience in his life, one which he waited for with great expectation, and reacted to accordingly. He got his treat after the fit was over, obviously. She was able to live a much more independent life once the dog was in residence.

I believe that the idea of training dogs to do this first came from one or two people noticing that their pet dogs seemed to be aware that a fit was coming on, but I would also imagine that there might be a lot of exaggeration about what pets might be capable of. The trained dog was quite something, though. As are guide dogs and hearing dogs, of course, and those dogs trained to fetch and carry for disabled owners.

Rolfe.

 

[Dragonrock]

The first I heard of this was in a "Chicken Soup for the Pet lover's Soul" story. The family got a dog because they heard that it helps reduce seizures by relaxing the person. The dog would warn the girl before she seized and the girl would put in her mouth piece and curl up on the couch. I had always wondered if that one was true.

 

[tracer]

Good point. I can see the possibility of only 'talented dogs' being able to do it though.

If it were Lassie, not only would she see the seizure coming half an hour in advance, she would pick up the phone and call a Trauma Response team.

Then she'd go back to her particle accelerator and continue writing her paper on detecting supersymmetric particles.

 

[Dragonrock]

Then she'd go back to her particle accelerator and continue writing her paper on detecting supersymmetric particles.

I thought she was working on constructing the stable wormhole?

 

[Goshawk]

The first I heard of this was in a "Chicken Soup for the Pet lover's Soul" story. The family got a dog because they heard that it helps reduce seizures by relaxing the person. The dog would warn the girl before she seized and the girl would put in her mouth piece and curl up on the couch. I had always wondered if that one was true.

Probably not, mainly because AFAIK (although I welcome enlightenment) they don't tell epileptics to "put in a mouthpiece so you don't bite/swallow your tongue", at least, not anymore. So it's either completely fictional, Urban Legend-ish, or dates from about the 1950s.

http://www.cnn.com/HEALTH/library/DS/00342.html

If you see someone having a seizure, call for medical help immediately and then follow these tips:

Gently roll the person onto one side and put something soft under his or her head.
Loosen tight neckwear.
Don't try to put your fingers or anything else in the person's mouth. The tongue can't be swallowed...

http://www.tdh.state.tx.us/news/acc1101.htm

"One of the myths about epilepsy is that people having seizures can swallow their tongue," Flores said. "It's not possible. Never put anything in someone's mouth who is having a seizure..."

And Google searches under things like "epileptic mouthpiece" are turning up zilch.

All those Chicken Soup things are generally acknowledged to be written by someone, anyway.

 

[tracer]

I thought [Lassie] was working on constructing the stable wormhole?

Well, yeah, in her spare time. The university isn't funding her wormhole research, so she's gotta pay the rent somehow.