WatchPat

Topics related to in home sleep testing (portable monitoring)

Moderator: labman2

WatchPat

Postby SleepTech01010110 » Tue May 26, 2015 6:12 am

So wondering what peoples thoughts are on this thing. My lab is trialing a few, and I'm not really a fan. Here's a few reasons why:

1. Sleep staging from finger
2. Apnea detection from finger
3. The reports we've received so far have 70-90% of the detected sleep time invalidated, appearing to leave only the spots around detected events for sleep time and pAHI calculation
4. Pt complaints over discomfort and heat of sensor
5. One patient insisted she didn't sleep, so we ran an in-lab PSG. WatchPat pAHI: ~15; In-lab AHI: ~2.5

Sure, one patient could have a bad sympathetic cardiac electrical heebijeeby magic signal, but I don't like the odds that that would be the one patient to complain and be retested.

If it works as well as they say, bravo, awesome level of innovation, amazing work. I'm just not sure I buy it yet, or that our lab should buy these things over a cheap lease of a normal HST that actually monitors airflow, the thing in the definition of the events we monitor.

Thoughts?
SleepTech01010110
Member
 
Posts: 118
Joined: Wed Aug 13, 2008 7:13 am
Location: See that river? Follow it south until it joins another. Go due west from that city 27.3 miles. ME:-D

Re: WatchPat

Postby somnonaut » Fri May 29, 2015 4:44 am

What about consumer sleep field. Like this thing?
http://apnea.cs.washington.edu/
It would drive the dog nuts screaming at 20Khz all night.
User avatar
somnonaut
Senior Member
 
Posts: 14097
Joined: Thu Dec 11, 2003 6:03 pm
Location: http://tinyurl.com/righteousdude

Re: WatchPat

Postby SleepTech01010110 » Wed Jun 10, 2015 10:21 am

somnonaut wrote:What about consumer sleep field. Like this thing?
http://apnea.cs.washington.edu/
It would drive the dog nuts screaming at 20Khz all night.


Sleep Diagnostics: there's an app for that.

Huh. Their data indicates a higher accuracy that what the AASM says for HST devices, iirc. Crap. We're out of a job, folks, lol.

Seriously? A smartphone app? Sonar? Goodness... I equally love and hate this thing :p
SleepTech01010110
Member
 
Posts: 118
Joined: Wed Aug 13, 2008 7:13 am
Location: See that river? Follow it south until it joins another. Go due west from that city 27.3 miles. ME:-D

Re: WatchPat

Postby picklish » Wed Nov 22, 2017 10:56 am

Is this device commonly used in the States? There's been a big push for it in the UK at the moment, but I'm unsure of it's validity, despite PAT being recently accepted by the AASM. - http://www.sleepreviewmag.com/2017/04/a ... p-studies/

I spoke to someone who said that they manually score the studies, rather than accepting the automated scoring that the device provides. But I didn't find out what parameters they use to manually interpret these studies.

I found a paper from 2008 which stated:

Manual scoring/editing of the Watch-PAT signals is possible and can be easily performed. As in manual scoring of conventional signals, i.e., effort, airflow/nasal pressure and oximetry that rely on simultaneous changes in all three channels, manual scoring of the Watch-PAT 100 signals can be accomplished using simultaneous changes in the peripheral arterial tone, pulse rate and oximetry. Identification of respiratory events can be accomplished based on the well documented intensive sympathetic activation accompanying respiratory events termination. Sympathetic activation causes attenuation of the PAT signal, indicative of vasoconstriction, coupled with pulse rate acceleration in addition to the typical changes in oximetry.
- https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2276832/

...but I'm not entirely sure what parameters to use - how much " intensive sympathetic activation" is required? Over what time base? What does that look like? What should I be looking for on the screenshot below to confirm whether the automated scored respiratory event at 03:52:20 is legitimate?
Attachments
watchpat.jpg
watchpat.jpg (375.05 KiB) Viewed 133 times
User avatar
picklish
Member
 
Posts: 27
Joined: Sun Jul 31, 2011 4:42 pm
Location: Scotland, UK

Re: WatchPat

Postby somnonaut » Mon Nov 27, 2017 4:48 pm

I have always had a problem with Watch-Pat that no one from the company would ever address. The unit relies on:
"manual scoring of the Watch-PAT 100 signals can be accomplished using simultaneous changes in the peripheral arterial tone, pulse rate and oximetry. Identification of respiratory events can be accomplished based on the well documented intensive sympathetic activation accompanying respiratory events termination."
But, this :activation" is many times not functional in patients with chronic OSA. I speak of those patients that have no cardiac response to obstructive apneas. The patients who have broken cardiac systems with no brady/tachy in response to OA. We have all seen them. What about those patients? Are the patients triaged for properly working cardiac systems prior to usage of WatchPat. Do these studies come back as false negatives?

As to the waveforms you post, is there any way to re-scale oximetry? Why is it over 100% anyway? What do the green "R" lines in the top overview represent? Respiratory events? Is actigraphy calibrated to waking bio-cal movements? Top box are 60 min apart while bottom tracings are at 00:01:40 (100 minutes) apart.
User avatar
somnonaut
Senior Member
 
Posts: 14097
Joined: Thu Dec 11, 2003 6:03 pm
Location: http://tinyurl.com/righteousdude

Re: WatchPat

Postby picklish » Tue Dec 12, 2017 10:53 am

I've also made the comment about patients who don't have a cardiac response to OA potentially resulting in false negatives to Itamar, will see if they give a response. It only took me a couple of minutes of looking through historic patient studies to find an example of this, it would be interesting to know how common it was.

To answer your questions, "R" on the hypnogram indicates respiratory events. I'm not sure what is happening with the actigraphy - I'm not used to looking at one night of actigraphy, but rather over two weeks or so. Looking at a full night overview its seems any actigraphy movements = wake, and resp. events are discounted at this time.

In terms of manual scoring, I was told they are going to provide some data and guidelines on this in 2018. When doing a visual inspection of the automated scoring I was told to look for the following:

PAT amplitude and pulse rate moving out of sync with each other (attenuation of PAT, increase of pusle), with an associated desaturation. However, it wasn't clear if all three events had to occur before the event was scored. Some examples:
Attachments
two events auto scored small.jpg
the pulse and PAT rates look quite different to each other here, but two events are scored. Desaturations are not marked (have to be over 4% in automated scoring)
two events auto scored small.jpg (155.43 KiB) Viewed 52 times
pat increase or no change, pulse increase or no change, desaturation - event small.jpg
This appears to be only a desaturation - is that sufficienct to score an event?
pat increase or no change, pulse increase or no change, desaturation - event small.jpg (141.36 KiB) Viewed 52 times
pat decrease, pulse increase, no oxy change - no event small.jpg
Does event scoring require desaturatiuon, if this isn't scored as an event?
pat decrease, pulse increase, no oxy change - no event small.jpg (148.87 KiB) Viewed 52 times
User avatar
picklish
Member
 
Posts: 27
Joined: Sun Jul 31, 2011 4:42 pm
Location: Scotland, UK

Re: WatchPat

Postby somnonaut » Wed Dec 13, 2017 1:52 am

GREAT WORK on the postings. Thanks. So, there is not specific "event" start and stop tag, just a generic "respiratory event is somewhere here."
I was just having a similar conversation with a colleague as to how during HSAT we would need EEG (arousals at minimum) to be able to perform Limb Movement detection as the physiological endpoint of PLMS is disturbance of sleep through arousals, no other physiological impact per se. This lead to discussion that we really should have a DEPTH of Arousal index which would include various inputs to gauge just how arousing the change in state is: EEG only, EEG+HR, EEG+Oximetry, EEG+Resp event. This would allow a holistic approach and carve out non-EEG based recording from trying to gauge EEG based arousals.

Something along these lines.

What say others?
User avatar
somnonaut
Senior Member
 
Posts: 14097
Joined: Thu Dec 11, 2003 6:03 pm
Location: http://tinyurl.com/righteousdude


Return to Home Sleep Testing (HST)

Who is online

Users browsing this forum: No registered users and 3 guests

cron