HYPOPNEA verification

[dataq1]

I’ve been closely monitoring my wife’s Oscar reports for the past 60 days, and the thing that I’m struck with is the relatively low number of obstructive or clear airway events. Approximately 85% of her AHI over this period is attributable to hypopnea events. *
 
In examining her flow rate curves, it’s very easy to pick out and verify the apnea events (clear, obstructive and unclassified, but it is much more difficult to pick out and verify the hypopnea events.

So, I’m seeking to verify the hypopnea detection process used by Resmed Airsense 11, to flag hypopneas.
Resmed (as do other manufacturers) claim to compare a 10 second segment of flow to a previous (suggested to be “normal” or “baseline” flow for the patient) segment. However there seems to be some dispute about the duration of the baseline segment.  

So, in working at verification I’m using the EDF files from the SD card.  
Any comments/thoughts or suggestions?  
 
*BTW, she maintains a stable Oxygen saturation and she complains that the use of the APAP has not changed how she “feels” over the past two years of use. She questions if the APAP is actually doing anything.  Her mean AHI (60 days) is 3.8 and her mean Hi (60 days) is 3.25. Her WatchPat study was pAHI16.3 and there was no reported hypopneas (not suprising as this Watchpat study did not monitor flow at all).

[staceyburke]

You will get a lot more responses if you post her Oscar charts (chart organization at the bottom of post). We give suggestions by looking at the charts.

Also at the bottom of the page is how apnea is classified.

[dataq1]

Stacy,

 I'm fairly certain that you you don't mean post all 60 of her charts, so I'll try to find one Oscar report that most closely resembles her average performance.

[img][/img]

now a closer examination at the area of 2AM is this:

[img][/img]

So you can see that the segment ending at 02:00:52 is flagged as a hypopnea.  (seen in light blue)
However succeeding segments (shown in peach) appear (to naked eye)  to be just as likely to be hypopneas, but are not flagged.

Again, to the eye, it would seem that either the Resmed algorithm missed those peach colored segments OR the segment in light blue should not have been flagged at all. But to evaluate it fairly, one of the very first things you need to know/learn is the duration of the baseline that is the basis of the comparison. Second thing to know/learn is exactly what is being compared, is it peak inhalation rate or peak inhalation volume ( or something else)*

That's what I hoping that someone of the forum has already worked out, or an engineer  that can guide me. 

BTW, apparently Resmed uses a reduction of 25-50% (of something) for 10 seconds to detect a hypopnea. 

*the peak inhalation rate and the peak volume can be two very different things

[Sleeprider]

Periodic breathing and high flow limitation is tough. She needs pressure support (EPR) and if that causes CA, she may need the trigger sensitivity setting of the Vauto or even an ASV. As a start, I would increase minimum pressure to 7.0, use EPR 3 full time and see what happens.

[dataq1]

Thanks for the reply. We are currently in a two week experiment* so I'm reluctant to change any setting at this time. Additionally, we will be meeting with her new sleep doc next week, and I'll mention your suggestion to him.

However much I appreciate your  suggested modifications, that was not the focus of my original posting. I am interested in learning:

 1) what criteria does the Airsense 11 actually use to detect a hypopneas - in particular what constitutes the duration of baseline that is used to calculate a % reduction, and   

2) does the Resmed algorithm (for hypopnea detection) use  the peak of the flow rate at each breath or the volume of that breath.

Initially I propose to utilize the flow rate data from the EDF files and attempt to apply various time durations for the baseline to define the baseline used to detect a specific detected hypopnea.

That is what I'm seeking thoughts/suggestions/comments about.


* I consider it a prudent practice to allow each experiment to run for at least two weeks to see an overall result. The obvious exception is if in the first few days that the experiment is a disaster or harmful.

[Crimson Nape]

dataq1 - Have you reviewed this document?
AASM clarifies hypopnea scoring criteria

How are you going to proceed after you obtain your answer?  I fail to see how critical the answer is as it relates to sleep apnea therapy.

- Red

[Sleeprider]

My suggestion to use a minimum pressure of 7.0 and EPR at 3 is intended as an experiment to improve flow limitation and has a greater chance of improving results and comfort, and a small risk of actually triggering CA events.  It helps us to learn whether the mechanism at play is central or obstructive, and allows us to come back with a more definitive optimization strategy, which could range from continuing with EPR, or removing EPR from therapy and using a fixed pressure to resolve obstruction.  The flow limitation shown here without EPR is impressive, and based on my experience would be reduced well under 0.1 (95% FL), RERA by 90%, and reducing hypopnea to by at least 50-75% with EPR 3.  Having this information in-hand before your appointment would be valuable, as compared to asking the doc for a guess.  You can decide whether the change is beneficial or not, but I don't see how anyone benefits from not evaluating this change ahead of the appointment, thus blinding everyone to a potential solution. What I see in this chart is classic for UARS and I have had great success in coaching people with this pattern to a more comfortable and effective therapy. You are in "a two week experiment", and I'm telling you to just try this proven method. The change will be amazing.

[dataq1]

@ sleeprider

Again, thanks for your suggestion, but 

You can decide whether the change is beneficial or not, but I don't see how anyone benefits from not evaluating this change ahead of the appointment, thus blinding everyone to a potential solution.

 WE promised the doctor to be compliant, and we agreed that we would run "his" experiment for two weeks. Her appointment is three days away.

[dataq1]

@ Red,
Yes i am aware (and have reread the AASM recommendations and acceptables).
I believe that these are written for the sleep medical community and not necessarily for the consumer XPAP community.
Regardless (and ignoring the desaturation requirements) the reccomendations suffer from:
 
1) there is no duration criterion for the pre-event baseline, 
2) no defination of the "arousal", is it pre-event or post-event, are they referring to an EEG indcated arousal (which I suspect) or are they referring to a "sigh" (lung expansion).
3) there is an inference that the peak signal excursion is the maxima of a breath flow rate, while completely ignoring the maxima of the breath volume. That should be clarified, as it stands it would seem to infer that the rate of inhalation has more significance that the volume of ventilation for the same breath. 

What would I do with such clarity?

Unfortuantely doctors (and patients) assess their therapy in terms of numbers, particularily AHI, as a central focus. 
Just as the polysomnographic technologist and or supervisor reviews and edits events that are not justifiable as an event by applying AASM standards, we, the lowly patients, should have the same opportunity to review and if necessary edit events that are not justified. 

Sleep professionals are counting on the reliability of a patient's XPAP report. While I've no qualms about apneas that can be clearly seen on flowcharts, hypopnea's are not so easily discernible. (Hence then putting much more emphasis of the accuracy of the device's ability to properly detect hypopneas).

So, what would I do? 
 If I'm satisfied that my wife's hypopneas are being detected reasonably accurately in most instances; I would do nothing. (but will have learned a lot)
OTOH, if I find that the detection is not reasonably accurate, I would discuss this with our doctor so that her "numbers" could be in viewed in proper perspective. .

[SarcasticDave94]

Just my opinion, but if you don't believe hypopnea are detected accurately, you're expecting the doctor to listen and modify judgement accordingly. However, this is the same doctor that you agreed to not self edit settings, as in it sounds like they don't trust your setting judgement. Conclusion, these don't seem to be consistent and rather counter each other in expectation. Your trust in some doctor that didn't want you to make therapy better by self advocacy edits conflict it seems.

Regardless, bottom line it's your therapy to handle as you see fit. Seems to me to be somewhat counterproductive to ask then not act. Again just my opinion.