Lots of OAs

[BlueDragon]

My wife is having a problem with taking her mask off at night because the pressure is "too high." Sleepyhead reports clusters of OAs, and indeed the machine is responding by increasing the pressure to 20 cmH2O, far greater than the 8 cm she starts with.

Does anyone have any thoughts on these OAs?

Her machine is an Airsense 10 Autoset and mask is a Airtouch F20. Pressure range is set to 8-20.

screenshot-20180827-065520.png

screenshot-20180827-065620.png

[Pugsy]

Is she asleep or awake when these clusters of OAs happen?

Is she on her back if she is asleep?
When we see times where the pressure goes along great when lower and then all of the sudden needs to go a lot higher we have to wonder what changed to cause the airway to be harder to keep open....usually only 2 things will do this...either supine sleeping or REM stage sleep or maybe a combination of both.

If she is awake...they don't count but since they mess with sleep we don't want it to happen. Maybe simply limiting the max so it can't go higher and wake her up.

If she is indeed asleep...figure out if it is supine sleeping and maybe try to stay off one's back (easier said than done) but if it is REM sleep not much we can do about REM.

[BlueDragon]

Thanks for your comments.

She's asleep, but she says that it is the increased pressure that wakes her up. For comfort, maybe we can set the max pressure much lower -- since going to 20 doesn't help, might as well stop at 12 or something like that.

We think she's on her back, but I would need to set up a video camera to be certain.

[Pugsy]

You can try limiting the max and see what happens but do understand that if she is on her back (or whatever it is causing the need for more pressure) then the clusters likely will get uglier. But then again maybe she won't roll over on to her back (assuming that is the cause).

I lean towards supine sleeping as being the culprit because most people will have more than one or two REM cycles during the night.
The first one normally comes on about 90 minutes after sleep onset and then as the night goes on they come around sooner and last longer. We might have 4 to 7 REM cycles a night depending on hours slept.

If it does end up that supine sleeping is the culprit and she spends much time at all on her back then you might want to discuss the various ways to try to stay off one's back.

[esel]

To me it looks like she did not take her mask off that night as the pressure slowly comes back down after the first set of OA's.

Interesting is that the leak rate shows up with the first set of OA's.

I have noticed in my graphs that leak rate, Tidal Volume and events may have a relation.

What I would do is reduce the max pressure to 15 and see what happens.

[palerider]

Thanks for your comments.

She's asleep, but she says that it is the increased pressure that wakes her up. For comfort, maybe we can set the max pressure much lower -- since going to 20 doesn't help, might as well stop at 12 or something like that.

Is it the increased pressure, or the series of obstructive apneas that's waking her up, and she's just blaming the pressure?

Also, you say "going to 20 doesn't help" but, in the screenshot you posted, it *did* help, the apenas stopped, and pressure trailed back down.

[BlueDragon]

Is it the increased pressure, or the series of obstructive apneas that's waking her up, and she's just blaming the pressure?

Also, you say "going to 20 doesn't help" but, in the screenshot you posted, it *did* help, the apenas stopped, and pressure trailed back down.

Good thoughts, but it is hard to know whether she is waking because of the OAs or the pressure, but she is blaming the pressure.

Also hard to know if going to 20 helped in the first episode or whether she just turned over on her side because of difficulty breathing. I would have to set up a video to find out.

[djams]

I have the same machine. Looking at the large pressure increases, they're very different from my machines. For instance, the first one (approx 23:35), it took the machine around 10 minutes to increase pressure from ~10 to ~19, and there's a very strange (to me) stair step pattern in the increase. In response to the same type of activity (snore and FL) I'm pretty confident that my machine would make this jump in < 1 minute.

Can you confirm that this is the result of setting "Response" to "Soft"? I've been meaning to try this setting out eventually, but haven't gotten to it yet, and most likely never will if this is the result of it.

I'd point out that once the machine finally got to the pressure target it was trying to reach, the OA's stopped (approx 23:45).

[palerider]

I'm pretty confident that my machine would make this jump in < 1 minute.

It wouldn't.

[BlueDragon]

Response was set to "standard." I have changed it to "soft" for tonight's test. I'm not expecting that to make much difference.

Looking at her original sleep study, of a total of 167 events, 146 were in the supine position and 135 were obstructive (only 11 supine events were hypnopneas). So it's clear that OAs while on her back are the major source of her AHI, at least in the test.

Regarding the OAs stopping when the machine topped out on pressure, don't assume causation from correlation. My guess is she responded to the difficulty breathing by turning back on her side. If the increased pressure stopped the OAs, I would have expected them to resume as the pressure decreased. But they did not recur for more than an hour, by which time the pressure was almost back to normal. More likely, I think, was another positional change.

As for the stair-step pattern, notice how the pressure steps at each OA. The OAs are about 1 minute apart, and the stepping looks to take 7 to 10 minutes. The machine appears to be bumping the pressure in response to each OA. The snore event is probably the gasp as breathing resumes, which is after the OA flag.

[djams]

I'm pretty confident that my machine would make this jump in < 1 minute.

It wouldn't.

I discussed this with Pugsy awhile back (she's of the same mind as you). But for me, it DOES (or did) happen, and pretty frequently. If I go back through days when when I was still sleeping supine, I see a lot of examples of this.

Rather than re-post the chart, here's a link to the post if you're interested: viewtopic.php?f=1&t=172376&p=1260882&hi ... e#p1260873

[djams]

Response was set to "standard." I have changed it to "soft" for tonight's test. I'm not expecting that to make much difference.

Thanks for the information. I'm actually very surprised to hear this. The difference in response between your chart and the one I linked above is amazing, really. I was virtually certain you're machine was in "soft" mode. I can see that both my snore and FL were larger, and that must be what causes the difference in response, apparently.

As far as the stair-stepping - I can see it it my linked chart too. Then when I came back and took another look at yours, I realized that it's a short overall time frame with a long response time. Hence, steps are visible. A DOH! on my part.

[Pugsy]

I discussed this with Pugsy awhile back (she's of the same mind as you). But for me, it DOES (or did) happen, and pretty frequently. If I go back through days when when I was still sleeping supine, I see a lot of examples of this.

Now I remember. Your machine did set a new land record IMHO for the amount of response in such a short time. It goes against everything I have ever seen, read or experienced but it's obvious yours did it.
I can't explain it so I won't even try. It does it and you see it happen on a fairly regular schedule. Maybe your machine is on some special steroids or something. For now I put you in the outlier category...not common response but it's obvious it does respond quickly in some situations. I will be watching for other similar responses. As far as I know ResMed hasn't made any special/different changes to anything to cause it to happen so all I can do is sit here and scratch my head about it.

For Everyone....just remember that there's more to any apap auto adjusting algorithm than just OAs or hyponeas...there are snores and flow limitations also and we don't always see much evidence of either of those items on the graphs or reports to where we can say....yeah, the pressure responded to so and so. The algorithm is a complicated combination of factors that causes the machine to do what it does. I don't worry so much about the "why" and I just trust the machine to do its job. I tend to not expend much energy on stuff I have zero control over and the "why" is one of those things. It is what it is.

[palerider]

I'm pretty confident that my machine would make this jump in < 1 minute.

It wouldn't.

I discussed this with Pugsy awhile back (she's of the same mind as you). But for me, it DOES (or did) happen, and pretty frequently. If I go back through days when when I was still sleeping supine, I see a lot of examples of this.

Rather than re-post the chart, here's a link to the post if you're interested: viewtopic.php?f=1&t=172376&p=1260882&hi ... e#p1260873

*documented*:
OA/hypopnea response: Increases pressure based on current pressure every 10 s of apnea: increment max 3 when pressure is 4. Increment drops linearly down to 0.5 when pressure is 20.

Flow limitation response: S9/Air10
: uses 3 breath average FL index. Increment typically around 0.6/breath for severely flow limited breaths. Lower increment if lower FL index, high leak or as pressure increases further above 15;

Snores: S8–S9: increment max 1/breath. Lower increment if snore is less severe, high leak or as pressure increases further above 10 S10: increment max 0.6/breath for a loud snore otherwise same as S9.

Your mistake is assuming that the response to obstructive apneas is the same as the response to flow limitations and snores.

[djams]

Now I remember. Your machine did set a new land record IMHO for the amount of response in such a short time. It goes against everything I have ever seen, read or experienced but it's obvious yours did it.
I can't explain it so I won't even try. It does it and you see it happen on a fairly regular schedule. Maybe your machine is on some special steroids or something. For now I put you in the outlier category...not common response but it's obvious it does respond quickly in some situations.

I've gotta tell you that I've grown to appreciate that fast response - believe it's important for me. I never have it at the beginning of the night since I'm side sleeping now (take a bow, Pugsy). This was when it would wake me up. So I'm all in on it.

And given that BlueDragon's machine was on "Standard' response, I will give "soft" a test just to see how it compares. Someday. I'll have to intentionally sleep supine to give it an honest go.

Don't mind being the oddball outlier at all - kinda the story of my life!