Flow limitations and auto-adjusting

[cathyf]

So what's the story with the ResMed algorithm and "flow limitations"?

What I can see from sleepyhead is that my machine is continuously going into hair-on-fire zoom-the-pressure-up mode and raising the pressure, and it appears that this is in reaction to pretty much constant flow limitations. But, at least in my case, the flow limitations are NOT a sign of impending apneas, but merely mean that I am asleep, and going more than a minute or two without a flow limitation only happens when I'm awake. On my sleepyhead graphs, every night I go hours and hours with continuous flow limitations and no apneas, with the machine "pegged" at whatever I have set my maximum pressure to, whether that's 10, 14, 17 or 20. When I have the top pressure limited to 10, I sometimes will get a couple of really ugly clusters that last for 5-30 minutes, while if I leave the machine wide open at 20 my AHI is really low and it's just the occasional one or two events which might not be real. In my sleep study, my non-REM AHI was only 14, while my REM AHI was 45. I can see that apneas are associated with places where the respiration rate pops up, which I think is REM sleep.

So why don't I just leave the machine wide open and spend most of the night at 20? There are definite trade-offs between the pressures and sleep quality for me. Besides the high pressures blowing the mask off and forcing me to crank the straps really tight, I deal with aerophagia, and not just the annoying/embarrassing belching and farting, but terrible gas pains that wake me up. So I have to limit my upper pressure to prevent the machine from waking me up. Because of the way the resmed algorithm seems to be relating to the FLs, the machine spends most of the night at whatever max pressure I set it at, rather than adjusting so that I only get the high pressures when I need them.

What I think I need is for the machine to ignore FLs, and only respond to actual events. So most of the time I would cruise along at my minimum pressure, which is NOT pumping air into my stomach. But then if I have an apnea or two, the machine can zoom the pressure up, leave it up for 20-30 minutes, then lower it back down and not raise it again until it sees another event. (Or what I call a "sub event" -- I have enabled the option in sleepyhead to flag events that are between 7 and 10 seconds, and those are pretty reliable indicators of a real problem in a way that FLs are not.) That would limit the amount of air in my gastrointestinal tract to something manageable, while still giving me the higher pressures I need during those relatively brief REM periods.

There is no way to get a machine that does this, right?

It seems to me that what is really needed is some artificial intelligence and machine learning here. We need machines that can learn things about our breathing over time and adjust the algorithm accordingly.

[Pugsy]

No auto adjusting mode machine will do what you are asking that I am aware of.

Flow limitations are part of what the algorithm is designed to try to prevent because they are normally considered warning signs of an impending airway collapse.

Snores and flow limitations are a big part of the driving force behind the machine's algorithm to hold the airway open.
It's trying to prevent the warning signs from progressing to the point that the airway ends up collapsing and you get enough of a collapse to earn an apnea flag of some sort.

[robysue]

cathyf,

You have my sympathies. The PR Dreamstation and System One Autos respond to flow limitations, but they are a bit less aggressive about it than the Resmeds are. Even so, I still find that I absolutely must cap my max IPAP or else it will pretty much inch its way all the way to the top without doing a dang thing to improve my sleep and often the unneeded excess pressure is disturbing to my sleep.

My own advice is this: Be aware of how your machine is designed to work. Set your max pressure high enough to keep the AHI under control, but low enough where the flow limitation pressure increases do not go into the range where your stomach is bothered. If you can't find a decent compromise that makes the stomach happy enough, but also keeps the AHI sufficiently controlled so that you feel decent in the morning, you might consider switching to a machine that has a less aggressive response to flow limitations.

[rkl122]

Perhaps side sleeping would help?

[cathyf]

This is what I mean by the "Flow limitation = asleep" and I'm pegged at the max pressure all night, and except for that little cluster, the FLs don't mean anything



So then I had a really ugly night which caused me to raise the pressure to 12:


And with the max at 12 it's doing the same thing -- reacting to FLs, but not needing to:

[cathyf]

Perhaps side sleeping would help?

I'm strictly a side sleeper. That's what causes the leaks -- I'm jostling the mask off against the pillow...

[palerider]

This is what I mean by the "Flow limitation = asleep" and I'm pegged at the max pressure all night, and except for that little cluster, the FLs don't mean anything

well, FLs mean you're not breathing well.

I'd like to see some zoomed in breaths.

[Pugsy]

This is one of those situations that we talk about when the machine wants to increase the pressure because it thinks it has a real reason to do it but maybe the reason isn't as big of a problem and sometimes trying to fix a problem causes more problems than the original problem causes.

Per the machine's algorithm it's going to respond to the flow limitations with more pressure...that's the job it is designed to do. Now if the flow limitations themselves aren't causing a problem and you are sleeping well and feeling decent for the most part then the flow limitations may not really need that aggressive of a treatment but the machine doesn't have any way of knowing what to do beyond whatever it's original algorithm is telling it to do.

You can do what you have been doing...let the machine pretty much stay at the max all night and since allowing more maximum creates problems with aerophagia then you pretty much have to make a compromise and decide which problem is the worse problem for you...an occasional ugly cluster in REM or risking the aerophagia with more maximum.
I could tell you to change over to cpap mode but essentially the machine is pretty much doing a fixed pressure anyway except for a couple of little downward blips.

Are you sleeping decently and feeling decently?

I get the sense you would like the machine to ignore the FLs and only respond to the apnea events that happen in REM since you sometimes need more pressure in REM and not so much the other parts of the night except for those pesky FLs. So you could use a much lower pressure for the biggest part of the night and only use the higher during REM where it seems to be needed more. I don't know of any machine out there that will do that. FLs are considered a prime warning sign for impending airway collapse for the bulk of the population. Not all the population but the bulk of the population....and these machines are designed to respond to what happens in the bulk of the population and the other part of the population just has to deal with it.

I don't know if another machine that isn't perhaps quite as aggressive will/would do what you want or not.
Is there any way you could try a different brand (like Respironics which is not quite as aggressive) to see if maybe it might be more apt to do what you want?
I have used both brands and do well with either but I never have really had much of a problem with FLs so I am not a good yardstick for comparing the response of ResMed to Respironics just for FLs.
My Fls reported on either brand have always been pretty much non existent.
Any way you could try something else? Talked to your DME about maybe trying a demo unit for a week???

[robysue]

cathyf,

To cut to the chase: Pugsy asks the most important and only relevant question concerning your flow limitations:

Are you sleeping decently and feeling decently?

Or perhaps in light of the fact that you've mentioned aerophagia: Are you sleeping decently and feeling decently when the pressure is low enough to not trigger the aerophagia?

If you are sleeping decently and feeling decently when your max pressure is set low enough to prevent the aerophagia, then you may be better off not worrying about the fact that your pressure is hugging the max value all night long. There's nothing wrong with that as long as you are sleeping well, feeling well, and the AHI is acceptable. Yes, you'll have the occasional bad night and you may see more bad clusters than someone else. But there's more to getting a good night's sleep than a great AHI.

Now if you are NOT sleeping decently and/or you are NOT feeling decent even when you do have a good AHI, then it could be those flow limitations are real and you just haven't increased the pressure enough to deal with them. But here's the thing: You have to also contend with the aerophagia, and a lot of people don't have to do that. It's easy to say "Increase the pressure" when you have never experienced severe aerophagia. And it's not particularly useful to be told "Increase the pressure" if you are already using as much pressure as your particular stomach can handle.

So tell us about how things are going in terms of the subjective quality of your sleep and the subjective quality of how you feel every day. Good enough when the aerophagia is under control? Or still pretty bad even on a night when the AHI is low and you think you slept at least ok?

[cathyf]

Or perhaps in light of the fact that you've mentioned aerophagia: Are you sleeping decently and feeling decently when the pressure is low enough to not trigger the aerophagia?

I'd say that in general I do feel pretty good except for the occasional isolated day when my AHI jumps for no discernible reason. And the clusters do tend to be short -- a couple of minutes at a time is most common, with the 10-15-20-minute clusters being more of an every-couple-of-weeks thing. The main problem I have is struggling with leaks all night. I wear an Amara View, and it only comes in one diameter (the s/m/l size is only about the depth) and my face is too small. Besides the aerophagia, I'm vain enough that I hate leaving the house in the morning with an obvious ring around my mouth, caused when I clamp down the mask tight enough to stay sealed at those high pressures.

Also there is something a little odd about the reported FL statistics. There are lots of places where the FL is being scored as zero although the waveform is really ugly, and also scored as both zero and very high (.5 or more) over the course of a single long OA or hyponea. Really? Wouldn't you think that "obstruction" would count as a total flow limitation, but sometimes it does and sometimes it doesn't. I'm kind of suspicious that the machine is reporting "0.0" sometimes when it's really zero, and sometimes when it's a "no data" condition.

[Pugsy]

Remember the definitions/criteria needed for OA/hyponea flag....X amount of flow reduction AND at least a 10 second duration.
It wouldn't be impossible for some of your flow limitations/reductions to be almost totally obstructed but if they only last 9 seconds they don't get a flag of any kind.
This fact could maybe be why the algorithm goes after FLs so much. They really aren't all that benign unless they are from the nasal mucosa being swollen and that's a totally different issue.

If you are using a full face mask because of chronic nasal congestion issues...maybe the FLs are from the congestion and not down in the usual area of the airway. The machine only measures air flow...not where the flow is reduced.

Have you tried a mask liner to help with leaks and the lines on the fact that the mask causes?
Have you tried/considered going simply with a lower fixed/maximum (because it's essentially doing that anyway) to see what would happen in terms of how you feel/sleep and maybe compromising and letting a few OAs slide?
Would maybe that compromise be something that is workable and be maybe worth it if you could give up the leak issues or the lines on the face?

[robysue]

Or perhaps in light of the fact that you've mentioned aerophagia: Are you sleeping decently and feeling decently when the pressure is low enough to not trigger the aerophagia?

I'd say that in general I do feel pretty good except for the occasional isolated day when my AHI jumps for no discernible reason. And the clusters do tend to be short -- a couple of minutes at a time is most common, with the 10-15-20-minute clusters being more of an every-couple-of-weeks thing.

I would say that all this indicates that it is reasonable to set the max pressure where your stomach can handle it without the aerophagia issues and just not worry about the fact that you are running at max pressure for most of the night due to the fact that machine is scoring a lot of flow limitations.

The main problem I have is struggling with leaks all night. I wear an Amara View, and it only comes in one diameter (the s/m/l size is only about the depth) and my face is too small. Besides the aerophagia, I'm vain enough that I hate leaving the house in the morning with an obvious ring around my mouth, caused when I clamp down the mask tight enough to stay sealed at those high pressures.

First a question: When you say "at those high pressures" what do you mean? The data you showed had max pressures of 10 and 12cm. Are you talking about when the machine is at 10cm? (As an aside, 10-12cm is pretty common around here and many people actually need a lot more pressure---as in they need the max pressure set in the 16-20 range to keep their AHI down where they want it.)

Now some comments:

1) It sounds as though you may want to try a different mask. The Amara View may just not be the mask for you. In looking at the description of the Amara View on cpap.com's web page, it looks like the big positive for the Amara View is the lack of a forehead support. If the idea of a forehead support is not something you are willing to consider, you might want to see if the AirFit F10 or the Quattro FX are better fits for your face. If you are willing to try a mask with a forehead support, you might find that the forehead support will take some of the pressure off area around your mouth. Right now the weight of the mask is being supported only by the area around your mouth where the air cushion sits. With a forehead support, some of the weight will be shifted to your forehead.

2) Fixing the leaks should lead to less waking up during the night, and less waking up might lead to significantly less aerophagia. Once you fix the leaks, you may find that your stomach can tolerate a bit more pressure. Or not.

3) Usually "clamping down" on a mask to the point where it leaves rings in the morning means that you are over tightening the mask. It looks as though the Amara View uses an air cushion to seal the mask. Once the mask is over tight, the air cushion cannot fully inflate and properly seal. JanKnitz has a good essay about Taming the Mirage Quattro that you ought to read. Although you are not using the Quattro, the tips she gives about how to properly get a seal on the Quattro without over tightening the mask apply to any FFM that uses an air cushion to seal the mask.

4) Adding a mask liner might help with the leaks. Consider your sleeping position as well. If you sleep on your sides or stomach, you may need to position your head at the side of the pillow so the mask can drape over the edge and not be pushed up by the pillow. You might want to look into the special CPAP pillows that are sold. They have cutouts specifically designed to let the mask hang down a bit in order to help maintain the seal.

Also there is something a little odd about the reported FL statistics. There are lots of places where the FL is being scored as zero although the waveform is really ugly, and also scored as both zero and very high (.5 or more) over the course of a single long OA or hyponea. Really? Wouldn't you think that "obstruction" would count as a total flow limitation, but sometimes it does and sometimes it doesn't.

The scoring algorithms for FL are NOT well understood around here because the manufacturers are pretty close mouthed about exactly what the machine is looking for. That said, my impression is that once an OA or H is being scored by a Resmed, the machine kind of quits worrying to much about properly scoring the amount of FL and so the FL during an event is basically whatever it was just before the event. In other words, if the machine was scoring little or no FL right before the event, the FL curve is likely to stay down near 0.0 during the event. Likewise, if the FL right before the event was scored up around .5, then the FL curve is likely to stay right around 0.5 during the event.

I'm kind of suspicious that the machine is reporting "0.0" sometimes when it's really zero, and sometimes when it's a "no data" condition.

That may be a reasonable assumption if you see 0.0 at times when the leak rate is close to or above the 24 L/min threshold. So it's reasonable to ask: Do the 0.0 areas of the FL curve correspond to areas with higher than normal leaks?

[rkl122]

First a question: When you say "at those high pressures" what do you mean? The data you showed had max pressures of 10 and 12cm. Are you talking about when the machine is at 10cm? ...............

Good question, because the data above also indicate that there's hardly any leak issue at all (assuming the redline setting is at default) - at least not one affecting therapeutic efficacy. When the OP says

I'm strictly a side sleeper. That's what causes the leaks -- I'm jostling the mask off against the pillow...

perhaps she's overestimating the leaking. The "jostling" of the mask could be causing the anomalous responses in the flow rate. As PaleRider suggests, it'd be worth seeing a couple of those incidents at the breath zoom level.

-Ron

[robysue]

perhaps she's overestimating the leaking. The "jostling" of the mask could be causing the anomalous responses in the flow rate. As PaleRider suggests, it'd be worth seeing a couple of those incidents at the breath zoom level.

It is clear that the OP's leaks are not enough to affect the efficacy of her therapy.

But ....

To the OP writes:

The main problem I have is struggling with leaks all night. I wear an Amara View, and it only comes in one diameter (the s/m/l size is only about the depth) and my face is too small. Besides the aerophagia, I'm vain enough that I hate leaving the house in the morning with an obvious ring around my mouth, caused when I clamp down the mask tight enough to stay sealed at those high pressures.

In other words, cathyf has been able to "successfully" prevent the annoying leaking by over tightening her mask to the point where it leaves an obvious ring around her mouth.

So my interpretation of what cathyf writes is is pretty simple: If she doesn't clamp down on the mask she gets leaks that are large enough to disturb her sleep---i.e. she struggles with fighting leaks all night. If she clamps down on the mask enough to end the leaks, she gets a noticeable ring around her mouth and the ring lasts long to make her feel self-conscious in the morning.

And here's the thing: If leaks are causing you to struggle with them all night long, they've got to be dealt with---even if they are well below the official large leak territory.

Finally cathyf's problem at this point is NOT "I feel bad and might it be from the FL?" Cathyf's problem is: "I have aerophagia and I either have to deal with leaks all night OR I have to clamp the mask down so tight that I get an obvious ring around my mouth." So in my humble opinion, there's not much to be gained from looking at the flow rate data to evaluate just how bad her flow limitations actually are.