Thoughts about the future of wide open 4–20 setting/detecting optimal APAP pressure

[alexander]

Edit: I’ve gotten many things wrong. Please ignore most of what I’ve written, and rather go to the second page.

In my country it seems like most get their machines set up with 4–20 as their pressure. My current understanding of optimal APAP pressure is that it should’t be too wide, not only because it takes too much time to reach the pressure needed in order to treat the apneas, but also because large jumps may disturb the quality of sleep. Also, the minimum pressure shouldn’t be so low that breathing becomes strenuous. Doesn’t this logic pretty much cover it for most cases – or am I over-simplifying things?

What I’ve been thinking about is all the people with 4–20 pressures. It’s not ideal, and it may work for some, but if the logic described in the last paragraph has some truth to it, then there has to be lots of people suffering from a treatment that is far from ideal. And these people that I’m thinking about might not have the ability to analyze their data and make adjustments. And there might not be resources available in the health care system to optimize their treatment.

So what I’m thinking about is something that many discussed in a previous thread that I posted, and that is AI, ML, and those kind of things. The quality of the data and algorithms are only going to improve. I’m just wondering what your feelings are towards this. Wouldn’t it be possible to have a machine that could start out with something similar to a wide open setting, and then gradually learn the optimal minimum and maximum pressure for each session? And then gradually change the pressures over time as the person gains or loses weight?

This might be a bit controversial, but still – isn’t this pretty much in line with where technology are moving? And couldn’t this sort of thing make more people wake up well-rested? I do think that there should still be an option for those wishing to lock minimum and maximum pressures to specific numbers. And I totally get that some are opposed to automagically controlled things (I used to be myself, but I’ve chosen to let go).

[raisedfist]

I believe things will become more automated as time goes on. Whether that makes them "smarter" I suppose will still be a topic of debate. For example: If I set up a VAuto with a min and max EPAP currently, even though I am awake, I have had the EPAP increase substantially over time for no reason even when I am just lying in bed awake.

[Goofproof]

You are correct, we call 4 cm to 20 cm the failure setting, because the machine doesn't go up fast enough to stop Apneas. Use The Software Oscar, to find the correct working apap pressure, 2 cm under and over your cpap pressure works the best. If you don't know your cpap pressure, I'd start 7 to 11, and post the results here. Many here think 20 is fine for the high, I do not, just because your cars speedo goes to 120 mph, doesn't mean you won't have to pay the fine! Jim

The fine can be the machine chasing leaks,

[jimbud]

You are correct, we call 4 cm to 20 cm the failure setting, because the machine doesn't go up fast enough to stop Apneas. Use The Software Oscar, to find the correct working apap pressure, 2 cm under and over your cpap pressure works the best. If you don't know your cpap pressure, I'd start 7 to 11, and post the results here. Many here think 20 is fine for the high, I do not, just because your cars speedo goes to 120 mph, doesn't mean you won't have to pay the fine! Jim

The fine can be the machine chasing leaks,

And aerophagia. (running from)
JPB

[Dog Slobber]

Leaving and APAP at 4-20 and sending a patient on their way, is in all probability negligent.

I'm also guessing it's a rarity, with the exception being home titration.

Having said that, I agree that next generation machines should take advantage of machine learning for establishing new minimums.

The ResMed AutoSet for Her algorithm can set a new minimum for a session. Establishing and revisiting new minimum pressures based on previous nights should not be a problem.

But why would we want AI and machine learning to establish maximum pressure? There is no reason to. Just as we should not be setting maximums, neither should the machine. If it needs to go higher it will, and shouldn't be handcuffed.

As always when talking about not setting a maximum, recognize that there are exceptions, such as aerophagia, where setting a maximum pressure is important. But that just re-enforces another reason why Machine Learning should not be used to establish a maximum pressure, only a minimum.

The device has no way of knowing you are suffering from Aerophagia during therapy and certainly not during the day while off the device.

[Goofproof]

Leaving and APAP at 4-20 and sending a patient on their way, is in all probability negligent.

I'm also guessing it's a rarity, with the exception being home titration.

Having said that, I agree that next generation machines should take advantage of machine learning for establishing new minimums.

The ResMed AutoSet for Her algorithm can set a new minimum for a session. Establishing and revisiting new minimum pressures based on previous nights should not be a problem.

But why would we want AI and machine learning to establish maximum pressure? There is no reason to. Just as we should not be setting maximums, neither should the machine. If it needs to go higher it will, and shouldn't be handcuffed.

As always when talking about not setting a maximum, recognize that there are exceptions, such as aerophagia, where setting a maximum pressure is important. But that just re-enforces another reason why Machine Learning should not be used to establish a maximum pressure, only a minimum.

The device has no way of knowing you are suffering from Aerophagia during therapy and certainly not during the day while off the device.

and Centrals and excessive leaks to to the higher unneeded pressures. The machine can chase snores making leaks which make more snores. That is the reason to rein in the high end to two, cm over the needed treatment pressure. Jim

[jimbud]

and Centrals and excessive leaks to to the higher unneeded pressures. The machine can chase snores making leaks which make more snores. That is the reason to rein in the high end to two, cm over the needed treatment pressure. Jim
[/quote]

Your point is well taken.
I have just dropped my high from 20 to 14.
Many false events have my machine raising the pressure needlessly.
I have put a cap on it to try out and see if I can reduce wake ups.
My flow rate is a mess some nights.

Good luck to me from me.
JPB

P.S. I snore no more.

[palerider]

In my country it seems like most get their machines set up with 4–20 as their pressure. My current understanding of optimal APAP pressure is that it should’t be too wide,

More properly, "the min shouldn't be that low". in the vast majority of the cases, the max pressure should be left at 20, there's no valid reason to lower it.

[Goofproof]

and Centrals and excessive leaks to to the higher unneeded pressures. The machine can chase snores making leaks which make more snores. That is the reason to rein in the high end to two, cm over the needed treatment pressure. Jim

Your point is well taken.
I have just dropped my high from 20 to 14.
Many false events have my machine raising the pressure needlessly.
I have put a cap on it to try out and see if I can reduce wake ups.
My flow rate is a mess some nights.

Good luck to me from me.
JPB

P.S. I snore no more.
[/quote]

14.5 cm to 17 cm for me, some light snoring but AHI from .5 to 2.0 max slight mouth breathing with nasal mask but leak vent rate under 38, well within range. As long snoring doesn't wake you up or shake the dust out of the rafters, I don't worry about it, Jim

[palerider]

and Centrals and excessive leaks to to the higher unneeded pressures. The machine can chase snores making leaks which make more snores. That is the reason to rein in the high end to two, cm over the needed treatment pressure. Jim

No, not at all.

[alexander]

Seems like the minimum pressure set too low is an issue, but that the AutoSet for Her is able to set a new within a session. It would still be beneficial if the detected optimal minimum pressure was remembered from the previous session.

Then there is the issue with the machine chasing leaks, aerophagia, and snores. In my case a higher pressure seems to increase flow limitation, which causes the machine to increase pressure which again increases flow limitation. But couldn’t improved data and algorithms solve this problem of chasing all of this? Or is it impossible?

[palerider]

Seems like the minimum pressure set too low is an issue, but that the AutoSet for Her is able to set a new within a session. It would still be beneficial if the detected optimal minimum pressure was remembered from the previous session.

Then there is the issue with the machine chasing leaks,

No machine, at least not anything made in the last decade, "chases leaks", in fact auto machines will *reduce* pressure in order to try to get the mask to seal up again.
Here's an example:

Please stop perpetuating falsehoods.

aerophagia,

Are you just making up things to have something to post?

and snores.

Well, you finally found something that's correct. machines increase pressure to eliminate snores.

In my case a higher pressure seems to increase flow limitation, which causes the machine to increase pressure which again increases flow limitation. But couldn’t improved data and algorithms solve this problem of chasing all of this? Or is it impossible?

Your imagination is causing you to think wrong things.

Flow limitations cause the machine to increase pressure, because the solution to having flow limitations (and snores) is to increase the pressure to open up the airway.

You have cause and effect reversed.

[alexander]

palerider, I’m not an expert. I see that people in this thread discuss the machine chasing a variety of events, and by chasing I assume that they mean that the machine increases pressure when it shouldn’t. So my understanding, which might be wrong, is that the current machines do in fact increase pressure in some cases when they should not. Is my understanding correct?

I have little knowledge about pressure increases due to snores (I don’t think I normally snore), leaks (I usually don’t have a problem with this), and aerophagia (I don’t swallow that much air). I was just summing up what was mentioned in this thread.

What I do have some personal experience with (but admittedly not much knowledge about) is flow limitation. If I set my maximum pressure to 20 then it will probably go to 13. However, most of my apneas and hypopneas are treated around 6.0–7.0. In my case, what I believe happens, is that increased pressure leads to increase flow limitation, which leads to increased pressure. I might of course be wrong, but what I do know is that with my current pressure settings I get restful sleep, and my AHI is usually around 1.

[Pugsy]

Snores and flow limitations are early warning signs that the airway is at least starting to or trying to collapse.
That is why auto adjusting machines in auto mode will increase the pressure....it's built in to the algorithm to try to prevent further collapse and ward off the full fledged OAs or hyponeas (which by the way during the actual event the machine does absolutely nothing because it can't accomplish anything. It simply can't push enough air to blow past the tissue obstruction) which is why prevention before the fact is the better solution than always playing catch up trying to fix something after the fact.

The machines today don't "chase" leaks at all. The pressure itself never increases because of a leak...the air flow needed to get to that said pressure might be increased a bit but more often with today's machines the pressure is actually reduced (thus the air flow needed to get to that reduced pressure) in an effort to allow the leaks to come under better control.
Air flow needed to get to a set pressure isn't the same thing as the pressure itself. Something people often get confused about.

The machine doesn't know if air in the gut is a problem...or if we are even asleep or not.
All it measures is air flow and responds to air flow issues per its designed algorithm.
Increased pressure will NOT cause increased flow limitations unless more pressure is affecting the nose in some fashion and causing congestion which the machine might flag as FLs but it's not the same kind of FL as what is in the airway. Again the machine doesn't know where the flow reduction/limitation is coming from....all it knows to measure is the amount of air we are moving.
So when we see increased pressure and increased FLs....it's the FLs causing the pressure to increase and not the other way around.
If the FLs weren't there the machine wouldn't do anything.

If the machine's way of doing something with its algorithm is causing a problem of some sort...then the solution is to address the problem some how...either change modes or limit the machine from doing what it is designed to do.
Sometimes when it goes trying to kill FLs with more pressure the end result (maybe aerophagia or bad leaks) ends up being a bigger problem than the FL was.

For some people the minimum of 4 will do a great job. For others it won't.
Some people have a problem even comfortably breathing at 4 cm...others have no problem.

Doesn't matter what settings are used as long as the person is sleeping good, feeling good and his OSA is under control.
If they are happy then I am happy and I could care less how they get it done.

[jimbud]

The fine can be the machine chasing leaks,

And aerophagia. (running from)
JPB

To alexander.

I meant:

The fine can be aerophagia.
Aerophagia can be caused by pressure.
Pressure caused by ?

JPB

I should have said it that way.
I was confusing you.
Apologies.