How is auto CPAP supposed to work?

[BeckyLC]

I thought the point of auto CPAP was to have a range of pressure, and the machine would increase to just enough to keep the airway open. Am I wrong? Here's the deal. I've been using CPAP since 2013, my range was set to 7-15, my sleep study AHI was 47 but with the machine it's usually been under 1. A couple years ago I bought a travel machine, an Airsense Mini. The default pressure was 5-20 I think, and it kept waking me up with the thing blowing full blast. I would have to restart it to make it calm down. After a few nights of this I googled how to change the settings, figured it must be defective, and set it to 8 to see what would happen. I slept fine, AHI was low, no snoring so I left it there.

A few months ago I got a Resmed 11 (had an S9 prior), and since the settings were easy to change, I started playing with them. When it was 7-15, my max was 14.8 every night. If I lowered it to 7-12, my max was 11.8 every night. Then several weeks ago I set it to 8 and left it there. My AHI every night is less than 1, half the time it's been 0, my leak is way less, and I haven't been waking up as often at night to fiddle with the nasal prongs. I'd wonder if the numbers it's telling me are accurate, but since I haven't been snoring, I tend to believe them. So do I not understand how it's supposed to work? If I don't need higher pressures, why would I always get them when a range is selected?

[chunkyfrog]

The key is having a tight enough range to be functional.
Default settings are the possible extremes, because we all need different pressures.
4-20 is crazy useless for most of us.
Gradually raising the lower number, while observing results, works quite well.
A few of us need to lower the max; but most do not.
Oh, and the p10 has PILLOWS, not prongs.

[Dog Slobber]

I thought the point of auto CPAP was to have a range of pressure, and the machine would increase to just enough to keep the airway open. Am I wrong?

Yes.

The Auto-titrating machines will increase the pressure to open your airway, but that doesn't mean that they will be operating at an optimal setting.

They should be set, so that the minimum pressure is close enough to your typical pressure. This way, when your airways do start to collapse they won't have to far to go to open it up again. And they won't need a lot of pressure to open them up.

[Pugsy]

When it was 7-15, my max was 14.8 every night. If I lowered it to 7-12, my max was 11.8 every night. Then several weeks ago I set it to 8 and left it there.

Where are you getting these max pressure numbers? Off the machine or using software to see the details?

It is very possible that the machines when in auto mode would increase the pressure because of flow limitations it was seeing.
Flow limitations aren't a part of the AHI though.

Might be interesting to see what the flow limitation graph looks like with the fixed pressure of 8 with that low AHI.

[Miss Emerita]

As Pugsy said, flow limitations may be driving your pressure up; there's no reason not to trust the numbers you're seeing. And if you're sleeping well with low AHI at a steady 8, stick with it!

[palerider]

I thought the point of auto CPAP was to have a range of pressure, and the machine would increase to just enough to keep the airway open. Am I wrong?

Yes.

Here's the deal. I've been using CPAP since 2013, my range was set to 7-15, my sleep study AHI was 47 but with the machine it's usually been under 1. A couple years ago I bought a travel machine, an Airsense Mini. The default pressure was 5-20 I think, and it kept waking me up with the thing blowing full blast.

"full blast" is what? It *never* 'blows' harder than it *needs* to.

A few months ago I got a Resmed 11 (had an S9 prior), and since the settings were easy to change, I started playing with them. When it was 7-15, my max was 14.8 every night. If I lowered it to 7-12, my max was 11.8 every night. Then several weeks ago I set it to 8 and left it there. My AHI every night is less than 1, half the time it's been 0, my leak is way less, and I haven't been waking up as often at night to fiddle with the nasal prongs. I'd wonder if the numbers it's telling me are accurate, but since I haven't been snoring, I tend to believe them. So do I not understand how it's supposed to work? If I don't need higher pressures, why would I always get them when a range is selected?

"auto" machines increase pressure whenever the machine detects flow limitations, snoring, hypopneas and apneas.

As soon as whatever caused the pressure to increase stops happening, the machine will start lowering the pressure back to the minimum pressure set, or when another breathing disturbance happens.

Without seeing any *DATA*, nobody can do anything other than guess.

[Dog Slobber]

"auto" machines increase pressure whenever the machine detects flow limitations, snoring, hypopneas and apneas.

As soon as whatever caused the pressure to increase stops happening, the machine will start lowering the pressure back to the minimum pressure set, or when another breathing disturbance happens.

I'm going to disagree with you about responding to hypopneas, at least with the ResMed AutoSet algorithm.

I don't think ResMed's respond to Hypopneas.

From one of the ResMed AutoSet patent descriptions:

https://patents.google.com/patent/US20110203588

A known algorithm that is used to automatically set patient pressure in APAP machines is called ResMed AutoSet. All in all, the AutoSet device, and its algorithm, is excellent for treating OSA patients. The ResMed AutoSet algorithm responds to three things: flow limitation, snore (audible noise) and apnoea.

Of course, other Auto machines might respond to hypopneas. But it makes sense that they don't given the nature of the hypopnea (Central or Obstructive) isn't necessarily known. And only hypopneas that are obstructive in nature should be responded to by a pressure increase.

[zonker]

I'm going to disagree with you about responding to hypopneas, at least with the ResMed AutoSet algorithm.

I don't think ResMed's respond to Hypopneas.

though i'm not sure it DID respond to the hypopneas or if it was just wiggling the jello. some responses are to nothing and the others i can't tell if they responded TO hypopneas or just nearby.

[Pugsy]

though i'm not sure it DID respond to the hypopneas or if it was just wiggling the jello. some responses are to nothing and the others i can't tell if they responded TO hypopneas or just nearby.

Don't forget the flow limitations that might be happening at the same time as the hyponea and it will for sure respond to FLs.

[zonker]

though i'm not sure it DID respond to the hypopneas or if it was just wiggling the jello. some responses are to nothing and the others i can't tell if they responded TO hypopneas or just nearby.

Don't forget the flow limitations that might be happening at the same time as the hyponea and it will for sure respond to FLs.

oh.


yeah, the responses do seem to match up with the flow limitations a bit better.



now, ds will see my response to him before he sees your response to me and....

<sigh>

[palerider]

"auto" machines increase pressure whenever the machine detects flow limitations, snoring, hypopneas and apneas.

As soon as whatever caused the pressure to increase stops happening, the machine will start lowering the pressure back to the minimum pressure set, or when another breathing disturbance happens.

I'm going to disagree with you about responding to hypopneas, at least with the ResMed AutoSet algorithm.

Well... given the image in the patent, and the dates, I think it's worth considering that the patent refers to prior versions, we know that the 'autoset' algorithm changed considerably in 2010 when the S9 came out.

However, all I'll say for sure is "Thanks, now I'm going to have to look at more charts." :sigh:

[palerider]

"You're full of it" (paraphrased) (or, more accurately, "You're 25% full of it"

However 2, the sequel:

This *seems* to indicate that the S9 (and later) do respond to hypopneas:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4481055/ see figure 3, D8 is the Resmed S9 Autoset.

As does this: https://www.ncbi.nlm.nih.gov/pmc/articl ... der-8-425/ which is from this:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4629962/

[Pugsy]

This *seems* to indicate that the S9 (and later) do respond to hypopneas:

Do you remember people with S8 reports with really elevated hyponea counts?
Some people thought that the S8 was super sensitive and recorded hyponeas where others wouldn't record hyponeas and they would tell people that ResMed just tended to over score hyponeas and "to just cut the hyponea count in half".
I never was comfortable just randomly cutting any number in half myself.
I think that the old S8 (and older auto adjusting algorithm) just simply didn't do anything about the hyponeas so no wonder the numbers were high.

There was a significant change in the auto algorithm with the S9 and low and behold the hyponea counts were greatly reduced...surprise surprise...the new algorithm seemed to want to increase to fight off hyponeas as well as the other stuff.

[GrumpyHere]

From the study cited by PR.

Documented proof that 4-20 cm setting works badly.

In their experiment, they measured treated residual obstructive from a simulated AHI 38.6

For a residual AHI of 5, the treatment efficiency has to be 87% and above.

Neither the S9 nor the Remstar is able to achieve that goal.

[palerider]

This *seems* to indicate that the S9 (and later) do respond to hypopneas:

Do you remember people with S8 reports with really elevated hyponea counts?
Some people thought that the S8 was super sensitive and recorded hyponeas where others wouldn't record hyponeas and they would tell people that ResMed just tended to over score hyponeas and "to just cut the hyponea count in half".
I never was comfortable just randomly cutting any number in half myself.
I think that the old S8 (and older auto adjusting algorithm) just simply didn't do anything about the hyponeas so no wonder the numbers were high.

There was a significant change in the auto algorithm with the S9 and low and behold the hyponea counts were greatly reduced...surprise surprise...the new algorithm seemed to want to increase to fight off hyponeas as well as the other stuff.

I hadn't thought about that, you make a good point.