Autoset - Rate of Time for Preasure Climb

Has there ever been a study to work out what the rate of preasure increases during an event? If you have a wide open Autoset and your preasure is at 4 when you have an apnea and the preasure required to stop the apnea is 12 how long does it take for an autoset to raise that high? I would surmise that its not instant, so how long?

If hospitals and clinics are sending people out with a wide open set then surley there is a case for negligence.

To my limited knowledge, your pressure would not approach 4...the machine will keep you at or near your average pressure. Most of us could not breathe at a pressure of 4. This is why I question why many here recommend that you set an autoset at 2-3 below and above average.
When I do use my autoset, I set the pressure at 8-16 with my average at 12. The report I looked at shows that my pressure never got below about 11. ???????

Its sometimes so hard to put across what you mean lol.

A machine (Autoset) is usually set at 4-20cm when you get it. So on your nights pap you don your mask and with a few minutes ramp it starts at 4cm and suddenly you are hit with an event. To clear that event (Apnea) your machine has to reach say 12cm (Example) of pressure to effect the apnea. How long will it take your machine to reach 12cm from 4cm? If it takes say 60 seconds to reach 12cm and your apnea lasts 50 seconds before your body kicks in then your machine isnt doing diddley squat to help you.

What I want to know is what incrimment of time does it take for the pressure to go up, ie 1cm = (Time)? also are other factors like a sluggish motor going to effect the time. The reason for this is I want to keep my pressure down but not too far as my machine doesnt have the time to reach my pressures to effect my events

I have a headache lol

Gazhacks wrote:Its sometimes so hard to put across what you mean lol.

A machine (Autoset) is usually set at 4-20cm when you get it. So on your nights pap you don your mask and with a few minutes ramp it starts at 4cm and suddenly you are hit with an event. To clear that event (Apnea) your machine has to reach say 12cm (Example) of pressure to effect the apnea. How long will it take your machine to reach 12cm from 4cm? If it takes say 60 seconds to reach 12cm and your apnea lasts 50 seconds before your body kicks in then your machine isnt doing diddley squat to help you.

What I want to know is what incrimment of time does it take for the pressure to go up, ie 1cm = (Time)? also are other factors like a sluggish motor going to effect the time. The reason for this is I want to keep my pressure down but not too far as my machine doesnt have the time to reach my pressures to effect my events

I have a headache lol

YES...You're giving me a pain too!

A while back I did read something about how long it took the PR S1 APAPs to get to a needed pressure but I don't remember where I read it. It isn't 1 cm a minute..more like 2 minutes because it stops and "tests" things before proceeding upwards. Sorry but I don't remember the exact time frame though, only that was well beyond the time that an event could/would come and go. I don't remember reading how the ResMed APAPS do their response thing or how long it takes. Not even sure that I have ever read it but I would imagine the information is out there but it may take some hunting to find it. Someone using SleepyHead (I know you can't) could possibly get an idea by zooming in on the time line and watching the pressure increase in steps and manually timing them. I have reports using a ResMed Auto available to look at but I had the minimum much higher (10 cm) so I can't go back and look to watch it climb from 4 cm to see how long it takes.

With a limited amount of "playing" with my PRS1 unit, it seems to not increase more than 1 cmH2O per minute.

It always starts at the minimum pressure setting when you turn it on, so if you're set at 4-20 and need 14, you'll have a long wait to get to full pressure.

Go and experiment with your own unit to see what it does. Put it on, and experiment with faking an apnea. You may find that making snoring noises is the easiest way to fake it out, or you may have to experiment with holding your breath or plugging up the hose.

If it really is that slow, what is the point of the autoset? Have I wasted 2 years of so called treatment because no one told me untill a wee while ago that I needed to raise my baseline pressure. All these thousands of people in this country that have been sent home without having their baseline raised at some point are probably saying to themselves, this isnt working!

I have just invested in a card and reader so will be able to look more closely at this, also people should be told about this and have their machines check and raised, Im going to up mine from 6 to ? maybe 12, as I have had readings of 12.6. Time to turn my autoset into an CPAP machine and just let the doctors complain at me, I will start talking lawsuits for the 2 years of non-therapy

Gazhacks wrote:If it really is that slow, what is the point of the autoset? Have I wasted 2 years of so called treatment because no one told me untill a wee while ago that I needed to raise my baseline pressure. All these thousands of people in this country that have been sent home without having their baseline raised at some point are probably saying to themselves, this isnt working!

I have just invested in a card and reader so will be able to look more closely at this, also people should be told about this and have their machines check and raised, Im going to up mine from 6 to ? maybe 12, as I have had readings of 12.6. Time to turn my autoset into an CPAP machine and just let the doctors complain at me, I will start talking lawsuits for the 2 years of non-therapy

10 minutes of apnea at the start of every night while the machine adjusts up is probably a lot better than 8 hours of apnea every night. It might even prevent most of the damaging effects of apnea. It's also probably better than having a manual CPAP set at too low of a pressure. Or a manual CPAP set at a higher pressure than you need, with the risk of the patient giving up on CPAP.

It does seem to me that most people would be better off if their APAP was set to a minimum pressure a little closer to what their needs are going to turn out to be.

Experiment with your own APAP and see what pressure it starts at and how fast it will go up for you. I presume your S8 shows the pressure on the display. Draw your own conclusions.

Gazhacks wrote: A machine (Autoset) is usually set at 4-20cm when you get it. So on your nights pap you don your mask and with a few minutes ramp it starts at 4cm and suddenly you are hit with an event. To clear that event (Apnea) your machine has to reach say 12cm (Example) of pressure to effect the apnea. How long will it take your machine to reach 12cm from 4cm? If it takes say 60 seconds to reach 12cm and your apnea lasts 50 seconds before your body kicks in then your machine isnt doing diddley squat to help you. (emphasis added)

Neither the Resmed S9 AutoSet nor the PR System One Auto are ventilators. And because APAPs are NOT ventilators, they do NOT try to force us to inhale. In other words, APAPs do not "clear" events by increasing the pressure. They wait for you to start breathing on your own and THEN they respond to OAa and Hs by increasing pressure after the event is over. The idea behind the pressure increase AFTER the apnea is over is to make it more difficult for your upper airway to collapse in the immediate future.

The whole, simple elegant idea behind xPAP therapy for OSA is the xPAP machine provides just enough pressure to KEEP your upper airway open. In other words, if you need 12cm of pressure, then 12cm makes it very difficult, but not impossible, for your upper airway to collapse, but pressures less than 12 cm may still allow your upper airway to collapse too easily at certain periods during the night (perhaps when you are in REM or perhaps when you are sleeping on your back, for example). And when this happens, the machine (including APAPs, BiPAPs, and Auto BiPAPs) waits for you to start breathing on your own---just the way you did when you had apneas before you were diagnosed. The difference is that with effective PAP therapy, your upper airway collapses at most a few times per hour and at most a few dozen times per night instead of happening 20-60 (or more) times per hour and hundreds of times per night.

Of course one of the underlying concerns (issues) in this whole discussion is the fact that it's not such great idea to run in wide open APAP mode if you are NOT someone like me who does well at low pressures. The problem is not as simple "It takes to long to increase the pressure from 4cm to 12cm when you have an apnea." The problem is more subtle. And it's tied to what it means to need 12 cm of pressure to effectively treat your OSA.

If you need 12cm of pressure, that does not mean that you need 12cm to open your airway up during an apnea. Rather a prescription for 12 cm means that you need 12cm to KEEP your upper airway open in the first place. And if you are using an APAP, when the pressure is still substantially lower than 12 cm---say when the pressure is between 4--8cms, you airway is still prone to repeatedly collapsing, which allows the apneas to occur. So what happens?

You go to sleep. And the apneas start. And the APAP raises the pressure in response to the events and your breathing improves (at least slightly.) And so the APAP starts to lower the pressure. But you are still so far below your effective pressure that it doesn't take much time for the NEXT cluster of events to start. And so it's the cycle of:

2. The person has an OA or two due to pressure not being high enough.
3. The APAP responds with a slight increase in pressure AFTER the event is over.
4. The APAP starts decreasing the pressure because the person's breathing is "better" than it was.
5. The person has another OA or 2 for the machine to respond to and we're back to Step 2

that can take a lot time to increase the base pressure up from 4cm to the 10-14cm that some folks need: Every time through the cycle, the base pressure is increased (somewhat), but it takes lots of cycles (and hence lots of apneas) to the pressure close to therapeutic pressure. And of course, once the pressure is close to what you need, your breathing stabilizes, which leads to the machine decreasing pressure as low as it can. Which starts the whole cycle over again.

By increasing the min pressure up closer to your 90% or 95% pressure level, you decrease (or better yet, eliminate) the amount of time at pressures that are clearly ineffective, which reduces the overall number of apneas, but still allows you to have less pressure overall, which may be more comfortable---particularly if aerophagia is a problem.

Gazhacks wrote:What I want to know is what incrimment of time does it take for the pressure to go up, ie 1cm = (Time)? also are other factors like a sluggish motor going to effect the time. The reason for this is I want to keep my pressure down but not too far as my machine doesnt have the time to reach my pressures to effect my events

The comments in this thread got the mathematician in me thinking. So I've gone back through my almost 17 months worth of data, about 12 months of which comes from using machines in Auto mode. This post reflects both information that I've learned by reading about the Resmed and PR Auto algorithms as well as what I've noticed in my own data.

Both Resmed and PR have discussions of how their Auto algorithms respond to events on their web pages.

• This paper concentrates more on the FOT algorithm for differentiating between CAs and OAs, but does contain the basics of the Auto algorithm as well.
• The PR Auto algorithm is discussed on these web pages.

Both the Resmed S9 AutoSet and the PR S1 Auto detect and respond to snores, flow limitations, OAs, and H's; both machines detect and do NOT respond to CAs; PR claims its System One machines can detect and respond to RERAs. PR's machines detect, but do NOT respond to periodic breathing.

Anybody who as looked at the pressure curves for both the S9 Auto Set and the PR Auto should realize that these machines have very different algorithms. Just based on looking at the many, many Resmed S9 AutoSet pressure curves and PR S1 Auto pressure curves posted here at cpaptalk, it's pretty easy to conclude that the S9 reacts more aggressively to events as they occur, while the System One reacts proactively by raising the pressure through a "hunt and peck" (H&P) algorithm that periodically increases the pressure to test whether there are distinct improvements in the shape of the flow wave curve even if there have been no events in a long, long while. The two machines also seem to have very different approaches to decreasing pressure: The S9 starts to slowly decrease the pressure as soon as the "events" that triggered the increase seem to be eliminated; the System One seems to use a reverse H&P---testing a decrease in pressure and if there is any deterioration in the flow wave, increasing the pressure back up to the current level.

So based on the pressure curves I've seen that have been posted by many, many users here, my impression is that the S9 tends to increase pressure faster (and perhaps higher), attempts to decrease the pressure sooner, but takes more time to decrease the pressure than the System One. As a result, there seems to be more variation in the pressure setting on the S9 than on a System One running in the same range.

Now for some additional observations based on my own data for the last 17 months using these machines.

For me, both machines are much more likely to increase pressure due to SNORING or FLOW LIMITATIONS than they are for (multiple) OAs or Hs. This may just be that my OSA is well controlled by CPAP at the pressures I've used. (And I am someone who needs relatively little pressure to control my apnea most of the time.)

Resmed S9 AutoSet Data
Most of the time I was using an S9 AutoSet in Auto mode, my settings were:

• min = 4cm
  max = 8 cm
  EPR = 3

In my data I can find numerous places where the S9 increased pressure from 4cm to 8cm in as little as 30-40 seconds. Most of the time these instances are in response to Flow Limitations and/or Snoring. (The more intense the snoring or flow limitation, the faster the increase in pressure.)

I can find places in my data where the S9 ignored an isolated H (i.e. did not increase pressure) when there was NO snoring or flow limitations going on. I can find places where the S9 responded AFTER a single OA ended by increasing the pressure by 1 cm in about a 5 or 10 second interval. I can find places where the S9 responded to "double" OAs (OSs separated by a single inhale) after the SECOND apnea ended with a rapid 3 or 4 cm increase in pressure. The increase in pressure typically takes somewhere around 10-15 seconds at most. I suspect (but obviously cannot prove) that the Resmed Auto algorithm takes into account the whole picture (snoring and flow limitations) when choosing to respond to an isolated OA or hypopnea. But if two or more OAs occur in quick succession, it responds at the END of the second OA by increasing the pressure by several cms in a 10 or 15 second window.

PR System One Auto BiPAP data
I use a S1 BiPAP Auto, not the S1 APAP. But the two System One Auto algorithms have a lot in common. In particular, both use the hunt-and-peck (H&P) algorithm. In the 11 months I've been using my BiPAP in Auto Mode, my settings have mostly been at:

• min EPAP = 4cm
  max IPAP = 8 cm
  PS = 4

This means that most of the time I've been using this machine, my EPAP has been restricted to 4--6cm and my IPAP has been restricted to 6--8cm. Not much wiggle room so to speak. But I have occasionally raised my max IPAP as high as 10 or 12cm and increased the PS appropriately. And many of my comments here are based on those periods where I was running in the wider range.

It's clear from the data I have as well as that posted by other BiPAP Auto users, the PR hunt-and-peck (H&P) algorithm for proactively testing pressure increases is applied only to the IPAP pressure.

For the most part, my EPAP only increases for vibratory snores (VS's). In SH, these are VS's NOT VS2's by the way. As pugsy indicates, the PR seems to increase the pressure by 1cm (over about a 10 second period) and then it waits before increasing the pressure again. When I'm snoring, I'll see an increase of about 1 cm in EPAP with a corresponding VS tick about once a minute in SH. Once my max EPAP is reached, of course, the snoring-induced pressure increases stop.

OAs are supposed to affect the EPAP on a PR S1 BiPAP. In my data, I seldom have more than isolated OAs. And it's clear that if there's no snoring going on, the PR ignores the isolated OA: In other words, when I have an isolated OA, there is no increase in EPAP at the end of the event. When I have clusters of OAs, the PR's response seems to depend on how close the OAs are to each other. If the OAs occur about three minutes apart, the machine seems to treat them as "isolated" and does not raise the pressure. If the OAs are less than three minutes apart, then at the end of the last one, the machine increases the pressure by 1cm. The only time I see my machine increase the pressure more than 1 cm for OAs is either for the OAs to be mixed in with VS (and the VS's increase the EPAP pressure) or if I have to clusters that more than three minutes apart (so both clusters increase the pressure by 1cm) AND the second cluster is close enough to the first cluster that the machine has not yet decreased the pressure back down it's initial setting.

My data indicates that my IPAP pressure is most commonly increased pro-actively by the H&P algorithm or for flow limitations. When absolutely nothing is happening, the H&P algorithm kicks in. The whole algorithm takes place in a 10 minute cycle. At the start of the cycle, the machine raises the pressure by 2 cm over a 4 minute period (i.e. by 0.5 cm/min). While performing this test increase, the machine is looking for improvements in the shape of the flow wave. If no improvements are found, then the machine lowers the pressure back down to the original level over the course of one minute, and then it waits five more minutes before starting the next H&P cycle. On the other hand, if improvements in the shape of the wave form are found, the machine will increase the baseline pressure and the H&P algorithm will use this new baseline for the next round. If the best wave form was at the peak of the trial pressure increase and the pressure is still below the maximum pressure setting, the machine will continue to increase the pressure setting for another 4 minutes at a rate of 0.5 cm/min. Judging from my data, the PR S1 uses a reverse H&P to determine when it is ok to lower the pressure and how much to lower it by.

IPAP is supposed to also be increased for FL, RERAs, and hypopneas. It is really tough in my data to tease apart increases that are only due to these events and those that are influenced strongly by the H&P part of the algorithm. As near as I can tell, however, it takes a pair of hypopneas in close proximity for my machine to respond to them if it's not already in the active part of the H&P routine.

As pugsy has indicated, the PR S1 seems to increase pressure by 1cm and then waits for a minute to see if further increases in pressure are necessary. While this may seems counter-intuitive to many on this board, it turns out that PR's approach does resemble what actually goes on during a titration study. (See Clinical Guidelines for the Manual Titration of Positive Airway Pressure in Patients with Obstructive Sleep Apnea.) When doing the titration study, the tech does NOT increase the pressure every single time you have an event. Rather, after increasing the pressure by 1cm, the tech is supposed to wait for five minutes before increasing the pressure again. The idea, as I understand it, is that the waiting period is to see if the new pressure level really is effective or not. And if not (i.e. more events occur) then the pressure is increased again.

Now to answer Gazhacks's basic question, "What I want to know is what incrimment of time does it take for the pressure to go up, ie 1cm = (Time)?"
My data indicates that it takes the S9 about 3 to 7 seconds to increase the pressure by 1cm. The rate is variable and it depends on what the S9 is responding to. On the PR System One, a 1cm increase takes around 10 seconds at most, but the System One will then stop and see what happens for a minute or so.

Of course, Gazhacks's real question seems to be, "How long is it going to take to go from 4 cm to 14 cm?
On the S9 under the appropriate set of circumstances, it might take as little as 30--70 seconds. (Keep in mind that for those of us who do not use ASV machines, that quick of a pressure increase could also trigger leak problems and comfort problems and could easily wake many folks up.) On a System One, it could take about 10 minutes to go from 4 cm to 14 cm since each "event' induced 1cm increase is followed by a waiting period. However, it's also likely that the H&P algorithm would kick in as well: And as long as improvements to the wave form are being seen, the H&P will continue to increase the pressure steadily upward. And that should shorten the overall time it takes the PR System One to reach 14 cm possibly by several minutes.

Now as well as my headache, I now have stomach ache as that took some major digesting

Talk about an eye openner, why is it that one question asked give you so many more questions but what an answer, I'm standing and aplauding that response, although alot is going to need some re-reading several times to understand it and alot of the abbreviations are alien but can make an educated guess.

I feel humbled just trying to respond to this. My education is growing quite fast concerning OSA. Firstly, I never thought for once that my machine wasnt a ventilator. Not sure why I did, I think thats why I was confusing myself. My thoughts were that the pressure is needed to be high to force you to start breathing but as you point out its not a ventlator At first glance the time it takes to raise itself is amazingly slow but now with the knowledge, understanding why it is, has solved that problem. Im still waiting for my data card to arrive so I can join the data crunches and start to look at my data more closely instead of relying on Oxibob (Nickname, Oximeter) I think Ive been living in the dark when it came to understanding my machine and not sure why I had a preconceived idea that it was something very different. (Also reading some of the links which I am now printing to digest later)

Can I put this down or vote for the "Response of the Year Award" I have even printed it out so not to lose it. I huge thank you for your time as that must have took some effort to write. Thanks