Need Resmed education

[-SWS]

Bench Evaluation of Flow Limitation Detection by Automated Continuous Positive Airway Pressure Devices
Frédéric Lofaso, MD, PhD; Gilbert Desmarais, PhD; Karl Leroux; Vincent Zalc; Redouane Fodil, PhD; Daniel Isabey, PhD; and Bruno Louis, PhD
CHEST August 2006 vol. 130 no. 2 343-349

Despite being a 2006-published bench study, it adequately supports one of Jeff's excellent rhetorical questions:

Don't different brands define, detect, and react to precursors differently?

Clearly they do. But we can characterize and factor some of those key differences toward making useful "fuzzy" associations or distinctions...

[jnk]

. . . rhetorical questions . . .

Well, my questions were sort of rhetorical, I guess. (Hopefully less in the sense of "pompous" than in the sense of "for effect." ) But I mostly asked them because I've wondered about them a lot. So I was just hoping to ride the wave of carbonman's thread to sneak my questions in to get your thoughts/considerations/answers. I sincerely meant them less as a mother-lode/gauntlet and more as raw paint material for further art. I learn from the way you frame your approach, so I was hoping to see where you would put the frame and just how you were going to stretch the canvas. But in trying to get you to expound in certain directions, I hope I didn't derail. Once you have time, I'd like to read more of your thoughts on carbonman's charts.

I would love the opportunity to try out a Remstar autobilevel one day myself. But I can't get much useful info out of studies with fake lungs and machines set wide open starting at 4 cm. That's just me. I am curious about real world application the way I would use an auto based on how different machines would detect, interpret, and react to my precursors. Because, to my mind, the autos these days seem to be more about reacting to the precursors than they are about reacting to the events themselves. Am I wrong on that?

No rush. Just trying to bump and give fuel for further discussion . . .

[sleepangel]

I'd get the NEW Resmed Vpap Auto 25!
Just got mine, Had a straight S8 escape but the pressure of 14 blew the ff mask off.
Auto 25 is packaged in same box, same features as the S8 Auto II but pressure goes up to 25, also several settings for inhale exhale.

Have to beg for the clinicians manuel so you can see your API,AI and HI, DME's dont give them out even when you beg!

lol

[robertmarilyn]

I'd get the NEW Resmed Vpap Auto 25!
Just got mine, Had a straight S8 escape but the pressure of 14 blew the ff mask off.

I had the same thing for one week...the Escape with the pressure locked in on 14...my mask was trying to fly away...and I was in so much pain from acid reflux and from having air forced into my stomach

Auto 25 is packaged in same box, same features as the S8 Auto II but pressure goes up to 25, also several settings for inhale exhale.

I wonder if very many people use a pressure of 25? That would be a big blast

Have to beg for the clinicians manuel so you can see your API,AI and HI, DME's dont give them out even when you beg!
lol

Righteo...they don't...but if you need one, just make a post asking for it and I bet it will appear in your mail box eventually

mar

[jnk]

I'd get the NEW Resmed Vpap Auto 25!
Just got mine, Had a straight S8 escape but the pressure of 14 blew the ff mask off.
Auto 25 is packaged in same box, same features as the S8 Auto II but pressure goes up to 25, also several settings for inhale exhale.

Have to beg for the clinicians manuel so you can see your API,AI and HI, DME's dont give them out even when you beg!

lol

The Autoset II is an auto for use for patients prescribed CPAP. The VPAP Auto or Auto 25 is an auto for patients prescribed bilevel.

[-SWS]

... I was hoping to see where you would put the frame and just how you were going to stretch the canvas.

I was initially hoping to go with a very little, minimally time-impacting canvas.

I hope I didn't derail. Once you have time, I'd like to read more of your thoughts on carbonman's charts.

No you didn't. But you should also know that waving very good rhetorical questions in front of the cpaptalk crowd is like waving highballs down at the local AA meeting. I sure don't have the will power to not discuss those excellent questions.

Because, to my mind, the autos these days seem to be more about reacting to the precursors than they are about reacting to the events themselves. Am I wrong on that?

Yes, on that note think of snore and flow limitation being A10 precursors to work with above 10 cm. Now think of snore, flow limitation, hypopneas, and apneas being four precursors a different algorithm chooses to signal-process and respond above 10 cm.

That's right about apneas and hypopneas being responded to as if they were precursors (which they are) to yet other pending apneas and hypopneas above 10 cm. Now think about Mar's response to A10, which will only work with snore and flow signal above 10 cm. There was no A10 response above 10 cm. But you succinctly pointed out that the AASM feels that APAP should be contraindicated for snoreless patients (including but not limited to UPPP cases).

Interestingly none of the manufacturers contraindicate APAP for UPPP. Probably because many, like Koppy, snore after sixth months or longer. But also because CPAP tends to work. But let's look at Mar's xPAP predicament. Her A10 algorithm had no snore or flow limitation signals to work with above 10 cm. That reduced her to running CPAP. But alas, she seems to need more straight pressure to consistently address her residual apneas and hypopneas than her poor LES closure can handle.

So under those circumstances, do we think it is plausible to consider what an APAP algorithm can do for her that can : 1) work with apnea and hypoponea signals above 10 cm, 2) may effectively address a good portion of the residual apneas and hypopneas at the expense of some minority of preliminary ones, while 3) keeping her mean pressure down, on average, throughout the night for the sake of pressurized air breeches through the LES.

Also, the question about whether it made any point at all to graphically demonstrate A10 not responding to A and H above 10 cm for lack of FL and snore in Mar's UPPP case. It absolutely makes sense to understand any and all failures. That's what we have done for other algorithms as well. We come around to understanding them so that we know why they do and why they don't work. Some patients who have never had UPPP will present prolific snore and FL signals, and others are guaranteed to fall short. I think the general rule in physiology is to expect bell-curve gradients. So for quite a few of us, this message board has been all about carefully looking at the details of algorithms.

Jeff, your rhetorical questions sure never sound pompous to me. But they are often very stimulating and very compelling IMHO.

[jnk]

... I was hoping to see where you would put the frame and just how you were going to stretch the canvas.

I was initially hoping to go with a very little, minimally time-impacting canvas.

I hope I didn't derail. Once you have time, I'd like to read more of your thoughts on carbonman's charts.

No you didn't. But you should also know that waving very good rhetorical questions in front of the cpaptalk crowd is like waving highballs down at the local AA meeting. I sure don't have the will power to not discuss those excellent questions.

Because, to my mind, the autos these days seem to be more about reacting to the precursors than they are about reacting to the events themselves. Am I wrong on that?

Yes, on that note think of snore and flow limitation being A10 precursors to work with above 10 cm. Now think of snore, flow limitation, hypopneas, and apneas being four precursors a different algorithm chooses to signal-process and respond above 10 cm.

That's right about apneas and hypopneas being responded to as if they were precursors (which they are) to yet other pending apneas and hypopneas above 10 cm. Now think about Mar's response to A10, which will only work with snore and flow signal above 10 cm. There was no A10 response above 10 cm. But you succinctly pointed out that the AASM feels that APAP should be contraindicated for snoreless patients (including but not limited to UPPP cases).

Interestingly none of the manufacturers contraindicate APAP for UPPP. Probably because many, like Koppy, snore after sixth months or longer. But also because CPAP tends to work. But let's look at Mar's xPAP predicament. Her A10 algorithm had no snore or flow limitation signals to work with above 10 cm. That reduced her to running CPAP. But alas, she seems to need more straight pressure to consistently address her residual apneas and hypopneas than her poor LES closure can handle.

So under those circumstances, do we think it is plausible to consider what an APAP algorithm can do for her that can : 1) work with apnea and hypoponea signals above 10 cm, 2) may effectively address a good portion of the residual apneas and hypopneas at the expense of some minority of preliminary ones, while 3) keeping her mean pressure down, on average, throughout the night for the sake of pressurized air breeches through the LES.

Also, the question about whether it made any point at all to graphically demonstrate A10 not responding to A and H above 10 cm for lack of FL and snore in Mar's UPPP case. It absolutely makes sense to understand any and all failures. That's what we have done for other algorithms as well. We come around to understanding them so that we know why they do and why they don't work. Some patients who have never had UPPP will present prolific snore and FL signals, and others are guaranteed to fall short. I think the general rule in physiology is to expect bell-curve gradients. So for quite a few of us, this message board has been all about carefully looking at the details of algorithms.

Jeff, your rhetorical questions sure never sound pompous to me. But they are often very stimulating and very compelling IMHO.

Another masterpiece post! Thanks, that helps me considerably. Hopefully, I'm not the only person in the forum that post will help.

[robertmarilyn]

... I was hoping to see where you would put the frame and just how you were going to stretch the canvas.

I was initially hoping to go with a very little, minimally time-impacting canvas.

You guys are such Renaissance men All arty and intellectual and "stuff" BTW, my parents owned their own custom picture frame shop for decades and I would work there part time...making frames and framing very expensive artwork

Since -SWS mentioned in his last post, some things about my situation, I added a link to my latest uploads from the last two nights.

viewtopic/t41490/so-very-tired-and-it-i ... ml#p367827

I changed the setting of my Resmed machine from CPAP pressure of 12.2, EPR of 2 on 051009 to CPAP pressure of 12.2, EPR of 1 on 051109. This ended up changing the median pressure from 10.6 to 11.4.

Jeff, your rhetorical questions sure never sound pompous to me. But they are often very stimulating and very compelling IMHO.

Another masterpiece post! Thanks, that helps me considerably. Hopefully, I'm not the only person in the forum that post will help.

I'm trying to learn from the questions you ask Jeff...I'm learning for the input of all of you...THANKS
mar

[Joel8]

[quote="Velbor"]Two aspects to the "age" of a ResMed machine: the first four digits of the serial number will give the year of manufacture, but probably more important is the number of hours on the blower. That can be found in the Clinical Menu, under Servicing information (not to be confused with the re-settable hours of use found in Results).

How do you access the Clinical or other menus?

Thanks,

Joel8

[dsm]

<snip>

Another masterpiece post! Thanks, that helps me considerably. Hopefully, I'm not the only person in the forum that post will help.

Yes - I agree - SWS can be dazzlingly eloquent

DSM

[jnk]

Two aspects to the "age" of a ResMed machine: the first four digits of the serial number will give the year of manufacture, but probably more important is the number of hours on the blower. That can be found in the Clinical Menu, under Servicing information (not to be confused with the re-settable hours of use found in Results).

How do you access the Clinical or other menus?

Thanks,

Joel8

This may help:

http://www.cpap-supply.com/Articles.asp?ID=130

Read their article, but buy from cpap.com (the sponsors of this forum).

[-SWS]

...here's that first graph again:



First Hour-and-a-Half- Essentially identical to CPAP at 12.5 cm. You can see the algorithm searching upward for any possible improvement in patient flow, but always returning to 12.5 cm pressure for lack of detected improvement. Resmed would not have trolled, but the trolling is not beneficial "treatment". It's just a pressure-search test.

First Three Isolated H Events- Respironics won't respond to single-occurrence H or single occurrence A. Resmed would not have responded to those either.

First Three FL Events- Resmed just may have responded earlier, thereby preventing that follow-up OA cluster by using preemptively-elevated pressure.

That Follow Up OA Cluster- Respironics hikes the pressure, but too late to prevent the OA cluster. That short pressure hike is pretty much the only thing that night---besides pressure-search tests----that distinguishes your night from CPAP at 12.5 cm. And that short pressure hike may not have truly accomplished anything---other than closing the barn door right after the OA animals escaped.

A belated bump from me.

And that short pressure hike may not have truly accomplished anything---other than closing the barn door right after the OA animals escape.

Even though it's too late, doesn't the pressure increase "shorten" and
lessen the effect of the oxygen loss and brain response due to the OA???
Decreasing the "avg. time in apnea".

That's a definite possibility for the second and third apneas embedded in that sequence of pressure steps, carbonman. There are multiple physiologic scenarios buried in that ambiguous sequence of tick marks.

Carbonman, you suggested that those pressure steps might "shorten" an apnea and thereby lessen its immediate and adverse effects on physiology. As DSM pointed out, the algorithm waits for an apnea to complete before the ensuing pressure step. But that second apnea occurs with the slight benefit of that first pressure increase; and the third apnea occurs with the combined benefit of that slight second pressure step added to the first. But alas, they were still apneas that unfortunately required active airway maintenance by you.

That hints at the possibility of a patent airway being achieved only marginally quicker with CPAP assisting your own neuromuscular efforts. But the argument for a hypothetically shorter apnea by assisted, yet still active, airway maintenance speaks of sleep necessarily being interrupted to regain that patent airway. How significant or marginal might the difference be between having to wake (or stage-shift) to open your airway at 12.5 cm versus 14.5 cm? Probably not a vast time difference, since we can all very easily and very quickly maintain airway patency when we are not asleep.

The length or duration of any given apnea just may have more to do with our own varying "neurological vigilance" and response time to chemoreceptor signals of suffocation---than the expediency of thrust another 2 cm offers a completely closed airway.

Each APAP pressure increase is intended to get you to a hopefully safer and higher pressure zone, where you won't have to wake or stage-shift to actively maintain a patent airway. Completely under-addressed apneas allow full airway closure. Partially addressed apneas tend to sag as hypopneas and lesser flow limitations. However, high and constant upper-airway resistive loading can supposedly manifest as suction-type hypopneas and flow limitations as well.

Another physiologic scenario worth noting is that your persistent OA events might have actually been central or transitional/homeostatic apneas in nature. If you scored only a couple or few during PSG pressure increases, they may not have been noted in your sleep study summary. Since pressure increases can perpetuate that scenario, Respironics employs their NRAH logic to limit pressure increases. That's what happens in the algorithm when you score NR flags.

whether another brand would have been riding at a higher baseline pressure to start with, in reaction to it's detection and reaction to precursors

Take a peek at the FL cluster directly in front of the OA cluster we just discussed. The Respironics algorithm simply probed up a bit by design. But the Resmed A10 algorithm probably would have flagged those as FL events as well. And if it did, the Resmed algorithm would not have simply probed. Rather, the Resmed algorithm would have stepped the pressure up for an extended period. And the Resmed algorithm may have sensitively scored additional FL along the way, preemptively hiking the pressure even more.

As I mentioned in my earlier post, that may have been enough to stave off that cluster of OA events---assuming they were obstructive versus central...

[carbonman]

SWS, thanks for your thoughts.

As I was writing a long diatribe response, it occurs to me that
all of this maybe or is, pure speculation.
On any given night, each individual will have a complex variance in
their neurological and chemoreceptor responses to each event.
How tired are you,
are you in somekind of pain,
do you have an upset stomach,
did you have alcohol or drugs,
are you depressed or anxious,
are you in your fear complex,
and any number of other conditions or
combination of conditions.
In response, each machine may or may not, react in a certain way,
that may or may not be beneficial.

All of this considered, brings me to the conclusion that,
data capability is crucial,
vigilance is crucial,
record keeping for trending is crucial,
and there are no absolutes.

So, the bottom line is, the proof is in the puff.

I will continue my search for a Resmed system....
and try it.

The more I learn and use cpap, the more I am fascinated by it.
My collection of cpap "stuff" continues to grow.
I got my battery last night.

Wonderful discussion. Thanks! all.

[-SWS]

Carbonman, you're absolutely right about myriad variables in physiology allowing for unpredictability. The PAP machine with an algorithm is but one "black box" or system pitted against a far more complex, far more varied, and far more mysterious system: the human body.

Predicting an end-to-end system response or treatment outcome is virtually impossible in most cases, where preclusive factors or information are not apparent. And I have vehemently argued that precise point in the past. However, I mentioned earlier in this thread that some degree of "fuzzy" or tentative analysis can occur toward this objective succinctly summarized by Mar:

to filter out certain machines that may or may not work

Looking at your charts, I don't see any particulars that would rule out a trial with A10. And statistically, that seems to be the case for the vast majority of people who occupy that A10-friendly part of the bell curve, regarding sufficient precursor signals. It's never a crime to screen for statistical outliers in physiology when a friend says they're getting ready to buy their cousin a pair of so-called "one-size-fits-all" waders. Because those waders are, in fact, "one-size-fits-most".

Then, of course, as you mentioned the discussion itself is plenty interesting. And hopefully it's educational for some message board members who are new to apnea or APAP algorithms in general.

A real live genuine diatribe? If so there's always Dabrowski's paradigm...