About APAP and algorithms - document from DeVilbiss

DeVilbiss Marketing wrote:For nasal interface users, the EPI is an indicator that a FFM, chin strap or oral nasal interface may be required.

Do you have documentation that "expiratory puffs" are clinically relevant?

If so, do you have follow-up documentation that the additional intervention (FFM, chin strap or oral nasal interface) results in either improved compliance or improvement in sleep quality?

How did you arrive at EPI Index of 30 as a significant value?

TIA for your response.

We do not have documentation regarding expiratory puffs or intervention, but the event and associated measurement is rooted in clinical experience. EPI is a metric that can be used to hone in on a potential cause for ineffective therapy. Excessive expiratory puffing reduces therapeutic pressure via “leak” out of the mouth. When expiratory puffing is present, the first line of defense is to eliminate the puffing from the mouth via chin strap or full face mask.

If there are no negative symptoms (excessive daytime sleepiness, high AHI) associated with an EPI of 30+, there should not be a problem. Again, the EPI metric can be used to identify potential problems.

DeVilbiss Marketing wrote:Achieving the lowest pressure to maintain airway patency is at the foundation of the DeVilbiss AutoAdjust algorithm.

The benchtop testing cited was conducted by Valley Inspired Products and published in Sleep Review http://www.sleepreviewmag.com/issues/ar ... -09_02.asp

Same study I believe but this link has the full information from it:

http://www.inspiredrc.com/McCoy%20et%20 ... vices1.pdf

DeVilbiss Marketing wrote: We understand that some may require a more aggressive response to apneas and hypopneas; as such we are the only manufacturer that accommodates adjustments to the definition of these events.

Are you sure about this? My Resmed machine most definitely gives a customized response to how severe my Flow Limitation is, and whether it is continuing or not etc.

It is a little tricky to tell on my charts as regards to apneas as MOST of the time when I have an obstructive apnea I am also having Flow Limitation issues (sometimes snoring but more often than not no snoring).

From the very few times I do not have Flow Limitation along with an obstructive apnea since starting APAP, it does look like the length of the apnea and whether there is more than one determines how aggressively Resmed responds to it.

It also considers when events are combined, like an apnea with snoring, or an apnea with Flow Limitiation, Flow Limitation with snoring, or all 3 together an apnea with snoring and Flow Limitaiton and agains responds accordingly accounting for how many factors there are AND how severe they are.

For example if it sees me have a 10 second obtructive apnea, it will ignore it (not sure I like this approach either. Looking closely it appears to score Flow Limitation much higher than apneas if it is looking at just one of those things by itself, I would hope they would reconsider this approach somewhat in future tweaking as I did find one night where my Flow Limitation was mostly fine and I had 5 ten or so second obtructive apneas in a row. Resmed took a NON-aggressive approach to this which allowed them to continue). IF however an obtructive apnea is combined with something else it responds (or if there is more than one but again it seems to score apneas on their own fairly low).

I have seen that other manufactures give a "canned" response but it sure doesn't look like Resmed is and good to hear Devilbiss isn't.

Jayjonbeach wrote:

DeVilbiss Marketing wrote: We understand that some may require a more aggressive response to apneas and hypopneas; as such we are the only manufacturer that accommodates adjustments to the definition of these events.

Are you sure about this? My Resmed machine most definitely gives a customized response to how severe my Flow Limitation is, and whether it is continuing or not etc.

DeVilbiss Marketing is talking about adjustments to the definitions of what the machine will regard as a hypopnea and an apnea.

As DeVillbiss Mkt'ing said, the other manufacturers (ResMed, Philips Respironics, etc.) don't have any way to change their definition of apnea or hypopnea. There's no way for the person who sets up those machines to change the definition of what those machines will regard as an apnea or as a hypopnea.

DeVilbiss Auto allows a way to change the definitions the machine will use.

So does this mean that the NEW Devilbis can now do what my OLD FRENCH built Goodnight/Tyco 420E (circa 2003) does and is ALSO adjustable for snore and amplitude like the devilbis ?

YIKES, That 420E was ahead of its time huh.

RG is correct.... the clinician's manual shows in the Advanced Sub-Menu the default settings for definitions of hypopneas and apneas, both % and duration. These are adjustable in this mode. %'s can be adjusted in increments of 5 and durations can be adjusted in increments of 2.

DeVilbiss Marketing wrote:Achieving the lowest pressure to maintain airway patency is at the foundation of the DeVilbiss AutoAdjust algorithm.

The benchtop testing cited was conducted by Valley Inspired Products and published in Sleep Review http://www.sleepreviewmag.com/issues/ar ... -09_02.asp and RT Magazine http://www.rtmagazine.com/issues/articl ... -08_05.asp and did not include a snore signal. As such we contacted VIP, provided them with a snore box and asked them to re-conduct the testing. Their results indicated that “the snore signal did have a significant impact on the response” and “yielded a more rapid pressure response,”

DeVilbiss Marketing wrote:... the event and associated measurement is rooted in clinical experience.

OK, that aggressive response when there is concommitant snoring is quite impressive, but my "clinical experience roots", as well as the several hundred Encore graphs tracking Vibratory Snores and their response to pressure tell me that snoring is easily controlled and quite responsive to pressure increase, while the flow limitation will persist for some time (indeed, in some cases, forever as NR-IFL).

I would therefore respectfully submit that the real-life response of the IntelliPAP algorithm would be much more similar to the original testing:



I would also object to logic behind NOA identification:

Non-Obstructive Apneas (NOAs)

Cause: Non-obstructive apneas may occur during OSA therapy if the PAP pressure reduces the normal carbon dioxide accumulation and degrades the CO2 stimulus response. NOAs also may be primary central apneas that cannot be properly defined by an Autotitrating device alone.

NOTE: Non-obstructive apneas in excess of 10 per hour could indicate primary central apneas, not NOAs or obstructive apneas, and clinical intervention may be required.

Definition: Amplitude is reduced to 5% or less of normal signal for approximately 10 seconds.

Response: No change in pressure. The AutoAdjust PAP tracks non-obstructive apneas, but does not respond.

Adjustment: Non-obstructive apnea response cannot be adjusted. A non-obstructive apnea’s non-response however can be overridden by changing the obstructive apnea percentage setting to 5 or less. This setting change will force the AutoAdjust to define non-obstructive apneas as obstructive apneas and respond accordingly.

because it says that severe reduction in flow must all be central in nature, and this most certainly cannot be so.

Indeed, central events may often have underlying obstructive component (see Mary Morrell's work), so FOT or pulse-wave technology seems like it would be far more accurate in central identification.

However, the logic used above re: snoring (flow limitation attack would remain sluggish) could be used here in re: OAs (pressure increase may not be desirable) since they both tend to disappear with application of pressure, and, after all, you do have that workaround:

A non-obstructive apnea’s non-response however can be overridden by changing the obstructive apnea percentage setting to 5 or less. This setting change will force the AutoAdjust to define non-obstructive apneas as obstructive apneas and respond accordingly.

rested gal wrote:

Jayjonbeach wrote:

DeVilbiss Marketing wrote: We understand that some may require a more aggressive response to apneas and hypopneas; as such we are the only manufacturer that accommodates adjustments to the definition of these events.

Are you sure about this? My Resmed machine most definitely gives a customized response to how severe my Flow Limitation is, and whether it is continuing or not etc.

DeVilbiss Marketing is talking about adjustments to the definitions of what the machine will regard as a hypopnea and an apnea.

As DeVillbiss Mkt'ing said, the other manufacturers (ResMed, Philips Respironics, etc.) don't have any way to change their definition of apnea or hypopnea. There's no way for the person who sets up those machines to change the definition of what those machines will regard as an apnea or as a hypopnea.

DeVilbiss Auto allows a way to change the definitions the machine will use.

Thank you for pointing that out. I read somewhere that some machines actually give a "canned" response to some events and I thought that was what was being referred to.

Having the ability to adjust the definitions is an interesting option depending on the original parameters and how far they are allowed to vary. People that are easily aroused could keep things slower, where as someone who always has issues in REM sleep and no arousals could hopefully speed up the HELP, though there will still be limitations and one big precursor missing completely from the equation.

I'd like to add some real life data to this interesting thread. My own nightly desats come in groups every 2 hours, probably in REM sleep, for a typical duration of 30 minutes. A typical Oximeter report for the first two hours of my night without CPAP looked like this



This cycle repeats more or less over the night.

Having done with mouth leaking and mask issues (using the Hybrid with no leaks) I can finally try how different auto CPAPs deal with this pattern.

The CPAP I'm currently using is the DeVilbiss IntelliPAP. My pressure range is set to 10-18 cmH2O.
NotMuffy, I know that 10 cmH2O may be too low to start with but I can't set the minimum any higher due to aerophagia (I tried).
(BTW for other discussion - the IntelliPAP reports expiratory puffs with a taped well-sealed mouth. I have no idea how it's possible)

I expect that once events start, the auto CPAP would quickly up the pressure to deal with the events, then slowly lower it when done, until the next round.

Sadly, the DeVilbiss IntelliPAP can't do that.

With its default settings, once problems start, the IntelliPAP would ramp up the pressure 2-3 cms over the course of 30 minutes.
This is a joke and does not deal with my breathing problems at all.

To increase sensitivity, as suggested here by DeVilbiss Marketing, I adjust the apnea % to 0% so that all events would be recognized as hypopneas. The IntelliPAP response to apnea and hypopnea is identical so apnea definition only changes the report but not the response. I also adjust the hypopnea % to 30% reduction for 6 seconds. These settings make the IntelliPAP as sensitive as possible.

What happens? indeed, the IntelliPAP raises the pressure up to 18 cms, much higher than the default settings.
However, it still takes the same time to get there, about 30 minutes, which is a useless solution to a 30-minutes hypopneas attack.

Nightly pressure with aggressive settings:



Although DeVilbiss claim that the AutoAdjust can raise pressure 1 cm every minute rate of change (a rate not very fast by itself), even with the most aggressive settings I have not seen it raise the pressure more than 1 cm every 3 or 4 minutes. That's awfully slow. By the time the pressure was raised from 10 to 18 the events are done.

Also, with aggressive settings the IntelliPAP identifies hypopneas while awake and raises pressure accordingly. I used 20 minutes ramp at the minimum pressure to avoid this problem. This is the first time I found use for the ramp function.

DeVilbiss could easily fix the AutoAdjust algorithm with a software upgrade, for example using an "aggressiveness" parameter or by recognizing flow limitations as everyone else do. I hope they do, because hardware-wise the IntelliPAP is very nicely built.

However, until they fix it, the AutoAdjust algorithm do not provide good therapy for people like me, Jayjonbeach and likely many other people who have periodic hypopneas that suddenly appear and disappear, say in REM sleep.

I plan to try out other auto CPAP and will update here.

You can adjust the timing for responses via the clinician's menu.....

Emilia wrote:You can adjust the timing for responses via the clinician's menu.....

In the clinician menu, Advanced submenu, there are four items defining events: apnea %, apnea duration, hypopnea % and hypopne duration but sadly nothing about the IntelliPAP event responses. I don't see it elsewhere in the menus nor in the documentation. How to adjust the responses?

yrnkrn WROTE:

> (BTW for other discussion - the IntelliPAP reports expiratory puffs with a taped well-sealed mouth.
> I have no idea how it's possible)

I get the same on an IntelliPAP with a taped mouth and a nasal pillow mask.

Often people with nasal masks, and a decent head of pressure, close off the back of their throats/oral cavity to prevent pressurised air getting into the oral cavity and expanding the cheeks.

I think the device is registering expiratory puffs when one relaxes the appropriate muscles and pressurised air is released into the mouth causing the cheeks to suddenly blow out and bulge.

I hope this helps

Best wishes

Ian Lowery