Mouth Expiration and Effect on SH Data

I originally posted this on another member's OP, titled " Data to tell if I am breathing out of my mouth"....my apologies to Mariposa13, I didn't mean to step on your thread. I will delete that post and start over with this new OP.

This topic about mouth breathing is very interesting for me. Mariposa13's, conversation, along with palerider's link to Jay Aitchsee's July thread on mouth breathing wave forms, has motivated me to ask for your opinions to my personal SH data.

My profile should show that I use a FULL face mask, and that is mostly because I exhale through my mouth. My novice attempts to understand all the SH data has led me to a lot of confusion over three issues:

1. my breathing wave forms shows the flattened exhalation that Jay Aitchsee has discussed. It is interesting that before I am asleep ( when I know that I am exhaling through my nose) my wave form shows the gradual exhalation "slope". However, when I am asleep ( and I assume I am mouth exhaling) the wave form changes to the "flattened" wave form.

2. my expiration time is always shorter than my inspiration time, which I believe is opposite the norm... my "95% Statistic" usually shows my inspiration time of 3.3 to 3.5 seconds and an expiration time of 2.9 to 3.2 seconds. I have attributed this reversed ratio to the fact that I exhale faster thru an open mouth than I do thru my nose.

3. my Dream Station software flags, on average, 8 to 10 FL's in an 8 hour period. The wave form that preceeds these FL's usually looks like the flattened (mouth exhalation) wave form shown in Jay Aitchsee's posts. I mention this mostly because the Dream Station software apparently responds to these FL's by bumping up the IPAP and EPAP pressure per my PS Min and Max settings...so, the pressure chart shows a lot of changes due to FL's and not because of HYP, OA or CA.

My questions are:

1. Do the comments and data I am reading in regards to Mariposa 13's nasal pillows and mouth breathing also apply to me (full face mask and mouth exhaling)?

2. Does my novice conclusion about my shorter expiration time and mouth expiration have any merit?

3. Can my "flattened" wave form be the cause for Dream Station to respond with increased IPAP and EPAP pressures?.....and is that a valid response?

I apologize is any of this has been discussed. My search skills are not that good. I wasn't sure what screen shots to post, but can do that if anyone wants more data.

Thank you,

Zack

Hi Zach,

First, let me say that I don't think you should put too much stock in the Inspiration time to Expiration time (I:E) ratio. It is a metric used in conjunction with ventilation and does not really apply to normal CPAP use. Additionally, there is some question about the accuracy of these times using Sleepyhead and Respironics machines as well as using the 95% numbers to calculate I:E. Please see this thread for a detailed and long discussion: viewtopic.php?f=1&t=113290&st=0&sk=t&sd ... 0#p1094368

I indeed expect the expiratory portion of the flow wave to be flattened when exhaling through the mouth when using a Full Face Mask as well, but not necessarily shortened in length (time). Primarily I think, as you imply, because there is less resistance when exhaling through the mouth. I have not yet captured the waveforms to confirm this, but I do have a good reference which I will add later by edit for support. Of course we know, that with a Full Face Mask, mouth leaks won't degrade therapy or result in leaks being scored by the machine.

I doubt your machine is responding to mouth exhalation to drive the pressure up (although, I am not very familiar with Respironics algorithms). I rather suspect it is responding to Flow Limitations.

I, like you, am a mouth exhaler when asleep and thought for years I couldn't wear a nasal mask. Some time back, I was forced to switch and found that, contrary to my expectations, I could. I also found that my pressure requirements to prevent obstructive events went down and my Flow Limitations were reduced dramatically. I have some mouth leakage, but even without any preventive measure like tape or cloth, it's not bad enough to degrade therapy. However, the leaks are disturbing so I've taken to wearing the headband described in the waveform thread.

I suggest you get one of the latest nasal masks, the ResMed P10 or the Respironics Dreamwear and try it. You might be surprised.

Jay Aitchsee wrote:Hi Zach,

First, let me say that I don't think you should put too much stock in the Inspiration time to Expiration time (I:E) ratio. It is a metric used in conjunction with ventilation and does not really apply to normal CPAP use. Additionally, there is some question about the accuracy of these times using Sleepyhead and Respironics machines as well as using the 95% numbers to calculate I:E. Please see this thread for a detailed and long discussion: viewtopic.php?f=1&t=113290&st=0&sk=t&sd ... 0#p1094368

I indeed expect the expiratory portion of the flow wave to be flattened when exhaling through the mouth when using a Full Face Mask as well, but not necessarily shortened in length (time). Primarily I think, as you imply, because there is less resistance when exhaling through the mouth. I have not yet captured the waveforms to confirm this, but I do have a good reference which I will add later by edit for support. Of course we know, that with a Full Face Mask, mouth leaks won't degrade therapy or result in leaks being scored by the machine.

I doubt your machine is responding to mouth exhalation to drive the pressure up (although, I am not very familiar with Respironics algorithms). I rather suspect it is responding to Flow Limitations.

I, like you, am a mouth exhaler when asleep and thought for years I couldn't wear a nasal mask. Some time back, I was forced to switch and found that, contrary to my expectations, I could. I also found that my pressure requirements to prevent obstructive events went down and my Flow Limitations were reduced dramatically. I have some mouth leakage, but even without any preventive measure like tape or cloth, it's not bad enough to degrade therapy. However, the leaks are disturbing so I've taken to wearing the headband described in the waveform thread.

I suggest you get one of the latest nasal masks, the ResMed P10 or the Respironics Dreamwear and try it. You might be surprised.

Thank you, Jay, for your time and explanation.

I looked more closely at my machine's pressure spikes, and you are correct that exhalation characteristics had nothing to do with it.. the spikes were caused by flow limitations, and rightly so.

I also looked closely at my SH flow rate and pressure charts for the last several weeks. I am very surprised that the algorithms for this Dream Station call for pressure spikes EVERY TIME that a FL is flagged, but do not respond with increased pressures to every CA, OA , or H that is flagged. Can you or anyone shed any light on this?

zack243 wrote: also looked closely at my SH flow rate and pressure charts for the last several weeks. I am very surprised that the algorithms for this Dream Station call for pressure spikes EVERY TIME that a FL is flagged, but do not respond with increased pressures to every CA, OA , or H that is flagged. Can you or anyone shed any light on this?

there are a number of things about the way the respironics auto machines operate that I think are kinda dumb. that'd just be one of them. another one is their pressure probes, allegedly to figure out whether the sleep can be improved with more pressure... however, they *only* do that when sleep is near perfect! as soon as your sleep is a bit disturbed, or uneven, when a little more pressure could help... then it quits, goes to lunch, does nothing. madness.

zack243 wrote: I am very surprised that the algorithms for this Dream Station call for pressure spikes EVERY TIME that a FL is flagged, but do not respond with increased pressures to every CA, OA , or H that is flagged.

Can you show us with a screenshot what you are seeing?
I am wondering if you are seeing the normal pressure probe thing and thinking it is the spike in pressure caused by the FL.

CAs....Unless you are using the 960 mode it won't ever respond to a CA with more pressure. These machine by design will totally ignore a CA/central.

OA and hyponea responses....they won't respond until after the fact and then only if certain criteria is met (and don't ask me exactly the criteria because I don't remember but it a combination of FLs, snore and how many OAs or hyponeas all within a certain time frame) will it respond. It won't ever respond in the middle of an OA or hyponea. This isn't just Respironics...it's ResMed and the other brands as well. They just sit and twiddle their little thumbs until you start breathing again and then they might decide "I might need to do something so this doesn't happen again" and that's when they start to evaluate all that other criteria that might have occurred at the time of that flagged apnea event.

FLs and snores are probably the biggest driving force behind pressure increases because they are usually the first warning signs that the airway is trying to collapse and get obstructed.

Pugsy wrote:FLs and snores are probably the biggest driving force behind pressure increases because they are usually the first warning signs that the airway is trying to collapse and get obstructed.

Yes, and I don't think that is a probably. It's for certain.

zack243 wrote:but do not respond with increased pressures to every ... OA , or H that is flagged. Can you or anyone shed any light on this?

True events of that nature end in an awakening. So, it doesn't make sense to increase the pressure when you awaken. What makes sense is to allow you to fall back to sleep and watch for FLs to start and then increase the pressure. Most OAs and Hs are indeed preceded by FLs (instability of airway patency).

Of course, the CPAP process does not use an EEG, so it can not be sure when you are awake and when you are asleep. (It can use the flow to estimate when you are awake, but this is not built into the auto algorithm.)

A small emergency came up. I apologize for not continuing my part of this conversation . I will respond tomorrow with some screen shots. Thanks to all who commented.

Zack

palerider wrote:
there are a number of things about the way the respironics auto machines operate that I think are kinda dumb. that'd just be one of them. another one is their pressure probes, allegedly to figure out whether the sleep can be improved with more pressure... however, they *only* do that when sleep is near perfect! as soon as your sleep is a bit disturbed, or uneven, when a little more pressure could help... then it quits, goes to lunch, does nothing. madness.

I understand your points and agree. Thanks for sharing that information....I was beginning to think there was a bug with my machine.

Pugsy wrote:
Can you show us with a screenshot what you are seeing?
I am wondering if you are seeing the normal pressure probe thing and thinking it is the spike in pressure caused by the FL.

CAs....Unless you are using the 960 mode it won't ever respond to a CA with more pressure. These machine by design will totally ignore a CA/central.

OA and hyponea responses....they won't respond until after the fact and then only if certain criteria is met (and don't ask me exactly the criteria because I don't remember but it a combination of FLs, snore and how many OAs or hyponeas all within a certain time frame) will it respond. It won't ever respond in the middle of an OA or hyponea. This isn't just Respironics...it's ResMed and the other brands as well. They just sit and twiddle their little thumbs until you start breathing again and then they might decide "I might need to do something so this doesn't happen again" and that's when they start to evaluate all that other criteria that might have occurred at the time of that flagged apnea event.

FLs and snores are probably the biggest driving force behind pressure increases because they are usually the first warning signs that the airway is trying to collapse and get obstructed.

Pugsy

Using your information, I can now better understand the PR programming. You are correct about CA's (with my model) not causing a pressure increase, and OA and H do not always result in pressure increases.....sometimes the OA and H will cause a pressure spike, but those events are always occurring at the same time as a FL. I have looked back through several weeks of SH data and the FL's always precede a pressure increase. Attached are three random screenshots that I believe prove your points. What do you think?

Midnight Strangler wrote:

Pugsy wrote:FLs and snores are probably the biggest driving force behind pressure increases because they are usually the first warning signs that the airway is trying to collapse and get obstructed.

Yes, and I don't think that is a probably. It's for certain.

zack243 wrote:but do not respond with increased pressures to every ... OA , or H that is flagged. Can you or anyone shed any light on this?

True events of that nature end in an awakening. So, it doesn't make sense to increase the pressure when you awaken. What makes sense is to allow you to fall back to sleep and watch for FLs to start and then increase the pressure. Most OAs and Hs are indeed preceded by FLs (instability of airway patency).

Of course, the CPAP process does not use an EEG, so it can not be sure when you are awake and when you are asleep. (It can use the flow to estimate when you are awake, but this is not built into the auto algorithm.)

Thank you. I totally agree, and still have so much to learn. Check out the screenshots I posted for Pugsy.

Jay Aitchsee wrote:

I indeed expect the expiratory portion of the flow wave to be flattened when exhaling through the mouth when using a Full Face Mask as well, but not necessarily shortened in length (time). Primarily I think, as you imply, because there is less resistance when exhaling through the mouth. I have not yet captured the waveforms to confirm this, but I do have a good reference which I will add later by edit for support. Of course we know, that with a Full Face Mask, mouth leaks won't degrade therapy or result in leaks being scored by the machine.

Jay

I am attaching two screen shots. I am not sure if these will add anything to your waveform information, but they may confirm some of your comments.
The first one is prior to falling asleep...I am certain that at this time I was awake and both inhaling and exhaling through my nose. The second screenshot is later that same night....at that time I am confident that I was asleep, and when I am asleep I am exhaling through my mouth. What do you think?

Next time you post images....omit the last two graphs (don't need snores and resp rate graph) so that the other 4 graphs aren't so tiny.

The pressure graph is so tiny it's really hard to see if the pressure probes are present or not but they don't appear to be happening in large numbers.
BUT I just now noticed your pressure range is so limited that they probably aren't going to show up that much anyway.
And I remember why the tight range to start with....the duct issue.

We've sort of tied the machines hands a little bit with your settings but it was done with a reason in mind so we accept it. This means the algorithm has also had its little hands tied a little bit.

Bear in mind that you may not always see whatever the machine is responding to but just be assured that it thinks it has a good reason to do what it is doing. Trust your machine.

Pugsy wrote:Next time you post images....omit the last two graphs (don't need snores and resp rate graph) so that the other 4 graphs aren't so tiny.

The pressure graph is so tiny it's really hard to see if the pressure probes are present or not but they don't appear to be happening in large numbers.
BUT I just now noticed your pressure range is so limited that they probably aren't going to show up that much anyway.
And I remember why the tight range to start with....the duct issue.

We've sort of tied the machines hands a little bit with your settings but it was done with a reason in mind so we accept it. This means the algorithm has also had its little hands tied a little bit.

Bear in mind that you may not always see whatever the machine is responding to but just be assured that it thinks it has a good reason to do what it is doing. Trust your machine.

First time with posting screenshots since I joined imgur. Before membership it seemed that the screen shots were always too large! Definitely tiny images now that I can choose the "large thumbnails" code. Sorry about that.

I don't know if your OS or browser gives you the capability, but these small screenshots can be enlarged quite a bit via the "zoom" function on one of the pull- down menus.

You have a great memory to recall my lacrimal duct issues and the resulting limitations on machine settings.

I agree with the self-imposed limits I have with my tight pressure range. As always, you are spot-on with your comments, and I appreciate your time to explain it.

Your large thumbnail image was fine. I could click on it and see the larger graph but the problem was with the organization of the SH graphs. When you try to get so many graphs on one screen shot it causes all of them to be so tiny and no amount of enlarging will help.

With your machine the main graphs that we like to see are the
Events
Flow rate
Pressure
Leak
and usually with just 4 graphs on the right (and don't forget the stuff on the left) we have the basics of what we need and if we need something else we can just ask.
No need to redo the images you did today though. Now that I see what is there and remember your history and the tight range....I don't need to see the pressure graph in better detail.

I am not so great with flow rate zoomed in for figuring out mouth breathing or not...so I will defer to Jay on your question about that.

zack243 wrote:Jay I am attaching two screen shots. I am not sure if these will add anything to your waveform information, but they may confirm some of your comments.The first one is prior to falling asleep...I am certain that at this time I was awake and both inhaling and exhaling through my nose. The second screenshot is later that same night....at that time I am confident that I was asleep, and when I am asleep I am exhaling through my mouth. What do you think?

Yes, I agree, the second set of waveforms has the characteristic shape of expiratory mouth flow, AKA mouth leak and sometimes mouth breathing. I think mouth breathing would more properly apply, though, if there was the presence of mouth inhale as well.

Here is an article that is interesting. It has nothing to do with sleep, but it does show the changes in the flow wave with different types of breathing: nose in, nose out; nose in, mouth out; etc. It also implies why it may not be a good idea to completely prevent mouth exhale as with tape.

http://jap.physiology.org/content/105/3/854