Rapid Breathing-Fractured Sleep-Out of Breath

[Fixit50]

Some additional settings seen in Oscar from last night:

[SingleH]

Thanks SingleH for the comments and insights. 

You said: “Regarding the first screenshot I can only assume those two large flow limitations are due to some kind of positional obstruction.”
Yes, this is very possible; I’ll keep an eye on these.

“Your flow limits appear to be well under control now as your 95% is zero.”
Yes, I’m delighted also to see this.  Aside from the PS, the other settings are virtually the same as what I used with the APAP so its pretty clear that the increase in PS from 3 to 4 is responsible for the improvement on flow limits. 

“Regarding the waveform, the improvement is clear with Bilevel and a waveform of sorts is now present rather than the chaos seen before with APAP.”
Well Put!  This tends to confirm my prior hypothesis that, at least in my case, the chaotic waveform was something that came about with the presence of flow limits during certain sleep states, likely REM, deep, or both.  The chinstrap, mouthguard, collar all got rid of the poor waveform by diminishing flow limits via anatomical adjustment.  The VAuto got rid of the poor waveform also by diminishing flow limits, but thru the use of Pressure Support rather than through the use of an anatomical changing device.  These observations enhance our understanding which is always nice.  

Last night I repeated the same settings and again used only the F20 mask.  Results are quite similar.
 
In adjusting Ti-min, is it also advisable to make only small adjustments?  Trigger and cycle are both set to medium which is how I received the machine. For the last two nights (mask only, no accessories) I got two CA’s each night.  Earlier on, I increased Timax from 2.0 to 2.2 because I noticed a small number of inspiration times that were higher than 2.0 and wasn’t sure if I wanted to cut that short.  Most inspiration times are lower than 2.0 however so there wouldn’t be a lot of breaths cut short by pulling Ti-min back to 2.0.  Any thoughts on that?  Perhaps I can try going back to the default value of 2.0.  As your insp time is typically well within 2.0, I believe reducing it to 2.0 wouldnt actually have any affect and therefore not have any CA altering effect. When I refer to cutting the insp time short by reducing Ti-Min, you will clearly see the Insp time flat line and level off. I definately wouldnt recommend do this as you want the machine to let you allow you to breathe as long as you want, it was just a suggesting to control CA's if they are really a problem. You want to maintain patient device synchronicity where possibly and cutting your breathing short would result in a drop in your Spont Cycle figures which are currently very good.
 
Also, the machine report shows I:E is 1:2.1 but I don’t really understand where that comes from.  Could it be the max ratio?  When I go to various parts of the flow rate, and directly measure the ratio, I rarely see more than 1:1.9 and also see ratios as low as 1.3.  So again, I’m a little confused about what the machine shows for that ratio. I note that looking at the stats section in Oscar implies I:E ratios and shows inspiration times that agree with what is shown in the machine's sleep report.  Also, the period for the Sleep Report is set to 1-day.  I have read comments on here that the I:E ratio is calculated from a short period of time like the last 20 mins of use or so, but I dont have definitive proof of that. The main things is 1:2.1 is a good figure . 1:2 is considered normal. It will vary and floating between 1:1.9 and 1.3 is fine from what I have seen.
 
The spontaneous cycle times are high for last night (98.5%) and the night prior.  I imagine this is the percentage of pressure transitions that were triggered by breathing effort as opposed to Timin or Timax limits.  Is higher better?  What would be a good number? 98.5% is great. Basically its simply saying that you are initiating the transiting from EPAP to IPAP your self. If the machine is not detecting your end of exhalation or cycle and the machine simply switched over at the end of TiMin then the spont cycle goes down. In the case of my Dad, to control his CA's I reduced his TiMax and his spont cycle which averaged around 70-90% dropped to about 40-60% because the machine was cycling before he wanted to.Whilst sleeping he can take very long inspirations (over 3.5s), but having a very long TiMax results in have lots of CA's. I need high PS to address his flow limits and waveform, but it causes CA's, I need high EPAP to address his apneas, but it causes more flow limits, I need a long Timax to maintain his schronicity, but it raises CA's. I'm trapped in a sort of three way paradox where changing one setting throws out one of his problems. You sound like someone who comes from a technical background, so I'm sure you can appreciate given the ability to adjust IPAP, EPAP, PS, TiMin/Max, Trigger and Cycle gives an endless number of permutation's through which I am working. I am however closing in reliable results now and the game changer was understanding how the body responds to oxygenation, CO2 washout and that causing CA's seems to have some long lasting effect in the brain that takes to time to normalise. 
 
Is Inspiration Time important?  What are we looking for here, and why important?
 
Below are some graphs from last night that may provide some insight.  Note that Tidal Volume is 600ml, so I’m a little concerned that increasing PS might make that too high.  Is 600ml getting too high or is there room to move here?  I’m not sure what the limits are, and I can imagine its individual dependent.  Last night shows 600ml, which is the highest I’ve seen.  I dont understand Tidal Volume well as yet. I can say that if your I:E is good around 1.2 on average and your inspiration time graph is not being cut off (flatlined) then I dont think there is anything to worry about.


Let’s zero in on a region where the waveform contains some aberrations.  The snip below shows five expirations all disturbed by two or three pulses.  Only in two of the five cases did these extra pulses cross zero and trigger an added inspiration pressure pulse in the mask.  Where the undesired added pulse in the mask is not seen, it did not prevent the disturbed expiration waveform, so it doesn’t give me a lot of confidence that this can be improved by changing machine settings.  Of course it can’t help to try lowering trigger sensitivity, but I’m not optimistic.  Any thoughts or comments on this point, or what might be done to clean up the waveform seen below? I would like to see the graph zoomed out a bit further similar to what you did in your before and after screenshot.  This is very similar to the issue my dad had. For all intents and purposes I call this a "mid inhalation expiration". In his case when he inhales he does not have a smooth linear expansion of his rib cage and as he breathes in it moves somewhat jerkily and you it almost looks like his chest dips in briefly before reaching the maximally expanded state before he exhales. One the waveform you see this as a dip below the zero line, which was present on his APAP. However with the Bilevel this is significantly reduced. Whilst he still does this when he is breathing during the day when he is sleeping, having the long Timin seems to have ensure the machine overides any muscular deficiency or poor co-ordination to just keep delivering air through the inhalation phase. If I watch him breathing I can see his chest wants to dip in a little bit as he inhales, but the machine ignores that as Timin sets a minimum inhalation duration, so it just pushes through whatever miscordination of the chest wall is ocurring.
As you can see from your mask pressure graph, the machine is thinking you are taking another breath and viewing this as one of these "mid inspiration exhalation" or essentially two breaths rather than one. The only thing I can suggest for this is increasing Timin and possibly setting cycle to low, so it minimises the detection of you cycling. Start with Timin first alone then add cycle later.

 

Overall it looks pretty good, so as I said before, I would stick with only small changes, as basically Events, I:E, Spont Cycle are all good, so there is not much to improve except the waveform.

[Fixit50]

Thanks SingleH for this comprehensive response!  I agree things look generally good with the possible exception that the waveform perhaps, can stand further improvement. 
 
Another issue you’ll see below, is the question about recovery breathing continues to return. 

Last night and the night before, I increased PS to by 0.2 to 4.2 with no other changes.  I did this to see if I could gain some improvement on the waveform.  Here’s a summary of key results.
·       The period of time for which the double expiration pulse is evident decreased from about 40 minutes to about 20 minutes or less.
·       There was a period of high, bursty Flow Limits on the first night following this change.  Not so clear if this was a result of the change in PS or just night to night variations.
·       Median Tidal volume increased from 600ml to 620ml.
·       Spont Cyc dropped from 98.5% to 92.5%.
 
One thing that has been a long-standing issue is the breathing recoveries I see.  These were a real problem with the APAP and they would follow a long period of rapid and chaotic breathing and often resulted in waking up short of breath.  With the VAuto Bilevel, they are still seen, even though the preceding breathing is not chaotic and appears healthy.  Now if a chinstrap or collar is used, these recoveries are cut down in magnitude by probably 33% (as seen by minute vent). 
 
I’m a little concerned about these recoveries, because they may indicate CO2 flushing or some other possibly unhealthy issue.  Also, with the larger recoveries, as is seen with mask only and no chinstrap, I often wakeup within a minute or two following the recovery, further suggesting some stress to my sleep.  Has anyone seen these recoveries, or have any ideas about the significance or causes of them?
 
Here below are some snips from night’s data including a zoomed in view showing an example of a recovery.  If needed, I can add some other zoomed snips.
           

[SingleH]

Regarding your observations:

In terms of PS, I think you need to leave at least a few days to monitor after making any PS changes as PS seems to trigger some change in the brain, I "think" it can temporarily affect the whole breathing balance, so its best not to draw too many immediate conclusions.
I observed Tidal volume changes, and depeding on the combination of parameters set, this figure was vary. Applying additional PS could affect this figure.
I have been unable to achieved a consisted PS number only approximate ranges where it might site between 80 and 95% for a period of time or 50-60% on another set of settings. Without you having left the machine on the same settings for an extended period of time to establish a baseline, its difficult to know if this is just a natural variation or a change in response to your setting changes

You noted a significant reduction in the recoveries by the change to APAP to Bilevel.  I would suggest you continue to experiment with PS to see if you can improve these further. As you mentioned wearing a collar and chinstrap resolves them, so it could be postural and in turn flow limit related and the application of additional PS might be a worthwhile test to see if it reduces them further. You are only at 4.2 so plenty of headroom to play with at the moment.

One question about your last image posted, I see in the screenshot your device settings list shows Timin,Max and trigger. Is that automatically appearing in Oscar, as in my Oscar (which is on the latest version) doesnt show anything after temperature enable.

[Fixit50]

Regarding your observations:

In terms of PS, I think you need to leave at least a few days to monitor after making any PS changes as PS seems to trigger some change in the brain, I "think" it can temporarily affect the whole breathing balance, so its best not to draw too many immediate conclusions.
I observed Tidal volume changes, and depeding on the combination of parameters set, this figure was vary. Applying additional PS could affect this figure.
I have been unable to achieved a consisted PS number only approximate ranges where it might site between 80 and 95% for a period of time or 50-60% on another set of settings. Without you having left the machine on the same settings for an extended period of time to establish a baseline, its difficult to know if this is just a natural variation or a change in response to your setting changes

You noted a significant reduction in the recoveries by the change to APAP to Bilevel.  I would suggest you continue to experiment with PS to see if you can improve these further. As you mentioned wearing a collar and chinstrap resolves them, so it could be postural and in turn flow limit related and the application of additional PS might be a worthwhile test to see if it reduces them further. You are only at 4.2 so plenty of headroom to play with at the moment.
Thanks for your help and comments.  I’ll plan to take changes slowly as you suggest, allowing time to establish a new baseline for the change. 

One question about your last image posted, I see in the screenshot your device settings list shows Timin,Max and trigger. Is that automatically appearing in Oscar, as in my Oscar (which is on the latest version) doesnt show anything after temperature enable.
I’m using the latest version of Oscar, version 1.5.1, downloaded and installed 10-days ago.  I didn’t make any changes to Oscar settings.  If you have the same Oscar version, I’d suspect the difference is a VAuto machine variant.  Date of manufacture, and country of sale, could result in different machine variants sold under the same name as “AirCurve 10 VAuto.”  Also check to see if Oscar displays TiMax, TiMin and Trigger underneath “Device Settings.”  You may have to scroll down using the scroll bar on the left to expose these.  If these are displayed on your screen, then the issue may be that Oscar has truncated them in the screenshot, if you took your screenshot using “View/Take Screenshot” from within Oscar.  Generally, I scroll until Oscar displays what I want to capture, and then I manually take the screenshot using the “snip tool” in Windows 10. This way, I know I can snip what I see on the screen. 

[SingleH]

Thanks for the info on the software version, I am on 1.5.1 also, and the aircurve is a recent model with support for the resmed my air app. I also have a 32" monitor so a fairly decent size.

As you can see I have nothing shown below Temperature enable.

   

[Fixit50]

I’m not sure what’s going on other perhaps than different machine software or machine variants. The UK, may not get the exact VAuto machine that the US does. Checkout your machine software under About. Mine shows SX567-0401. Under Statistics in your screenshot, I see I:E Ratio. My I:E Ratio is not shown in Oscar. I have to go to the machine for that. Also, the I:E Ratio you see on Oscar is 69, which seems too high to believe.

Would you mind zooming in at 1:10am on your screen shot? Use a 4 to 5-minute timespan with the spike in Tidal Volume in the middle, and add Minute Vent to the displayed graphs so both TD and MV can be seen. I’d like to better see what looks like the kind of breathing recovery I see with some regularity. Thanks.

[SingleH]

I actually have never seen IE before in Oscar so that must have been a recent update as looking back at old screenshots it wasnt present. It is however completely wrong on any day I look at it, so maybe this is a work in progress update.

The aircurve unit was actually purchased and shipped from the US, as its just too complicated to get your hands on one in the UK. Anyway no big deal about the missing data, I will figure it out.

I have posted a screenshot of the event your requested plus a few more. As I mentioned before I have seen some similar looking "things" to what you highlighted but whatever they are they are not logged as events. The common thing I see, that is unusual, is that the machine seems to have applied a rapid spike in pressure very quickly after the initial breath. Look at the mask pressure graph and see the strange rough double spike in pressure at 1:09. Im guessing maybe the machine is confused by the erratic breathing. If you look on your pressure graph when these events happen do you see the same thing?

Going back through his data I can many examples of these things, and usually his breathing flow rate graph is basically all messed up and the mask pressure graph is too and always with one of these mask pressure disturbances in the middle. 

           

[Fixit50]

Thanks for posting the screenshots showing what I’ve been referring to as “recoveries.”  I think these are very similar to what I see.  Some of mine have approached 30 l/min at the peak of Minute Vent, so they can represent being badly out of breath.  I wish I understood the origin of these things better but I really don’t.  Best I can come with is CO2 flushing.  If that were the case, I would expect that if breathing preceding the recovery is healthy, then there should be no reason for elevated CO2 levels leading to a recovery.  The problem is that I’ve seen that case.

I can’t zoom in on what you’ve posted but from the distance it looks like mask pressure is just following the zero crossings in the flow rate.  I expect the machine to transition, or to begin the transition from expiration to inspiration pressure on positive zero crossings in the flow rate (and visa-versa of course).  It looks like its doing that or something pretty close.  If you zoom in closer, you might be able to disprove that.  On my data, it looks like the machine pressure is reasonably following the flow rate as expected. 

Thanks posting these snips.

[SingleH]

As you point out the increase in tidal volume and minute vent increases at around the time of the breathing abnormality and yes upon zooming it it looks like the pressure is just following the breathing pattern.

Regarding your theory on C02 Flushing, one thing I can say in my Dads case is he was getting a lot of CA's with a high PS and a long Timax. I had to dial back his Tmax significantly to cap his Inspiration time which in turn reduced CA's. My understanding is the shorter inspiration time, also reduces CO2 accumulation. If you look in my full screen screenshot back two posts you can see the recoveries happening during periods when his insp time was not flat lining, and I will now post some additional screenshots where you can see he has events where he is trying to breathe for longer and is also getting these events. Whether this follows your theory or not I'm unsure.


[attachment=58768]     [attachment=58770]