EERS Experiment Data (sherwoga)

[pholynyk]

The flow rate example in post 109 is a cautionary example to those who suggest that respiration rate can be found by using a hysteresis band to define inhale and exhale time. Those are pretty wild swings across zero.
It occurred to me that a Fast Fourier Transform of the flow rate might give a more accurate idea of the fundamental frequency of the breathing, while the strength of other frequencies might be an indication of flow limitations.

This is all just off the top of my head, so take it as something for further consideration, rather than as fact.

[sherwoga]

I used my ResMed AirCurve 10 VAuto for the first time last night.  My sincerest gratitude to each member at this board that was so very helpful in my acquiring this machine, as well as, to have enlightened me along the way. This truly has been an empowering experience and I think that is consistent with your mission as stated on the home page under "About Apnea Board" located at the bottom of the page.  And I think it appears on every page.
   and feeling very empowered!

Settings:
Mode         VAuto
Max IPAP   16
Min EPAP   5
PS             5.4
TI Max       2.0 (default?)
TI Min        0.3 (default?)
Trigger       Med (default?)
Cycle          Med (default?)
Mask          Full Face
EERS          12 inches


Logic of these settings (such as there is any logic):  
I was trying to get this machine set up pretty close to the way my AirSense 10 Autoset was set up as I used last on Feb 6, night before last and for multiple nights before that.  A big change made on purpose is the setting for pressure support.  Slowriter had recommended here in Post 100 of this thread,  a pressure support value of 4.  Some weeks ago JoeyWallaby had shared his settings and flow patterns here in Post 37 of this thread. Joey used a setting of 5.4.  I am of the belief that in experimentation if you want to see an effect you need to be as "aggressive" as you think the process can tolerate.  I felt 4 was a bit conservative and that Joey's shared data gave me a safe aggressive setting.  Had I not been able to tolerate this much pressure support, I felt it would be obvious quickly and that I could lower it during the night if needed.  

I had a lot of dialog with Geer1 in many recent posts about the additional settings on the AirCurve 10 for which there are no equivalent settings on the AirSense 10, these being, Trigger, Cycle, Ti Max, and Ti Min.  I was really concerned that cardiogenic oscillations that were so prevalent in my breathing pattern obtained on the AirSense might really compromise the functionality of these four settings.  I think I have all of these set at "default" values, but have added a question mark to each entry above.  I'm fairly sure from the clinician manual the default for trigger and cycle is Med.   But my machine had 21 hours of use on it as received, i.e., it had been used at least in a demonstration setting.  So these settings (values found on the machine as received) might have been changed and the clinician manual is not very definitive.  As you can see from the results below, it does appear, based on sleeping through only one night, to have been a wasted worry.  I am still in need of help understanding and maybe modifying these settings.  See questions later.

Observations, Results and Questions:

           

Flow limits are almost gone   .  The 95% and maximum flow limits reported in the detail screen are 0.01 and 0.1, respectively.  Those are easily the lowest I've seen.  That is what this substitution of machines was all about.  If this result persists this whole effort including investment in the new pump will have been rewarded admirably.  

Cardiogenic oscillations (assuming Geer1 and I have these named correctly) are almost completely absent.  I say almost because I can see occasionally an oscillation somewhere on the flow pattern.  But I can't find anywhere where the presence of said oscillations cause a breath to be counted more than once because of multiple double crossings of the zero line in the flow pattern.  Hence there was less need to worry about the extra AirCurve 10 parameters discussed above.

Flow patterns look good to me.  I'm including a pop out of these for your review and comments. My pop out for last night is combined with one of my earlier flow patterns showing dense cardiogenic oscillations and one showing characteristics of flow limitations. 

   
 
Respiration Rate chart is the least noisy I've seen.  The machine reported a RR of 13 and OSCAR refined that number to 12.8 in the details screen. The RR stayed very close to 13 throughout the night with only two or three brief periods of much variation at all.  Note that I have often seen "apparent" values maxing out at 50 breaths per minute for 10s of minutes.

Apneas are very acceptable, but not as low as I've seen.  They appear to be relatively clustered and not to be SWJ (sleep wake junk).  I even wonder if they might have been positional.  And there are long periods throughout the night where both apneas and flow limits appear to be completely absent.  Does that mean I slept better during those periods?  I'm assuming/hoping so.  Also, I am gratified that the rather large increase in pressure support didn't bring on a significant number of central apneas.  I'm assuming that the EERS can be credited with keeping them in check.  Anyway I will be wanting to follow this closely and maybe to eventually check out the impact of reducing EERS to 6 inches.  I'm not interested in going to 18.  

Tidal Volume went up to 800.  I haven't shared my analysis of the data from my experiment, but for some output parameters, including tidal volume, I have reduced the data to mathematical models.  The pressure support of 5.4 really requires me to extrapolate quite a bit and any scientist or engineer worth his salt knows extrapolation to be a dangerous practice.  I certainly wouldn't make a business decision based on such extrapolation.  But if you allow me that flexibility I would have predicted a value of 811 cc for tidal volume, most of the increase coming from the impact of pressure support.  For now no one can challenge that conclusion or my data analysis, because it isn't published.  I hope to find a way to share those models at an appropriate level of detail.  But this observation is reassuring that the experiment may turn out to be useful.  Regardless, the reality is that OSCAR and the machine reported a Median Tidal Volume of 800.  Is that bad?  If so, then I still can't pronounce the EERS to be the tool I want to use to keep central apneas in check while simultaneously using higher pressure support to keep flow limits in check.   If I don't need to be concerned about tidal volume than publication of my results could turn out to be quite useful to others that might consider using the EERS to attenuate the negative effect of increased pressure support (that being the onset of CA events).  And in my case there may well be room for optimization.

E:I ratio is nowhere reported in the OSCAR Detail screen, but was reported on the machine itself in the clinician mode sleep report as 1:1.8.  I know from the manual that that value is critical to setting the Ti Max and Ti Min values optimally.  Can this ratio be calculated manually from other parameters reported in the detail screen?  If so, I haven't figured out how.  Can you help me use that value to reset Ti Max and/or Ti Min?

And Spont Cycle was reported on the machine as 93%.  It seems intuitive to me that this value is also related to the Ti Max and/or Ti Min.  Is 93% good?  Should I strive to make it better?  If so, which way do I need to change Ti Max and/or Ti Min?  

Comfort was okay to very acceptable.  I can honestly say I wasn't uncomfortable, but I do think I sensed a bit of effort breathing toward the end of inspiration as my diaphragm was changing directions to start expiration.  So a pressure support of 5.4 might be a little too high.  I don't plan on changing right away though.  Perhaps even that sense of effort will quickly dissipate with use.  Also, this could be ameliorated with an optimized Ti Max or Ti Min.  Thoughts?

[sherwoga]

The flow rate example in post 109 is a cautionary example to those who suggest that respiration rate can be found by using a hysteresis band to define inhale and exhale time. Those are pretty wild swings across zero.
It occurred to me that a Fast Fourier Transform of the flow rate might give a more accurate idea of the fundamental frequency of the breathing, while the strength of other frequencies might be an indication of flow limitations.

This is all just off the top of my head, so take it as something for further consideration, rather than as fact.

I like what I think you are proposing.  I'd love to see a deconvoluted flow pattern.  

Do you think that if you had the real or even an assumed heart rate, you could simply subtract a sine wave, or something close to it, from my "squirrely" pattern to generate a reasonable, corrected flow pattern?

[slowriter]

Glad to hear this is looking promising, and agree that the data does look better on the VAuto.

A couple things:

And Spont Cycle was reported on the machine as 93%.  It seems intuitive to me that this value is also related to the Ti Max and/or Ti Min.  Is 93% good?  Should I strive to make it better?  If so, which way do I need to change Ti Max and/or Ti Min?  

It probably means you can bump your Ti Max a bit; maybe to 2.4?

Comfort was okay to very acceptable.  I can honestly say I wasn't uncomfortable, but I do think I sensed a bit of effort breathing toward the end of inspiration as my diaphragm was changing directions to start expiration.  So a pressure support of 5.4 might be a little too high.  I don't plan on changing right away though.  Perhaps even that sense of effort will quickly dissipate with use.  Also, this could be ameliorated with an optimized Ti Max or Ti Min.  Thoughts?

You still have some reported clear airway events. Some of those appear to be while you're awake?

You might try changing trigger to high or very high and see if that helps with the remaining CAs?

BTW, I believe default Ti Min is 0.5.

[slowriter]

A tip on trigger, cycle, Ti: you can view the sleep report in real time on the unit.

So one way to adjust these is while awake, playing with the settings to see how they impact comfort and such.

[Geer1]

Did you drink 6 beer the first time you drank alcohol just because someone else you knew drinks 6 beer? You want pressure, PS and EERS to be minimum required. You should start out low and work your way up, instead you started at the maximum settings you should require and some of your results are questionable because of it...  

Regarding some of your questions.

I'd like to see what the flow rates look like during those apneas (flow rate and mask pressure), I don't trust the event flags.

I assume the machine is reporting I:E. Value should be around 2:1 for E:I. It is easy to calculate in OSCAR(don't trust the reported E&I times though), find some breaths and measure time for expiration(negative flow rate) and inspiration(positive flow rate). You need to look at examples from your different breath wave forms as the value will vary through the night.

[slowriter]

Did you drink 6 beer the first time you drank alcohol just because someone else you knew drinks 6 beer? You want pressure, PS and EERS to be minimum required. You should start out low and work your way up, instead you started at the maximum settings you should require and some of your results are questionable because of it...  

5.4 PS is hardly "maximum." I see nothing wrong with him going that route.

EERS might be another matter, since it increases TV, and you certainly don't want more than you need.

I'd be curious about the CAs with 6-inches of EERS, and trigger of high or very high.

[Geer1]

There is a reason EVERY titration protocol is to start with low settings, increase them slowly as needed and then return to a lower setting if there are no obvious improvements. These settings were a shot in the dark, and were based on data from another user who is still struggling to find good settings... A user who has been averaging PS of just over 4 on ASV and who had ugly results at PS of 6 due to the same issue (deciding to increase in a big jump rather than small steps). 

Many users see a significant difference in flow limitations from changing EPR of 3 to PS of 4 and Sherwogas flow limitations were already responding well to changes in EPR. I bet they would be minimal at PS of 4 and I do believe 5.4 PS is the maximum he will need.

Just my opinion though and I have already voiced this opinion strongly enough so I will let it rest now.

[sherwoga]

...

And Spont Cycle was reported on the machine as 93%.  It seems intuitive to me that this value is also related to the Ti Max and/or Ti Min.  Is 93% good?  Should I strive to make it better?  If so, which way do I need to change Ti Max and/or Ti Min?  

It probably means you can bump your Ti Max a bit; maybe to 2.4?

I increased Ti Max to 2.4 as you recommended and the Spont Cycle did increase to 98.5%.  I don't understand this.  The chart on page 6 of the ResMed AirCurve™ 10 VAUTO Clinical Guide (the clinician's manual) is difficult to understand.  I do observe column one lists example respiration rates and column two converts those RRs to seconds per breath.  I think later columns are supposed to be taken at probable settings for Ti Min  and bracketing settings for Ti Max corresponding to the RR.  The two Ti Max values correspond to differing I:E ratios, but not in a very intelligible manner.  

My RR was 13 for both nights putting me between rows 1 and 2.  My I:E  was 1:1.9, not much different from the night before at 1:1.8.  At no point in rows 1 and 2, or later rows for that matter, does the recommended Ti Max get bigger than 2.0, contrary to your recommendation.

What is the logic behind your recommendation of increasing Ti Max to 2.4?  It did appear to work.  So I'm not being argumentative.  I'm just trying to understand.  And is it worth it to keep inching Ti Max up, say to 2.5 tonight, 2.6 tomorrow night...?  Or should I be satisfied with Spont cycle at 98.5%?
...

You might try changing trigger to high or very high and see if that helps with the remaining CAs?

I did not change trigger for last night, as at this stage of getting used to my new pump and the measurement systems that are different relative to those of the AirSense 10, my experimentation philosophy reverts back to doing one factor at a time (OFAT).  I just don't have enough experience with this new pump to be changing more than one variable at a time and I would want to do that in the context of a designed experiment anyway.  Nowhere ready for that!  Your answer to my question above about further increasing Ti Max will dictate how soon I will try this change.

BTW, I believe default Ti Min is 0.5.

I did not change Ti Min either.  The chart seems to indicate though that 1) the I:E ratio has no or little impact on this setting and 2) I could be operating with Ti Min at 1.0 until RR exceeds 20.  Ti Min is, per the text accompanying the chart, designed specifically for patients with "weak inspiratory effort".  I don't believe I fall into that category and so suspect my current value of 0.3 is unnecessarily low, but that the low setting is of no consequence.  It likely would be for the target patient.  Am I overthinking this?  

Other observations are:

1. Tidal volume (Med) the same at 800 cc.
   
2. Flow limits again look good to me with far fewer observed throughout the night (than when I was using the AirSense).  Only significant observation is that the one period of large flow limits did cause the Flow Limit max to increase to 0.36 (from 0.1), but I consider this to be an outlier as the 95% Flow Limit was 0.00 down from 0.01. I guess I'm saying that the max value for flow limit seems a poor measure of overall performance.  Further, I suspect that is true for most, if not all, of the output parameters.  However, and surprisingly, the Max Flow Limit measurement did show a significant effect in my designed experiment   .
   
3. AHI down from 3.23 to 2.63.  Probably not a significant change.  
   
4. CAI down from 1.55 to 0.92.
   
5. OAI up from 1.42 to 1.58.  
   

Seeking your input for tonight to know whether or not to increase Ti Max further or change the trigger sensitivity to High or Very High.  Also, can I just ignore Ti Min for the time being?  

Another change that is obviously needed is to reduce the EERS length from 12 to 6 inches.  But I am planning on making that change much later as I really expect to increase CA's when I do that.  And you seem to be anticipating further lowering of CA's by increasing trigger sensitivity.  

Will gladly post screenshots if you think my doing so is needed.

[slowriter]

All the Ti Max does is specify the max time for the machine to stay in IPAP before it switches to EPAP.

If it's a bit short, it will at times switch before you're ready; when you're still breathing in.

The spont cycle % is telling you how often that doesn't happen; that you are driving that switch.

Therefore, if it's below 100, it suggests you can extend it a bit.

Again: you can experiment yourself while awake. Set it back to 2.0, for example, and take a deep breath, and notice what happens.

In the end, I think these tweaks are really about comfort though.

I don't really know any downside for it being too long; maybe sleeprider can clarify.