Tweaking Bilevel

[ahuman]

Should I give a fixed pressure a try? What settings should I start with?

[ahuman]

I'm testing EPAP 10, IPAP 14, PS 4 just to see. From limited data, curve shape can still be not smooth at times and there's a very slight (similar to the 9/14 setting charts posts) sinusoidal envelope at times during the night versus completely level wave tops.

Are some people unable to get to nicely shaped curves throughout the night? Any other suggestions?

[PeaceLoveAndPizza]

How are you looking at the waveforms? Post a few 2-3 minute zooms for folks to review.

Not every breath is perfectly shaped. You are looking for a general trend of smoother transitions from IPAP to EPAP. There will always be something limiting the flow that will show up given the granularity of the readings and how the smoothing algorithms work. 

To your fixed pressure query, what you are using (9-14 PS 4) is similar to a fixed pressure mode, just with 1 cmH2O of headroom for the machine to use if needed. Nothing wrong with trying a fixed pressure if you want to give it a go.

[ahuman]

Possibly some of this is useful?

[PeaceLoveAndPizza]

They are all the same with some flow limited transitions, but most are quite good. 

You may want to try adding a bit more pressure support to see if it makes any difference. Maybe something like:

Mode VPAPauto
EPAPmin 9
IPAPmax 15
PS 4.2

Depending on how things look you can add a bit more PS (0.2 increments) or drop it back by a similar number. As things settle in you can change the min/max as desired.

[ahuman]

I'll give that a shot. I did experiment with more PS before, but the pressure range for this machine was less well defined so it might not have given expected results. Is the idea that more pressure support can best target flow limitations? Wouldn't 14/10 PS 4 give even more pressure than 15/9 PS 4.2 (except at the very high end) or am I missing something important about the pressure support number itself and how that provides stability to the air flow curve? How will I know when the PS is too high?

Without measuring REM fragmentation, is there an easy method to assess whether it could be worth trying ASV as it sounds like some people might need that to get to optimal stability in the air flow curve: https://www.youtube.com/watch?v=2zm-Bthcd9w . How hard is it to get sleep docs to let one try ASV? I didn't have trouble with the bi-level script. Other than machine cost, is there any downside to ASV versus sticking with bi-level?