Flow Limits mid Clear Airway Events?

Hi there,

This is my first time reaching out for help. I have been reading here for several months and have learned SO much. I so appreciate the contributions of new and seasoned posters alike.

I have been looking at my data and have noticed flow limitations in the middle of several of what my machine flags as clear airway events. Just wondering if anyone else has run into this. I'm not concerned with the number of centrals I have per night (it varies anywhere from 7 to 20), but I do think their presence might make me not feel well in the morning and I'm just wondering if my machine is flagging these events correctly, as its "non response" to CA could affect my therapy. My AHI runs between 1.5 and 3 depending on how long I sleep, but I am still just not feeling well during the day, and I wake up often with headaches. My sleep hygiene is poor... I'm a night owl and take advantage of being able to sleep in on the weekends, and I also have a hard time with sleep maintenance - no trouble with onset. I can't post pictures right now, but will get some up later. Just wanted to see if anyone had heard of flow limitations during clear airway events.
Thank you!

Seems like there shouldn't be any such thing as a flow limited Clear Airway, but perhaps the two algorithms work independently and the definition of not obstructed enough to be an obstructed apnea is still restricted enough to be flow restricted? Seems like a bug, but I really don't know.

A quick check of my Sleepyhead data found a flow limited CA. The upload is a bit fuzzy - the built in screen grab function in the unstable beta is, well, unstable and I squeezed a regular screen grab down to save space since the image isn't critical.

Yep, that's what I'm seeing. It's the same in Rescan too. Flow limit(s) of varying degrees during FOT determined clear airway. Maybe, like you said, the flow limit isn't big enough or long enough, or occurring at a point to upset the running average to override the OA vs CA FOT formula determination.

Hartemed wrote: ↑

Tue Aug 07, 2018 11:38 pm

Yep, that's what I'm seeing. It's the same in Rescan too. Flow limit(s) of varying degrees during FOT determined clear airway. Maybe, like you said, the flow limit isn't big enough or long enough, or occurring at a point to upset the running average to override the OA vs CA FOT formula determination.

When I zoom out, it's clear that in spite of the flow limitation the APAP pressure does not increase. Apparently priority is given to the CA designation over the flow limitation, whereas the hypopnea does get a pressure increase.

Here's the ResMed patent application, including the logic flow charts for determining flow limitation, FOT CA detection and whether to increase PAP.

https://patentimages.storage.googleapis ... 730886.pdf

Hartemed could you do something for me please and get a screen shot of what you are seeing and post it.
Zoom in similarly to what Stom did with his screenshot but just not quite so much.
I want to see a few more breaths.

Mainly because I want to see what your breathing looked like before the CA/Central flag along with the little blip that you are thinking is a Flow Limitation. So I can see what you are seeing please.

Stom's looks like awake breathing or post arousal breathing prior to the area where there is a little blip sort of in the middle of the flat line called a central/CA. I don't think it is a flow limitation per se....I think it might be simply a small little quick breath and not necessarily a restricted airway breath.

It's hard to know for sure and later when I have time I am going to try to duplicate with my AirSense machine and see if it is what I think it might be and duplicate it.

Here ya go Pugsy...hope it works and looks ok. As far as a double breath goes, I might be doing that. My centrals and obstructions never really look flat, and it always confused me, but once I learned about FOT, I attributed the waviness and fluctuations mid-event as the machine measuring airway patency. I might be trying to "recover" mid event, but I don't know. There are so many variables.

Stom- thanks for the patent. I read it and in between my eyes glazing over at times, I gained a very rudimentary understanding of what the machine is doing and how it determines airway patency. For anyone who is interested, the relevant parts to my question are sections 14C-16 (near the end of the document) Basically after 4-6 seconds of apnea, it delivers 1cmH20 pressure and manipulates the data and feedback, adjusting for heartbeat flow etc. to determine a mean induced flow signal. If categorizes the airway as "open" if it comes up with >0.03l/min and airway "closed" if its <0.03l/min. In theory, based on the patent and the formulas used to determine flow and patency, it appears you can have a "closed" central apnea. Not sure how the machine responds to something like this, or how/if it flags it though. I was way over my skis trying to make sense of the modulating, demodulating, adding amplitudes vectorially etc etc to get more than a very basic understanding of what the machine is doing.

Hartemed wrote: ↑

Wed Aug 08, 2018 5:49 pm

I gained a very rudimentary understanding of what the machine is doing and how it determines airway patency. For anyone who is interested, the relevant parts to my question are sections 14C-16 (near the end of the document) Basically after 4-6 seconds of apnea, it delivers 1cmH20 pressure and manipulates the data and feedback, adjusting for heartbeat flow etc. to determine a mean induced flow signal. If categorizes the airway as "open" if it comes up with >0.03l/min and airway "closed" if its <0.03l/min.

Or... this: https://www.youtube.com/watch?v=4GW97Xk06N8

Hartemed wrote: ↑

Wed Aug 08, 2018 5:49 pm

Basically after 4-6 seconds of apnea, it delivers 1cmH20 pressure and manipulates the data and feedback, adjusting for heartbeat flow etc. to determine a mean induced flow signal. If categorizes the airway as "open" if it comes up with >0.03l/min and airway "closed" if its <0.03l/min.

Thanks. I've read a number of descriptions of FOT, including at least on published paper specifically on it. While I understand the forced oscillation part, I've never actually understood how the FOT-based scoring for patency works, nor how it can work with different mask types having different amounts of dead space.

Stom wrote: ↑

Tue Aug 07, 2018 11:49 pm

Here's the ResMed patent application, including the logic flow charts for determining flow limitation, FOT CA detection and whether to increase PAP.

https://patentimages.storage.googleapis ... 730886.pdf

Thank you stom!

Pugsy, any thoughts on that picture I posted? It might just be a no effort to breath with an obstruction. I might be a person who is having post arousal centrals due to inadequate pressure. I raised my pressure last night and only had 4 centrals (one with a flow limit in the middle part of it on the graph) and one apnea (which came out of the blue with no real warning). Still feel tired today, but that's nothing new. I only wore the machine for about five hours or so.

I have another question that I can't seem to find an answer to...
Does anyone know how the machine responds when a person's baseline zero flow rate changes/fluctuates? I can often see fluctuations in the negative flow rate (ie an exhalation stays below zero with small increases and decreases in negative flow, that appear to be small breaths) Does the machine convert a negative + negative flow rate to a positive and count those breaths? Some are so small in comparison to the preceding flow, I wonder what the machine makes of them, if anything.

Hartemed wrote: ↑

Thu Aug 09, 2018 4:29 pm

Does the machine convert a negative + negative flow rate to a positive and count those breaths? Some are so small in comparison to the preceding flow, I wonder what the machine makes of them, if anything.

If you mean the respiration rate, the machine doesn't count breaths, other than for it's moving averages.. sleepyhead is coming up with the RR.

I don't see a flow limitation on that one screen shot where there are 2 CAs/centrals.
It's a bit iffy in terms of asleep vs arousal breathing. Not clear cut or easy for me to say which it is.

As to your second question. I have no idea. Sorry.

Pugsy wrote: ↑

Wed Aug 08, 2018 3:11 am

It's hard to know for sure and later when I have time I am going to try to duplicate with my AirSense machine and see if it is what I think it might be and duplicate it.

For curiosity's sake I did try triggering flow limit and CA this morning before I shut down my APAP. Just partially closing my airway for a while while breathing very slowly was enough to do it. Dunno what to make of it, but it doesn't happen more than a couple of times in my sleep data.

Stom wrote: ↑

Thu Aug 09, 2018 5:17 pm

For curiosity's sake I did try triggering flow limit and CA this morning before I shut down my APAP. Just partially closing my airway for a while while breathing very slowly was enough to do it. Dunno what to make of it, but it doesn't happen more than a couple of times in my sleep data.

Flow limitations aren't something we can fake I don't think and a small blip on the flow rate graph isn't a flow limitation either...at least by anything I was ever taught. I was taught flow limitations are something we see on the breath itself and the shape of the breath.

Pugsy wrote: ↑

Thu Aug 09, 2018 5:43 pm

Flow limitations aren't something we can fake I don't think and a small blip on the flow rate graph isn't a flow limitation either...at least by anything I was ever taught. I was taught flow limitations are something we see on the breath itself and the shape of the breath.

I think someone could totally fake flow limitations while awake if they practiced it, especially if they could see a nasal flow meter in real time while they practiced it.* It's just the rate you breathe in over time, which is totally something we have voluntary control over while we are awake. There is no technical or anatomical reason I can think of that would prevent someone from modeling those sleep scoring flow patterns to one degree or another while awake.

The blip on my flow limitation graph, the little grey bar, is specifically the ResMed machine flagging a flow limitation. It's just math comparing the current breathing amounts and inspiration curve to previous ones. No reason I can think of that the machine couldn't be tricked by someone who is awake and consciously breathing.

But I do think it would be harder to fool a manual PSG scorer who eyeballs the flow rate -time graph than machine scoring. The machine doesn't have intuition for "something seems fishy" based on the exact shapes and details of the inspriation curves on the flow rate time-graph.

In contrast, I'd totally say people shouldn't be able to play the violin, but that's only because I think the amount of practice needed play one well is crazy! Seems to me it would be a lot easier to control a breathing pattern than to learn to play a fretless instrument. But I certainly couldn't fake flow limits in my sleep. That's not voluntary.

*Penn of Penn & Teller has said that one way magicians are able to fool people is by spending an *insane* amount of time practicing skills, and that audience members just can't comprehend someone going to that much trouble to perform a trick. That's what comes to mind when you write "Flow limitations aren't something we can fake". They might not be *easy to fake well*, but humans can do amazing things if they practice.