AHI < 1.0 now, but still tired? Is it UARS, RERA, arousals or FL in your FR peaks?

[2SleepBetta]

cathyf, a bit of follow up.

In re the top and bottom of your third graphic in your most recent post: 

I see your point about similarity of FR curves vs difference in FL. Unless expanded views of one to two minutes showed sufficient explanatory small FR and FL variations  (and there are some indications of small FL changes), it almost seems, given the stickiness of your FL values, that the FL scale-# holding register could  require a substantial drop in the sliding window-area value from the earlier stored scale #. Then the register would be reset to enable a fresh look at changes in (sliding window) air volume delivery. That delivery (biorn indicated, as I understand him ) is based on a 2-second-wide look back window for determining the FL flag value (if any). I do believe zoomed views of 1 - 2 minutes of FR and corresponding FL would clear away our questions--if enlarged so that an enclosing rectangle for a single breath would  require a 1/2" wide x 1" box, or (better) a larger box, on screen.

I don't know what to make of your "FR falls from the top of stairs" vs indicated FL:

Similar to my comment above, I think--confirmation bias flag flying here-- that expanded views of the inspiratory curves would show reasons for most all the FL's displayed--there being at least one main exception from about 05:44:00 to 5:45:45--it's those first two FR drops to the bottoms of the stairways, particularly, those along with the drop after the peak at about 5:45:10. The upstairs moves would be mostly ignored (until flow was above the axis), as I understand, but it seems greater FL's would show for the drops.

As a beginner fiddling with this unfamiliar topic, I have come to believe my visual assessments are not at all dependable; one has to (I must) measure variables and their accumulations. My slow developing understanding has come to realize that the signal represented by the FL flag is dependent not only on the variations in the inspiratory FR amplitudes and shapes (the most obvious things) but also on the more subtle accompanying relative durations of inflow and outflow.  It's a matter of all those variables and the critical differences in fill levels of successive sliding window-boxes which the Resmed algorithm uses. 

I did paste a couple of 1 min. to 2 min. views into MS publisher and stretched them way out to get a better feel for your variations in FR form and size at large variations in FL, but those checks were not enough to conclude what I've written, not more. 

A late-in-PAP-life learning example from working on assessing a FL after an OSCARed sigh in a FR currve last night:

It always puzzled me  why (amid clean breathing) a small FL so often occurred during exhalation immediately after one of those "Norwegian" sighs--those two stage inhalations that together with their exhalation afterward create huge repeating spikes in many of our FR curves. The drop, I concluded, from measurements, is because of the prolonged exhalation which more than offsets the two stage (two step) inhalation. The sigh's sliding window, however wide, had sufficiently less fill than the predecessor sliding window. Therefore, a FL flag.



Musing, yes, but no help here . I do hope you will use the c-collar, get a VAuto, get an oximeter and get a device (a camera or accelerometer) to show your sleeping positions. My Autoset FL were bad but nowhere equal to yours. The VAuto cut them waaay down--with pressure support of just 4cm--vs my 3 cm EPR with the Autoset--to the point FL are few above about 0.20 and FL are mostly sparse along the nightly time line.


A later relevant research addendum for cathyf's consderation:

A research paper applicable to patients with low AHI but inspiratory flow limitation: Quantifying the magnitude of pharyngeal obstruction during sleep using airflow shape.
shape

https://erj.ersjournals.com/content/erj/...2.full.pdf

[KingKongBingBong]

So is there actually any way to eliminate RERAs lol. i might speed up my jaw surgery if not

[Sleeprider]

So is there actually any way to eliminate RERAs lol. i might speed up my jaw surgery if not

Please see your other thread http://www.apneaboard.com/forums/Thread-...ting-RERAs

[2SleepBetta]

Parts of the report linked below are covered in the ApneaBoard's Wiki, but the compreshensive treatment of CPAP type machines and their differences justifies posting the link and a few comments, here, now, and later. In my limited understanding, it treats at a fairly deep level almost all brands and types of machines most regularly discussed here at AB, including a lot of detail about algorithms.

A look here disabused me of the notion that Resmed Autosets, maybe their AirCurve VAuto, might reflect in flow rate (FR), tidal volume (TV), and minute volume (MV) a reduction for breathing deadspace. If I read the study correctly that deduction is only made by devices that are designed to deliver fixed MV or TV. Pictured, below, from the long paper is the heading from Table 2 comparing Auto CPAP device attributes for Resmed, Respironics and DeVilbiss devices. 

Table 2 coverage of flow limitation handling, and more, are likely to be of interest to flow limit sufferers who are digging deep in dealing with that mostly foster or forgotten child of sleep medicine and insurance. Table 2 mentions four elements involved in the Resmed detection and scoring of flow limitations with FL flags. I believe the AB Wiki includes much of that.

[attachment=33610]

[2SleepBetta]

Part 1
     Recommended: Scan the research paper on nasal airflow restrictions and measurement  at [url= https://www.atsjournals.org/doi/full/10....1005-034RN] https://www.atsjournals.org/doi/full/10....1005-034RN[/url]. This part of this post presents a few snippets and a couple of images  from that linked paper. I offer a few words attempting  to highlight what struck me most.

There are a couple of do-it-yourself tests. Most sufferers I am aware of would want to make a judgment about whether their sleep MD or advisor is giving them the guidance they need. The paper will help upper airway sufferer-readers do that. The ApneaBoard Forum has lay experts--not me--who can provide informed guidance and suggest possible answer to most all questions after they see the set of graphics they will call for. This ApneaBoard Forum, in my  opinion, would never have been needed if sleep medicine deficiencies and medical insurance had met needs dramatically better than is common. 

The text image, the first below, has these section headings:

-Nasal airway resistance is about half of total airway resistance (for normal breathing 2SB asks?)
-The nasal cycle usually lasts 4 to 6 hours
-To assess airflow one needs to understand the nasal valve (see second graphic picture and the related Cottle maneuver). The nasal cavity from the nostril to the nasal valve is the area of greatest flow resistance.
-It has been postulated that objective assessment of the nasal airway patency ("adequacy"?) can never predict the actual subjective sensation of nasal patency
-The Cottle maneuver is one anyone can do to check their nasal valve. Further, one can check whether or not a decongestant opens the congested airway. If it does not open the airway it is a sign of structural obstruction rather than nasal congestion.

Part 2
     More on the topic of upper airway resistance

Below there is a sampling from among recent threads that have dealt with, or are currently dealing with, difficult upper airway restrictions, flow limitations, and unrestful sleep as well as with the meaning of the Resmed devices' flagging and not flagging of flow limitations. There is an emerging success story and a continuing difficult case of airway restriction. These linkages are here instead of others only because I have spent more time and thought on them recently as I was preoccupied with matters related to this thread in the posts at the bottom most URL.

In some instances graphics have been deactivated as members ran out of their ApneaBoard storage space. Nevertheless, scanning the posts will yield better understanding of this little-treated subject of flow limitations--as distinct from apneas-- in sleep medicine. I either link to the first post in the threads or where posts seemed to become focused more directly on flow limitations.

The fourth link is focused currently on the extent to which a Resmed FL flag indicates drops in essential air flow, that is, drops in Tidal Volume. Some flags are believed  to reflect the machines' sensitivity--as part of its pressure regulation up and down--to certain wave shapes, inspiratory wave flatness and respiratory rate changes more than the flag reflects a local loss of TV.

http://www.apneaboard.com/forums/Thread-...#pid389057  - deals with nature of FL and how scoring it can be  summarized better

http://www.apneaboard.com/forums/Thread-...ow-limited - deals with a difficult case that appears to be a success story (after using the Resmed AirCurve10 VAuto to treat flow limitations

http://www.apneaboard.com/forums/Thread-...#pid400219 - deals with a difficult case of upper airway resistance and is motivation for sharing, at the top section of this post, the link to and the images and text snippets from the site https://www.atsjournals.org/doi/full/10....1005-034RN
 
http://www.apneaboard.com/forums/Thread-...#pid394885  - deals with trying to discover  the extent to which flow limitation flags, FL,  are determined by drops in tidal volume, TV, or by inspiratory wave shape, wave flatness or respiratory rate changes, RR changes.

         

[2SleepBetta]

----First, a correction: please, accept my apology.

Ouch! Just  saw my most recent obvious and belatedly acknowledged blunder and lapse: omission of the link I promised two posts above where I included only the heading of Table 2. That table compares and contrasts technical differences among several AutoCPAP machines and is linked-to,  this time, here https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3792383/ .

That table is only a taste of the NIH paper about diagnosing and treating sleep apnea with those and other more sophisticated machines. I checked back at that post above to find the (absent) link after spending too much time and effort in errant directions  a sooner revisiting of that table (about algorithms) would  have better informed, stopped or redirected.  

I had been laboriously analyzing and graphing breath by labored breath of my most SDB-illustrative sleep breathing. The visual correlations of flow limitations and higher duty cycles and I/E ratios have been disappointing to date. But the Sao Paulo study, below, may explain why. I intend to post more about that later, either here or in my closely related "Call for VBA help...[re FL]", an AB Software Forum thread.

----Second, research done to help identify levels of flow limitation which may or may not need treatment.

This is--a duplicative/redundant? 'hope not--link to a "Sao Paulo study" of the range of inspiratory flow limitation (IFL) in a large sample of persons without identified sleep breathing disorders or other breathing related health issues.  

The research conclusion:

"In summary, our study demonstrated that only 5% of normal individuals present with > 30% of the total sleep time with IFL, establishing a value for IFL below which one statistically cannot invoke OSA in otherwise healthy individuals. Despite the observation that some individuals of the “normal” group presented up to 56% of IFL, the 95th percentile of a distribution (e.g., 30% for % IFL) is conventionally used to statistically establish a likely boundary between health and disease. This is further supported by the increase in % IFL found in patients with mild OSA (AHI 5-15 and no symptoms) and OSAS (AHI 5-15 with clinical symptoms). Thus, values of up to 30% of IFL should be considered “within normal limits” and only values > 30% can support the likely presence of abnormality. However, studies of clinical outcomes related to the % IFL need to be done to confirm these findings."

The link to the research:  https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3792383/

[cathyf]

I'm rather bemused by my 2014 ApneaLink report where it records flow-limited breathing. (I'm also bemused that my APAP is a flow-limitation-treatment machine, but none of my in-lab sleep studies mentioned them at all.)

   

The part where it says that the normal level of flow limited breaths without snore is <60%, and the normal level of flow limited breaths with snore is <40%.

So if you have 59% flow limited without snore, and 39% flow limited with snore -- i.e. 98% flow limited -- that's normal, LOL!

[2SleepBetta]

----First, a correction: please, accept my apology.

Ouch! Just  saw my most recent obvious and belatedly acknowledged blunder and lapse: omission of the link I promised two posts above where I included only the heading of Table 2. That table compares and contrasts technical differences among several AutoCPAP machines and is linked-to,  this time, here WRONG.

Gotta own up to it! My "correction" only compounded the initial blunder in using the same link as was used for another purpose in the post. Nothing to say but "Sorry".
The Table is attached and, better yet, here is a link to the entire piece where it is found https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4629962/ .

   

[2SleepBetta]

The attachment attempts to correctly display (from an Excel spreadsheet chart) graphs of about a 6-hour portion of a member's sleep data taken from data files with Resmed Autoset (RMA) time stamped flow rate (FR) and Flow Limit (FL) data. The graphic is extremely busy but with notes on the graphic and, if necessary, a limited review of the thread where the graphs were worked up, I believe persons deeply interested--those as lost as I am in our sleep metric weeds--can decipher it. Questions and constructive criticisms are encouraged and welcome. The intent is to show a real, highly flow limited example of most of the key relationships among those presented by OSCAR for us.

The graphs in the graphic, top to bottom, are intended to show:

1.  The red, rust and  yellow colors: Show agreement and disagreement of tidal volume losses (red and rust TVd) indicated by numerical integration of FR data (in  dark green)  to determine tidal volume (TV, in light green) vs. indications of  RMA flagged flow limitations  (FL, yellow and rust) . These are plotted descending from the top graph axis. In theory the red graph would mesh exactly with the dark green graph lower down to form a solid red and green rectangle overall, no cracks between red and green. Where the red graph shoots above the upper axis, there were one or more large breaths that exceeded the fixed 0.5 L (assumed "normal") baseline TV I used (as subtrahend) for determining TV drops. (TVd = 0.5 L -TV as TV was determined from integrating FR data).

 My position is that we are most interested in how much a FL flag tells us about our Tidal Volume losses (TVd, red and rust), drops caused by restrictions of airflow we call flow limitations ("fL" is my shorthand for all flow limitations, other than apnea, flagged or not). We are not particularly interested (most of the time) in FL shown only because particular wave shapes were encountered in our sleep. RMA FL indicate one or more among mere "offending shape" detection, respiratory rate (RR) change, duty cycle (dC) change or occurence of a TVd. We mostly care not, but might look at inspiratory tips of FR curve waves to see FL explanatory deformations when we are troubled controlling our fL.

2. The light green and dark green colors: Show ordinary FR (dark green) and TV (light green), both plotted on the same zero-axis. Note: it is obscure, but TV values range from the 0-axis up to 0.5 L or more.

3.The grey, orange and blue colors: Show breathing duty cycle (dC = Ti/Total = Ti/(Ti+Te, gray), inspiratory time (Ti or I, orange) and expiration time (Te or E, blue). Relationships among those items as well as the more erratic Ti/Te =I/E (not shown), which our RMA show on screen (e.g., as "1:1.8"), can tell us how hard we are working to breathe in sleep. Unusually high values of Ti, I or dC indicate a high work level fighting fL, which may or may not be flagged as FL. I learned, from experts here at Apnea Board, to look at I and I/E ratios in cases were AHI is low but sleep is unrestful. The work of breathing against fL wakes many of us up with frequent mini arousals we mostly do not remember. 

Here is a link to a a late post in my thread where there is more explanation of the graphic and how it came to be done: Call for Excel VBA help: to support effort to clarify/understand FL | Apnea Board. 

In a few words, another thread member sheepless started gave rise to debate about what RMA FL mean. My linked thread (and the attachment) have answered that, in part at least, but did not produce the improved FL summary sheepless hopes will be developed. Neither does my thread convincingly answer his question about the relative importance of a low level long duration FL vs. a shorter but higher severity level FL. Pending availability of better percentile or other answer types: Despite reasonable criticism of it, I still believe a total of  all 2-second severity level one FL values would shed some more light for comparisons of our FL night to night. For example, a 4 second FL of severity level 4 would add 16 FL units to the accumulating total and stacked and stair-stepped FL would need proper handling.  

   

[sheepless]

2sb, understanding this complicated subject would take more time than I'm willing to devote, so I, for one, really appreciate all your efforts!

in the interest of incorporating duration (in addition to severity) in the flow limitation metric, I wonder what the oscar team thinks about your comment above, quoted below. should it be added to the list of possible future oscar enhancements to be evaluated by the team?

"Despite reasonable criticism of it, I still believe a total of all 2-second severity level one FL values would shed some more light for comparisons of our FL night to night. For example, a 4 second FL of severity level 4 would add 16 FL units to the accumulating total and stacked and stair-stepped FL would need proper handling."