Inspiration expiration ratio

Ron, here's the two sample charts from above and I think it illustrates the probable problem with your I:E.

The top chart is fairly normal and the computed I:E is about 1.5/2.25 or 0.66 (sorry I lost the times shuffling these images around)

The I:E in the bottom chart is unknown because the expiration portion crosses 0 at least 3 times, the one the machine uses will determine the I:E.

From left to right, the first arrow starts the expiration and the last arrow should start the inspiration. If so, the I:E would be about the same as the one at the top.

However, I'm betting it takes one of the first to cross which would increase your I:E considerably. If it took the first to cross and used the preceding inspiration, to the right of the green line, it would definitely be greater than 1.

Now in the subsequent waves. it can be seen that your expiration curve has that little bump at the top where it goes positive. Technically, I suppose ending expiration and starting inspiration only to fall back negative. How does your machine treat that? I don't know, but my guess is it causes an error in the I:E computation. We can see by looking that your true inspiration doesn't start until a bit later.

So now the real question is not what is your I:E, but what is that little bump? Sleeprider referred to it as rebound something, I think. I'm still leaning toward mouth expiration. Is it significant? At this point, I don't know.

This phenomenon, which looks a little like taking a small breath during exhale, has been discussed in this forum before by our multi named, used to be, resident expert (muffy, not muffy, Sludge, et al.) along with diagrams, etc. Maybe someone remembers where that was?

Is it possible this is caused by machine early IPAP trigger?

So, our used to be resident expert, AKA Mollete at the time, referred to these "bumps" as I called them, as BLL (Bunch of Little Lumps) in this post
He goes on to define them as Ballistocardiographic Artifacts. Which sounds bad until the word Artifact is considered. Apparently, these are caused by the movement of blood through the arteries which can cause a small change in the respiratory pressure or flow. From a quick glance at some of the posts in the forum which mention them they can cause some CPAP machines to sense Inspiration beginning sooner then it actually does which would lead to incorrect computations of Expiration Time, which in turn would lead to incorrect I:E ratios using those times. Again from a quick glance at the posts, these BLL do not seem to be cause for concern, but do cause concern among some who discover them in their data.

Jay Aitchsee wrote:.............
He goes on to define them as Ballistocardiographic Artifacts. ...............Again from a quick glance at the posts, these BLL do not seem to be cause for concern, but do cause concern among some who discover them in their data.
..........

Thank you Jay, that looks like an interesting thread. I will read it more carefully. I do have arterial plaque - maybe that exaggerates this particular phenomenon?

At the moment, I'm less concerned about I:E inversion per se, and more concerned about whether there are abnormalities in my flow that may not be artifacts.

I was going to wait a few days til I've accumulated more data with the ffm, but I'll present some data from last night. As you suggested, last night, I switched to the FFM, turned off AFLEX completely, and switched strictly to CPAP, so the pressure was fixed at 9.5 the whole night. The first thing I notice is a marked increase in leaks, snores, and in RERA's. There's one bathroom break in there, and that cluster in the second session I believe occurred during restless wake time. The junky flow traces around apnea events are still there (but not shown.) In addition to the full night, I've included zooms on two moments when I'm pretty sure I was asleep with no nearby event. Notice the saddle points or flatness on some of the inspiration peaks. My impression is these are more pronounced than they were with the Dreamware mask, but I'm not sure. Do you think these are that same BLL phenomenon, or something else? Technically they are flow restrictions, right? Can they be mask dependent? Since I think the majority of the apnea events are junk-related, I'm thinking I'll lower the pressure tonight. I need to work on the leaks, no?

But I did awaken a couple of times, supine and with dry mouth. Wasn't the purpose of this expt to test for mouth breathing? It appears to still be there. Should I be making additional changes?

http://imgur.com/a/REtK4

BTW, that bump in respiration around 2:40 - I almost always get that, followed by lesser ones later. I attribute that to dreaming hence probably REM stage. The interesting thing is, there are usually few or no apneas during those bumps - the opposite from what I'd have expected from my reading.

Jay thx for your continued interest,

-Ron

Hey, Ron, I think this has been an interesting thread.

This Ballistic thing is quite interesting and there's lot's more in that other link. I think what is happening, referring to the flattening in the peaks, is that the heart rate and breathing rate are not the same so we see the BLL appear to move in position. Another thing is that the heart rate is going to be around 70 BPM and the breath rate around 15, so harmonics probably come in to play as well. Anyway, if you look at your example and the one from Mollete's posting you can see bumps seem to move along the wave flattening the top of the Inspiration wave as they go. I don't think these are Flow Limitations and your machine didn't flag them as such.

I'm not used the Respironics displays so a lot of this stuff is "noise" to me. Rera, VS2, the leak presentation, etc. I do know that the 24 L/m leak line is a ResMed and not a Respironics metric and I know Pugsy would say if the machine is not flagging a lot of Large Leaks, you're probably OK and yours isn't. Basically you would subtract the static leak from the leak line to get your actual which I think is what SH is trying to present with that bottom line, which looks pretty decent.

Yes, on the Full Face. I was interested in seeing what that would do to the bumps. Since we now Know the cause, we know why it didn't do anything. No, on the Full Face stopping intentional mouth breathing, It will stop unintentional mouth leaks, where the cpap pressures causes air to be expelled through the mouth, but it won't stop intentional exhale through the mouth and the dry mouth that goes with it. It looks to me you are still exhaling through your mouth because of the deep plunge and fast rebound of the expiratory curve. Nasal expiration would produce a slower rise taking on a more triangular shape. Although, I think the two FFM samples looked a little better than the Nasal samples. But then, I didn't put them on the same page and take a critical look.

However, there was one big change that I noticed. Your AHI was about the same, but it shifted from CA to OA. Which could mean that you suffered fewer disturbances with the FFM. Since OA and H went up and CA went down, the next move would typically be to increase the pressure a little to see if the OA didn't come down. But, since you said you thought those were occurring while you were awake and one night is not a decent sampling, I think you should spend at least a couple of nights getting accustomed to the FFM before making more changes; Or, lower the pressure a little if you think you would be more comfortable, but I think it's fairly common that FFM pressures need to be a little higher than nasal to sustain the same level of OA suppression.

You're a numbers guy. Maybe you should make a spread sheet to keep track of some of the key stats to compare between pillows, FFM, and pressures.

Jay, I'm starting to read about ballistocardiography. Fascinating. [Reminds me of the trouble I had trying to meditate - I kept feeling I was being interrupted by my own heartbeat. I'm thinking of founding a new branch of medical science: ballistocardionarcissism... ]

Since we now Know the cause...

Er, let me interject a dumb question here. If these machines are sufficiently sensitive to detect modulations in the air flow caused by the cardiovascular system, why doesn't everyone's graph show them? I doubt I'm the only one here whose heart is beating.... (Which BTW, in my case occurs about 52 times/min.)

-Ron

rkl122 wrote:Jay, I'm starting to read about ballistocardiography. Fascinating. [Reminds me of the trouble I had trying to meditate - I kept feeling I was being interrupted by my own heartbeat. I'm thinking of founding a new branch of medical science: ballistocardionarcissism... ]

Since we now Know the cause...

Er, let me interject a dumb question here. If these machines are sufficiently sensitive to detect modulations in the air flow caused by the cardiovascular system, why doesn't everyone's graph show them? I doubt I'm the only one here whose heart is beating.... (Which BTW, in my case occurs about 52 times/min.)

same reason that not everyone's meditation is interrupted by their own heartbeat... people are different.

rkl122 wrote:Er, let me interject a dumb question here. If these machines are sufficiently sensitive to detect modulations in the air flow caused by the cardiovascular system, why doesn't everyone's graph show them? I doubt I'm the only one here whose heart is beating.... (Which BTW, in my case occurs about 52 times/min.)

Ron,
Remember at the outset of this thread, I said I was no expert? Well...

Anyway, I vaguely remembered something about this distortion in the waveform, but not anything specific, so I had to hunt it up. I haven't had much more of a chance than you to read up on it, I just glanced through some of threads and posts to see if it was something that one should be concerned about and didn't see that it was. Someplace, the discussion centered around just that. That this caused a problem in the results so some manufactures used filters to either pass or block certain bands of frequencies when producing the flow wave. For example, a band pass filter that would only allow frequencies below 50 Hz should filter out most heart related components of the wave but allow the respiratory components or frequencies nominally around 15 hz through.

Now, I have to caution that the above is a lot of fill in the blanks from picking up a few words of discussion while skimming a bunch of posts. I have no idea what the actual methodology is, or even if there is one, for that matter. It's a subject for further research.

And, as I see PR has just posted, "people are different"

I could have made this a lot shorter by saying, I don't know, or I'm not sure.

Ron, I'm sure you've given some thought to the fact that your bradycardia could be another reason that your sleep is disturbed and also could be an underlying cause of fatigue during the day without being directly related to sleep apnea. Honestly, I don't know how one rules that out.

Interesting thread. When I see my sleep doc in a few weeks I'll make sure to ask him about my inspiration/expiration data, which according to some in this thread are unusual. My inspiration/expiration ratio over the last two weeks (since I bought the fancy new blower) is in the 1.5 to 2.0 range; my old machine didn't show the data.

Jay Aitchsee wrote:".....

I could have made this a lot shorter by saying, I don't know, or I'm not sure.

Thx Jay. Maybe the refs will help us understand better.

Ron, I'm sure you've given some thought to the fact that your bradycardia could be another reason that your sleep is disturbed and also could be an underlying cause of fatigue during the day without being directly related to sleep apnea. Honestly, I don't know how one rules that out.

Yup it's on my list for the docs. I'm not sanguine about the cardiologist though. He's been trying to control my systolic hypertension for years, and now suddenly I learn on my own that CPAP therapy is more effective than all those medicines combined. Doh. But yeah my highest priority is to deal with those very symptoms.

Anyway, thx again Jay. Will stick with the ffm for a bit and bump this if I learn something new.

Pneumophile - I've noticed several postings that show that inverse ratio. Maybe it's more common than I first thought. Let's post if the docs provide insight.

-Ron

rkl122 wrote:... why doesn't everyone's graph show them? I doubt I'm the only one here whose heart is beating

Hah! So, I looked at my results from last night and there they were! Not quite as pronounced as yours, but there. My heart rate is a little on the slow side too, about 60, my RR was about 15 and there were the BLL, about 4 per breath cycle. They seem easier to see or more pronounced during expiration or if one happens to hit top dead center of inspiration.

Very interesting.

In case you're wondering: my AHI is typically low, < 1, nearly all CA, my machine doesn't report I:E, but the I and E average times don't show a tendency toward inversion.

Here's a snip from last night that shows some "wild" breathing similar to that you've seen in your own. I'm putting it up to demonstrate that yours are not necessarily abnormal or uncommon. Here's my interpretation: I was sleeping pretty good, nice even flow waves, but in large leak (not mouth, but mask), which, I think, finally disturbed me enough to wake or arouse me. I fixed it while taking a few irregular breaths and went back to sleep. Now this was a rather isolated event without producing a CA or OA, so I don't consider it significant. But, If I'd had a bunch of them, maybe a different story.

BTW, notice the BLL. I don't know what the RR and PR rates are in this snip, probably about the same, 15 and 60, but the pulse rate looks a little more irregular here than in the first sample I noticed. I'll have to check, out of curiosity. Probably due to the disturbance.

Pneumophile wrote:Interesting thread. When I see my sleep doc in a few weeks I'll make sure to ask him about my inspiration/expiration data, which according to some in this thread are unusual. My inspiration/expiration ratio over the last two weeks (since I bought the fancy new blower) is in the 1.5 to 2.0 range; my old machine didn't show the data.

Your treated AHI isn't too bad. Are you experiencing a lot of clusters of those events? Are they primarily CA, OA, or A mixture?

I notice you and Ron have the same, or at least similar machines, PR System One REMStar Auto CPAP with A-Flex and PR System One REMStar 60 Auto CPAP (PR 560).
Mmm...? Small sample size, but I wonder...

Good thought Jay. How well have these devices been compared - accurately and scientifically - to see if there are significant differences in detection and reporting of events? Do the manufacturers have to show FDA comparative data (equivalence)? I've spent some of my career working with FDA on drugs or potential drugs, never on devices.

Jay Aitchsee wrote:

Pneumophile wrote: Your treated AHI isn't too bad. Are you experiencing a lot of clusters of those events? Are they primarily CA, OA, or A mixture?

Primarily OA but a mixture.

Pneumophile wrote:Good thought Jay. How well have these devices been compared - accurately and scientifically - to see if there are significant differences in detection and reporting of events? Do the manufacturers have to show FDA comparative data (equivalence)? I've spent some of my career working with FDA on drugs or potential drugs, never on devices.

Yes, the manufactures have to go through a screening process for FDA approval and yes, there have been some "bench test" studies done comparing various machines. There are links to them in this forum, but I don't have them immediately available. The bottom line, I think, is that some machines are better than others at certain things, but no clear winner, though as I recall, reading them reinforced my preference for ResMed which I have only because it's the only brand I've ever used.

There's lots of high level and sometimes humorous conversations going on in this thread regarding Ballistocardographic Artifacts. It's from the early days of the S9 and these guys were able to modify software to see things not available before: I think at least two of them are Respiratory Health Professionals and some of the graphics are from Polysomnography. I've thrown in a few quotes.
viewtopic.php?f=1&t=61466&start=45#p576977

In this context, ballistocardiographic artifacts refer to "heartbeat pulsations" sitting in the flow channel as signal noise. They are also referred to as "cardiac oscillations" or "cardiogenic oscillations".

Agreed. My understanding is that those might even be a few "inspiratory hitches" a.k.a. "false starts" during inhale. Respironics refers to them as such and even filters them out in at least one of their patent descriptions.

As previously demonstrated, filtering "hides" (or sometimes "exposes") a great deal of data. Heart Rate Variability is especially interesting.

I think it is very possible, as I mentioned above somewhere, that these Artifacts confuse machine Inspiration and Expiration measurements resulting in erroneous reporting and calculation of I:E. I also think its possible the effects of the Artifacts are more pronounced in the presence of expiratory mouth breathing. Further, I think it is very possible that the manufactures of these machines don't care that these statistics may be in error in lower level (non ventilating machines) because typically, the statistics are not used in non ventilating scenarios. It is only, I believe, in the more advanced models with ventilating capabilities where these particular components of breathing are monitored with some degree of control. It's just that some of the lower level machines have the ability to measure them and through the happenstance of software, primarily Sleepyhead, we are seeing these statistics completely out of the context in which they were designed to be used.

A good example of that is my own S9 Autoset. It has the capability to measure many variables including RR, It, Et, VT, and VM, as evidenced by their presence in SleepyHead. However, the only one of these variable statistics made available to health care professionals through ResScan, ResMed's Official Software for my S9, is Minute Volume (VM). When I saw my doctor Friday, she was amazed that I had the ability to see Tidal Volume (VT).

I think we have to be careful when we start to analyze some of the variables and statistics available to us that we do so with a clear understanding of their meaning and intended use. A good example along those lines, I think, is the ability to recognize CA. How many posts have there been to this forum asking a question similar to, "OMG! I had x number of CA last night, am I going to die"? At least hundreds and probably thousands.

This starting to sound like a rant, so I'd better quit

Pneumophile wrote:.........How well have these devices been compared - accurately and scientifically - to see if there are significant differences in detection and reporting of events? Do the manufacturers have to show FDA comparative data (equivalence)? I've spent some of my career working with FDA on drugs or potential drugs, never on devices.

You guys may be interested in http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4629962/ , a relatively recent comparative review of APAP algorithms (click on Table 2), plus other interesting stuff, by manufacturer. There is also reference to filtering of the ballistocardio flavor - search on the word "cardiogenic" - but nothing quantitative.

A word of caution (my own): several key refs are to company patents. While that suggests the related technology is incorporated in the relevant product line, it doesn't guarantee it. Patent refs also suggest to me that the researchers (assuming they're reasonably competent) failed to acquire disclosures directly from the companies or from the open literature. I searched briefly for FDA approval documentation; nothing so far. I know that NDA's are confidential, at least for a while, so same is probably true for DME applications.

-Ron