resmed, respironics, AHI and facts?

[dkdc]

Seems some people think Resmed machines over report hypopneas and some people think Respironics machines under report hypopneas.

I have seen Velbor's chart, and read some discussion - and I don't know what to think. Is my 7-10 average HI on resmed machine actually ok? Or is something wrong that I should pursue with the doctor and RT? Maybe I just don't know the data and terms - but it seems I am getting pretty vague answers on possibly an important issue. (and the fault may be entirely my own lack of study, or maybe not)

It seems like someone, somewhere in the production of these machines and in government testing - would have to have compared each Respironics and Resmed model to what a sleep lab gets for data. In fact, doesn't almost every sleep lab USE one of these companies machines and couldn't they compare the results with their sophisticated measuring equipment?

There has to be some reasonably definitive answer on this question, I would think in my newbie ignorance. Maybe I am talking beyond my knowledge base. Maybe this is all big business and the resultant politics that we don't get clear info? Maybe I am overwrought and under-slept and the issue is not the problem I think it is?

By the way, my Apnea Index was 0.5 before cpap - so saying that I just should watch the AI means I might as well quit using the machine. My AI was fine before all these tubes and face markings and trying to sleep comfortably without being able to turn my head too much! The HI must be important as well - my oxygen was too low, my sleep was too poor, etc., etc. It is all there on the test which cost thousands of dollars to run.

Thanks for any help in getting enlightenment. Sorry if this sounds to heated of a question - I am a little frustrated.

Velbor's chart: viewtopic.php?p=356255#p353046

[dkdc]

Interesting post and link on the subject

viewtopic.php?p=324087#p324240

link is http://www.resmed.com/au/assets/documen ... 0906r1.pdf

[jnk]

Home machines give data estimates for you to see if your numbers are trending upward or downward. No home machine's estimates are directly comparable to the data from a sleep study, which measures things no home machine can measure.

You get a sleep study. You get a prescription for a certain pressure. You run your machine at that pressure and see what numbers it gives you. If you start feeling bad again, you try slightly higher or slightly lower pressures for a few weeks and see what effect that has on how you feel and what numbers the machine gives you to compare those numbers with the numbers you had before you changed your pressure. You use the numbers as an indication of whether you went in the right direction. If the numbers trend worse, you probably went the wrong direction. If the numbers trend better, you may have gone the right direction. If neither direction helps your numbers and how you feel, your pressure may already be optimized and there may be no reason to fiddle with it. That result may mean your problem lies elsewhere. But you don't for one minute believe that the numbers your machine gives you are comparable to a real sleep study or comparable to the numbers from any other brand of machine.

Bottom line is that all the home-machine numbers are good for, really, is judging whether pressure should go up, go down, or stay the same, when it comes to straight pressure. (The only variation on that is to use the numbers to get the range of pressures just right for you when using an auto.) They can also give a heads-up if there is suddenly an increase in the number of estimated events. But if the only choices of action are (1) raise pressure, (2) lower pressure, (3) leave pressure the same, all you need is some trending data in the way of estimates from your machine and a way to rate and record how you feel. In that context, there really is no such thing as overestimates or underestimates of events.

It is good to have a machine that gives you efficacy data. But the value and purpose of that data should not be overestimated. And that, in my opinion, is the real point when it comes to the idea of overestimation versus underestimation--seeing the true value and reason for the data home machines provide.

In your case, if AI is not useful trending information for you (in that you already have AI consistently below one), then you are ready to move to the next step and to use AHI as your meaningful trending information with your ResMed. The point with that brand, in my opinion, is that you have to ignore HI at FIRST and get AI below one FIRST, and THEN your AHI becomes more meaningful with that brand. In your case, if you never had AI above one, you were ready to use AHI as trending information from the start. But only as trending information--not as a diagnostic tool for guessing how severe your breathing problem is or was. That is what sleep studies are for.

jeff

[Slinky]

WELL SAID, jnk!!!! EXCELLENT explanation. How come I didn't think of that?? *wicked grin*

[Pugsy]

By the way, my Apnea Index was 0.5 before cpap - so saying that I just should watch the AI means I might as well quit using the machine. My AI was fine before all these tubes and face markings and trying to sleep comfortably without being able to turn my head too much! The HI must be important as well - my oxygen was too low, my sleep was too poor, etc., etc.

Hypopneas are indeed very important. If they weren't no one would bother to score them. Have enough hypopneas close together and you will get desats and symptoms just like with an apnea of longer duration.

I rarely have very many hypopneas flagged by my Respironics machine. Maybe one of these days I will get a ResMed machine to play with and see just how many it may call for me. I am just curious. I don't obsess over minute differences in numbers. I look at the overall trend and compare it with how I feel.

We seem to want or need a numbered line in the sand to validate our therapy when in fact this line in the sand is so hard to draw. It is as variable for each individual as any line in the sand during a wind storm. JNK made some very excellent observations.

[jnk]

WELL SAID, jnk!!!! EXCELLENT explanation. How come I didn't think of that?? *wicked grin*

Thanks, Slinky.

After 15 months or so of my reading your posts, maybe some of what you have taught me is STARTING to sink in.

[dkdc]

But only as trending information--not as a diagnostic tool for guessing how severe your breathing problem is or was. That is what sleep studies are for.

Hmm - so it is all kind of guesswork unless you get another sleep study. If your trends or machine data are good - no need for another study, hang in there and see if you feel better. If your trends are not good take a crack at changing something and see if it improves the trend. Or if you start to feel much better or much worse over a sustained period of time, that is another meaningful guide.

So, one should ignore the warnings that a certain machine over or underscores anything. It is all BS. Seems like a different view than one usually gets on these boards

[dkdc]

We seem to want or need a numbered line in the sand to validate our therapy when in fact this line in the sand is so hard to draw.

Maybe I should post my "trend" - it tends to go up and down. I still think the numbers are important to get, and to know what to aim for. It can take months to feel better. If the numbers aren't right one can spend 3-4 months just waiting to feel better and never get there.

I guess the point is not to hang on any one night's data - or any week's data - but maybe to put some stock in the month to month data?

[dkdc]

I am in a cranky mood today - sorry if I am making this sound adversarial !

I am still in the tired if I CPAP and tired if I don't cpap stage. Hopefully that will change soon.

[jnk]

. . . so it is all kind of guesswork unless you get another sleep study. . . .

It should not be guesswork. It should be systematic, scientific, to the extent it can be, which requires awareness of the limitations of the exactitude of the data at hand.

Sleep studies can be used to verify efficacy of treatment. But the value of a lab titration is to give you a starting point as far as what pressure to use. Home-machine efficacy data can help in the tweaking of pressure after that. Skillfully done, using the data from most any brand of machine can be useful for preventing any further titration sleep studies that might be done for the sole purpose of finding a different treatment pressure to try.

. . . If your trends or machine data are good - no need for another study, hang in there and see if you feel better. If your trends are not good take a crack at changing something and see if it improves the trend. Or if you start to feel much better or much worse over a sustained period of time, that is another meaningful guide. . . .

I agree with that in principle and in general, although sometimes there MAY be a reason for another study. From what I've read, patient self-titration can be more effective than a titration study as far as finding a good pressure, or range of pressures, that would allow the patient to feel his or her best, once the patient has a safe starting point to work from in the light of information from a lab titration. But there are still reasons to get another sleep study if PAP therapy is not making a dent in daytime sleepiness or other symptoms. My statements were made from the perspective of a patient trying to get the best PAP therapy possible without having to have repeated studies to find the optimal pressure for nightly treatment over time.

. . . So, one should ignore the warnings that a certain machine over or underscores anything. It is all BS. . . .

I would not ignore anything, myself. Those discussions have their share of bull, since anyone can participate in those discussions. (Even a newbie like me when I'm feeling brave.) But I would take those discussions in context. Technical discussions by those with detailed knowledge about the differences in algorithms and treatment approaches among machine brands are some of the most enlightening and helpful discussions that take place on this board or anywhere, in my opinion. Understanding the differences between brands and the reasons for them are important things to know. I reread those discussions over and over again, and still haven't got all the meat off the bone. I recommend to all who want to be fully informed that they dig into those discussions with fork and knife and learn all there is to know. It may take me a while, but I'm still trying, myself.

HOWEVER, I would not extend application of those discussions beyond what they are meant to be about. The fact that there are still mysteries to be worked out about the machines is fine. That doesn't make any of the machines, or discussions, worthless. But they are all limited (machines and discussions both), and we newbies can't become overly obsessed about the limitations to the point that it distracts us from how very useful that limited information is that these machines (and discussions) provide us.

. . . Seems like a different view than one usually gets on these boards . . .

My views are not always standard, and many people disagree with them.

But understand that professionals may use home machines in a different way from the way we as patients usually use our own machines in tweaking our own treatment pressures.

To a professional, the data from one machine may be of more value to that professional when it comes to using home machines for home sleep studies and home-based titrations, depending, for example, on what the expectations of a sleep-study tech is and the doc the tech works for. It is very important the professionals understand what may be an overestimation or an underestimation in comparison to a PSG. Discussions about that on some boards can be heated discussions, because some techs are apparently now being asked to take the limited information from home machines and then to try to extrapolate that into some semblance of a PSG-type diagnostic study or titration. That is not an easy thing to do (if even possible), and so the frustration is understandable. Occasionally, a patient my find himself in the position of having to do the same thing, too, if insurance nixes the need for a lab study. If so, I believe the key is to understand the differences between brands to the extent that it helps us make small treatment-adjustment decisions for ourselves (hopefully with the support of our doc and RT, whenever possible) without unnecessarily condemning any brand's choices for scoring events for trending purposes.

In my opinion, that is.

I am in a cranky mood today - sorry if I am making this sound adversarial !

I am still in the tired if I CPAP and tired if I don't cpap stage. Hopefully that will change soon.

Don't worry about sounding adversarial. That is welcome on this board. It is understood that many of us are sleep deprived and grouchy at first. You are asking good questions that get to the heart of the matter. Sorry you are having a rough time. I too hope you soon feel better. And I hope this discussion may be at least a little helpful to you. And I hope you are working with good professionals in figuring out your health.

jeff

[Pugsy]

I guess the point is not to hang on any one night's data - or any week's data - but maybe to put some stock in the month to month data?

I would think it would depend on where the "trend" seems to be going.
Example myself: I started therapy with the auto unit 7 min and 12 max. No software initially. Machine showed AHI average between 5-9 each night and leaks appearing to average within acceptable range. So not hugely bad data but more importantly I felt awful and pretty much exactly like I did pre-cpap. So for 10 days my "trend" was those numbers.

I got the software and saw that while I wasn't have a huge number of events per hour (each hour) but I was having some pretty significant groupings of events on a 90 minute to 2 hour schedule. This correlated to what I already knew that I was much worse in REM stage sleep. I would have maybe 2 hours with maybe 1 or 2 events than I would have an hour with maybe 10-15 events. So while my overall average wasn't that bad those times in REM sleep with all those events were enough to impact how I felt overall.

So I tweaked things and increased my minimum pressure to 10 cm and let the max be wide open. Thinking that with 10 cm my chances of preventing those groupings of events would be better because the machine could respond more quickly. This was successful. My AHI dropped from averaging 6-9 to 3-5. I have had a few nights where the AHI has dropped to 1 or less and I usually feel really good the next day but an AHI of 3-4 still lets me feel decent. When I have a 5 or 6 AHI I don't feel the greatest but still much better than pre-cpap.

I wouldn't have waited a month at the 6-9 AHI I saw initially but I don't think that seeing that number for one or two nights warrants immediate changes. Remember I also was feeling totally like crap.

Fast forward to now after nearly 5 months and I am slowly seeing other "trends". The REM grouping events seem to be breaking up. I don't see nearly as many dark green bands on my reports. I still seem to average 3-4 AHI per hour but I don't see as many times where the events were so close together and so frequent that they scored as dark green bands. So I think the trend is still improving for me even though the AHI pretty much remains the same. With the exception of the nighttime pee breaks, I haven't seen any overall miracles. Instead I see a gradual improvement when I can get over 7 hours of sleep. That has been hard because of other factors.

The software was priceless in terms of enabling me to see a possible cause as to why I wasn't feeling any improvement. Each person is going to have their own "line" where they will see improvement (even if gradual) and we cannot expect my line to be what someone else needs. We cannot expect Respironics numbers to compare with ResMed or any other machine either for that matter. On my sleep study I had mainly apneas and not many hypopneas, so it stands to reason my data will be similar.

If a person had a majority of hypopneas on the sleep study then they likely will also tend to score more hypopneas on any machine. It is hard to compartmentalize the ResMed data because we see and hear that "magic" less than 5 AHI number and it seems like so many ResMed users tend to report a little higher HI. Especially if they also had a majority of hypopneas during their sleep study. There are ResMed users that have AHI less than 5 and even less than 2. So where do we draw the "line"??? What number does it take to validate what we perceive as optimal therapy? If Respironics users didn't "seem" to be reporting less HI, would that make it easier? How many of the hypopneas that any machine flags are really maybe simply turning over in bed and maybe briefly holding our breath? Remember the machine is stupid in that it doesn't know if you are awake or not or you maybe hold your breath for some other reason. It reports whatever it senses based on how it has been told to report.

I still say to look at the numbers and how they factor into the big picture. If we feel crappy, look to leaks first, then look to overall sleep quality (enough hours of sleep? fragmented?) and/or then look to pressure needs. Best numbers in the world don't mean much when I only get 5 hours of sleep.

If we feel great, then wonderful, probably data off the LCD will be sufficient to validate things in our minds.
If we don't feel so great then we wonder....what if?

[jnk]

Well said, Pugsy. I really like the way you put things.

[Pugsy]

Well said, Pugsy. I really like the way you put things.

Thanks Jnk and I also like the way you put things. I think people tend to want to have the numbers be the only means of validation of their therapy and unfortunately it isn't that simple. As you point out, these machines simply do not match up to the data that is obtained during a full sleep study.

I always have to chuckle a bit when someone reports an AHI of 2.3 (or whatever low number) and wonder if they need to tweak things a bit. No mention is made of how they feel, if there were leaks, or how many hours of sleep they got. This is probably why doctors don't push data because they feel some people will obsess over a simple set of numbers.

One of these days I will get a ResMed machine and give it a try. Since I know from my sleep study that I don't normally have many hypopneas I might be a good experiment to see if ResMed really flags them a whole lot differently. My Respironics machine consistently gives me HI numbers of 1 or less.

I still say that if at all possible the software use is priceless. Especially if someone is not feeling so great despite everything else being "acceptable". But heck, if therapy was ideal and a miracle happened that patient likely wouldn't be poking around here would they? For those that are "what iffing" I say get the software, it is the only way to possibly shed some informative light on what is going on and even then it has its limitations but we do the best we can with what we have. Is it perfect? No. But it is better than nothing.

[jnk]

I just tried a Respironics BiPap Auto for a few weeks and found that it reports my AHI below 2, often near 1, at the same pressures at which my ResMed reports an AHI of 6 or 7.

I wish my ResMed software reported vibratory snore, because for me, that seemed like useful trending info.

DMEs should offer free one-month rentals of other brands based on frequent-breather hours, or something.

[Guest]

Great post about the relavance of data jnk, I cringe everytime I hear somebody chasing that percieved nirvana of 0 AHI.

I also wish manufactures would quit using the term Hypopnea, a hypopnea is SO much more than what a PAP machine measures.


AKCPAPGUY