Challenges of Diagnosing Complex/Central/Mixed Sleep Apnea

This post started out as a reply to one of Pugsy's comments (below) but it ended up exploring a topic that may result in a useful discussion, so i thought it may be worthwhile to start a new thread rather than have it buried within another one. Please feel free to contribute your experiences and opinions.

EDITED TO CLARIFY: The content in this post is intended to refer to the Diagnosis & Treatment of Central/Complex/Mixed Sleep Apnea issues and not clear-cut cases of standard OSA which are relatively more straight forward to diagnose & treat. (The title of the thread has been amended to make this clearer).

From: viewtopic.php?f=1&t=80934&start=15&sid= ... 93#p737618

Pugsy wrote:Treating of either is really fairly clear cut once the proper diagnosis is made.
The problem is trying to make the diagnosis from what we see on a home machine within the limitations that it has...
but the main thing is getting the proper diagnosis to start with.

Yes you're right, there's defnitely lots of room for improvement in the diagnosis of Sleep Apnea related disorders (as there is with most other medical conditions).

A) Sleep Studies. I'm still trying to figure out if sleep studies are completely accurate or can be wrong at times. The general consensus is that Home Studies are of limited value as they don't track sleep staging. However, the patient is in their normal sleep environment which IMO should make the test a bit more relevant. A lab test provides more data but i'm wondering if the relevancy of the data may be a bit 'distorted' as the patient is in an unfamiliar environment (and most likely uncomfortable vs being at home in your own bed) with someone literally watching you as you sleep (this must be unsettling for a lot of people). Even if this equivalent of the 'white coat syndrome' is not significant in affecting the results, i wonder if a series of the same tests done once per week over a month (or over a few consecutive days) would yield the same results. Since PSGs are costly, most people can only have one test or two at most (if that), and if the 'environment effect' does play a role, then the primary way of diagnosing may be less than optimal.

B) xPAPs. Home machines are another source of data which the more technically savvy xPAP users rely on to refine the diagnosis of their condition. Although it's useful in observing trends and responses to different settings, many experienced users in here have noted the limitations of these devices as a diagnostic tool. In fact, the usefulness of the data produced is questionable as each brand uses different algorithms and standards in detecting events and interpreting them. Imprecise science to say the least, but it's better than nothing i guess.

C) Knowledgeable Experts. Sleep medicine is a relatively new niche where patients may/often get different advice from different specialists who are limited to their specific experience on the topic and biased by their primary area of focus (ie ENT surgery, pulmonology etc). This certainly makes getting an accurate diagnosis a challenge. There's a thread (in fact there's probably quite a few) in here where one member mentioned how he/she travelled all across the US to see various Sleep medicine experts which produced inconclusive/unsatisfactory results. And that's in a country which is supposed to have the best medical practitioners. Safe to say, the diagnoses in other countries will likely be scatty as well. Regardless of where you are, it probably boils down to seeing the right doctor who has the 'right' knowledge, experience, attitude, has sleep apnea themselves and is a xPAP user too.

D) Interpreting Data. Anyone who uses software to analyze their xPAP data for a period of a few weeks or more will know that it's fairly time consuming if you're interested in ferreting into all the details to get as clear a picture as possible. If a doctor were to get an accurate idea of their sleep apnea patient's xPAP data, it would take at least 30 mins to 1 hour, assuming that they have the relevant software and are experienced in interpreting the data. However, since most of them do not even have the software/experience and tend to focus on keeping each consultation to between 20-30 mins, it's a mystery how they can possibly have the best chance of identifying the patient's unique sleep health issues.

So, how can the current options for Sleep Apnea diagnosis be improved? (Here are a couple of ideas to get things going. There must be lots of other possibilities so please share your thoughts below.)

1) xPAP Data Analysis Software/Service … Even though the data from home machines may have their inaccuracies, there's no doubt that they're useful to observe trends and to give a broader picture of how a patient is responding to therapy. Communities such as this one serve as a valuable source to help patients who would like to learn more about their CPAP data, but it's probably fair to say that we're in the minority here, as most patients do not (i) have a data capable machine (ii) are not technically savvy enough to get the data into a software on a computer and (iii) may not understand the concepts sufficiently or have the analytical abilities to interpret the data effectively.

To solve this (apart from machines with no data capabilities), it'd be great if there was a data analysis service where patients could upload/send their data and have it do the analysis which many of the more tech savvy patients try to do themselves when asking for advice in online forums. This could be done as extra features in the existing softwares but a service would be more useful to more people who don't have the ability or inclination to mess around on a computer. This could then free up patients to note/record more data about their sleep (ie was their nose very blocked on a particular night and other non-sleep related events) so that a more accurate diagnosis can be reached.

2) Home Sleep (Apnea) Monitoring System … Although home machines will never be as accurate as a lab test, it'd be great if they could come up with a home system that integrated the existing PAP machines with a (i) sophisticated oximeter* (ii) device like the ZEO to give data on sleep staging (even if it's not 100% accurate) (iii) accelerometer or something similar to detect supine/non-supine sleep positioning and (iv) anything else that might be useful in giving us a clearer picture and is feasible to incorporate.

* "one of the new pulse oximeters with not just heart rate and SpO2 percentage measured but “perfusion” (amount of blood in the finger) measured as well … Perfusion would tell, rather indirectly, if the event aroused you" … from Todzo's comment in viewtopic.php?f=1&t=79766&start=45

If we could get as much of the effectiveness of a lab test PSG into a patient's normal sleeping environment and allow for constant monitoring each night rather than only in a one-off test in a lab, we'd certainly be a lot closer to getting accurate data. This data could then be uploaded to a professional data analysis service in order to produce higher quality information to assist doctors in providing accurate diagnoses.

Most apnea is bleeding obvious from just an airflow measurement. No subtleties of O2, EEG, etc. I know several people I knew had apnea just from listening to them breathe in their sleep.

Any kind of sleep test will flag these people as 100% sure-fire apneacs.

Some people do require a more refined test with full PSG, and may fall into a gray area.

Proper treatment does need a good data capable CPAP machine and followups, but for most of us apnea and CPAP is not subtle or unclear. Fairly simple CPAP treatment will make an enormous improvement in our long term health outcome.

Some need more complicated treatment like bilevel or ASV.

I'll toss in some variation allowable under the sleep sciences scoring criteria, specifically around hypopneas. It is very possible to score someone as positive using one criteria and negative under the second. Both are allowable but lead to completely different outcomes. What is a patient to do in that case?

It just shows how evolving this science is. From what I have seen, it is very difficult to score a PSG. The zeo website posted a study that showed fairly wide variance between two techs scoring the same PSG. I still can't believe a diagnosis of a chronic disorder is based on what seems like iffy data at best, but I guess it is the best option we have. Hopefully better diagnostic tools are on the horizon.

I see several different stages where we need this discussion:
1) initial diagnosis
2) Ongoing Therapy effectiveness (pretty much solved with SPO2 meters and full data machine, toss in a zero for a pretty full package)
3) very complex cases.
For mass screaming an app is probably the best bet…. I think that the IPhone Snore application with a few tweeks could get most initial screening. You could find a way to strap it to your chest and record breathing, so you could pick up the percentage of apnea people that do not snore (if they even exist…)
Second stage would be a full data machine and SPO2 meter would handle most "screening" that is needed. (how many of us have put a son or daughter on the machine for a night to see if that funny snoring is apnea or not??) .
For sleep stage monitoring all you really need is a full data machine, and probably an SPO2 meter. You could add in a Zero. But there are other simple things like how do you feel that can pretty much tell us if we slept good last night.....
The challenge is getting the insurance company to accept initial diagnosis without a full sleep study and complex cases..... There we are still having problems.....
Even with home studies, there are problems getting”normal” nights sleep. They tried a home study, because my machine data is out of sync with the sleep titration study. I found that the home study was a machine strapped to my chest, and I could not sleep on my stomach like I normally do. I slept very lightly and the apneas were lower than "normal". (the best single night that I have had in 100 days). I am now waiting for the analysis to see what position I was sleeping in. The night after the study
I would re-design the home machine with the following to allow the least disruption to a normal night sleep:
1) a WIFI SPO2 meter to avoid any wires
2) wifi bands…. Strap them on, no wires to get tangled.
3) a wifi camera to record position or a small wifi ball cage switch to record sleep position The camera could also pick up restless leg and other movements. You could also get his with a small accelerometer that could be put on each leg.
4) A zero to record sleep depth (missing from my home study)
For people without a computer, you could include a simple hot spot type device that picks up and records the reading from all of the devices.
I would do wifi to try and keep everything synced up. If we could get the clocks synced to less than a quarter second, we could probably get by with separate devices that record that data and merge it in the morning. With USB interfaces, it would be easy to plug everyone in to get charged and sync up the clocks before the study started.
The real challenges come in scoring and Insurance Company acceptance f the data.

padster

I am glad to see that you are giving this subject so much thought and taking the time to post about it.

May I suggest that in your future thinking and posting you clearly delineate between diagnosis and treatment as they are jumbled up in what you just posted?

For instance,

Home machines are another source of data which the more technically savvy xPAP users rely on to refine the diagnosis of their condition.

It is not common (that means a few exceptions do exist) to use data from CPAP machines for diagnosis. Most CPAPers are already fully aware that they have sleep-disordered breathing (SDB) and need CPAP to treat it. The data from their machines is used to verify and optimize the treatment of their condition.

themonk wrote: From what I have seen, it is very difficult to score a PSG. The zeo website posted a study that showed fairly wide variance between two techs scoring the same PSG. I still can't believe a diagnosis of a chronic disorder is based on what seems like iffy data at best, but I guess it is the best option we have. Hopefully better diagnostic tools are on the horizon.

I don't place much value on the accurate scoring of PSGs for most patients. For most patients the importance of a diagnosis is only to know that he has SDB. This should be done utilizing a minimum of resources as resources are limited. The resources should be focused on effective treatment. Archangle has said it well,

Most apnea is bleeding obvious from just an airflow measurement. No subtleties of O2, EEG, etc. I know several people I knew had apnea just from listening to them breathe in their sleep.

Any kind of sleep test will flag these people as 100% sure-fire apneacs.

Some people do require a more refined test with full PSG, and may fall into a gray area.

Now there are certainly exceptions and these patients are important also. These are patients who have borderline cases and patients who have SDB but for a variety of reasons get a negative on their first sleep test. We don't want to overlook these patients - they need to be diagnosed and treated.

In general I disagree with the premise of this discussion if it indeed is calling for vast more amounts of resources for diagnosis. I actually propose the opposite - pull away resources from diagnosis of the obvious cases. A good sleep doctor can diagnose many cases with a short interview of the patient and his bed partner and an examination of his jaw and airway. The precious resources could then be saved and send the patient straight to the lab for a CPAP titration. During the titration the technician will be easily able to note that indeed the patient has SDB and the CPAP pressure had to be raised above the floor (4 cm) to control the condition.

The resources saved from the first titration night can be used to:
- diagnosis the borderline cases and more-difficult-to-diagnose cases and
- spend more time setting the treatment for the difficult-to-treat cases.

The first thing that went through my mind when reading the original post was there would need to be a greater need for the service suggested, i.e., reading and evaluating a home machine print out. The number of people that follow their data must be WAY in the minority. This became astonishingly clear to me when the RT at the DME I use, said I was the only one of all of his several hundred patients that track their data or had even shown an interest in doing so, and out of the 10 people I am acquainted with that are on CPAP, again, none are interested but me. I thought that when I told these 10 people about being able to monitor and possibly improve their therapy, they would be interested in learning how but that hasn't been the case at all. Amazing!!

It's alarming to read that two techs can score a sleep study differently. Makes me grateful that my neurologist/sleep doc did the heavy lifting and wrote a lovely two page report delineating his findings on top of the analysis and charts produced by the sleep lab. He then sat for a hour explaining it all to me. Not that any of it stuck in my sleep deprived little pea brain. I believe that the technology is eventually going to develop to the point where all the vital information will be transmitted electronically to a computer with analyzing software, which can be set up to video a night's sleep thus allow accurate synchronizing and evaluation of data. However, to be able to afford the R&D, it needs to be commercially credible and at this point in time, I just don't see a large enough market for it.

What we receive here on the forum is some pretty darn good advice based on experience. When a problem defies the knowledge and experience of our most learned participants, referral to a physician is always recommended. Note that advice can vary so each one of us is responsible for choosing which path to follow. Choosing the right path depends on each of us as individuals becoming as educated as possible in every facet of this strange affliction we share. Being responsible for our own health and well being - that's the hard part, and the part we will faced with carrying until someone builds a better mousetrap.

Thinking out loud, again...

Suz

archangle wrote:Most apnea is bleeding obvious from just an airflow measurement. No subtleties of O2, EEG, etc. I know several people I knew had apnea just from listening to them breathe in their sleep.

Any kind of sleep test will flag these people as 100% sure-fire apneacs.

Some people do require a more refined test with full PSG, and may fall into a gray area.

Yes, you are right. It is not very challenging to detect if someone has stopped breathing for 10 seconds or more. The original post and title should have been better written to make it clear that i was referring to people who fall into the gray area and patients with Central related Apneas and/or lots of Clear Airway events detected by CPAP machines which leads to inconclusive information regarding their condition.

archangle wrote:... but for most of us apnea and CPAP is not subtle or unclear. Fairly simple CPAP treatment will make an enormous improvement in our long term health outcome.

As the gold standard, CPAP treatment has surely helped the majority of Sleep Apnea patients to sleep better, improve their health and saved lives along the way. I may have been mistaken in thinking that there's still lots of ways to improve how it records & interprets events, and how more data may assist in giving a patient and his/her doctor a better idea of how the treatment is progressing, and an ongoing update of their condition.

ChicagoGranny wrote:

padster

May I suggest that in your future thinking and posting you clearly delineate between diagnosis and treatment as they are jumbled up in what you just posted?

For instance,

Home machines are another source of data which the more technically savvy xPAP users rely on to refine the diagnosis of their condition.

It is not common (that means a few exceptions do exist) to use data from CPAP machines for diagnosis. Most CPAPers are already fully aware that they have sleep-disordered breathing (SDB) and need CPAP to treat it. The data from their machines is used to verify and optimize the treatment of their condition.

Yes, certainly, thank you for highlighting this. Apologies for the sloppy phrasing and choice of words. To clarify, when mentioning the use of home machines to "refine the diagnosis" ... this is an example of what i mean ... a sleep study diagnoses a patient with OSA and no CSA, but once PAP is used to clear the obstructed airway, the pressure causes Central events which are noted (rather than diagnosed) by the CPAP data.

I think i may be one of those few exceptions [ie had a home sleep study indicating OSA but no Centrals, and CPAP data that showed no Obstructive events (at 4cm pressure) but lots of Clear Airway events (which may or may not be Centrals)], which may explain why i may have got the wrong idea that lots of others are also analyzing their CPAP data to try and figure out whether they've got OSA, CSA, both, and/or why their CPAP data seems to be in conflict with their Sleep Study diagnosis.

ChicagoGranny wrote:I don't place much value on the accurate scoring of PSGs for most patients. For most patients the importance of a diagnosis is only to know that he has SDB. This should be done utilizing a minimum of resources as resources are limited. The resources should be focused on effective treatment.

If the SDB is regular OSA then the extra information that's obtainable from a PSG may not be so necessary, but if it's not such a clear-cut case and Central Apneas are involved, i thought that getting as much data as possible would be useful in producing a detailed and accurate diagnosis.

ChicagoGranny wrote:In general I disagree with the premise of this discussion if it indeed is calling for vast more amounts of resources for diagnosis.

I actually propose the opposite - pull away resources from diagnosis of the obvious cases. A good sleep doctor can diagnose many cases with a short interview of the patient and his bed partner and an examination of his jaw and airway. The precious resources could then be saved and send the patient straight to the lab for a CPAP titration.

The resources saved from the first titration night can be used to:
- diagnosis the borderline cases and more-difficult-to-diagnose cases and
- spend more time setting the treatment for the difficult-to-treat cases.

I agree. There's not much point in dedicating lots more resources to improve the diagnosis of obvious cases of Obstructive Sleep Apneas which can be relatively straight forward to identify from snoring, recessed jaws etc and via the sleep studies used at the moment. It makes much more sense to improve ways to diagnose and treat the more challenging cases.

themonk wrote:I still can't believe a diagnosis of a chronic disorder is based on what seems like iffy data at best, but I guess it is the best option we have. Hopefully better diagnostic tools are on the horizon.

Perhaps the iffy data is sufficient in diagnosing patients who obviously have OSA (as ChicagoGranny noted), but for all the other cases that are not so easy, there seems to be lots of room for improvement. Was hoping to spark up a chat about some of these possibilities in this thread.

Anything called a "sleep test" really should, uh, measure SLEEP, shouldn't it, and not just breathing? That's why the lab/center tests are the gold standard and are considered by many to also be the most cost-effective way to address sleep complaints.

Your average sleep doc, as a general rule, couldn't care less about treatment-machine data. He gives a diagnosis, he prescribes a pressure, then he's pretty much done. He leaves the details to the DME RT to work out. If we want to use the data for trending to find our most comfortable and most effective pressure, we can. That doesn't mean it needs to be done by every patient or for every patient. It is an option.

Home-treatment machines are not designed to give sleep-diagnosis info. So they don't. We may pretend to ourselves they do. But they don't. The fancy report sheets may LOOK like diagnostic results, but they aren't. Don't be fooled. Home-machine data is VERY valuable for checking leak and for trending. It isn't meant to be used any other way that I am aware of. We use the reports to figure out what pressure(s) or comfort feature(s) to try if we continue to have problems, as a way to troubleshoot the therapy, just in case that makes us feel better. That doesn't make what we do clinically significant for the majority of PAP patients, I don't think, and it certainly doesn't make what we do a form of diagnosis. A high AHI as reported by a home machine may, or may not, be significant. I believe that a high residual AI should be investigated in any case in which improvement is possible, but getting the details of what the AI meant would require a lab to decipher once trending data has been used to get AHI as low as possible.

It doesn't much matter if sleep labs vary in the details of the numbers--as long as the data is good enough to point to whether PAP should be tried. Most patients with obstructive apnea also have some centrals and mixed events, so those events don't mean much beyond reinforcing the need for PAP unless they point to problem that should be treated with something beyond PAP. Yes, if AI remains high and sleep complaints continue, other things should be tried, maybe even ASV--but that is true regardless of the detailed interpretation of breathing squiggles, whether interpreted by man or machine. There is no need for the data to be more detailed or accurate in circumstances in which the increase in data detail or accuracy does not change the actual treatment decision that needs to be made. That is a primary rule of medical testing. One may want more detailed home-treatment-machine data to satisfy curiosity or to assure a patient, but that doesn't mean it is needed medically. If the machines designed to treat obstructive apnea do not solve a problem we have with centrals, then a machine designed to specifically treat central problems may be needed. But more detailed home-machine info won't necessarily help that process along.

Only one man's opinion as a patient.

If I can't sleep well and my machine shows a high AHI and I can't lower it on my own, I would want my sleep doc to know that. But I wouldn't irritate him by shoving a bunch of printouts at him from my treatment machine unless he asked to see them.

But hey, that's just me.

You know it wasn't even until the most recent advancement in technology that we even got the added central data to our software reports from our cpap/apap machines. Prior to the advancements we had reports that really only lumped everything in the obstructive apnea or hyponea basket. 3 years ago when I first started therapy only the high dollar BiPap ASV machines would flag a central. We had no idea if we were having centrals or not...real or not.

So in the last couple of years we get machines that flag centrals or clear airway events based on air flow only...it isn't perfect but it is better than what we had 3 years ago. I think that technology is continuing to evolve and there is a very good chance that things are being worked on right now that will enable the machines to do a better more refined job at delineating the type of events. We also have to remember that home machines right now are meant for treatment and not diagnosis. We tend to try to use the machine to establish a diagnosis and to some extent we can but we can't go beyond the abilities of the machine to see stuff it doesn't have the ability to evaluate. Until they figure out some way to evaluate the additional criteria needed to adequately establish if a central is "real" or not or if it warrants concern we are just stuck with what we have now. The sleep lab study is still thought of as the best way to really see what is going on and a means to accurately score events to establish a correct diagnosis. It isn't perfect either. I have found that there is really very little in the field of medicine that would be considered "perfect" in all situations. There is always room for improvement and in general improvements are continuing to happen...just sometimes they aren't happening fast enough to suit us.

So I wouldn't be surprised to learn that things are being worked on right now that would help you with your questions.
Sleep apnea diagnosis and treatment has grown by leaps and bounds considering how long the treatment has been out there.
The last few years...growing even faster as they refine the machines and what they can do and what they can report.
Makes me wonder what the machines available 3 years from now will show us. Maybe they will be of more use in the diagnostic realm and not limited so much to treatment realm.

Pugsy, in addition to helping determine if there are central apneas, the research in support of the ASV community resulted in technologies such as Resmed's Easy Breathe technology. We now (only a few years later) take these things for granted. As you note, I wonder where things will be in a few years.

JohnBFisher wrote: I wonder where things will be in a few years.

You know when I look back on things just in the last 3 years since I have been on the machine....I am totally amazed at how things have progressed. You have seen even bigger changes and all have been for the better.
I like to compare it to computers...remember when adding 4 MB RAM required taking out a small loan? Now we have cell phones that have more storage and more memory than a whole generation of computers. Used to it took a couple of years to be out dated...now...just a few months. I remember when having a computer with 1 GB of storage was a luxury. Now we have little SD cards that will hold that much and a lot more. Even my cell phone has 8 GB. Technology is advancing at a run away pace and it gets faster instead of slower.

The changes we want with our cpap/apap/bilevel machines will come. It may not be as quickly as we want but it will come.
Science, medicine and human nature will dictate the need along with the market and who knows what we will see in a few years. Heck, the official gold standard of the sleep lab for diagnosis may fall into the dinosaur category. Maybe they will fine tune some sort of technology like the Zeo and be able to incorporate it into cpap data.
I am fairly certain that what we think we might want...has already occurred to the powers that build this machines and someone somewhere is working on it because they know that people will buy it. If there is a way to make more money...companies will build it.

The way things work now is that the big money is in the big business of diagnosing and prescribing. Since the manufacturers have to cater to that cartel in order to stay in the market and keep the consumer market from, uh, "influencing" the price of the machines too low, I doubt anything will be manufactured on purpose that could be seen as attempting to take the diagnostic process out of the hands of the pros any time soon.

Until the treatment machines are sold off the shelf, without an Rx, at the local drug store, the interests of the medical businesses will supersede the needs and desires of the patients. When the machines are on the shelf at Walmart next to the room vaporizers, then, and only then, will the desires of the patients trickle up to manufacturers unadulterated. Then market forces will make it in the best interests of manufacturers to sell directly to customers. The way things work now is that we aren't really the customers--the DMEs are the customers. Most PAP patients don't even know what brand of machine they own, let alone what hidden features it has.

My guess is that if it becomes too popular for patients to access the data from the machines, the DMEs will attempt to force the manufacturers to look for ways to keep us from getting to it. In fact, I believe efforts are likely underway to make the real customers, the DMEs, happy in that exact way.

If dangerous drugs can be sold OTC with no Rx, so let it be so for PAP machines. In no time the price would be down to $150 a machine for full-data APAP. Call them "anti-snore" machines if that helps. Companies would then make MUCH more money with volume than with inflating prices as is the case now.

One of my many off-the-wall opinions.

jnk wrote: I doubt anything will be manufactured on purpose that could be seen as attempting to take the diagnostic process out of the hands of the pros any time soon.

Ahhh, but what if they do make something that makes the diagnostic part of things easier and better and more accurate for use at the sleep lab setting? I think it could filter down to home machines. The pros will want to use the best so they can justify their fees and have less work in earning them and maybe eliminate the human body sitting at the controls all night. Would save on salaries.