• Site Supporter

CPAP Data - what does it mean

This area is for CPAP Mask and CPAP Machine Related Questions used in the treatment of Sleep Apnea.


CPAP Data - what does it mean

Postby rlirwin » Fri Nov 19, 2010 6:33 pm

I have a Phillips Remstar Auto with a humidifier. My sleep test showed my AHI to be about 11 for the whole night and 19 during REM sleep. I was able to purchase the software so that I could see the data, but I'm not really sure what it means. The AHI on the reports typicaly runs between 1.5 and 2.5 - once in a while a little higher. It also gives me numbers for open airway apnea, closed airway apnea, hypopnea, 90% pressure and more. My doctor prescribed the machine because I was complaining about fatigue, and a have afib. Which of these numbers are important, and how do I tell if the machine is doing what it's supposed to do? My fatigue is a little better, but the improvement is not as much as I'd hoped.
rlirwin
 
Posts: 6
Joined: Tue Sep 07, 2010 1:57 pm

Postby Mrs Rip Van Winkle » Sat Nov 20, 2010 1:14 am

Main thing is to keep the AHI ...apnea/hypopnea index...under 5. the lower the better. Your typical 1.5 to 2.5 is good.
I'm only a sufferer, not a medical pro. What I post are my thoughts as a sufferer, not that of the ASAA. As a moderator on these forums I oversee the posting rules. This is the internet, always discuss what you read with your medical team.
User avatar
Mrs Rip Van Winkle
 
Posts: 4246
Joined: Thu Jun 08, 2006 11:54 am
Location: Nature Coast, Florida
Machine: M Series Auto BiPAP
Mask: Masks Vary
Humidifier: Yes

Postby CrohnieToo » Sat Nov 20, 2010 12:19 pm

How long have you been using your PR S1 Auto?? It takes many of us a while to pay off the sleep debt we've built up before CPAP therapy.

Keep an eye on your Leak rate as well as your AHI. If your Leak rate is excessive most of the night you can NOT rely on the AHI data to be correct.

Mrs Rip is right. We like to see the AI under 1.0 and the AHI under 5.0.

Did you ask for a copy of not only the doctor's dictated results from your sleep evaluation and CPAP titration studies but also the full scored data summary reports w/condensed graphs from both studies? You may well have another sleep disorder in addition to OSA that had the possibility of effective PAP therapy also reducing or eliminating it. If CPAP therapy wasn't able to reduce it then you may need additional assistance such as medication, etc. One instance that comes quickly to mind is leg movements: PLMs or RLS.

Did you ask for a copy of your equipment order (script)?? There is a possibility that your DME provider didn't set your CPAP to the scripted pressure by mistake. Its not likely but it has been known to happen. AND you should have a copy anyway to keep in your personal files and take a copy w/you when you travel just in case your CPAP or accessories are lost, stolen or broken.
ResScan 3.10 - Resmed S8 ResLink & oximeter
ConTec CMS-50D+ oximeter - Philips EverFlo 5L Oxygen Concentrator
PR SystemOne BPAP Auto w/Bi-Flex & Humidifier - EncorePro 2.2.14.0
User avatar
CrohnieToo
 
Posts: 7943
Joined: Mon Mar 20, 2006 9:18 pm
Location: Mid-Michigan
Machine: Resmed S8 VPAP Auto
Mask: Resmed Quattro FX Small
Humidifier: Resmed H3i
Year Diagnosed: 2006

Postby poliziano » Mon Nov 22, 2010 8:15 am

I have a related question on this topic.
How am I supposed to interpret the decimal part? If 5.0 AHI means 5 AHI occurrences during a specific time frame, what does a 0.5 (or whatever number preceded by 0) mean? Are those fractions - partial events?
I know I sound thick. Can anyone please explain?
Thank you.
poliziano
 
Posts: 3
Joined: Mon Nov 22, 2010 8:08 am

Postby CrohnieToo » Mon Nov 22, 2010 8:34 am

0.5 means 1 apnea/hypopnea per every 2 hours of sleep whereas 5.0 means 5 apneas/hypopneas per hour of sleep as you surmised. Nope, the 0.? does NOT mean "partial" events.
ResScan 3.10 - Resmed S8 ResLink & oximeter
ConTec CMS-50D+ oximeter - Philips EverFlo 5L Oxygen Concentrator
PR SystemOne BPAP Auto w/Bi-Flex & Humidifier - EncorePro 2.2.14.0
User avatar
CrohnieToo
 
Posts: 7943
Joined: Mon Mar 20, 2006 9:18 pm
Location: Mid-Michigan
Machine: Resmed S8 VPAP Auto
Mask: Resmed Quattro FX Small
Humidifier: Resmed H3i
Year Diagnosed: 2006

CPAP data interpretation

Postby am0665 » Sat Dec 04, 2010 5:24 pm

Just a few comments from a newbie, also very interested in the data.
I suspect that your AHI statistics, if like mine, are not on the same scale for lab results and home results: the lab may use only A&Hs during certain stages of sleep (maybe only half the night), which gives a higher score than home PAP, which knows only the total recording time.
I second CrohnieToo's question to you: it is my impression that home PAP data is meant to show a trend on certain useful measurable variables; the final word may still come from another lab sleep study... at least for me, rated as a "severe OSA" case.
Moderate / Severe OSA (2 tests offCPAP).
CPAP ResMed S9 AutoSet EPR, Hi5 Humidifier.
ResMed Masks tried: nasal pillows Swift FX, nasal mask Mirage Activa LT, FF mask Mirage Quattro or FX.
Home CPAP data analysis: ResScan 3.16 sw, oximeter, pulsemeter
User avatar
am0665
 
Posts: 93
Joined: Sat Nov 27, 2010 3:20 pm
Machine: ResMed S9 Auto
Mask: ResMed nasal mask Mirage Activa LT
Humidifier: ResMed Hi5
Year Diagnosed: 2010

Postby robysue » Mon Dec 06, 2010 2:54 pm

I'll start with some really important key definitions so you know what these numbers actually mean.

The AHI number stands for Apnea/Hypopnea Index and it is computed by the machine or the software by dividing the total number of recorded apneas and hypopneas by the total time the machine ran for the given data period. The AHI represents the average number of events the System One detected per hour of run time. If you're looking at overnight data in Encore, that's the number of apneas and hypopneas divided by the time the machine ran over night. If you're looking at seven day data, it's the number of apneas and hypopneas in the last seven days divided by the total number of hours the machine was running in the seven day period. If you're looking at thirty day data, it's the number of of apneas and hypopneas in the last 30 days divided by the total number of hours the machine was running in the 30 day period.

The closed airway apnea index is computed by the number of obstructive (closed airway) apneas that were recorded during the time period divided by the total number of hours the machine was running in that time frame. So the closed airway apnea index represents the average number of closed airway (obsructive) apneas the System One detected per hour. I use a ResMed instead of a System One, so I don't know the abbreviation that PR uses for the closed airway apnea index.

The open airway apnea index is computed by the number of central (open airway) apneas that were recorded during the time period divided by the total number of hours the machine was running in that time frame. So the closed airway apnea index represents the average number of closed airway (obsructive) apneas the System One detected per hour. I use a ResMed instead of a System One, so I don't know the abbreviation that PR uses for the open airway apnea index.

The HI number is the Hypopnea Index, and the HI is computed by the number of hypopneas that were recorded during the time period divided by the total number of hours the machine was running in that time frame. So the HI number represents the average number of hypopneas the System One detected per hour.

You should double check your user manual to find out whether OAI stands for OBSTRUCTIVE APNEA INDEX (i.e. the closed airway index) or whether OAI stands for OPEN AIRWAY INDEX. I point this out because the usual language on these boards and in the stuff about sleep apnea you find on the web and in the professional literature, most people use OAI for the OBSTRUCTIVE APNEA INDEX, not the OPEN AIRWAY INDEX.

Likewise, you should double check your user manual to find out whether CAI stands for CENTRAL APNEA INDEX (i.e. the open airway index) or whether CAI stands for CLOSED AIRWAY APNEA INDEX. Again I point this out because the usual language on these boards, in other stuff on the web, and in the professional literature, most people use CAI to stand for the CENTRAL APNEA INDEX and not the CLOSED AIRWAY APNEA INDEX.

It may make it clearer to look at some examples. Since the PR System One's LCD screen only reports 7 and 30 day averages, I'll look at what the 7 day averages mean.

First, lets suppose that you slept with the System running for about 7.5 hours per night for seven nights. That means that the machine was run about 7 x 7.5 = 52.5 hours. And suppose we're looking at the 7 day average data in Encore Viewer and on the LCD for the numbers it shows. Now lets suppose your Leak Rate data looks like this:

Median = .10 L/sec
90% = .25 L/sec
Max = 1.5 L/sec

And lets suppose your pressure data looks like this:

Median = 7.5
90% = 9.8
Max = 10.2

And finally lets suppose that your event data looks like this:

AHI = 0.9
Open airway AI = 0.1
Closed airway AI = 0.5
HI =0.3

First we can compute the number of events of each type you actually had during the entire week long period without going into Encore Viewer:

(0.9 events per hour)x(52.5 hours) = 47.25, which equals about 47 or 48 events that for the week.
(0.5 closed airway apneas per hour)x(52.5 hours) = 26.25, which equals about 26 or 27 closed airway (obstructive) apneas for the week.
(0.1 open airway apneas per hour)x(52.5 hours) = 5.25, which equals about 5 or 6 open airway (central) apneas that week.
(0.3 hypopneas per hour)x(52.5 hours) = 15.75, which equals about 15 or 16 hypopneas that week


You could divide each of these event numbers by 7 to get an "average number of events per night" if you wanted to, but at least on my S9, I don't tend to have the same number of events each night: Some nights I have as few as 0--5 events all night with an overnight AHI = 0.0 to 1.0 and other nights I have as many as 10--20 events during the night with an AHI of 2.0 to 3.0 (I tend to sleep between 5 and 7 hours each night.)

Now let's look at what the rest of those numbers actually mean.

Let's look at leak data first. On the PR you have to keep in mind that the leak rate data includes both the INTENTIONAL exhaust flow for the mask at the given pressure AND the unintentional leaks. You have to look at the owners manual for your mask to find a chart that indicates what the INTENTIONAL exhaust leak rate is at the pressure you are using and SUBTRACT that number off your leak data to find out whether and how bad your mask is leaking. Note that the INTENTIONAL exhaust flow is designed to prevent you from rebreathing CO2.

The median leak rate = .10 L/sec means that for 50% of the time the System One was on during the week, your leak rate was LESS THAN or EQUAL TO .10 L/sec. And so for 50% of time the machine was running, your leak rate was also GREATER THAN or EQUAL TO .10 L/sec. So in the 52.5 hours you slept with the machine during the week, for 26.25 hours you had a leak rate of LESS THAN or EQUAL TO .10 L/sec.

The 90% leak rate = .25 L/sec means that for 90% of the time the System One was on during the week, your leak rate was LESS THAN or EQUAL TO .25 L/sec. And so for 10% of that time, your leak rate was GREATER THAN or EQUAL TO .25 L/sec. Now it's important to realize that 10% of one hour is equal to six minutes. Since you slept for 52.5 hours that week, this means your leak rate was GREATER THAN or EQUAL TO .25 L/sec for a grand total of (6 minutes)*(52.5 hours)=315 minutes during the entire week, which is about 5 1/2 hours over the course of the week that your leak was GREATER THAN or EQUAL TO .25 L/sec . Without more detailed information about how often you're leaking at smaller rates it's hard to tell when the leak data is big enough to really be a problem.

The max leak rate=1.5L/sec indicates that at some point when the System One was sampling leak data, it detected a leak that was as large as 1.5L/sec. It may have been a very short lived thing---when you broke the seal to scratch your nose for example. Or it could be that you've had several big leaks during the week where, the leak pretty quickly went from in the neighborhood of .25 L/sec to 1.5 L/sec and stayed there. But you know for sure that you could not have leaked at a rate of 1.5L/sec for more than about 315 minutes during the entire week because of the 90% leak rate number and the time the machine was on.

The pressure numbers are similar:

The median pressure = 7.5 means that for 50% of the time the System One was running during the week, it was at a pressure that was LESS THAN or EQUAL TO 7.5 cm. And for 50% of the time, the System One was running at a pressure that was GREATER THAN or EQUAL TO 7.5 cm.

The 90% pressure = 9.8 means that for 90% percent of the time the System One was running during the week, it was running at a pressure that was LESS THAN or EQUAL TO 9.8 cm. And so for 10% of the time that week, the System One was running at a pressure that was GREATER THAN or EQUAL to 9.8cm. Again, it's useful to realize this means that the pressure was GREATER THAN or EQUAL TO 9.8 for about 315 minutes over the course of the full week.

The max pressure = 10.2 means that at some point(s) during the week, the System One reached a pressure of 10.2cm. You know (for sure) from the 90% data that the System One was running at a pressure of 10.2 for less than about 315 minutes out of that week.

It's a real shame that the System One does not report overnight numbers on the LCD. In my opinion the overnight numbers are easier to make sense of since the time frame is both shorter and represents "one night" instead of many nights.

But hopefully this helps some of the newbies who are trying to make sense of the data.
current settings Min EPAP = 4, Max IPAP = 8 and Rise time = 3

8/1/2010 sleep study results:
AHI = 3.9 [AHI = (#OA +#CA + #H w/desat) per hour]
RDI = 23.4 [RDI = (#OA +CA + #H w/desat + #H w/arousal) per hour]
Dx: Moderate OSA
User avatar
robysue
 
Posts: 1330
Joined: Tue Aug 24, 2010 6:47 pm
Location: Buffalo, NY
Machine: PR System One BiPAP Auto
Mask: Swift FX for Her
Humidifier: System One Heated Humidif
Year Diagnosed: 2010

Re: CPAP data interpretation

Postby robysue » Mon Dec 06, 2010 3:51 pm

am0665 wrote:Just a few comments from a newbie, also very interested in the data.
I suspect that your AHI statistics, if like mine, are not on the same scale for lab results and home results: the lab may use only A&Hs during certain stages of sleep (maybe only half the night), which gives a higher score than home PAP, which knows only the total recording time.

The sleep lab most likely computed an overall AHI for the time you were actually ASLEEP without the CPAP on (if it was a split night study) as well as separate breakdowns for AHI in REM vs AHI in NREM sleep and AHI in the supine position (on your back) and AHI in non-supine positions (not on your back) because for many people, their AHI's tend to be higher (sometimes substantally higher) in REM and in the supine position. They may also have computed a separate RDI (Respiratory Disturbance Index) that included not only the apneas and hypopneas, but also other respiratory events that led to short micro arousals (and disturbed sleep structure), as well as RDI's for REM vs. NREM and supine vs. non-supine.

Another thing about the apneas and hypopneas scored in the sleep study is that they were scored based on multiple criteria coming from multiple sensors. There's a reason why they have all the wires on you during the sleep study. Some of the things that go into scoring apneas and hypopneas in the sleep lab are

    breathing effort (measured by the belts around your chest and abdomen) vs. flow of air into your nose (measured by sensors on/in your nose and mouth area). These measurements both spot the main criteria of apnea/hypopnea (not breathing) and determine whether the event is an obstructive (closed airway) event or a central (open airway one)

    O2 saturations measured by the O2 sensor on your finger---the usual definition of hypopnea requires a "desaturation" event---a drop in the O2 reading as well as a reduction of breathing.

    arousals and awakenings related to respriatory events---the arousals and awakenings are measured through changes in your brain wave patterns. Respiratory Event Related Arousals (RERAs) are sometimes added to a so-called RDI and the diagnosis of sleep apnea can be based on an elevated RDI.

    REM vs. nonREM events are determined by detecting rapid eye movement through both the EEG and the sensors by your eyes.

And remember your AHI or RDI during the diagnostic part of the sleep study NEEDS to be elevated before they will diagnose sleep apnea in any form. The intention is that by putting you on xPAP of some sort, your AHI numbers will be reduced to something below 5.0. Ideally, the AHI under treatment should be pretty close to 0 on most nights, but effective treatment is usually defined as AHI is consistently less than 5.0. The titration study (second part of a split night study) is designed demonstrate that CPAP therapy can reduce your AHI numbers to less than 5.0---ideally in REM sleep in the supine position since that's when most people have their most numerous apneas/hypopneas.

Now, home xPAP machines cannot possibly determine apneas and hypopneas with as much precision as a sleep lab simply because the xPAP machine can only measure estimated air flow into your nose by looking at how the pressure numbers it can determine are changing. The companies that make data capable machines have spent much effort in designing quality algorithms to make sure that their machines "work" in the sense of correctly identifying events that would likely be scored as an apnea or hypopnea in a sleep lab based on just the flow data. Does it work all the time? Certainly not. And different companies have different algorithms. So it can be meaningless to compare Person A's AHI on a ResMed S9 to Person B's AHI on a PR System One. And of course, if RERAs were a major part of your events during your sleep study, you have to remember that there's no way an xPAP can determine whether you've had a three-second arousal based on something that is respiratory related, but does not clearly meet the machine software's pre-programmed in definition of hypopnea.

So it's really not appropriate to directly compare xPAP computed AHI to sleep lab computed AHI. That's like comparing apples and oranges. However, the xPAP related AHI's do give some indication of whether most of your apneas and hypopneas are being prevented (machine AHI consistently lower than 5.0 and you feel good) or not. And that's why the companies as well as folks on these boards tend to say that home xPAP data is best looked at as trend data: In the early going, are your home AHI's going down and staying down? And in the long run, are your AHI's reasonably stable and consistently low enough for you to feel better?

So since the diagnostic study's AHI numbers represent your UNTREATED apnea and your home xPAP's AHI numbers represent an approximation of your TREATED apnea, and since the home xPAP's ability to detect apneas and hypopneas is limited to one and only one piece of data, it's extremely reasonable to assume that your home xPAP machine's AHI should be a whole lot lower than the lab's numbers on the diagnostic sleep study that led to your being diagnosed with sleep apnea. So if your home xPAP AHI numbers are as bad as your diagnostic sleep study AHI/RDI numbers were, then there's some solid evidence that the xPAP is not (yet) effectively treating your apnea and that's a good reason to call the doctor's office back. It can take time to adjust of course, and so in the early going, some people do have home xPAP AHI numbers that are pretty high, but if treatment is really working, those numbers should start to drop and trend toward being below 5.0 in a reasonable amount of time.

I second CrohnieToo's question to you: it is my impression that home PAP data is meant to show a trend on certain useful measurable variables; the final word may still come from another lab sleep study... at least for me, rated as a "severe OSA" case.


It's really rare for someone to have a third sleep study done to confirm whether their current CPAP/APAP therapy is working or whether the home xPAP data is correct. The "proof" that CPAP is an appropriate therapy is done during the titration study when you're put on CPAP and the sleep tech adjusts the pressure (usually up) in response to your events that night. The lab's AHI and RDI numbers for the titration study will likely include an overall figure as well as sub figures for each pressure level used during the night. And for each pressure level where you had REM sleep, there will be both a REM AHI/RDI and a NREM AHI/RDI. Some of these specialized AHI/RDI's may be computed on as little as 10 minutes of sleep, but they still represent (# of events)/(time in hours).
current settings Min EPAP = 4, Max IPAP = 8 and Rise time = 3

8/1/2010 sleep study results:
AHI = 3.9 [AHI = (#OA +#CA + #H w/desat) per hour]
RDI = 23.4 [RDI = (#OA +CA + #H w/desat + #H w/arousal) per hour]
Dx: Moderate OSA
User avatar
robysue
 
Posts: 1330
Joined: Tue Aug 24, 2010 6:47 pm
Location: Buffalo, NY
Machine: PR System One BiPAP Auto
Mask: Swift FX for Her
Humidifier: System One Heated Humidif
Year Diagnosed: 2010

portable at home titrating study?

Postby truckerdad57 » Tue Dec 07, 2010 7:51 am

What about the concept of using a portable Type III ambulatory study done in the patient's home as a confirming study of APAP or titration?

The better at home sleep studies can be used to confirm APAP effectiveness.

It would remove the variable of changes in sleep due to the strange lab environment. Also you can get 2-3 nights of sleep data on a portable study for $ 3-400 which includes DR interpretation of the portable study data.

No one has a sleep lab study done on CPAP or APAP to confirm effectiveness or compate APAP AHI data to lab data due to the costs of lab studies.... You will see re-titrations after 3-5 years of PAP use or after major changes in weight or other medical history changes....

Just a thought....
(just a truck driver with sleep apnea)
Co-coordinator Truckers for a Cause Chapter of A. W. A. K. E.
awake.truckersforacause.com
Do not substitute information from here for professional medical advise.
truckerdad57
 
Posts: 1074
Joined: Thu May 22, 2008 12:01 am
Location: Batavia, IL
Machine: CPAP tester, you name it I've got 1
Mask: tend to like nasal pillow
Humidifier: YES - must have
Year Diagnosed: 2002

Re: CPAP data interpretation

Postby am0665 » Fri Dec 10, 2010 1:21 am

robysue wrote:It's really rare for someone to have a third sleep study done to confirm whether their current CPAP/APAP therapy is working or whether the home xPAP data is correct. The "proof" that CPAP is an appropriate therapy is done during the titration study when you're put on CPAP and the sleep tech adjusts the pressure (usually up) in response to your events that night. The lab's AHI and RDI numbers for the titration study will likely include an overall figure as well as sub figures for each pressure level used during the night.


I hope robysue's detailed reply didn't overwhelm, I sure appreciated it.
One significant aspect that wasn't mentioned: as much as I wanted to be cooperative, I found both studies considerably sleep disturbing, hence the duration of qualifying sleep was maybe only 60% of total recording time. This is just one aspect that creates a large difference to AHIs on home CPAP.
Being around 2 weeks into my home CPAP, it was my understanding that a 3rd sleep study was in the plan. Especially if the titration study with various masks doesn't succeed in creating enough sleep for useful statistics and decisions.
Then there is the issue of a patient moving to another area. I am told the new team will not accept just the detailed paper report, but insists -- the insurance provider willing -- on having their own sleep study ... maybe a split night with and without CPAP.
Finally an observation on my own home CPAP data so far (ResScan 3.11 on ResMed S9): the A+H counter clicks away for many breathing "events" which the PSG lab study may not qualify as A or H: like breathing muscle relaxation just before going to sleep, or obstructed sleep disturbances due to nose congestion.
Moderate / Severe OSA (2 tests offCPAP).
CPAP ResMed S9 AutoSet EPR, Hi5 Humidifier.
ResMed Masks tried: nasal pillows Swift FX, nasal mask Mirage Activa LT, FF mask Mirage Quattro or FX.
Home CPAP data analysis: ResScan 3.16 sw, oximeter, pulsemeter
User avatar
am0665
 
Posts: 93
Joined: Sat Nov 27, 2010 3:20 pm
Machine: ResMed S9 Auto
Mask: ResMed nasal mask Mirage Activa LT
Humidifier: ResMed Hi5
Year Diagnosed: 2010

Postby HRR » Wed Dec 15, 2010 2:23 pm

I too appreciate the info, and would also appreciate a little more. Everything I read says the goal, or a good result, or whatever, is an AHI below 5.0. What about the other data. Robysue, maybe you could give us some idea of what are considered good, desirable, etc., levels for the other items of data.
HRR
 
Posts: 1
Joined: Wed Dec 15, 2010 2:00 pm

Re: CPAP data interpretation

Postby Bill Bolton » Wed Dec 15, 2010 5:49 pm

robysue wrote:Ideally, the AHI under treatment should be pretty close to 0 on most nights, but effective treatment is usually defined as AHI is consistently less than 5.0.

No. Ideally it only needs to be under 5.0, which is the level which is defined as normal sleep.

There is nothing particularly special about any specific AHI score below 5.0.

Cheers,

Bill
Bill Bolton
 
Posts: 252
Joined: Tue Aug 21, 2007 12:21 am
Location: World Traveller

Postby robysue » Sun Dec 19, 2010 11:57 am

HRR wrote:I too appreciate the info, and would also appreciate a little more. Everything I read says the goal, or a good result, or whatever, is an AHI below 5.0. What about the other data. Robysue, maybe you could give us some idea of what are considered good, desirable, etc., levels for the other items of data.


Pressure data is dependent on the patient and there's no real correlation between how much pressure is required to keep a particular patient's apnea under control and the severity of the apnea during the diagnostic sleep study from what I've read. [Apparently pressure needs depend on the anatomy in one's neck.]

Leak data, on the other hand, will give you some insight into whether you've got problems with mouth breathing or mask fit problems.

On the PR System One, the leak data includes both the mask's intentional leak rate and all other unintentional leaking. The intentional leak rate is the leak rate that's built into the mask by the manufacterer to insure that you don't rebreathe too much CO2. The intentional leak rate varies by pressure setting. You need to find your mask's user guide and read it carefully. Somewhere in the user guide there should be a chart or a graph showing the intentional leak rate or exhaust flow as a function of pressure setting. Use this chart to figure out what the expected, intentional leak rate for YOUR mask at YOUR pressure setting is. If you leak rate graph in Encore Viewer (or your 90% Leak Rate on the LCD) is pretty close to the expected leak rate for YOUR mask at YOUR pressure setting, then that means you have an UNINTENTIONAL leak rake of close to zero, which is very, very good! As for how high your UNINTENTIONAL leak rate can be before it starts to compromise therapy, I really don't know for sure. My guess is that for most people, though, the unintentional leak rate will start to interfere with comfort (and hence need to be dealt with) before it starts to interfere with therapy (and definitely needs to be addressed). If my leak rate were frequently approaching that Green line in Encore Viewer, I'd be worried about leaking and trying to figure out what was triggering the leaks.

Editing to add a piece of semi-relevant information: On the ResMed machines, therapy is reported as potentially compromised when the UNINTENTIONAL leak rate figure is:

Leaking at a rate of 24 L/min or more for 25% of the night

I think that in practice, most people find leaks very uncomfortable and have to addressed long this level of leaking is reached.
current settings Min EPAP = 4, Max IPAP = 8 and Rise time = 3

8/1/2010 sleep study results:
AHI = 3.9 [AHI = (#OA +#CA + #H w/desat) per hour]
RDI = 23.4 [RDI = (#OA +CA + #H w/desat + #H w/arousal) per hour]
Dx: Moderate OSA
User avatar
robysue
 
Posts: 1330
Joined: Tue Aug 24, 2010 6:47 pm
Location: Buffalo, NY
Machine: PR System One BiPAP Auto
Mask: Swift FX for Her
Humidifier: System One Heated Humidif
Year Diagnosed: 2010

Postby am0665 » Sun Dec 19, 2010 6:16 pm

robysue, would you have some additional information on the comments you made above?

robysue wrote: Pressure data is dependent on the patient and there's no real correlation between how much pressure is required to keep a particular patient's apnea under control and the severity of the apnea during the diagnostic sleep study from what I've read. [Apparently pressure needs depend on the anatomy in one's neck.
... Pressure data is dependent on the patient and there's no real correlation between how much pressure is required to keep a particular patient's apnea under control and the severity of the apnea during the diagnostic sleep study from what I've read. [Apparently pressure needs depend on the anatomy in one's neck.]

Would you have any references to pressure calibration strategies during home CPAP? ... toward a semi-knowledgeable OSA sufferer partnering with the physician and DME toward continuing his / her own home evaluation of how to achieve optimal results by changing various parameter, such as pressure.
In my case, having undergone a basic PSG study, then a CPAP titration study (during which the discomfort and the mask changes allowed only 3 hrs of sleep), I was provided just a one page summary of "scoring" for each, in which the results looked conflicting. For all I know, the different conditions, different sleep techs, different scorers may result in significant differences in results. The resulting prescription by the physician for auto-CPAP contained a pressure range between 8 (for side sleep) and 14 (for supine sleep); which was later changed to constant 9, which the physician called "somewhat undertitrated". [/quote]

robysue wrote:The sleep lab most likely computed an overall AHI for the time you were actually ASLEEP without the CPAP on (if it was a split night study) as well as separate breakdowns for AHI in REM vs AHI in NREM sleep and AHI in the supine position (on your back) and AHI in non-supine positions (not on your back) because for many people, their AHI's tend to be higher (sometimes substantally higher) in REM and in the supine position. They may also have computed a separate RDI (Respiratory Disturbance Index) that included not only the apneas and hypopneas, but also other respiratory events that led to short micro arousals (and disturbed sleep structure), as well as RDI's for REM vs. NREM and supine vs. non-supine.
Another thing about the apneas and hypopneas scored in the sleep study is that they were scored based on multiple criteria coming from multiple sensors. There's a reason why they have all the wires on you during the sleep study...

If I may paraphrase with an example: say the lab study shows that you were ASLEEP 50% of the time deep enough to develop any apneas, and say you got an AHI score of 40 (severe OSA); then say your home CPAP device gives an AHI of 20; note that this smaller score is measured on 100% of recording time ... no matter whether you were awake or in some sleep stage. So if one does the math, that's 40 A&H events on the average per SLEEP hour in both cases!
My short PSG results do mention % of night in each stage, but don't disclose the subdivision of A/H/C events by sleep stage. I have requested the detailed results one month ago, still waiting to see them. I'd also like to see not just an average for the night, but also the distribution: 95 percentile, max hourly, and the local detail corresponding to the special events.

robysue wrote:Leak data, on the other hand, will give you some insight into whether you've got problems with mouth breathing or mask fit problems.
On the PR System One, the leak data includes both the mask's intentional leak rate and all other unintentional leaking. The intentional leak rate is the leak rate that's built into the mask by the manufacturer to insure that you don't rebreathe too much CO2... On the ResMed machines, therapy is reported as potentially compromised when the UNINTENTIONAL leak rate figure is:
Leaking at a rate of 24 L/min or more for 25% of the night [/i] I think that in practice, most people find leaks very uncomfortable and have to addressed long this level of leaking is reached.

Conversely, it the leak rate is minimal, then one can go on to other sources of trouble on sleep discomfort or high AHI.
Moderate / Severe OSA (2 tests offCPAP).
CPAP ResMed S9 AutoSet EPR, Hi5 Humidifier.
ResMed Masks tried: nasal pillows Swift FX, nasal mask Mirage Activa LT, FF mask Mirage Quattro or FX.
Home CPAP data analysis: ResScan 3.16 sw, oximeter, pulsemeter
User avatar
am0665
 
Posts: 93
Joined: Sat Nov 27, 2010 3:20 pm
Machine: ResMed S9 Auto
Mask: ResMed nasal mask Mirage Activa LT
Humidifier: ResMed Hi5
Year Diagnosed: 2010

Postby robysue » Mon Dec 20, 2010 11:42 am

am0665 wrote:robysue, would you have some additional information on the comments you made above?

robysue wrote: Pressure data is dependent on the patient and there's no real correlation between how much pressure is required to keep a particular patient's apnea under control and the severity of the apnea during the diagnostic sleep study from what I've read. [Apparently pressure needs depend on the anatomy in one's neck.
... Pressure data is dependent on the patient and there's no real correlation between how much pressure is required to keep a particular patient's apnea under control and the severity of the apnea during the diagnostic sleep study from what I've read. [Apparently pressure needs depend on the anatomy in one's neck.]

Would you have any references to pressure calibration strategies during home CPAP? ... toward a semi-knowledgeable OSA sufferer partnering with the physician and DME toward continuing his / her own home evaluation of how to achieve optimal results by changing various parameter, such as pressure.

I'm not a doctor. And no I don't have any references to professional literature. I've done some reading here and elsewhere on the web on how titration is done in the lab. Google CPAP titration and you'll find a wealth of information pointing to what the sleep techs face when trying to do their jobs.

In my case, having undergone a basic PSG study, then a CPAP titration study (during which the discomfort and the mask changes allowed only 3 hrs of sleep), I was provided just a one page summary of "scoring" for each, in which the results looked conflicting. For all I know, the different conditions, different sleep techs, different scorers may result in significant differences in results. The resulting prescription by the physician for auto-CPAP contained a pressure range between 8 (for side sleep) and 14 (for supine sleep); which was later changed to constant 9, which the physician called "somewhat undertitrated".

From what I've read, the techs can get a surprising amount of data (far more than we think they can get) from as little as two hours of actual sleep time spread out over the 6 or 7 hour period of "in bed time". For what it's worth, I share your frustration: I've done three sleep studies myself in the last 4 months: The diagnostic one, a straight CPAP titration, and a bi-level titration. The best I've done is a bit more than 3 1/2 hours of sleep. The last study I only got about 1 1/2 of sleep all night long. But I do think that the AASM (the accrediting organization) takes interscorer reliability as a very serious issue and learning to score PSGs is a long and apparently tedious and difficult process. I'm not sure your results would be significantly different based on who's scoring them if the scorers are fully qualified. As for why your original prescription was changed from what seems like a reasonable range (APAP 8--14cm) to fixed CPAP at 9cm, that's a question for a physician, not an internet board. If you often sleep on your back, however, I think I'd agree with the doctor who said called the prescription of 9cm "somewhat undertitrated"


robysue wrote:Another thing about the apneas and hypopneas scored in the sleep study is that they were scored based on multiple criteria coming from multiple sensors. There's a reason why they have all the wires on you during the sleep study...

If I may paraphrase with an example: say the lab study shows that you were ASLEEP 50% of the time deep enough to develop any apneas, and say you got an AHI score of 40 (severe OSA); then say your home CPAP device gives an AHI of 20; note that this smaller score is measured on 100% of recording time ... no matter whether you were awake or in some sleep stage. So if one does the math, that's 40 A&H events on the average per SLEEP hour in both cases!

It's not this simple mathematically. You can't simply average averages.

In the lab, they are looking at your EEG readings in 10 second increments to determine if you are really asleep during that 10 seconds. And sleep depth is not the issue---if you are alseep (in stage 1, stage 2, stage 3/4, or REM) and you're not breathing or struggling to breath, that's a problem. If you're awake and you're not breathing---that's not sleep disordered breathing and it's not scored as such in the lab. So they know with 100% accuracy which respiratory events occur when you are asleep and which occur when you are awake. And if you're not breathing when you're awake, that's not a problem with sleep disordered breathing. In other words, EVERY SINGLE apnea that is scored in the lab happened when you really were asleep. There were NO so-called false apneas---a detected interruption in your breathing scored as an apnea that actually occured when you were awake. A significant amount of the the work in scoring apneas and hypopneas is indeed in making sure that each and every scored apnea and hypopnea is a real one that actually did happen when you were asleep and met multiple criteria for scoring it. Moreover the arousals measured in the lab do not require that you wake all the way up to WAKE---only that you go from a deeper stage of sleep to a lighter stage of sleep. Sleep efficiency in the lab (% of time in SLEEP as opposed to WAKE) is usually reduced for most people during their sleep studies and for many people it's substantially reduced from what they get at home.

Now, at home as you've pointed out, the machine doesn't know when you're awake and when you're asleep. You've made TWO severely flawed assumptions in order to make the claim that your AHI is 40 in both the lab and at home.

First, you've assumed that you were asleep only 50% of the time when you are in bed at home with the machine running which is highly unlikely---it is your own bed and you're not hooked up to all the wires after all. I think you're vastly overestimating the amount of time that you're geniunely in an EEG WAKE state when you're in your own bed with the machine on. But you are right that "Machine time does not equal sleep time" and that does introduce a real tendency for the machine numbers to underestimate the real AHI.

Second you've assumed that the home CPAP detected no false apneas. But it's not actually that uncommon for some people to have irregular breathing patterns (including apnea-like events while NOT asleep given EEG evidence) while they are transitioning into stage 1 sleep and then have their breathing settle into a regular pattern. It's also no uncommon for people to arouse slightly to turn over in bed and possibly miss a breath or two, which with EEG evidence would not be scored as a hypopnea or apnea in the lab, but will be scored as such by a home CPAP machine. This problem, on the other hand, would cause the home CPAP to slightly over estimate the true AHI.

The net results of these two types of built-in limitations in how a home CPAP machine can determine hypopneas and apneas are these:

1) The fact that the home CPAP cannot determine sleep state means that the reported AHI is likely an underestimate of the real AHI in the sense that the machine AHI is computed by dividing the number of events by the hours the machine is run instead of by the total sleep time as established by the EEG data, which is not existent. However, I believe that once a patient's apnea is reasonable well managed and the patient is sleeping reasonably well with the machine, the machine's AHI is a decent rough approximation of the real (and fluctuating) AHI simply because for most people, the time the machine is running won't be more than an hour or so greater than total sleep time (if we're generous). As an example: Let's say you have 12 apneas and hypopneas on a night where the machine was running 7 hours and you actually slept 6 hours. The machine AHI is 12/7 = 1.7 and the true AHI = 12/6 = 2. Not really that different.

2) The fact that the home CPAP may detect false apneas means that occasionally the home CPAP will overcount the number of apneas, and hence could overestimate the true AHI in an extreme case. Again, let's look at an example. Suppose that the home CPAP records a total of 12 apneas one night, but one of them is a false apnea that happens before you're fully asleep. So there are really only 11 apneas that should be counted. And lets suppose that you slept for 6 hours. So your true AHI should be computed as 11/6 = 1.8. And lets suppose that the machine ran for 7 hours. So the machine recorded a total of 12 apneas (including that one false one) in 7 hours of run time (including the one total hour of run time when you weren't fully asleep). So the machine's AHI is 12/7 = 1.7. Which actually compares rather favorably to the true AHI of 1.8 after all.

My short PSG results do mention % of night in each stage, but don't disclose the subdivision of A/H/C events by sleep stage. I have requested the detailed results one month ago, still waiting to see them. I'd also like to see not just an average for the night, but also the distribution: 95 percentile, max hourly, and the local detail corresponding to the special events.

Most PSG's do break down NREM and REM AHI, but they don't break it down between S1, S2, and S3/4. Sometimes it's a matter of the truly tiny amounts of time involved: If you were only in, say, S1 for a matter of 5 minutes all night long, and had 10 events during S1 all night long, then the report would have to list the S1 AHI as 120. Now, saying the S1 AHI = 120 is technically correct, but not really all that meaningful in this context in my humble opinion given that you only had 10 apneas in S1 sleep that night.

Does your PSG contain the summary graphs? If it does, the information about how many events per hour and what type of events per hour you seek is likely there. In mine, there's a little tick mark for each event at the time it occured. So it's really easy to see when the clusters occured and by comparing the event graph to the sleep stage graph, it's even easy to see roughly how many events occurred in which stages of sleep (and in which time I was in a particular stage of sleep).

robysue wrote:Conversely, it the leak rate is minimal, then one can go on to other sources of trouble on sleep discomfort or high AHI.
Agree 100% with this! Low leak rates indicate the problems (if any) lie somewhere other than mask problems.
current settings Min EPAP = 4, Max IPAP = 8 and Rise time = 3

8/1/2010 sleep study results:
AHI = 3.9 [AHI = (#OA +#CA + #H w/desat) per hour]
RDI = 23.4 [RDI = (#OA +CA + #H w/desat + #H w/arousal) per hour]
Dx: Moderate OSA
User avatar
robysue
 
Posts: 1330
Joined: Tue Aug 24, 2010 6:47 pm
Location: Buffalo, NY
Machine: PR System One BiPAP Auto
Mask: Swift FX for Her
Humidifier: System One Heated Humidif
Year Diagnosed: 2010



  • Site Supporter

Return to CPAP Mask, Machine, & Monitoring Treatment Questions

Who is online

Users browsing this forum: No registered users and 1 guest

  • Site Supporter