My 1st night - what do I make of the downloaded data?

I just want to say, by way of introduction, I *love* this board. I've suggested to my docs and technicians that they tell patients about it.

So after lots of profitable reading here, and after winning a game of chicken with my insurance company, I took delivery yesterday of a Puritan Bennett 420E, H2O humidifier, Swift mask and SilverLining software. I had the luxury of an unusually long night in bed. I can't say a long night's sleep, but I'm determined to make this work.

I have lots of questions. Please pardon an awfully long post.

Basics: I'm a 40-something male, not overweight, with apnea diagnosed in sleep studies at a modeate 15 AHI overall but a severe 45 during REM. My untreated sleep is highly fragmented during dreaming, and I don't get much deep (S3, S4) sleep. Hospital recommendation was CPAP at 12 cm pressure. Taking advice here, and with doctor's okay, I set the machine for APAP ranging from 8 to 14. (Or maybe not. I tried to set 8-14 with a ramp beginning at 4, but it looks like I actually set the desired overnight pressure at 4-14.)

I've spent a couple hours now puzzling out the not-so-fab SilverLining manual, and I'm not sure how to read some of the data I collected. For those of you with sufficient endurance, I'm appending a bit of the summary information to this post.

So some questions:

*. WHAT TO DO WHEN I GET OUT OF BED. Three times last night I took the mask off to get out of bed for a couple of minutes each. I didn't know how the machine would treat off/on events, and I wanted data for the whole night, so instead of switching off the air flow I just pressed "ramp." This dropped the pressure down to 4, and I laid the mask at my bedside that way, putting it back on a few minutes later. My question is, is this the best way to capture good compliance data for the night? I'm thinking that my method exaggerated leaks, since leaks were 100pct for about 10 minutes, and may have been recorded as some kind of event.


*. LEAKS. I don't know how to interpret the leak data. I had some leak trouble when sleeping on my side, but none to speak of on my back. Average leak recorded overall was ~0.5 liters/second, low 0.3, high 0.8. My average pressure overnight was 8 cm, and leak flow not surprisingly increased with higher pressure. But if I'm reading the charts correctly, the maximum possible leak (a 100 pct leak) at pressure of 8 would be 1.2 l/s and a maximum possible leak at pressure of 14 would be 1.6 l/s. If I'm leaking 0.5 out of a possible 1.2, and 0.8 out of a possible 1.6, doesn't that seem high?

*. EVENTS. I'm also not sure what to make of the data on irregular events:
EVENT INDEX/h
----------------------------------------------
Apneas 1.5
Apneas with cardiac oscillation 4.6
Hypopneas 2.7
Flow limited hypopneas 1.6
Flow limitations - runs 23.1
So, looking at these:
1. Can anyone explain what it means by cardiac oscillation, and how the machine thinks it can measure a cardiac event?
2. Difference between hypopnea and flow-limited hypopnea?
3. If I add the first four lines, does that not give me an overall AHI of 10.4? That's better than what I got in the sleep lab, but not dramatically so. What should I make of that?
4. What are flow limitation runs, and what do they mean to me?

*. EFFICIENT PRESSURE. Through the night, I spent more than 10 percent of the time (each) at pressures of 4, 5, 6, and 7, nearly 10 percent (each) at 8 and 9, hardly any time at pressures 10-13, and about 25 percent of the time at the maximum pressure I set, which was 14. Altogether the software reported that my minimum efficient pressure was 14. What does that mean? As I said earlier, I accidentally set the range as 4-14 instead of 8-14. Do these data mean that I should leave the bottom end at 4, since something like 40 percent of my night was between 4 and 7? Do the data also mean, in combination with the index of apneas and hypopneas, that I might be better off with a high setting that is higher than 14?

*. OBSERVED PRESSURE. Sometimes I woke up to find the pressure steady at 14, and it stayed that way for as long as I kept watching (several minutes) even though I was awake and breathing well. Does that mean there was something wrong with the auto-set detection? Shouldn't it have been dropping down while I breathed normally?

Thanks so much for helping me learn.

-I'm thinking that my method exaggerated leaks, since leaks were 100pct for about 10 minutes, and may have been recorded as some kind of event.

You're right in that thinking.

*. WHAT TO DO WHEN I GET OUT OF BED If you get up in the night, turn the machine on, and then off, it will start recording again after 15 minutes - the time it puts aside to let you settle down. You'll see that break on the next morning's compliance screen,
I decided to hit the ramp when I wake and the pressure bothers me, but to turn the machine off if I get up.

*LEAKS - I read the data like that too - haven't found anything to say that's the way to think about it, but it seems reasonable.

1. Can anyone explain what it means by cardiac oscillation, and how the machine thinks it can measure a cardiac event?

- Kind of frightening when you see it for the first time, isn't it?
Don't worry. Somtimes you stop breathing without your airway beind obstructed. Under those condition, the machine can sense the air in your open airway oscillating at the rhythm you hearbeat. As a matter of fact, I can sometime hear that in my mask's exhaust. These breathing interruption can occur for many reasons, including tossing and turning in bed and are called "central apneas" because they are caused by your central nervous system.

2. Difference between hypopnea and flow-limited hypopnea?

The latter come at the end of flow limitations, or are concurrent with them, the former appear just like that. Flow limited hypopneas are those refered to by the IFL2 switch.

3. If I add the first four lines, does that not give me an overall AHI of 10.4? That's better than what I got in the sleep lab, but not dramatically so. What should I make of that?

see my reaction to your pressure question.

4. What are flow limitation runs, and what do they mean to me?

PB gives you quit a bit of info on those in the 418p manual. You've got it on your CD: open SL3, click on the little black question mark, and you will have a choice of manuals to read. The 418p was the 420E's predecessor, and it's manual explains a lot. What they mean to you is that when they occur, the machine raises the pressure, in order to preempt apneas. This reaction is controled by the famous IFL1 switch. Based on the results you reported, it seems to me you should leave this switch as it is.

*. EFFICIENT PRESSURE. The most efficient pressure, by default, is the pressure at which, or below which, you spent 90% of your time. Since you spent 25% at 14, that't what the machine reports. The fun thing SL3 is that you can change the percentage you're asking about. If you were to ask about 70% of the time, you'ld get the next highest pressure.

As I said earlier, I accidentally set the range as 4-14 instead of 8-14. Do these data mean that I should leave the bottom end at 4, since something like 40 percent of my night was between 4 and 7? Do the data also mean, in combination with the index of apneas and hypopneas, that I might be better off with a high setting that is higher than 14?

It's the other way around: to me the data indicate that your low pressure might be too low, since you've got hypopneas (the machine doesn't raise the pressure in reaction to those) hypop + flow limitations, and flow limitations. I wouldn't touch the high limit, but I'd raise the low limit to where you had planned to have it. One of the drawbacks of SL3 is that it doesn't let you see how many events occured at each pressure. It is very possible that many of your events occured at the lower pressur range, and then the machine raised the pressure.

*. OBSERVED PRESSURE. APAPs are meant to identify sleep breathing patterns - the waking ones are different, so this may very well be a case of the watched pot. They also do not change pressure immediately - I think they wait about 15 minutes before they decide to lower the pressure.

For more info, do a search for posts by -SWS (include the minus) about the 420E.

Good luck!
O.

Wow. That's a huge help, thank you. I'll do a search for those posts and check the 418 manual. I'd never have thought to look in the other manual, since I'm using the 420E.

I'm going to have to play a bit with the Swift to get the best comfort/leak compromise as well.

Well, the second night leaves me with more questions than answers.

I changed two things: I set the APAP range from 8 to 14 (instead of 4 to 14), and I used the large Swift mask insert instead of the medium. The large was more comfortable and seemed to stop leaks more easily.

The results appear to be worse:

Event Index/hour
Apneas 0.8
Apneas with CA 6.7
Hypopnea 4.8
w/ flow limit 1.4

AHI altogether, if it in fact is supposed to be all these four added, is higher than the first night -- 13.7.

The leak numbers continue to be, it seems to me high: average leak of 0.8 liters/second, with low of 0.6 and high of 0.9.

I could really use help understanding all this.

*. Is AHI in fact the sum of the first four indexes? If so, and if there's not a big measuring error, this is not nearly good enough. But I'm skeptical of the numbers, because three sleep lab tests showed nearly no central apneas and no heart arythmias (just the expected increase in heartbeat during obstructive apneas, when my oxygen desaturated).

*. What's an acceptable volume of leakage? I'm not clear how to read the leaks line drawing (under SilverLining's pressure tab, third display down), but the measured leak of 0.9 at 12 cm of pressure corresponds to a "max leak" of 1.4 l/s at that pressure. Does that mean 1.4 l/s represents a 100 percent leak? Yikes.

*. Is it possible that high leaks would exaggerate the machine's detection of apneas and hypopneas?

As for the leaks themselves:

*. When I put on the Swift and adjust it as best I can, I can feel no leak at all from my nose or from any part of the mask or hose. (Jostling it, to test this, produces very clear leaks.) However, there is a very substantial outflow, regardless whether I'm exhaling, from the vent holes at the front of the mask. So much so that it hardly varies when I inhale and exhale. Is this normal? I'd suspect a defective nasal pillow insert, but this was true of both the medium and the large pillows.

*. While awake, and as far as my bed partner can tell when I'm asleep, I'm not mouth breathing. But I can't be too sure of this. I do have a deviated septum on one side and a history of mouth breathing while asleep. How can I tell for sure whether I'm mouth breathing with this mask, and whether that is affecting the results?

*. What else could account for such high leak numbers? Is there any possibility that there's something amiss with the narrow-gauge sensor line inside the main hose? It seems to be connected properly to the humidifier.

Any other theories?

I'd sure appreciate any and all suggestions from you folks. Thanks.

O,

Now I have a question that maybe you can answer, the woman with all the answers...(truly amazing how much you know). I was always under the impression that a hypopnea was an at least 50% obstruction of your airways for at least 10 seconds or greater with oxygen desats. If that is the case, what is a flow limitation? Is it an obstruction of the airways of less than 50% or one without desats? I'm a bit confused now.

n0trab:
Unless you feel absolutely miserable, I would give it at least a week before making any changes.
Your mind and your body have to learn to breath and sleep in a new way, they have to get used to this strange way of breathing and sleeping.

Last night the machine was recording all your breathing experiments - and I'm pretty sure that while trying to find the source of the leak tonight, you stopped and listened. The PB doesn't know you're awake and listening, it just "knows" you're not breathing, and your airway is unobstructed - so it records an "apnea c/a". If my personal experience is anything to go by, you'll find them decreasing once you get used to the process, and start sleeping.

*Leaks: I believe you're ok as long as they are lower that the top limit. The vents in the mask have to let out air, and according to my logic (not knowledge) the air rushing out should'nt decrease when you inhale because the machine is supposed to keep that pressure constant.

*Your AHI: Study the details screen: Look at the hypopneas and note if the machine raises its pressure when they occur. If you keep on having those hypopneas, and the machine does not react by raising it's pressure, raise your low pressure gradually, since they could mean the splinting of you airway just isn't enough, and what you're getting is a "almost collapse" due to the pressure being too low.

I don't really think the resluts are measuring error on the machine - I think its more a real variability in the data + a sampling error. You've had three nights with xPAP: one in the lab, two at home - they are not necessarily representative.

Lori and n0trab:
A lot of what I know comes from reading -SWS's excellent postings - try a search for "flow limitation" with -SWS (don't forget the minus) as author.

O.

O,

There are many topics on which I can have a discussion and sound, and actually know, like I know what I'm talking about. But after reading that post from -SWS, this is not one of them. I couldn't understand what I was reading. All I know is that last night, I had two apnea events all night and no hypops or snore index, yets my flow limitation was high, about 9 of them. What this all means? I don't know. I don't like not understanding what is happening to me and usually I can get a graps on what people are saying. This thread was way too technical for my totally unscientific mind to comprehend.

Thanks for the link, but it really didn't help me understand anything. Guess I'll just do the research myself and see what I can come up with for the anti-scientists like myself.

I just re-read this post and I am not trying to give off any air of annoyance or aggravation at all towards you, O. I hope you don't take it that way. I am just saying, "Inquiring minds want to know," but this is far too scientific for me. Thanks for trying, though. And, yes, I am admitting a deficit in my own intelligence. Science was always my worst subject in school. Now, ask me about English and grammar...which is now part of my profession, also, and I can keep up with the best of them...

If I undestood the subject correctly, a flow limitation - as measured by an APAP - is a change in the shape of the air flow, that indicates something is limiting it.

This limitation can be:
1.The result of sagging tissues, and in that sense be a an avoidable semi-obstruction (my terminology). There will be more flow limitations of this kind when the pressure is low, less when it is raised.

Or
2. The result of the way your airway is built. If it is a result of the way your airway is built, it not be improved by any change in pressure.

My "limited flow cycles" are alway close to 40% of the cycles, regardless of the pressure, and the way I read it, it's something structural that can't be changed by raising the pressure.

Hope this is correct, and makes it clearer.

O.

Aw, O! You are too sweet. You really didn't have to go into detail. I could search the Internet myself and find info on this. Thank you so much.

My flow limitation index seems to always have some good figure, regardless if I'm on high or low pressure, or if I've had one event all night or five. I don't understand it. And yet my ENT says he can see absolutely no anatomical reason for my OSA. And he truly is excellent. Been using this practice for 15 years and they are very well know to have a wonderful reputation where we are.

I guess as long as my AHI's are 0.2 and 0.3 even with this FL index, I shouldn't complain. I just really like to get a handle on my condition and when I don't understand certain things, it nags at me until I do.

Again, as usual, O, you are overly helpful and kind. Thank you again for taking more of your time to explain this. You, and others like yourself, are what makes this site what it is, a blessing.

ozij wrote:n0trab:
Unless you feel absolutely miserable, I would give it at least a week before making any changes.
Your mind and your body have to learn to breath and sleep in a new way, they have to get used to this strange way of breathing and sleeping.... You've had three nights with xPAP: one in the lab, two at home - they are not necessarily representative.

osij, Thanks very much -- I am probably too much a tinkerer by temperament, and this sounds smart. I guess I'm no more miserable than usual, so I'll see how it goes after I settle in a bit.

ozij wrote:n0trab:
*Leaks: I believe you're ok as long as they are lower that the top limit.

Well, from another thread (viewtopic.php?p=34899#34899) I learn that leaks are more commonly reported in liters/minute, not liters/second (as SilverLining does). So, converting to liters/minute, I see my high leaks have been something like 54, which seems to be higher than desirable but maybe not a catastrophe. My average leak is around 48.

Last night, at a couple of points on the cusp between waking and sleeping, I noticed an escape of air from my mouth. Not enough to "decompress" completely, but there was definitely a break in the seal. I have a feeling I'm going to have to address this somehow.

Any idea how I can tell, as I go along, how much I'm mouth breathing (or leaking) and whether I need to go down the road of the tape/mouthguard/full face mask?

n0traub,

I had the same problem with knowing I had SOME leaks through the mouth, but didn't know how much. The only way I discovered how much it was was to tape my mouth one night. My leak rate went from 32-43 to 18-19. There was my answer.

Unfortunately, unless you have someone nice enough to sit and watch you all night and see how long/often you mouth breathe, I don't think there is another way, short of knowing your mask/hose leaks, to determine how much of your leak rate is due to mouth breathing.

n0trab wrote:The results appear to be worse:

Event Index/hour
Apneas 0.8
Apneas with CA 6.7
Hypopnea 4.8
w/ flow limit 1.4

AHI altogether, if it in fact is supposed to be all these four added, is higher than the first night -- 13.7.

It does look that way.

The leak numbers continue to be, it seems to me high: average leak of 0.8 liters/second, with low of 0.6 and high of 0.9.

The reported leak rate will be a function of the pressure as the mask will always "leak" some air out the the ports.

You have the same setup I do but are running a higher pressure, so I'll have to project my results.

My swift is reporting leak rates around 0.4 l/s (or 24l/m for people not using PB machines) at 5-6 and 0.45 at 7. I don't have my latest data with higher pressures here at the office so I'm not sure what my experiments at 10 reported, but I know it was higher.

The point is, stating a leak rate without saying what pressure you were at could be meaningless. A leak rate of 0.9 at higher pressures doesn't sound out of line with what I'd expect to see. In SL3 go to the "Pressure" tab and click on the histogram (bar chart) in the middle of the screen. You'll get a breakdown of your leaks at different pressures (low, average and high) plus a plot of leaks per pressure.

I could really use help understanding all this.

We're trying! (We're very trying! )

*. Is AHI in fact the sum of the first four indexes? If so, and if there's not a big measuring error, this is not nearly good enough. But I'm skeptical of the numbers, because three sleep lab tests showed nearly no central apneas and no heart arythmias (just the expected increase in heartbeat during obstructive apneas, when my oxygen desaturated).

It appears that AHI is the sum of the first 4 numbers from your "Synthesis" report.

Remember that a reported "Apnea/CA" isn't necessarily a CSA. For that matter, reported apneas might not even really be apneas!

The machine will report an apnea anytime you don't take a breath when it is expecting you to. And it'll report "CArdiac Oscillations" if you merely hold your breath.

Without the oximetry data and sleep state (were you really asleep?) which you get during a PSG the machine is just making intelligent guesses. This isn't to say that you should ignore what is reported, but you have to use it in context; that is knowing and understanding that it could be reporting more apneas than really happened.

Personally I tend to discount early data (the first day or two after I make a change) because I could still be adjusting to it. Give it time and look at the average of a number of nights. If something unusual happens some night (e.g. rainout or allergies disturbing you), ignore those nights also.

*. What's an acceptable volume of leakage? I'm not clear how to read the leaks line drawing (under SilverLining's pressure tab, third display down), but the measured leak of 0.9 at 12 cm of pressure corresponds to a "max leak" of 1.4 l/s at that pressure. Does that mean 1.4 l/s represents a 100 percent leak? Yikes.

Yes, I think that is exactly what it means. Someone needs to find out what the Swift normally flows at 12 for us to know if you are leaking or not.

*. Is it possible that high leaks would exaggerate the machine's detection of apneas and hypopneas?

One night on my recent trip I was sick in the middle of the night (food allergy to something I ate it seems). In my haste I just pulled my mask off and left everything running. That data shows an incredible spike in the leak rate (as you'd expect) but also shows quite a few apneas during that time that I didn't even have the mask on!

You can draw your own conclusion.

As for the leaks themselves:

*. When I put on the Swift and adjust it as best I can, I can feel no leak at all from my nose or from any part of the mask or hose. (Jostling it, to test this, produces very clear leaks.) However, there is a very substantial outflow, regardless whether I'm exhaling, from the vent holes at the front of the mask. So much so that it hardly varies when I inhale and exhale. Is this normal? I'd suspect a defective nasal pillow insert, but this was true of both the medium and the large pillows.

Normal.

*. While awake, and as far as my bed partner can tell when I'm asleep, I'm not mouth breathing. But I can't be too sure of this. I do have a deviated septum on one side and a history of mouth breathing while asleep. How can I tell for sure whether I'm mouth breathing with this mask, and whether that is affecting the results?

Not sure. Anyone?

*. What else could account for such high leak numbers? Is there any possibility that there's something amiss with the narrow-gauge sensor line inside the main hose? It seems to be connected properly to the humidifier.

A leak of 0.9 at 12 using the Swift doesn't sound unreasonable. Can anyone else confirm Swift leak rates at pressures above 10?

If this info helps make sense of where your numbers stand:

My leaks have hovered in the 0.5 to 0.8 l/s range on average or 30 to 48 l/m. Here and there the leaks are in the 1.0 to 1.5 range - but so far infrequently - I'm guessing I momentarily have shifted into a "leaky position" or I've adjusted the mask while sleeping for these jumps.

My flow limited runs have been in the 24 - 27 range.

My AHI is roughly 2.4

My acoustical vibrations average 5.0 but seem to vary widely night to night. ie, night one is 8.3 then night 2 is 1.6 then night 3 is 5.7, etc

A leak of 0.9 at 12 using the Swift doesn't sound unreasonable. Can anyone else confirm Swift leak rates at pressures above 10?

I am at 10 cms and my leak rates lately on Encorepro (lpm) are usually 18-19; however, the past couple of nights, and one was with a broken hose but I didn't know it, have been about 32-42. I have to go check and see if the hose is still broken after its fix yesterday. Should have gone back to 18-19 since my Tegaderm patches are working great, no leaks.

I did send back the mold for the Dr. Sue guard on Friday and am praying it comes within the week so I can see if it works for me. Then no more having to be silent after the patches are on (it's killing me...love to talk, as you can tell).

Sleepless on LI wrote:I am at 10 cms and my leak rates lately on Encorepro (lpm) are usually 18-19

Wow! Your leak rate at 10 is lower than my leak rate at 7! Haven't you added paper tape gaskets to your Swift? Maybe I need to try that also.