Leakage way Up, but UA & CAs coming Down.

Progress (?) Report: >> Would you be happy with this? (I am -- I think, but I'm also surprised.)

ResScan Statistics (AI Events/hr)
12-7-11 to 1-16-12 Obstructive = 2.3, Central = 7.7 Leakage (95th %-ile) = 4.8 (OA and CAs down from much higher Sleep Lab Titration values.)
1-17-12 to 2-02-12 Obstructive = 0.6 Central = 4.6 Leakage (95th %-ile) = 35.0 (Could mouth breathing during REM account for the higher "leakage"?)

I'm feeling more and more alert, more sociable, and sharper although the improvement has been a subtle, but steady and sustained. There hasn't been any dramatic eurika moment even at the beginning two months ago. Just a quiet creeping improvement. Yes, I've been experimenting with different masks over the entire period, but am rather sure the leakage being reported by the software is far more volatile than what has been going on with the masks. And despite the belated big increases in reported "leak", the Obstructives and Centrals gradually keep coming down.

Anybody else experience this sort of phenomenon?

Sleep doc doesn't seem to give much importance to the "leak" numbers right now, and maybe not in the future either, so long as they stay under 40. The techs, however, seem to focus far greater emphasis on leakage, not only wanting to generally shoot for under 15, but under 5 and the closer to zero the better. With my Titration pressure level having initially been determined at a relatively low 8 (H20) and nobody seeing any reason to change it, does that low value also minimize the significance of "leakage" as high as 35 or even 40 if that trend persists?

Would appreciate any input as so much emphasis, maybe too much, seems to be focused on leakage level issues, masks, etc.

Well, the mask is the interface where man meets machine so if it's a leaky connection that can't be good. The time period your OA and CA came down your leaks came down, too. I like my leaks as close to 0.0 as possible. If your therapeutic air splint is leaking too much it can't do it's job.

Papit wrote: Leakage (95th %-ile) = 35.0 (Could mouth breathing during REM account for the higher "leakage"?)

Yes, mouth breathing can be blamed for the higher leak numbers. Prolong sustained leak above 24 L/min with ResMed machines can/will cause the machine to report fewer events because the machine loses sensing ability. Like give you a false sense of security about that low AHI.

Instead of going by the 95% leak number which just means that for 95% of the time you were at OR below that number...look at the leak line itself and see how much time you spent near or above that 24 L/min line. The 95% number can be falsely elevated by taking the mask off with machine running (machine will report it as a maximum huge leak and thus impact that % number).

Your AHI numbers could very well be the real deal if you spent the bulk of the night below the 24 L/min line and the % number is elevated due to short lived spikes. I would use the software and check each night instead of basing things on the % number. The median average and/or leak line inspection would be what I would be looking at.

Short lived spikes to the red line or above..wouldn't be the end of the world. Now if half the night was spent there....different story.

"Yes, mouth breathing can be blamed for the higher leak numbers." . . .

" . . . if you spent the bulk of the night below the 24 L/min line and the % number is elevated due to short lived spikes. I would use the software and check each night instead of basing things on the % number. The median average and/or leak line inspection would be what I would be looking at."


I'm hoping that the preponderance of significantly elevated "leak" indications on the graphs which are occurring in the mid to latter half my of recent sleep nights are, as my sleep doc speculates, due to new, more frequent and longer REM sleep periods. I notice that I'm gradually beginning to dream more than I have for a very long time. Only one or two very brief (a couple minutes each) REM periods ever occurred during my first sleep lab and second (titration) lab sleepovers.

I went back and checked the nightly detailed leak graphs. As you suspected, most of the occurrences of leaks that went above the 24 L/min 'threshold marker' line were momentary spikes rather than sustained leakages. Again as you suspected, when I looked at the median (average) leak number each night when the 95%ile figure was way up there (30 to 50), the median was down around a relatively nice 7 or so. So I guess your conclusion that the reported nice looking Obstructive apnea index and Centrals index can indeed be looked at "real McCoy" numbers.

The median Leak statistic for the 1-17-12 too 2-02-12 period is 7.2 with the 95 %ile at 35.0. The corresponding statistic for 12-07-11 to 1-16-12 is reported as 0.0 with the 95%ile at 4.8. Bottom line thought seems to be that REM-induced mouth breathing that gets reported as high 95%ile "leak," but acceptably low Median "leak" numbers is desireable and good to see. Am I reading you right?

Papit wrote:Bottom line thought seems to be that REM-induced mouth breathing that gets reported as high 95%ile "leak," but acceptably low Median "leak" numbers is desireable and good to see. Am I reading you right?

Pretty much.

I have never been one to dwell on the leak numbers themselves because they are just numbers and the story they tell may not be the full story. I prefer to look at the leak line each night and if it is generally acceptable overall then I know I don't have to worry about the actual leak number.

My OSA is worse in REM sleep. Significantly worse and I also will often see some leak spikes during a time frame that is most likely REM stage sleep. I would say 5 nights out of 7 I will see an increase in leak right around the 90 to 120 minute mark after sleep onset which pretty much when first REM cycle normally occurs. Sometimes a few spikes in the wee hours of the morning when our REM sleep comes quicker and lasts longer. Why more leak at that time vs the other? I don't really know why. Never have been able to figure it out. All I know is that I see it often.

Why more leak at that time vs the other? I don't really know why. Never have been able to figure it out. All I know is that I see it often.

My leak doc explains it this way. He says that what appear to be leaks during REM sleep is often (?typically?) actually open jaw mouth breathing that tends to occur for some of us as a result of pronounced relaxation of our muscles (including face and jaws in deep REM sleep). Sounds very plausible.