Mirage Quattro acceptable leak rate - resolved (I think)

[RandyJ]

From my recent data (see my other posts if interested) I assumed that if my leak rate went above 24L (the red line on the Rescan graph) that I was leaking too much.

However, after searching and reading various threads here, it seems that if my average pressure is around 17 cmH20 and the vent rate for my mask (the Mirage Quattro) at 16 cmH2O is 48, then if I am showing a 95 percentile leak rate of 34, I am actually ok and within acceptable parameters for my therapy.

Also, if this is acceptable, can I assume that all data (AHI, AI, etc) is valid?

Can anyone confirm this?

Thanks!

[nanwilson]

You are doing good, the leak rate on the quattro is high, my pressure is at 11 and the charts states a leak in the high 30's is acceptable. Check out padacheek.com for mask liners, the one Karen makes for the quattro is worth its weight in gold, but in fact is actually pretty cheap.

[Otter]

However, after searching and reading various threads here, it seems that if my average pressure is around 17 cmH20 and the vent rate for my mask (the Mirage Quattro) at 16 cmH2O is 48, then if I am showing a 95 percentile leak rate of 34, I am actually ok and within acceptable parameters for my therapy.

The catch is that Resmed machines subtract the intentional leak rate of the mask before reporting the 95% leak rate. Five percent of the time your total vent/leak rate is about 82L/min or higher. So no, your leak rate is high enough to confuse the machine.

I suggest you start a new thread about your mask and leaks. Be sure to put the name of the mask in the title. Or, if you already have such a thread, bump it to the top of the forum again.

[Pugsy]

ResMed already subtracts an unknown fixed amount for full face mask (per the mask setting) before ever reporting your leak. You would need to add that unknown amount back into your reported leak amount to compare it with the intentional vent rate at your pressure to have an apples to apples comparison.

You would need ResMed to also report a Total leak rate so that you could do some math to try to isolate that unknown factor.

Now if someone can figure out the number ResMed subtracts prior to reporting the leak you could know for sure if your reasoning is sound or not.

[Otter]

I believe the firmware subtracts based on the PQ curve for whatever type of mask you selected, so the leak numbers should be reasonably accurate at all pressures. The 24 l/min standard is for unintentional leak, which is what the 95th percentile number represents. Hence, you should try to get your 95th percentile leak rate down below 24, lower if you can.

[RandyJ]

You are doing good, the leak rate on the quattro is high, my pressure is at 11 and the charts states a leak in the high 30's is acceptable. Check out padacheek.com for mask liners, the one Karen makes for the quattro is worth its weight in gold, but in fact is actually pretty cheap.

Thanks, nanwilson - I just got a mask liner from padacheek yesterday and used it last night. I had the impression that I didn't have much if any leak, but reported leak rate on the S9 was 34 (which is what I have been getting without the liner).

Will keep trying...

[Jay Aitchsee]

Randy, I agree with Otter that your leak rate is way to high. If you've told your machine the proper mask, then the leak rate reported is the leak rate of the mask only. One thing to watch, it's fairly common for these parameters to be presented in different units. For example, L/sec and L/min.
From my own expereience, I've found that many times these leak spikes will occur with a reported apnea or hypopnea, calling into question the validity of the event. At times, the machine will report an apnea with a high leak when I know I'm awake, for example in the morning when I take off my mask. By definition, those are not apneas. As well, the high leaks will wake me up during the night, often with an apnea reported around the same time. Now, I don't know if those are apneas or not. My advice is to work hard at reducing your leaks to near zero for most of the night.
Jay

[RandyJ]

thanks otter, pugsy & jay

wishful thinking, I guess will keep working on & also review data last night for hour by hour to see if any trends emerge...

padacheek mask liner did not seem to affect leak rate either positively or negatively.

[Otter]

Folks, are you talking about this chart?
http://www.resmed.com/us/assets/documen ... sa_eng.pdf

That's not the acceptable leak rate, that's the expected flow rate through the mask when it's sealed. The chart shows you what the s9 subracts when it reports your leak rate. The 24 l/min limit applies no matter what mask you use or what pressure you're at. Hence the red line in ResScan.

Randy, have a look at your leak line in ResScan and see about how much of the time it's over 24 l/min. The 95th percentile at 34 could mean that your mask is perfectly sealed 95% of the time. This seems unlikely, but a look at the graph will give you a better idea how bad the leak problem is.

I hope your new mask liner solves the problem.

[Jay Aitchsee]

Randy, have you tried the "mask fit feature" of your machine? I assume the S9 works the same as the S8. Just hold the "on" button down for a few seconds and the machine will come on at the highest pressure it is currently set for, i.e, the cpap pressure or the max of the apap pressure as appropriate. It will then report your leak rate or mask fit as "excellent", "very good", etc.
I find that to be helpful in trying new masks, seals, liners etc.

Jay

[RandyJ]

Randy, have you tried the "mask fit feature" of your machine? I assume the S9 works the same as the S8. Just hold the "on" button down for a few seconds and the machine will come on at the highest pressure it is currently set for, i.e, the cpap pressure or the max of the apap pressure as appropriate. It will then report your leak rate or mask fit as "excellent", "very good", etc.
I find that to be helpful in trying new masks, seals, liners etc.

Jay

jay - yes, on the S9 I get a big green smiley face that says my mask fit is good, and I don't perceive any leaks. I haven't had a chance to load last night's data to look at graph hour by hour but will do so and report back.

thank you!

[Otter]

My advice is to work hard at reducing your leaks to near zero for most of the night.
Jay

Good advice.

It may seem like you're at the foot of leak mountain now, Randy, but don't worry, you'll get there. Part of keeping your leaks low is learning how to adjust the straps on your mask, and this really is more of an art than a science. It takes time to develop this skill, and time to learn which tricks (mask liners, moleskin, anti-leak straps, washing your face, etc) work for you.

If you really can't get the thing to seal, it's possible that it simply doesn't fit your face. Many people go through several masks before they find one that really works. Once you do, though, it's like magic.

In the meantime, keep hosing. Even if your therapy is somewhat compromised by the leaks, it's still much better than not using your machine at all.

Oh, and check to make sure you don't have holes in your tubing, and that it's connected to the machine properly.

[Pugsy]

wishful thinking, I guess

I understand where your reasoning was taking you. With the higher pressures on your mask there is a higher vent rate.
The problem is we just don't know exactly how ResMed comes up with the number that they subtract (don't feel bad, Encore Pro 2.4 lets us choose Unintentional leak too and we aren't told what that magic number is either) before it reports the leak. Let's suppose it happens to be 35 L/min for FFM choice. If we add that amount back into your reported 34 L/min we still would get Total Leak of 69 L/min which is considerably higher than the 48 L/min vent rate your mask literature offers at your pressure. This amount would be more than I would be happy with.

Instead of dwelling on the 95% number or median or maximum, I would opt to choose to dwell on the leak line itself.
Assuming proper fit when starting out the night and no leaks, look at where the leak line itself starts. Use that line as your baseline. It might start at zero line or it might start a little higher. Watch for spikes that are large and prolonged. The goal would be to have those times at a minimum. You might have a nasty leak for an hour but a pretty much straight line the rest of the time. I would then figure that for the majority of the night I did well.
No if I look at that leak line and it is huge and all over the place encompassing large amount of time, then I didn't do so well and I would then suspect the data shown. You likely won't have a perfectly straight leak line. Little zippers or blips are okay. It is those large spikes for long periods of time that can impact therapy.

[Otter]

The problem is we just don't know exactly how ResMed comes up with the number that they subtract

I'm pretty sure they use a chart like the one I linked above, and that it's based on testing sealed masks at different pressures. Is there doubt about this? How else would they do it?

[Pugsy]

I'm pretty sure they use a chart like the one I linked above, and that it's based on testing sealed masks at different pressures. Is there doubt about this? How else would they do it?

S9 Mask choice simply says "full face mask" is that true? ResMed offers more than one full face mask. What happens with a non ResMed full face mask? The machine thus has no way to know for sure which mask is being used. They don't all have the same vent rate. They may come close but they aren't all identical. So where does the number subtracted come from? We aren't ever told exactly how it is determined. Is here some hidden sensing factor based on pressure and leak at the onset of masking up? What happens when people use APAP and the pressure varies?

I don't know how they do it. We can assume all day long but I have never seen a written explanation. Prior to S9 mask choice users were told to simply use "standard" when non ResMed mask was used. Where did they come up with that magic number to subtract then? It was commonly told that the magic number for standard selection was 24. I saw it mentioned everywhere but from ResMed itself. No one ever pointed to actual documentation that I ever saw.

I use Respironics machine. We aren't ever told exactly how the machine comes up with its magic number either.
Depending on software choice we can see total leak or unintentional leak. I often see Total leak average being much less than my stated vent rate for my mask. I am not talking a little difference, I have seen average of 26 L/min when my vent rate at my 10 min pressure is around 34 L/min. This tells me the numbers themselves are not much better than a WAG and I choose to just monitor the leak line itself. The line doesn't lie. Numbers may not lie but they don't always show us the whole truth. They are too easily manipulated.