Leak question

[Ineedanap]

I'm sorry if this is a repeat question. I did a search and didn't see it answered. My leak numbers say
Median: 3.6, 95th% 13.8 and Maximum 154.2 What do these numbers mean?

I know I have some leak issues which feel like they've improved immensely since working on fit but the numbers only show real improvement for the days I had a headband holding the middle down and had the mask so tight there were still marks on my face 3 hours later. I know my mask doesn't fit properly and am working with my worthless and hopeless DME provider for one that fits better. That's a story that's been played out here plenty

[robysue]

I'm sorry if this is a repeat question. I did a search and didn't see it answered. My leak numbers say
Median: 3.6, 95th% 13.8 and Maximum 154.2 What do these numbers mean?

Since you are using an S9 Elite, the numbers represent your UNINTENTIONAL leak rates.

In other words, ResMed and ResScan subtract off (an estimate for) the intentional leak rate for your mask at your pressure setting. So all the leak that is reported is UNINTENTIONAL or EXCESS leak---the kind you want to minimize.

Median: 3.6 This means that for 50% of the time the S9 was running, your unintentional leak rate was AT or BELOW 3.6 L/min. It also means that your unintentional leak rate was AT or ABOVE 3.6 L/min for 50% was running. If these are overnight figures, that's 50% of the night with leaks AT or BELOW 3.6 L/min and 50% of the night AT or ABOVE 3.6 L/min.

95%: 13.8 This means that for 95% of the time the S9 was running, your unintentional leak rate was AT or BELOW 13.8 L/min. It also means that your unintentional leak rate was AT or ABOVE 13.8 L/min for 5% was running. If these are overnight figures, that's 95% of the night with leaks AT or BELOW 13.8 L/min and 5% of the night AT or ABOVE 13.8 L/min. And how much is 5% of the night? Well, 5% of one hour is three minutes. So you can get an idea of an upper bound on how long your worst leaks (the ones above the 95% leak figure) lasted by multiplying the usage (in hours) by three minutes/hour. For example, if your machine was running for 7.3 hours and the 95% leak figure is 13.8, that means your leak rate was AT or ABOVE 13.8 L/min for a total of (7.3 hours)*(3 minutes/hour) = 21.9 minutes

Maximum: 154.2 This means that at some point when your S9 was sampling the leak rate it detected a massive leak of 154 L/min. Note that we know you could not have been leaking at this rate for more than 5% of the night because your 95% is lower than 154.2. When the max leak rate is this much higher than the 95% leak rate it could simply be that the S9 happened to sample the leak rate just as you were making a major adjustment to the mask and you knew the seal was broken because you'd pulled the mask away from your nose in order to reseat the mask. Or it could be that you had a 15 minute leak that was really, really bad. Only the leak line tells which case is more likely.

I know I have some leak issues which feel like they've improved immensely since working on fit ...

Ok----let's talk about how much you need to worry about these leaks from a purely practical point of view.

Leaks come in three basic varieties:

• Leaks that are LARGE enough and LONG enough to adversely affect your therapy. These obviously have to be dealt with. Are your leaks in this category? Well, the Resmed engineers believe the machine is capable of compensating for leaks up to 24 L/min as long as they're not too long lasting. The infamous Mr. Red Frowny Face shows up when the Resmed engineers believe your leaks are both LARGE enough and LONG enough to compromise both your therapy and the data the machine records. Mr. Red Frowny Face shows up when your 75% leak rate is AT or ABOVE 24 L/min. In other words, the Resmed engineers believe that if your leak rate is AT or ABOVE 24 L/min for 25% (or more) of the night, there's a real leak problem that the machine cannot cope with. And at that point, you really do need to do something to address the leaks.
• Leaks that are LARGE enough and/or LONG enough to adversely affect your comfort. These are leaks that are either not large enough or not long enough to clearly adversely affect your therapy and the data. But if a leak wakes you up, it's a problem. If you wake up with dry eyes from leaks causing air to blow in your eyes, that's a problem. If you wake up with a dry mouth, that's a problem. If you can't get to sleep in the first place because every time you move in bed the mask springs another leak that makes you more alert, that's a problem. And if you can't get to sleep because the leaks simply annoy you and keep you from drifting off into slumberland, that's a problem. In other words, if the leaks are causing you discomfort, they're a problem and they need to be dealt with so that you'll be more comfortable, and hence, sleep better.
• Leaks that are NOT large enough and/or long enough to adversely affect your therapy AND do NOT cause you any discomfort. My advice on these leaks is to not lose sleep over them. No one is going to have 95% leak rates equal to 0.0 L/min night after night after night. If that 95% leak rate is below 10 or 15 and you're sleeping through the leaks and not waking up with any problems caused by the leaks, then don't get too worried about working excessively hard to eliminate the last bit of leaking.

Since you write:

I had a headband holding the middle down and had the mask so tight there were still marks on my face 3 hours later. I know my mask doesn't fit properly and am working with my worthless and hopeless DME provider for one that fits better

I'd say that working on mask fit and comfort is a higher priority than eliminating leaks from this mask that doesn't fit very well to begin with. Do some mask shopping. Your nose (and eyes, forehead, face, and neck) will appreciate it.

[DoriC]

Thank you Robysue! Another masterpiece that answers ALL my questions about LEAKS!!

[RandyJ]

For what it's worth, I have achieved acceptable leak rates (most of the time less than 5L/min) with the Mirage Quattro FF mask with a Padacheek liner and anti-leak strap.

By all means shop for a new mask (I am currently trying to find a nasal mask and have tried 3 so far), but if you like this one, there might be a way to make it work, if you work at it.

Good luck!

[robysue]

Thank you Robysue! Another masterpiece that answers ALL my questions about LEAKS!!

Your welcome!

From the friendly neighborhood mathematician, robysue

[scottjf8]

For what it's worth, I have achieved acceptable leak rates (most of the time less than 5L/min) with the Mirage Quattro FF mask with a Padacheek liner and anti-leak strap.

By all means shop for a new mask (I am currently trying to find a nasal mask and have tried 3 so far), but if you like this one, there might be a way to make it work, if you work at it.

Good luck!

YMMV but I've found the Quattro FX to be much more comfy and less leaky. The Mirage hurt my nose something fierce and the FX doesn't at all.

Add to that a Padacheek and you're in business.

[Ineedanap]

Thank you Robsue!
Answers like these are one of the reasons I never would have made it without this forum. As time passes I think less and less of the office that handed me a mask and machine and told me to be sure to use it every day. I really don't see how people can possibly make this work without a place like this to turn to.

I don't think my mask is the right one for me at all and wanted to try the FX. I signed up for it at the mask library, a nasal mask came today, oops I tried it for a nap, it was REALLY small, unobtrusive, didn't leak. It is very clear why the sleep center pushed the nasal masks. I had to spend a lot of time trying to ignore the feeling that I wasn't getting enough oxygen and fighting the urge to open my mouth and gasp for air. I'm going to try it again to see if it gets easier.
Again, thanks to everyone who has ever answered a question so that I was able to find the answer.

[Grouch]

I'm piggybacking on this thread. I've tried to read and understand what's been said so far, but I'm not sure what the units are for my leak data. I'm only two days in on using CPAP, so I may not have my mask adjusted correctly, even though I've watched a couple of videos. I'm using a Swift FX mask. The Sleephead leak graph looks awful to me, but my other stuff is really looking much better. AHI the night before was 11. Now it's 1.74.

Should I be concerned about leakage?

[Pugsy]

Sleephead leak graph looks awful to me, but my other stuff is really looking much better. AHI the night before was 11. Now it's 1.74.

Should I be concerned about leakage?

First of all the PR S1 machines report Total leak (vent rate plus excess leak) and ResMed machines report only excess leak so don't try to compare your leak numbers with your machine to a ResMed machine's leak.

Respironics doesn't flag a large leak until around 90 L/min. It has to be a pretty good size leak to get the Large Leak flag. Your leak line is all over the place but the most it ever gets to is 66 L/min. Your baseline of no leak is around 31 L/min. So you are having some good sized leaks but not technically at the level where therapy might be impaired.
It will always show at least this baseline because that is the intentional vent rate (it may vary a bit with pressure variations...more pressure equals more vent rate..) so don't be alarmed.

They warrant some work but this was only your second night and we don't expect perfection even when we have been at this for a long time. Leaks are always going to be there to aggravate us. You will want the leak line to be more level with not so many peaks into the 60's. Your leak line isn't beautiful and does probably need some work but it isn't hugely horrible. Try to look for less large spikes and of shorter duration. It will get better with time. For night number 2 this leak is acceptable. I have seen much worse from people doing this a lot longer than you have.

[Grouch]

Pugsy,

Once again you've come to my rescue. I appreicate the detailed explanation. It makes a lot more sense now. Ironically my first night leak graph was better - mostly in the low 30's. Two things were different last night. I'd added cheek softies, so the mask might have not stuck in place quite as well. I took them off tonight. I fell into a deep sleep almost immediately, so I might not have been as aware of the mask sipping as I was the night before. And you're right, I've had no large leaks reported.

I've tried readjusting the mask again. And I've gone up to the medium size pillow from the small. The tech fitting me said I was in between sizes and I might want to go up to the medium. Will see what, if any, improvements I can get tonight.

[Ineedanap]

oops sorry typo. RobYsue

Edited to add:
Just couldn't do the nasal mask. I am comfortable with a full face mask and consciously tell myself to breath through my nose but for the first 10-20 minutes in, my sinuses are still swollen due to allergies and breathing through my nose just doesn't provide enough air! I know I can use sprays and take an allergy pill earlier (I take it at bedtime) but it just doesn't seem like a good trade off for me. I was so relieved to go back to my FF I'm glad I was able to borrow one from the mask library. Now I know for sure its not for me.

[Grouch]

I've tried readjusting the mask again. And I've gone up to the medium size pillow from the small. The tech fitting me said I was in between sizes and I might want to go up to the medium. Will see what, if any, improvements I can get tonight.

The medium sized pillow was a disaster. Ouch, ouch, OUCH! Woke up at 3 am and my nose was really sore. Had to do my first pillow change in the middle of the night. That wasn't quite how I'd envisioned it. The medium size didnt' seem to help with leaks, either. While I couldn't feel any when I first went to bed, I could when I woke up at 3.

Have been examining my leak graphs. They are all low and almost flat, when I first put on my mask. The larger peaks come later on - maybe an hour or so after starting CPAP . This leads me to think that I have figured out how to successfully adjust my head gear and pillow, but the headgear on my very minimal Swift FX may not be quite sturdy enough to hold it in place later in the night. Am thinking of trying out the Swift LT, which they used in my titration study. The sides of it look a bit sturdier.

The red lines are after I've adjusted my mask. The gap is where I was changing pillows.

[ronbien]

Version 0.8.6 of Sleepyhead allows you to set an intentional leak value so it gets removed from the graph.
It can be set in File > Preferences and under the CPAP/Visual tab which should be the default.
If you set an intentional leak value then the graphs show just leaks above that value. I believe 24 would be a good starting point for most masks.

Good Luck!

[robysue]

Version 0.8.6 of Sleepyhead allows you to set an intentional leak value so it gets removed from the graph.
It can be set in File > Preferences and under the CPAP/Visual tab which should be the default.
If you set an intentional leak value then the graphs show just leaks above that value. I believe 24 would be a good starting point for most masks.

The mathematician in me just can't let the emphasized statement stand without a response.

If you are going to use this feature in SH so that only unintentional leaks show up in the report is it absolutely critical that you consult the guide that came with your mask to find the intentional leak for YOUR mask at YOUR pressure levels.

For example: The intentional leak rate for the Swift FX ranges from about 20L/min (at 4 cm pressure) to something like 50L/min (at 20cm pressure). At 8cm, the intentional leak rate is 29L/min. And at 10cm (a middle level pressure value) the Swift FX's intentional leak rate is about 32L/min.

Most FFM have intentional leak rates GREATER than the leak rates for nasal and nasal pillows masks because there is more air in the mask itself. The so-called total face masks have intentional rates even greater than the FFM if I recall correctly.

So DON'T GUESS about the intentional leak rate for your mask. LOOK IT UP in the owner's manual.