Leaks - how much is too much?

Hey Trout - the next time you want to talk about how good you're doing be sure to do it via PM. When you post to the world your machine and mask read it and they'll see to it you take 4 steps backward. Every time I see real progress I turn around quickly to see who's looking. Keep up the good work.

jimnsc wrote:Hey Trout - the next time you want to talk about how good you're doing be sure to do it via PM. When you post to the world your machine and mask read it and they'll see to it you take 4 steps backward. Every time I see real progress I turn around quickly to see who's looking. Keep up the good work.

I didn't think my machine had a wifi adapter and ability to read the forum, but you never know...

Pugsy wrote:Yes, that leak is a problem. Roughly half the night your were very close to that red line or above it where ResMed says it probably can't handle things.

If that continues, I'll call the DME about a chin strap. Can't wait to see how they react when I talk about my graphs!

Sheriff Buford wrote:Trout: I was thinking your 95 percentile was the pressure setting the machine was putting out 95% of the night. Apparently after reading the other posts, it was the leak rate... (95% of the night). So it does look good.

As far as the things in the trunk.... you wouldn't mind if I looked would you...?

Just to clarify the correct mathematical meaning of "95 percentile" of anything: In a given set of data (numbers) the 95% is the number for which 95% of the data is AT or BELOW that number. (And hence 5% of the data is AT or ABOVE that number.)

So "95% Leak Rate = 3.2" means that for 95% of the time the S9 was running the detectable UNintentional Leak rate was AT or BELOW 3.2 L/min and 5% of the time the S9 was running the UNintentional leak rate was AT or ABOVE 3.2 L/min. And how to easily determine how long the "large" leaks (those above the 95%) lasted? Well, 5% of one hour equals three minutes. So you can just multiple the usage hours by 3 to get the total time in minutes that your leak rate was Above the 95% leak rate.

For APAPs, the 95% pressure level is similar: If the 95% pressure is 8.7, then for 95% of the time the S9 was running, the pressure was AT or BELOW 8.7 and the pressure was AT or ABOVE 8.7 for 5% of the night.

today's math lesson brought to you by robysue, the friendly math prof

I realize that this discussion is about leaks, but you appar to have had 1 central and 1 hypopnea for the entire night even with all those leaks.

That's incredible!

What do you get on a good night without the leaks?

John

robysue wrote:today's math lesson brought to you by robysue, the friendly math prof

Thanks for putting this in context robysue, it really helps! So, last night I was above 28 l/m for close to three hours.

Suuuuuuper.

70sSanO wrote:What do you get on a good night without the leaks?

So far--and take this with a grain of salt because I'm a newbie with only a week under my hose--I usually have the same results, maybe one or two more, maybe less. I also have a night with zero events. On the 17th though, I was less leaky and had 5 hypopnias, 1 OA, and 1 CA. Any way you slice it, my events are really few and far between when I'm masked up.

Kilgore Trout wrote:

robysue wrote:today's math lesson brought to you by robysue, the friendly math prof

Thanks for putting this in context robysue, it really helps! So, last night I was above 28 l/m for close to three hours.

If your 95% leak rate was 28 on the LCD, it is IMPOSSIBLE for you to have been ABOVE 28 L/min for three hours:

(time in Leak ABOVE 95% leak rate in minutes) = (3 minutes per hour)*(usage in hours)

And 3 hours is 180 minutes. So we get:

180 minutes = (3 minutes per hour)*(usage in hours)
180/3 = usage in hours
60 = usage in hours

And there's no way you could sleep 60 hours in one night. Now, if your MEDIAN leak rate (which is shown in ResScan) was 28 L/min, then that's a horse of a different color and QUITE possible physically speaking: If the Median Leak rate = 28 L/min, that means that you were leaking AT or BELOW 28 L/min for 50% of the night and you were leaking AT or ABOVE 28 L/min for 50% of the night since "median" means the MIDDLE data point---the one where half the data is above the value and half the data is below the value. So a MEDIAN Leak Rate = 28 for a six hour sleep period means the leak was AT or BELOW 28 L/min for half the night and AT or ABOVE 28 L/min for half the night. And half of six hours is three hours.

However, in eyeballing the graph you posted, it looks to me like your 95% was indeed about 28 L/min on that night. Since you had the machine running approximately 6 hours, that means you were leaking AT or ABOVE 28 L/min about 3*6=18 minutes (not three hours=180 minutes). More closely eyeballing the same data shows that you were AT or ABOVE the RedLine (24 L/min) for a TOTAL time of about an hour and fifteen minutes. That means you were AT or ABOVE the Red Line for about (1.25/6)*100% = 21% of the night. In other words, 24 L/min is pretty close to your 80% leak rate: You were leaking AT or Below 24 L/min for 80% of the night and you were leaking AT or Above 24 L/min for about 20% of the night. That's pretty close to Mr. Red Frowny Face's standards for showing up, but still just barely in Mr. Green Smiley Face's territory.

Since you were leaking At or Above the RedLine for a significant chunk of the night, the recorded data may not be accurate. In other words, it's quite possible that the leaks were both LARGE enough and LONG enough to cause the S9 to not properly detect respiratory events during the LARGE LEAK periods.

robysue wrote:If your 95% leak rate was 28 on the LCD, it is IMPOSSIBLE for you to have been ABOVE 28 L/min for three hours:

I have no idea where I got that extra zero from; you're completely right. Blame it on Friday brain?

Pugsy wrote:

Kilgore Trout wrote:*whew* thanks for the reassurance! What numbers should I be mindful of?

In regards to leaks. Remember ResMed machines already subtract the intentional vent/leak rate so what you see is all excess leak. ResMed says it can handle up to 24 L/min before it questions therapy and event scoring.
Of course you want to stay away from that 24 L/min.

Hi Pugsy and everyone,

I finally got to log on again. Not being able to connect with the other CPAP users here has been tough! I just checked my machine and saw that my AVERAGE leak for the last YEAR was 24L/min. Not sure of what to make of it. Maybe it's just as bad as not using a CPAP?

Maybe I need to find a new doc with a decent tech? I'm so tired all the time...

DeeCPAP wrote:Hi Pugsy,
I finally got to log on again. I checked my machine and saw that my AVERAGE for the last YEAR was 24L/min. Not sure
of what to make of it. Maybe finding a new doc with a new tech?

This 24 L/min that's off the machine itself?
The leak number shown on a ResMed machine LCD displace is NOT the overall average...it is what we call a 95% number and all that means is that for 95% of the time your leaks were at OR BELOW the 24 L/min. 5 % of the night you were above that number. No way to know how high without using the software.

95% numbers are easily skewed to the high side with just a little bit of time spent with large leaks...
So I don't get all in a panic over a 95% number that is around 24 because I know that there's a good bit of time spent well below that red line.

Pugsy wrote:

DeeCPAP wrote:Hi Pugsy,
I finally got to log on again. I checked my machine and saw that my AVERAGE for the last YEAR was 24L/min. Not sure
of what to make of it. Maybe finding a new doc with a new tech?

This 24 L/min that's off the machine itself?
The leak number shown on a ResMed machine LCD displace is NOT the overall average...it is what we call a 95% number and all that means is that for 95% of the time your leaks were at OR BELOW the 24 L/min. 5 % of the night you were above that number. No way to know how high without using the software.

95% numbers are easily skewed to the high side with just a little bit of time spent with large leaks...
So I don't get all in a panic over a 95% number that is around 24 because I know that there's a good bit of time spent well below that red line.

Yes, it was off the machine itself. I'm still way to tired than I should be. I had a home sleep study and my blood oxygen dipped down to 82% for a little while. and average was 90%. I was thinking it might be interesting to take my blood oxygen level overnight while wearing the machine, but not sure how to get it done. It's a feeling that something's wrong. I was also recently diagnosed with glaucoma and enlarged optic nerves which may have something to do with sleep apnea according to this article:


https://www.aao.org/eye-health/news/stu ... leep-apnea

Another possible problem my eye doc is now tracking is macular degeneration:

http://macularhope.org/sleep-apnea-affects-eyes/

DeeCPAP wrote: I was thinking it might be interesting to take my blood oxygen level overnight while wearing the machine, but not sure how to get it done

You would need an overnight recording pulse oximeter.
http://www.pulseoxstore.com/overnight-pulse-ox