Where's the leak?

[ZeeMan]

After 3 masks that leaked worse than a congressional committee, my current Fisher & Paykel Forma mask seems to be an excellent fit. I feel around the edges for leaks each night, and find none. The instructions say that that at my pressure - 9 PSI - it should bleed 29 LPM +/- 15%. That would allow a maximum of 33.3 L/min. The CPAP machine initially claimed it was measuring leakage in the lower 40s. So where's the leak?

After removing the humidifier module, the leakage dropped to the 36-37 range and has stabilized there. My DME supplier says that use of a humidifier loosens the mask seal, so this finding does not imply leakage in the humidifier module. Its perfectly normal for the leakage to drop that much when you remove the humidifier, so they say.

I pressure checked my hose by covering one end, submerging it (other than a few inches), and blowing hard in the unsubmerged in. I repeated the test with the other end slightly out of the water. In both cases, there were no bubbles. That clears the hose of wrong doing.

My OSA had me take the mask back to the DME supplier for a leak check. They had me wear it while they applied their blower. Its display said "EXCELLENT FIT". That would appear to clear the mask of wrong doing.

So that leaves the CPAP machine as the sole suspect. Does it leak internally, or is it just bad at measuring leakage? My DME supplier says they could send it back to the manufacturer for testing, but it would take 6 to 8 weeks and they have never found a problem on any machine this DME supplier has sent in. I'm considering sending it in anyway, just to see if I get different results on a loaner machine.

Question 1: Is it normal for a CPAP machine to report leakage of 27% above the mask's specification (ie, 37 L/min vs the 29 the mask is rated for)? I'm wondering if the mask spec of 29 should be read like "EPA Estimated Mileage... actual mileage may vary.

Question 2: I've been told that the AHI numbers can't be trusted if the leakage is high. Can I start believing the machine's AHI values now that the leakage is a mere 27% high?

Thanks in advance for your comments.

[dave21]

After removing the humidifier module, the leakage dropped to the 36-37 range and has stabilized there. My DME supplier says that use of a humidifier loosens the mask seal, so this finding does not imply leakage in the humidifier module. Its perfectly normal for the leakage to drop that much when you remove the humidifier, so they say.

I've never heard of a humidifier causing a bad leak but the air does have to flow through the humidifier so there's always room for doubt here.

Question 1: Is it normal for a CPAP machine to report leakage of 27% above the mask's specification (ie, 37 L/min vs the 29 the mask is rated for)? I'm wondering if the mask spec of 29 should be read like "EPA Estimated Mileage... actual mileage may vary.

Most mask manufacturers provide leak rates, I would check with the manufacturer. If I look at ResMed's full face mask they report 32 litres/min @ 8cm or 37 litres/min @ 10cm as acceptable. I don't know the machine you're using, whether the percentage your seeing in your stats are actually above the mask rating or whether it's disgarding the mask rating and just saying it's noticing a 27% leak?

Question 2: I've been told that the AHI numbers can't be trusted if the leakage is high. Can I start believing the machine's AHI values now that the leakage is a mere 27% high?

Yes, the machine measures the AHI (AI=full blockage of upper airway, HI=partial blockage of upper airway), so if you then simulated by this by pulling your mask away from your face to create a leak, would mean that the machine cannot accurate detect whether there is an obstruction or not, so the higher the leak rate the less accurate the AHI will be but only if you get way outside of the tolerant level.

[ZeeMan]

Thanks for your helpful response, Dave21.

Would mask leakage tend to inflate the measured AHI or lower it?

I got to wondering if the problem could be that the machine is blowing a higher pressure than I think its set to, so I just pulled out the mask's spec sheet again. To get the 37 L/min the machine is reporting, the pressure would have to be 17 according to the spec sheet. If it was really blowing 17 instead of 9, I suspect I'd know that. So another dead end.

I feel I should be able to solve this mystery with a little thought and analysis. Reality is that I'm still so tired when I get home, I just don't have the energy to give it much more effort. The AHI says I should feel great, but I don't. I feel stalemated. Its discouraging.

[jules]

can you post the daily graphs?

[ZeeMan]

Thanks for asking, Jules, but apparently not. After a lot of trial and error, I finally looked in the FAQ and found that I'd need a publically accessible web server. I don't happen to have one of those handy, and I'm too tired to think about coming up with one.

I saw my sleep doctor, and he's loaned me a RemSTAR C-Flex for a week. This thing is supposed to determine the pressure I need and automatically adjust to match it. It has a decal that says "Feel the Difference". I was sure hoping I would, but so far I don't. That suggests: 1.) my problems are probably not the fault of my own CPAP machine, and 2.) I may not have an optimal pressure that fixes everything. He also perscribed a mild stimulant that I'll be experimenting with this week.

I also saw my internist. He's checking me for thyroid level and lots stuff like that. I failed to take notes this time. None of the possiblities sound life threatening, but neither are they heath conditions you'd want to buy if they were offered on ebay. I would have preferred for this to be resolved by a simple repair or adjustment to the CPAP machine, and I'm disappointed that it might not go that way.

So I don't have more data to explore just now, or anything else fun. I just felt I should post something out of respect to Jules who took the time to respond to my earlier post. Thanks again, Jules.

[ZeeMan]

I can and did post my cpap data without the use of an external server. Its in my avitar!

Interesting part is the AHI trend line. It was in the mid teens as recently as 2 months ago, and then it began a slow decent that eventually bottomed at 2.1. Last 7 day average before I switched to the loaner machine was 2.9. I suspect the movement between 2.1 and 2.9 is noise related to my never ending quest to find a good sleep position.

What's most intriguing is the slow descent from 14 or wherever it was down to 2.1. Does anyone have a theory on that? My weight has actually been increasing somewhat during the period that AHI was gradually dropping, so that's not it.

My sleep doctor basically blamed the CPAP machine and said there are no journal articles in the trade literature supporting the reliability of AHI displays on CPAP machines. That slow decline over a couple months just doesn't look like a software or sensor problem, so I'm less confident that my machine caused that trend line. But what would cause it? And if my AHI really did drop to 2.1, why do I still look and feel lousy?

[jules]

sign up for a photobucket account and upload you data there - your avatar is not big enough to look at - okay?

once you have uploaded the data, you can click on IMG option in photobucket and copy and paste that into here - if you get lost on that post again and someone will help you

[ZeeMan]

Ages ago, I started this thread because my therapy was going poorly and my CPAP was reporting really high leakage numbers. I could not find the leak.

The mystery was solved when I used two different loaner machines for a week each. Both reported no significant leak. When I went back to my own machine, it went on reporting high leakage numbers. I think my machine lies. Anyway, mystery solved.

Sorry to take so long following up. I'm still struggling, and I'm working extra hours. Just hard to keep up with things. Hopefully I'll have an auto pap soon and things will get better.

[dave21]

That's good to hear, thanks for the update! It's possible that the machine is faulty, you might want to get the machine services by the manufacturer. They'll probably charge for it, but it would be a way of hopefully fixing it.

Thanks
Dave