Machine not registering leaks?

The last few nights, I wake up with what feels to me like significant air leaks. I don't think I'm imagining them because in the morning, my eyes are dried out and bloodshot. I also have to adjust the mask during the night and normally pull it away from my face and re-sit it to adjust it which briefly causes air leaks. I would think this would register as a leak. However in the morning when I check the machine, it shows zero leaks. My AHI for the week shows as 1.0 and for last night, 1.2. The AI for both day and week shows as zero. I'm still using the Med. comfort gel mask until I go in on Wednesday to exchange it for the Small Quattro. The mask setting was set by DME to Mir Full. Last night was a bad night with what felt like many leaks. I noticed my pressure has been pretty steady the last 9 days @ 12 and last night it was 14. Because the machine is set to Auto, is the machine adjusting the pressures to offset the leaks and therefore showing as zero on the readout? If I get the software, will it reveal more detailed information?

I'm going in on Wed. for the new mask but don't want to even admit to the DME that I've gotten into the results menu -- I'm not yet an "illegal tweaker" but feel like I'm an illegal reader!
Am I worrying for nothing?

SnoozyWoozyCat wrote:I'm not yet an "illegal tweaker" but feel like I'm an illegal reader!
Am I worrying for nothing?

We can all chip in to your defense fund.. I think you are worrying about nothing but then I am (or was till my changes were sanctioned by doc) and illegal tweaker. I think the biggest fear DME's may have is not that you see the data but the realization that if you know how to get the secret data viewed, then you know how to access the clinician menu.

Regarding the leaks and zero average showing up on the machine. It actually takes quite a bit of leak either large leak for a sustained length of time or a moderate leak for even longer to impact the average you see on the machine. Briefly fiddling with the mask won't make any significant impact on the overall average.

One morning, very early, I woke up to find I was having issues with the nasal pillows. Air blowing everywhere so I kept trying to get them to seat comfortably. I couldn't seem to get it right. Adjusted head gear several times, tilt of the pillows several times, pulled them out several times. All over the span of nearly 10 minutes. So I had many, many episodes of air blowing all over the place. I have the software. None of this showed on the graphs and the overall average was still right at my expected leak/flow rate which equates to zero leak average. In other words, don't worry about that either..

Pugsy wrote

One morning, very early, I woke up to find I was having issues with the nasal pillows. Air blowing everywhere so I kept trying to get them to seat comfortably. I couldn't seem to get it right. Adjusted head gear several times, tilt of the pillows several times, pulled them out several times. All over the span of nearly 10 minutes. So I had many, many episodes of air blowing all over the place. I have the software. None of this showed on the graphs and the overall average was still right at my expected leak/flow rate which equates to zero leak average.

I'm surprised by your observation.

The machine leak data (recorded every 30 seconds) provided to my EncorePro reporting software is contiguous from blower on to blower off and stored in the EP created SQL tables as time-stamped 30-second reporting intervals. Whenever the leak rate changes (remember that it only changes in steps of 7 LPM) by at least one step either up or down from one 30-second interval to the next, I observe an up or down step in the leak curve displayed by EP or my own Excel Charting program. When I fiddle with my mask to stop a leak and my fiddling spans more than one 30-second reporting interval, I see leak spikes. I'm surprised that you don't.

I noticed that you have EncoreViewer. My guess is that EV may be processing the machine recorded data that it receives as input differently than EP (note that it receives the exact same input as EP). It must be "averaging the data" that it reports over a longer duration of time than the machine's 30-second reporting interval. I don't own EV so I can only speculate as to why you are not seeing leak curve spikes at the times you are fiddling with your mask.

cinco777 wrote:When I fiddle with my mask to stop a leak and my fiddling spans more than one 30-second reporting interval, I see leak spikes. I'm surprised that you don't.

I noticed that you have EncoreViewer.

I actually have Encore Pro 1.8 now along with Viewer. I don't remember exactly which day it was that I did all the fiddling or I would go get some copies to post to look at. I just remember thinking that I probably am going to have some awful leaks doing this and I was so surprised when I didn't. I don't have the desire or inclination pick every graph apart so I only gave it a passing glance. I don't have the patience.. So I didn't make any special notations about that night. Should it happen again or if I happen to remember, I will post it. If I remember correctly I only saw a tiny rise in the leak and certainly not a large spike. Mostly maybe a 3 L/min rise. I didn't go back to sleep though. Finally said to heck with it (well, that is the cleaned up version) and got up. So I do know for sure that I did fiddle with things for quite a while but maybe the time off my face was so minimal it just didn't register.... I don't know. All I know is what I felt, did and saw on the graph.

Mostly maybe a 3 L/min rise.

On a Respironics machine, the recorded/reported/plotted Leak values are 4, 11, 18, 25, 32, 39, 46, 53, 60, ... The machine recorded/reported leak value is a step function due to the measured leak granularity (7) of our machines. If you look closely at your leak charts, you may be able to discern that the difference in your plotted leak values are always multiples of 7 LPM.

The system leak averages - not the plotted leak values - that are reported by EPA & EPA (and likely EV) are weighted averages of all the 30-second time-stamped intervals in the sleep session (blower on to blower off). The weighted averages can be any number: 34.16, 51.18, whatever, based on the weighted average calculation using all the recorded/reported leak values: 4, 11, 18, 25, .... in that sleep session.

I hope this clarifies how our Respironics machines record & report leak values, and how EP & EPA (and likely EV) plot the leak graphs (and calculate the avg system leak).

Please note, as I have posted in prior threads, that the LCD system leak averages (7-day and 30-day) are simple averages. In case anyone is not clear about the difference between simple and weighted averages, I posted an example in an earlier post on LCD#s accuracy.

cinco777 wrote: The system leak averages - not the plotted leak values - that are reported by EPA & EPA (and likely EV) are weighted averages of all the 30-second time-stamped intervals in the sleep session (blower on to blower off). The weighted averages can be any number: 34.16, 51.18, whatever, based on the weighted average calculation using all the recorded/reported leak values: 4, 11, 18, 25, .... in that sleep session.

I hope this clarifies how our Respironics machines record & report leak values, a

I was recalling from memory a visual thing..... Trying to maybe recall the tiniest of pressure notation on a graph. I never bother with the leak averages except to eyeball them in passing. I have no interest in dissecting the data. I will leave that to you and Velbor and the others that really enjoy that stuff. I do not. It is not my cup of tea.

I only mentioned what I personally saw and did. I did fiddle with the mask off and on for well over 5 minutes and there were no huge glaring leak spikes nor seemingly any alterations to the leak average. I use APAP, my flow rate varies each night as the pressure may vary just a bit.

So I know that it is possible to quickly re-adjust a mask repeatedly for a few minutes and not show any grossly wild leak lines. It was a few seconds each time out of 5 to 8 minutes and the 5-8 minutes was out of 7 1/2 hours.

I cannot supply the empirical data that you like because I just don't have it. You will just have to accept that I reported what I saw and that is all I can do. I don't remember what night it was. Maybe my "3 L/min" guess was off, probably was. I can say with absolute certainty that there no leak spikes showing that were anywhere remotely near a large leak.

I posted a somewhat detailed description of Leaks in an attempt to clear up any confusion on how our Respironics machines and software record, report, and graph the leak values, and how EncorePro, and I suspect EncoreViewer, calculate the average leak value for reports. I thought it might be helpful to some members to have a better understanding of the underpinnings of the leakage graphs. If someone doesn't need this level of detail, just ignore my posting.

cinco777 wrote:I posted a somewhat detailed description of Leaks in an attempt to clear up any confusion on how our Respironics machines and software record, report, and graph the leak values, and how EncorePro, and I suspect EncoreViewer, calculate the average leak value for reports. I thought it might be helpful to some members to have a better understanding of the underpinnings of the leakage graphs. If someone doesn't need this level of detail, just ignore my posting

I thought it was aimed at me because you included a quote from me. If you had not included the quote from me I wouldn't have gone off on the long tirade of why I am not geared along those lines. I think the way you all can dissect this stuff is wonderful but while not above my level of comprehension, right now I have had some personal stuff going on and I just simply can't put forth the effort needed to correlate all the details.

I am really, really stressed right now and my plate is full. So no hard feelings I hope. Sometimes I can be such a bitch and not mean to. Then of course sometimes I mean to but not this time.

Okay, this is about as scientific as I can get right now. Woke up too early again. Lay in bed and decided to fiddle with things to see what showed up on my graph. Bear in mind it is 4:30 AM and dark. I reseated the nasal pillows 10 times. Pulled them away from face and put them back with approx 30 seconds between each reseating. Roughly 5 minutes or so total time.

Below is the Encore Pro report. I ended the experiment by turning off the machine before final removal of the mask and got up. I did look at same report from Encore Viewer with the overall average leak reported at 27.00 L/min and with Encore Pro it seems to be 26.88. So tiny difference there. The graphs are identical visually though.

Pugsy wrote

Okay, this is about as scientific as I can get right now. Woke up too early again. Lay in bed and decided to fiddle with things to see what showed up on my graph. Bear in mind it is 4:30 AM and dark. I reseated the nasal pillows 10 times. Pulled them away from face and put them back with approx 30 seconds between each reseating. Roughly 5 minutes or so total time.

Below is the Encore Pro report. I ended the experiment by turning off the machine before final removal of the mask and got up.

Thanks for posting some details on your "mask fiddling" experiment and the resulting EP leak graph. However, this is not enough information for me to determine what is taking place re the mask leak rate and what was recorded by the machine and passed on to the EncorePro reporting software. This is not a critique of your efforts just my stating that I don't have enough information to reach a supportable conclusion. OK.

When I monitor and analyze my personal sleep data and reports, I have more data/reports/tools at my disposal than the average user. My motivation for developing my own charting program (which does more analyses than charting) was that I found many discrepancies in the EP and EPA reports. For one, I couldn't figure out how some numbers were derived/calculated as there seemed to be a magical component in some of the numbers. To date, I've found close to 20 "things" that weren't right. Based on my findings, I can easily understand the often expressed member confusion with the graphs, charts, numbers, etc. reported by the machine (LCD display) and the reporting software (EP, EPA, and EV). They confuse me too! I'm curious and I'm trying to figure things out and, when I do, provide an explanation. I'm not altruistic about my efforts - I just enjoy figuring things out and making sure that 1 + 1 = 2.