Hi from new member (tried taping my mask)

Ok - shaved again last night, and when I masked up last night I made a point of laying awake for a while paying attention to the mask. I noticed that it was constantly on the verge of leaking, and also that the airflow out of the vents in the front of the mask remained consistently high. The only thing I could think of doing was to tighten the straps a bit, which seemed to stop this leakage.

I have noticed that the machine will try to maintain the target minimum pressure within the mask at all costs - if there is a leak, then it will have to increase flow rate to do this (V=IR). In my case this seems to be causing the machine to rev like an F1 race car due to the level of leakage (which generally started after I fell asleep, which is usually instant).

I think I have invented a new CPAP sin, which I will dub "strap winging". I had tightened my mask, then loosened it, and suspect that the combination of stubble and loose straps was behind the crazy leakage (which renders the number meaningless, as Pugsy has kindly pointed out). Last night was the first night with both a clean shaven face (right before bed) and a slightly tighter strap setting. The leaks seem to have been better managed throughout the night, although Sleepyhead still points out that the mask is leaking way too much.

So I am back to 15+ territory as demonstrated in last night's data:



I have been feeling better, but think this might be mainly diet and weight loss (not that I'm complaining). I'm now down over 5lbs, am enjoying my food, and have had blood ketones as high as 1.6 (I think I used to spend a lot of time in an insulin fog...) I'm also just about fully off the SSRI and no longer take Concerta (which still has a post-clearance 'crash' phase even though it's time release and not as bad as actual ritalin in that respect).

The realization that at least some of my numbers have been meaningless leads to some important questions:

• * what should I do with my titration? correct the mask leaks then start over at 7 cmH2O?
  
  * is it possible that I have my pressure set too high now, and that a lower pressure might actually give a better AHI? (alternate version of this question below)
  
  * as minimum pressure is increased (during titration) from the initial low starting point, does AHI decrease to a minimum value and then start increasing again as min pressure goes past the 'optimal' value? or does AHI reach a minimum and then just stay there with no additional change as pressure is increased further? (I'm trying to understand what would happen if minimum pressure is set too high)

My next task after clicking 'Submit' will be to order some mask liners...

All I can say is I don't know what I would do without the guidance from you guys!

Did we ever change your SleepyHead to the correct red line number for your machine? I don't remember.
SleepyHead is showing 27% of the time over the red line last night which looks to be about right when I look at the one large leak block. So maybe I already had you change that red line number (SH defaults to ResMed's 24 L/min red line and that won't work so great for SH statistics in terms of % of time in large leak for Respironics users).

Just in case I didn't tell you...or other Respironics users reading this thread are seeing some time spent in large leak on the statistics that isn't supported by the events graph large leak flagging.
SleepyHead defaults to the ResMed 24 L/min red line and SH compares the leak number to that 24 when it gives the % of time over red line statistic and it's not accurate for Respironics users.
Respironics users need to go to Preferences/CPAP tab and change the 24 to something more in line with what Respironics machines have and it can't be exact because they never give us an exact number but something around 75 L/min would get them in the ball park with probably a little safety cushion.
Or you can just turn that feature off in SH by removing the little check mark.

I discovered this little tidbit by accident one day when I notice SH said I was having massive leaks but I wasn't seeing the large leak flags on the events graphs to support it...wondered what the heck.
I have both ResMed and Respironics machines and when I was using the Respironics is when I would get SH telling me that I was going to hell in the leak department...then it dawned on me that red line number can't be the same for both machines. Once I adjusted the red line number my statistics for leak time over the red line became more in line with the flagged large leak times, which BTW with Respironics I rarely have and with ResMed I almost always have at least a tiny time frame where I hit large leak.
Unfortunately there is no way for SH to divide up that red line number so that it is correct all the time for people like me who might use 2 brands of machines on occasion. Changing it will change all past data also.

So short of keeping 2 profiles separate with one being for the ResMed machine and one being for the Respironics something is going to be slightly incorrect when combining the brands in one profile. Since my primary machine is the ResMed I have elected to leave it at 24 L/min red line and when I use the Respironics machine and it tells me I went to hell in the leak depart I just ignore it.
It's easy enough to take a quick look at the events graph and if there aren't any large leak flagged time periods..then we know the statistics aren't accurate.

If you want to experiment with a liner before you get them, find an old t-shirt and cut a gasket from it. Make the inner hole much too small and enlarge it with fittings.

Or you could try one of these ....



Though he doesn't look to thrilled about wearing that thing.

and imagine the skepticism his wife has

DreamStalker wrote:Or you could try one of these ....



Though he doesn't look to thrilled about wearing that thing.

and imagine the skepticism his wife has

I like it - is it soundproof?

leptic wrote: I like it - is it soundproof?

Don't know. I use the Swfit LT -- It's been working for me for over 8 years now. I did have to get new head gear straps for it about 6 months ago because the original started fall apart.

Ok - I did an experiment and tightened the straps to minimize leaks (subjectively it did feel like the leaks were a lot less, although not zero).

Here is a period with looser straps, huge leaks:



This is with tightened straps, lower (although not zero) leaks:



Am I correct to suppose that the apparently low numbers (e.g. 3.33) are garbage due to the massive leakage, and that the 20+ values more correctly reflect reality? I suspect I was wearing the straps looser than needed - they probably broke in a bit initially and I never changed it. In the latter night they are snug but not uncomfortably so.

My concern now is that the 'titration' I thought I was doing was probably meaningless...

leptic wrote:Am I correct to suppose that the apparently low numbers (e.g. 3.33) are garbage due to the massive leakage, and that the 20+ values more correctly reflect reality?

Unfortunately you are correct. Those nice low AHI numbers weren't representative of what was actually going on.
It's why we harp on getting the leaks under control. When they totally miss flagging of events it can give someone a false sense of security that their therapy is going well.
Short lived large leaks aren't the end of the world but your times in large leak along with how deep in large leak territory you go pretty much makes the AHI worthless.
Half the night it is accurate (on the June 25 report) but the other half is totally unreliable.

Now about the June 26 report...I see a break in therapy around 4 AM with some ugly clustering for about 45 minutes before that break and about 45 minutes after that break...were you awake/semi awake or asleep during each of those 45 minute periods? If you were awake then those events don't count in the AHI and you would need to mentally remove those awake/semi awake events from the AHI evaluation and if that is the case then the AHI isn't nearly as bad as first thought.

Getting the mask to fit comfortably and without leaks all night is the hardest part of this therapy.

When it comes to mask fit, the thing to keep in mind is that manufacturers design their masks to fit the “average” person but who is average? I’m certainly not but I digress. There are three basic types of masks – a full face like the one you seem to be using, a nasal cushion mask, and a nasal pillow mask (yes I know there are other variants but the three I mentions are the most common).

Another thing to keep in mind is that a full face mask has a greater amount of surface on the face to seal and consequently, there are more places for the seal to leak. Also, the full face mask headgear is designed so that your skull is the stabilizing platform for the mask and yet your lower jaw articulates and is not part of that stable platform ... hence leaks are prone to occur around your lower jaw as you relax while asleep. Furthermore, the headgear strap tension changes depending upon your head to neck position ... just try and do a neck flexion and extension (forward and sideways) with a full face mask on and notice the change in tension straps. So unless you sleep in one position all night long, those tensions are going to be changing all night as you toss and turn and behold ... mask leaks.

As for nasal cushion masks, for me they did not work -- although they did not have the lower jaw and strap tension issues. For me, the cushion seal around my sinuses would cut off blood flow circulation causing nasal sinus congestion and restricted airflows. For others this may not be an issue.

That left me with the nasal pillow options. There are different designs to choose from here as well and so I tried several until I found the one that fit and worked best for me (Resmed Swift LT). Then the problem was that I had mouth leaks. But I finally figured out how to fix that though it took a few months of experimentation – (finally managed to seal my mouth from air pressure by training my tongue to stay planted on roof of my mouth).

So yes, the mask is the hardest part to get down, but once that is achieved, the rest of the therapy approach is quite easy ... optimizing machine pressure(s), sleep hygiene, rebalancing hormonal regulation through diet and exercise, etc.

Pugsy wrote:

leptic wrote:Am I correct to suppose that the apparently low numbers (e.g. 3.33) are garbage due to the massive leakage, and that the 20+ values more correctly reflect reality?

Unfortunately you are correct. Those nice low AHI numbers weren't representative of what was actually going on.
It's why we harp on getting the leaks under control. When they totally miss flagging of events it can give someone a false sense of security that their therapy is going well.
Short lived large leaks aren't the end of the world but your times in large leak along with how deep in large leak territory you go pretty much makes the AHI worthless.
Half the night it is accurate (on the June 25 report) but the other half is totally unreliable.

Thanks - this is scary because it suggests I may never have had an effective night of therapy even though it's been around a month. I suppose this is in keeping with the notion that it can take a few months to dial everything in.

Now about the June 26 report...I see a break in therapy around 4 AM with some ugly clustering for about 45 minutes before that break and about 45 minutes after that break...were you awake/semi awake or asleep during each of those 45 minute periods? If you were awake then those events don't count in the AHI and you would need to mentally remove those awake/semi awake events from the AHI evaluation and if that is the case then the AHI isn't nearly as bad as first thought.

I just can't remember if I was semi-awake during those period... as I've mentioned my sleep has always felt extremely fragmented. I don't know if this fragmentation is due to real sleep apneas causing semi-arousals, or if something else is creating a semi-awake state in which the breathing is not accurately parsed by the CPAP. I would say that, during these periods, I am definitely not in volitional control of my breathing. Rather it would be more the kind of reflexive breathing I'd associate with sleep.

The diet changes I've outlined in my other thread have been a real paradigm shift, so it's possible the increased wellbeing I've been reporting is mainly due to diet/weight loss/hormonal (insulin). I've also greatly reduced my load at work and have come off a couple of meds might have been counterproductive.

Pugsy wrote:

leptic wrote:getting the leaks under control

One more question - what are the units in the Leaks plot, and what do the two curves really mean (i.e. total leaks vs. leak rate)?

Are the lower numbers in the 50's from the second night ok, or still too high? Is there a certain maximum leak rate that's generally accepted as acceptable across different CPAPS and masks?

I think I'll try the nose mask again today during the day and see how it works out at night.

DreamStalker wrote:Getting the mask to fit comfortably and without leaks all night is the hardest part of this therapy.

When it comes to mask fit, the thing to keep in mind is that manufacturers design their masks to fit the “average” person but who is average? I’m certainly not but I digress. There are three basic types of masks – a full face like the one you seem to be using, a nasal cushion mask, and a nasal pillow mask (yes I know there are other variants but the three I mentions are the most common).

Another thing to keep in mind is that a full face mask has a greater amount of surface on the face to seal and consequently, there are more places for the seal to leak. Also, the full face mask headgear is designed so that your skull is the stabilizing platform for the mask and yet your lower jaw articulates and is not part of that stable platform ... hence leaks are prone to occur around your lower jaw as you relax while asleep. Furthermore, the headgear strap tension changes depending upon your head to neck position ... just try and do a neck flexion and extension (forward and sideways) with a full face mask on and notice the change in tension straps. So unless you sleep in one position all night long, those tensions are going to be changing all night as you toss and turn and behold ... mask leaks.

As for nasal cushion masks, for me they did not work -- although they did not have the lower jaw and strap tension issues. For me, the cushion seal around my sinuses would cut off blood flow circulation causing nasal sinus congestion and restricted airflows. For others this may not be an issue.

That left me with the nasal pillow options. There are different designs to choose from here as well and so I tried several until I found the one that fit and worked best for me (Resmed Swift LT). Then the problem was that I had mouth leaks. But I finally figured out how to fix that though it took a few months of experimentation – (finally managed to seal my mouth from air pressure by training my tongue to stay planted on roof of my mouth).

So yes, the mask is the hardest part to get down, but once that is achieved, the rest of the therapy approach is quite easy ... optimizing machine pressure(s), sleep hygiene, rebalancing hormonal regulation through diet and exercise, etc.

Thanks - I suspect I've been spinning my wheels without enough consideration of the mask/leaks. The good thing is that your dietary advice seems to have had a big impact! If I can get sleep down, I'll be leaping tall buildings (so again thanks to you, Pugsy for all the support and advice!!)

Leaks are L/min and with the Respironics machine the top leak line is total leak which is the mask's expected vent/intentional leak plus any excess leak. So that number is always going to be up there somewhere.
Large leak territory line in the sand will vary because the intentional leak rate will vary with pressure...more pressure means more vent rate.

The bottom leak line is a close approximation of the excess only leak but Mark (who is the father of SH) had trouble with the calculations involved in reporting the excess only leak line with the Respironics machines because of that auto thing...when pressure increases the vent rate increases.
Plus Respironics never gives us an exact "number" for either total or excess only leak for when large leak territory begins. So that bottom leak line is going to be a close approximation of excess only leak but it's pretty much meaningless because we are never told what numerical value in terms of excess leak is something to avoid.

About the best we can do is maybe get lucky and figure out where the machine flags a large leak and when it doesn't. Like in the past I found times where I hit 85 L/min total leak and didn't get a LL flag but at 90 L/min I would get a large leak flag. So it's safe to assume that at whatever pressure I was using (assuming the same pressure) that large leak territory was up around 90 L/min.

Did you ever change the leak red line threshold in SleepyHead from the default 24 L/min? That's what the statistics leak % comes from. SleepyHead defaults to the ResMed 24 L/min excess only leak and that's not the same number that Respironics uses....remember we are never given any set number since it varies.
I can't seen any large leak flags on the events graph on the June 26 report yet I see 1.68 % of the time spent over the red line threshold. If you actually had a large leak per the machine...we should have seen a large leak flag. Now maybe I just can't see it on the image but if there are no large leak flags showing on the Events graph....that statistics % number should be 0.0. We could change it to something closer to what the actual excess leak line in the sand is...if we knew what it was but we just don't know.
I do know that Respironics machines have a much greater cushion for before flagging a large leak than ResMed machines. I have used both brands...with Respironics I hardly ever see any sort of large leak flag but with ResMed I almost always see a short time above 24 L/min.

Since the bottom leak line is only a close approximation for excess leak only and since we don't know exactly how Respironics machines figure large leak...we have to be happy with relying on the LL flags on the Events graph.
To make the statistics % be more accurate....go to Preferences/CPAP tab and either turn that statistic off or change it from 24 L/min to something at is probably closer to the real line in the sand for Respironics and that's probably going to be up around 40 to 50 excess only leak but it's just a guess. We simply don't know how much excess leak is too much for Respironics and a LL flag gets earned.

So....look at the events graph for evaluation of leaks as the ultimate evaluation point and not the statistics % number.

Those 2 times where I wondered if you were awake or not...you have a bunch of centrals/CAs along with the OAs and the machine won't respond to the CAs and it looks like that some of the time it isn't responding to the OAs which makes me wonder if all this was awake/semi awake breathing irregularities getting flagged by mistake.

Could you do me a favor and zoom in on a CA cluster shortly before and after that 4 AM mark...go to the events tab and then the CA category...click on it and look to see the various times where the CA was flagged.
Pick one with a couple before and after it in close proximity time wise. Click on that one and watch the flow rate graph changed to a much more zoomed in image...get me a screen shot of that image.
Do one right before the break in therapy and after the break in therapy. Hopefully you can get a zoomed in section with 2 or 3 CA flags on it.

Thanks for info on leaks!

Pugsy wrote:Could you do me a favor and zoom in on a CA cluster shortly before and after that 4 AM mark...go to the events tab and then the CA category...click on it and look to see the various times where the CA was flagged.
Pick one with a couple before and after it in close proximity time wise. Click on that one and watch the flow rate graph changed to a much more zoomed in image...get me a screen shot of that image.
Do one right before the break in therapy and after the break in therapy. Hopefully you can get a zoomed in section with 2 or 3 CA flags on it.

Ok - here's the last one before 4am wakeup that has several CA's in a row:



and the first after 4am with multiple CA's:



I must say it hardly feels like I'm the one doing you a favour here still floored by the generosity on this site.

The absolute value of the leak rate is less important than the variability of the leak rate throughout the night ... in other words, you want your leak graph to show as close to flat line as possible. Obviously you do not want the leak to be flat line off the scale because the mask has fallen off your face and as Pugsy noted, you do want to be in the ball park of the estimated vent rate for the mask (each mask is different based on machine pressure setting ... your mask manual should have come with a graph to estimate vent rate based on CPAP pressure).