Different masks = Different numbers ??

I'm using a pressure of 6-9.6 cm on my APAP but get vastly different results using different masks.

Even though i have zero leaks and I've tried all EPR levels i notice the following trend.

With my Swift LT i get an AHI 4 - 9

With my Quattro i get an AHI of 10 - 25

I also have the correct mask setting on the machine. So I'm wondering why the numbers are so different ?

What mask was used during your titration??

The Swift has a length of hose off the nasal pillows that is narrower at the pillows than the standard hose from CPAP to mask.

Does the VPAP machine work with a Breeze mask?

It can, but some of the features may be affected. The Breeze system has a very small inlet hose, which creates very high resistance to flow. This high resistance can cause the mask alarm, Smart/Start/Stop feature, and compliance data functions to be compromised. The treatment pressure (at the mask) may also be slightly less. If none of these features are important to the user or clinician, then the effect is unimportant.

And, it is not unusual for 1-4 cms higher pressure to be needed w/a full face mask over a nasal cushion mask.

So - here you are using a mask w/a small inlet hose and switching to a full face mask.

I had one of the traditional nasal masks during my titration, sorry i can't remember the name but it was not a pillow type mask.
I was titrated at 7cm. I actually had no events at 6cm but no REM sleep was observed at that pressure.
I understand what you are saying Slinky. I wonder if upping my lower pressure setting be the best thing to do?

I'm NOT the one to ask. I'm on a VPAP Auto and am getting drastically different data between my Simplicity simple nasal cushion and the Zzz-Mask full face and haven't quite decided just what to do yet myself. These doggone VPAPs have so many more feature/options to "play" with ...... see what others more experienced have to say. They'll be along to read this thread in due time.

Uncle_Bob wrote: .... I'm wondering why the numbers are so different ?

You're not the only one wondering. With an AutoPAP, compensating for different airflow and pressure requirements to achieve similar respiratory results with different masks (though why there should even be different pressure requirements requires better explanation than I have seen - theory says that the algorithmic determinants of the pressure setting originate in the airway, not in the mask) should be part of the machine's normal job. But it just doesn't seem to happen quite that way; result differences appear more pronounced than simple pressure differences.

I've posted this sort of diagram in the past. Here are my latest 59 days, all using a ResMed S8 Vantage in Auto mode, pressure 10 - 20cm:



The lines are linear regressions of the scatter-plot data points.

ResMed Activa nasal mask, 26 days, mean AHI 6.7, mean AI 0.3, and mean %-time-in-apnea 0.10%
ResMed Ultra Mirage full face mask, 21 days, mean AHI 9.8, mean AI 0.9, and mean %-time-in-apnea 0.34%
Fisher & Paykel 431 full face mask, 12 days, mean AHI 10.8, mean AI 0.9, and mean %-time-in-apnea 0.36%

The blue box is an arbitrary "target". 88% of Activa nights are in the box; 43% for UMFF, and 42% for F&P 431.
The mean contribution of AI to total AHI is 3.1% for Activa, 9.1% for UMFF, and 8.5% for F&P 431.

Mean 95th centile pressures are not markedly different: 11.5 for Activa, 12.6 for UMFF, and 12.9 for F&P 431.
Mean nightly median pressures are even closer: 10.7 for Activa, 11.0 for UMFF, and 11.3 for F&P 431.

So, the two full-face masks behave fairly similarly (though not identically), but rather differently from the nasal mask.
Results with a Respironics M-Series AutoPAP are similar.

Why, indeed! Velbor

And supports those sleep doctors who say xPAP data isn't as good as PSG data or particularly accurate, etc., etc. - and are too ignorant to recognize that the xPAP data IS valuable for detecting TRENDS which PSG can NOT due do to the expensive of multiple repeated PSGs every night over, say, a one week period.

Slinky wrote:And supports those sleep doctors who say xPAP data isn't as good as PSG data or particularly accurate, etc., etc. - and are too ignorant to recognize that the xPAP data IS valuable for detecting TRENDS which PSG can NOT due do to the expensive of multiple repeated PSGs every night over, say, a one week period.

Or to state Slinky's important point in a slightly different way, night-to-night variability can be potentially so high that, while PSG can give PRECISE data for a single night, that data may not always provide an ACCURATE picture of the SPECTRUM of respiratory distress experienced by a person over time, or provide an optimum therapy plan. Data collection with xPAP may not be precise (for a variety of reasons, including but not limited to mask effects), but it may well be the only practical way to obtain a comprehensive view of what's happening over time, and to decide how to best manage it. PSG remains invaluable for diagnosis, and as one approach to provision of a starting point for therapy.

Velbor, how do the leakage rates compare among the masks? And also, have you kept a log to see if your observed #'s correspond to how you feel? Cos its really how you feel and not numbers that matter

One could argue (my sleep doc does, I'm not so sure) that with AI's less than 1, you're really measuring about the same. He's made the argument to me that its not worth looking for differences if AI's are less than one, and I shouldn't be concerned with HI's as long as in single digits.

Velbor wrote:...

Great phrasing Velbor! Will copy it for times when I get frustrated with my slow recovery or when I've fought masks leaks all night.

GaryG wrote: Velbor, how do the leakage rates compare among the masks? And also, have you kept a log to see if your observed #'s correspond to how you feel? Cos its really how you feel and not numbers that matter

Seal leak has been negligible with all three masks. Mouth leak, in my case, is significant with the Activa nasal mask, but does not compromise therapy. For the dataset presented above, as reported by ResMed's methodology (unintentional leak only, lps units),

mean nightly 95th centile leak: Activa is 0.36, UMFF is 0.04, and F&P 431 is 0.15
mean nightly median leak: Activa is 0.04, UMFF is 0.00, and F&P 431 is 0.00
maximum leak reported with the Activa: 0.64

While the Activa 0.36 lps 95th centile value may seem high, what it's also saying is that values above 0.38 lps are present for overall less than 5% of the time. Accepting ResMed's "alert" value of 0.40, I find this reassuring and acceptable, particularly in the face of an almost irrelevant 0.04 overall nightly median leak. The maximum value of 0.64 is also reassuring, as the ResMed "leak alert alarm" triggers at 0.70. (I also use the Activa with a Respironics M-Series AutoPAP, and have never experienced any "Large Leak" reporting.) The mouth leak does not disturb my sleep, and given the superior efficacy (and superior comfort) of the Activa nasal mask, I use it preferentially.

As to "how I feel" (what we data dorks sometimes call "subjective global introspection"), I am in the fortunate, or unfortunate, situation of having few daytime symptoms regardless of therapy (or lack thereof), so I have no correlations to make. Even using tools such as the Epworth scale in an attempt to quantify "how I feel," nothing useful appears. My only recourse is to retreat to my comfortable, beloved statistics! (Though Mindy may rightly scold me, again, for not having provided any standard deviations in any of my posts in this thread.)

GaryG wrote: One could argue (my sleep doc does, I'm not so sure) that with AI's less than 1, you're really measuring about the same. He's made the argument to me that its not worth looking for differences if AI's are less than one, and I shouldn't be concerned with HI's as long as in single digits.

I agree fully. That's why I've set the borders of my "arbitrary target box" at an AI of 1.0 and a effectively a "single digit" AHI of 10.0

Regards, Velbor

Bless you, Velbor, for clarifying so well!!! Youse a keeper, guy!

I was told, when I bought my Quattro, that the larger the mask--area that needs to seal--the larger the leak is likely to be. That makes sense intuitively, but I am still sleep deprived, so my logic brain cells could be misfiring.

r

Velbor wrote:The mouth leak does not disturb my sleep, and given the superior efficacy (and superior comfort) of the Activa nasal mask, I use it preferentially.

Your leak numbers are great. And now that you reminded me how comfortable the Activa LT is, I may take mine out and give it a try again.