Algorithm change in the ResMed AirCurve 10 ASV

[terp1984]

I really hesitated to post this thread.  If I had no issues with my CPAP machine and I read what follows, I would be highly skeptical.  However, I believe I can make a sound engineering argument regarding my claim, so please bear with me and read this with an open mind.
 
When the ResMed AirCurve 10 ASV became available, it was stated that the algorithm was the same as the S9 VPAP Adapt.  Although the software was different (e.g., different menu structures, etc.) the algorithm within that software was supposedly unchanged.  Until recently, my personal experience seemed to confirm that.
 
I have reason to believe that it changed in January of this year.  It was then that I began having a lot of trouble with my full-face mask leaking (ResMed Quattro FX).  For more than 12 years, I experimented with at least 15 different masks.  I finally settled on one combination which consistently gave me a virtually leak-proof night.  I use that mask with an 1/8” thick solid gel (cut to size) and a REMZZZ liner.  I also smear my face with lanolin before putting on the mask.  In January, when my severe mask leakage began, even after using all new parts, I could not stop the leakage.  It wasn’t just a subtle leak, it was extreme.
 
Looking carefully at my SleepyHead data, I noticed that the pressure was significantly higher than normal.  The 95% pressure when from an average of about 15.5 to 22.  The max pressure went from an average of about 18 to 25.  The only way I could get my mask to stop leaking was by using skin glue.  In addition to it being a royal pain to clean in the morning, it was also very expensive.  Instead of the mask cushion lasting about a month, the glue deteriorated the cushion such that it only lasted three to six nights.
 
At that time, I figured there must be some sort of physiological change to my body that caused the pressure requirement to increase.  I tried sleeping on my side, which is something I had not done in over 20 years.  That did reduce the pressure need slightly, but not enough to prevent the mask from leaking.  In addition to that, it nearly destroyed my back, which is why I stopped sleeping on my side years ago.
 
I also went to my local DME provider to try and get a better mask.  When I showed them the list of 15 masks I had used, they said, “Wow, we only have one mask you have yet to try”.  This was a F&P Vitera.  There was a time when one could try a mask and if it didn’t work bring it back and pay nothing.  They stopped that policy at my DME provider.  I had to buy this mask out of pocket before even trying it.  It did not work.
 
In April, I went on a 10 day road trip.  I always use my old S9 VPAP Adapt while traveling.  That has proved to be very convenient.  I just keep it packed and ready to go, so I don’t have to pack up my AirCurve.  After the first few nights, I noticed my pressure was consistently back to normal (i.e., to what it was before I had the mask leakage beginning in January).  I then switched to using my mask without skin glue for the remainder of the trip and never had a leak.
 
When I returned home, I compared the data for the 10 day road trip to the average for the AirCurve for all of 2023.  Those values are identical.  Drilling down into the daily results in January, I noticed that on January 12, the pressure suddenly increased.  Prior to that date, the average 95% pressure was around 15.5 and the average max pressure was 18 (same as my 10 day road trip).  From January 12 on, the 95% pressure was around 22 and the max pressure was 25.  Those are huge differences and certainly enough to cause my mask leakage issues.
 
I believe that somehow (probably via 4G), the software (with updated algorithm) to my and other AirCurve 10 ASV machines were updated on January 12, 2024.  Perhaps not all on that same day.  Maybe sometime in January, the update process was started by changing a certain number of units each day until all were changed.
 
I reported this to my sleep doctor and my theory that the algorithm was changed in January and his immediate response was “no way”.  He speculated that perhaps there is a malfunction with my particular machine.  After showing him the data and pointing out that I have used three different  AirCurve 10 ASV machines this year and they all behaved identically, he was convinced (see my unrelated problem in thread “Humidifier heater issue” for further explanation).  He said he would contact his local ResMed rep and try to find out more.
 
Meanwhile, much to the chagrin of my doctor, I have switched to using my S9 exclusively.  What a relief it has been to not needing the skin glue to prevent mask leakage.  The doctor suggested changing the settings to limit the max pressure.  I did that.  However, even if the max pressure is forced to be no higher than 18, the 95% pressure is almost as high and my mask still leaks.  15.5 to 18 may not seem like much, but apparently it is enough to cause severe leakage in my mask.
 
I will be curious to learn what comments, if any, ResMed will have.
 
So, what do you think?  Is my theory plausible or am I smoking something?

[Sleeprider]

I wonder if the flow/pressure sensor calibration is off. The idea of a software push is plausible, but Resmed does not usually make a secret out of such things. It would be interesting to compare the before and after charts, and maybe even compare the S9 VPAP Adapt, to see if we can pick out what is pushing up the pressure.

[SarcasticDave94]

Would you be ready to use OSCAR charts in place of SleepyHead? It's 100% free and better than SH, more accurate, etc.

Given charts we can help with settings, if you want the assist. If not, that's fine. I've used the 10 series ASV with the Fisher and Paykel Vitera myself, and it can take IPAP 25 without glue.

[terp1984]

After being ignored by my local DME provider for several months, last week I received a call from a rep with AdaptHealth (the parent company of that DME provider).  He had been evaluating my situation and was also in contact with ResMed about it.  ResMed claims there have been no changes to the ASV algorithm for S9, AirCurve10 or AirCurve11 and they all use the exact same algorithm.  I suggested that if that is true, then perhaps there was a firmware update that caused these symptoms.  He agreed about that possibility although had no opinion about that likelyhood.
 
He had the data from the AirCurve machine that was in use in January and acknowledged that there was a huge jump in pressure beginning on 1/12/2024.  He suggested there may be a bad sensor in that machine causing this problem.  I pointed out that two subsequent AirCurve machines were behaving exactly the same way in addition to the fact that the S9 was still working normally.  He said he did not have access to those machines’ data and therefore could not confirm my claim.  So, I sent him my SleepyHead data (yes, I know that software is old, out of date and so forth, but it works and does what I need) of all three AirCurve machines plus data from two S9 machines (one I used last April and a new one I have been using since May).
 
In a follow-up conversation this past week, the AdaptHealth rep confirmed receipt of my data and also confirmed that those data correspond to all my claims.  He plans more discussion with ResMed and will get back to me.
 
I have a hard time believing that I am the only one who is experiencing this problem with the AirCurve 10 ASV.  Perhaps it is just more noticeable for me due to my unusually high pressure requirements.
 
I have a request to users of AirCurve 10 ASV machines:  Please check your data before and after January, 2024 and see if those pressures are different – thanks.

[SarcasticDave94]

It may be many other possibilities, but the one I'm thinking goes like this.

It could be your old S9 is wearing down and not blowing as much as it should. When you're trying the new machines, you perceive it as blowing too hard.

You can still show us the charts, regardless of OSCAR or sleepyhead.