Anybody who has looked at a Resmed's pressure curve in either ResScan (going back to the S8's and S9's) or SleepyHead/Oscar (S9's, AirSense 10's & 11's) knows all about that saw toothed curve. It's not a surprise at all: Resmed has not significantly changed much about their Auto algorithm since the introduction of the S9 with its FOT algorithm to determine whether the airway is open or obstructed. If I recall correctly, before FOT was introduced, Resmed machines did not increase pressure for clusters of OAs if the pressure was already above 10 cm. Resmed also redefined how hypopneas were scored with the introduction of the S9s; I remember a lot of talk back in 2010 about how much more aggressive the S8s were in scoring Hs than the S9s.sptrout wrote: ↑Sat Mar 18, 2023 1:43 pmIf you look at the ResMed graphs on page 5 you will see a very disturbing pattern. (BTW - - All machines in the Study were set "wide-open" 4-20). Both patterns look like the blade from your favorite saw; sharp vertical rises followed by a rather fast decline. The levels are far above what was necessary to correctly respond to any detected breathing change and could easily cause sleep disruption from either the sharp rise in pressure in the user's face, or the increased likelihood of leak problems.
Some of us have argued in the past that Resmed's auto mode is too aggressive in responding to events and that those sudden pressure increases can cause problems with either sleep disruption or increased leaks. Others have praised Resmed's aggressive algorithm as being responsive enough to really break up clusters of events before they get started. It's a point that's been argued about before and no-one's opinion is "better" than anyone else's opinion.
Also worth pointing out: We continuously recommend against running a machine wide open---i.e. with a pressure range from 4--20. Most people do much better when their minimum pressure is set to something close to what an in-lab titration study showed they need or something pretty close to their 95% pressure level if they don't have an actual titration study. And some of us also regularly recommend that setting a maximum pressure that is less than 20 can be a good idea if you have reason to believe that sharp pressure increases may be causing you to wake AND you've already got your min pressure set up where it needs to be set.
You need to define what you mean by "fairly well." Because not everyone here will agree with your definition of "fairly well" regardless of how you define it.Note that the "for Her" model was better than the standard model, but still poor, or really just plain bad. The other brands of APAPs had their own problems, but 2 or 3 did fairly well.
Three comments:One other item that I found interesting was the APAPs reaction to the first 50 minutes of each test. The first fifty minutes simulates the user being awake and breathing normally, therefore, the APAP machines should just sit at 4ccH2O waiting for the user to fall asleep and start having breathing problems.
1) In some people, real wake breathing is notoriously irregular compared to sleep breathing; in others? Not so much. We have no idea what the "simulated wake breathing" looked like in this study. But if it didn't look like yours, then there's no way to guarantee that a machine will respond to your particular wake breathing the way these machines responded to the simulated wake breathing.
2) A simulated sleep latency time of 50 minutes is way, way, way too long. Anybody who is regularly taking 50 minutes to get to sleep after turning the light out has a problem with sleep onset insomnia. Yes, some people are likely to slap the machine on their face and read for an hour in bed or (shivers) watch tv or (more shivers) spend time looking at their phone. But if you're not even trying to get to sleep and the pressure does wind up increasing, that's not really a problem unless you are uncomfortable. If you are uncomfortable, you can always turn the machine off and back on to reset the pressure while you are reading in bed. Once you turn the light out? Ideally you should be drifting off to sleep within 15-20 minutes at most.
3) Many, many, many people do not feel comfortable breathing at 4cm. And when you're not comfortable, your wake breathing is even less likely to match whatever was used as "simulated" wake breathing in this study.
An APAP cannot tell (for sure) whether you are awake or asleep, and the vast majority of them have been programmed to assume the patient is asleep if they're using the machine. Yes, some of them now have so-called "smart ramps" that attempt to detect the onset of real sleep breathing by looking for something like 2 or 3 minutes of very regular breathing that has the characteristics of sleep breathing. But if a patient has a serious problem with apneas occurring just as real sleep is established, then that person's breathing might never settle down enough due to repeated OAs, CAs, Hs, FLs and RERAs that start just as soon as the person is asleep. This is a very serious potential problem if the minimum pressure is left at 4cm and the patient needs at least 8 or 9cm to control the apnea. Do we really want a machine refusing to respond to something that might be significant FL breathing, along with Hs and OAs when you've dropped off to sleep just a few minutes after turning the machine on just because the machine is now programmed to assume that you are still awake until a real, regular normal sleep breathing pattern has been detected for at least two or three minutes?On the 10 graphs this 50 minute "awake" time is easily seen on the left side of each plot; a straight line at pressure level 4. If a machine is responding correctly it should be doing nothing except providing a level 4 pressure.
Three comments:However, look at the various machines, some go straight to a very high pressure and others climb but not as high. Most of the machines try to start correcting a non-problem; a couple remained at or near 4. As Nick mentions in his video if using one of the machines with fast rising pressure you may want to use the "ramp" feature to help control these rockets.
1) There was a similar bench test done even earlier than this one with much the same results, but with only 5 machines: Some machines seem to be overly aggressive; others barely reacted to the simulated events that should have triggered pressure increases. But in the earlier study, no-one had the crazy idea of trying to see if the machines would be "tricked" by simulated wake breathing over a 50 minute period. (See Auto-Adjusting Positive Airway Pressure from 2009.)
2) If the minimum pressure is 4cm, the ramp is disabled on many (most) machines: You can't "ramp" up to 4cm. So in order to use the ramp, you can't be running at 4-20.
3) If you do have a minimum pressure set, ramping up from 4cm may or may not be a good idea in the first place: Many people feel like they're suffocating at 4cm because too little air is coming into the mask. Next, if the ramp is a traditional ramp, it starts increasing the pressure almost as soon as you turn the machine on because the machine is programmed to steadily increase the pressure in a linear fashion until the minimum pressure is reached at the end of the ramp time. If pressure increases while awake are "troublesome" to you, then you might just notice the constant increase in pressure if you're lying awake while the ramp clock is ticking down. (I know I always did and that made it doubly hard for me to relax enough to get to sleep during my rough early days.)
What do you mean by "control the 'awake' pressure"? Most of us turn the machine on, turn the light off, and go to sleep. If you're asleep within 10 minutes of turning the machine on, pressure increases before you get to sleep should not be a big issue. If you are reading in bed or otherwise using the machine while not actively attempting to go to sleep AND you are bothered by the pressure increasing while you are awake, you can always turn the machine off and back on.I have two main takeaways from this Study.
1. Make sure that you control the "awake" pressure since this could be reoccurring each night and even maybe multiple times per night if you wake for one reason or another and have to restart your machine.
If you have severe SA issues, you should NEVER run the machine wide open: You absolutely need that minimum pressure set up high enough prevent your airway from collapsing in the first place. While you might need to run the machine at 4-20 for a week or so at home if your insurance won't pay for an in-lab titration study, once you've got a weeks worth of data, you need to use it to set your min pressure setting no more than a couple of cm less than your 90% or 95% pressure level. Anything less than that and you're inviting lots of events to happen before the machine can respond by increasing the pressure to a more appropriate level. Moreover, anytime the pressure drops below the level needed to prop your airway open, you're likely to see additional events piling up before the machine has a chance to respond by increasing the pressure to an appropriate level.2. Do Not leave your APAP machine pressure set to "wide-open" unless you absolutely have to due to severe SA issues.
Good for you for finding a pressure range that works for you in terms of your data. The important question, however, is: How do you feel when you get up in the morning? If the answer is something like Pretty Good, more power to you---you've optimized your APAP therapy.If you do leave it wide-open you can see what happens with your machine. I have 4 ResMed 10 machines and I have seen the sawtooth pattern many times before I reduced the maximum pressure considerably. I now have my machine set for a low of 6 (could be lower I expect) with a high of 9. My AHI is nearly always less than 1, mostly between 0 - .5. each night (from OSCAR; my 30 day AHI average as of this morning is .32). Before I reduced the maximum pressure from I believe it was 15, I was having several CAs each night; almost certainly caused by excessive pressure. I seldom have CAs at the reduced pressure.
But there are people who do have problems with RERAs, FLs, and snoring continuing to make them feel bad even though they have a near perfect AHI each night. And one giveaway that they may have the max pressure set too low is if the machine is constantly bumping up against that max pressure when there is something obviously wrong with the breathing pattern---i.e. you're hitting max pressure while there is are severe flow limitations or snoring that is present. For these people, they may need to do some additional dial wingin' to find the best pressure range for them to use.
By using a pressure range of 6-9cm, you pretty much doing what we usually tell people:I hate to say this: But, according to Nick's recommendation the maximum pressure does not need to be much more than the "95% Pressure" as shown on the OSCAR Statistics page. I use the "7 day 95% Pressure" reading of 8.94 to set my max pressure to 9.0. As I mentioned already, the results are great. Low AHI, no sleep interruptions from unnecessary high pressure, and no leaks.
- Set min pressure 1-3 cm less than the 95% pressure if you're comfortable breathing at that pressure.
- Set max pressure 1-3 cm above the 95% pressure; if you set it just above the 95% and you don't have too many events and you are feeling fine when you wake up, then that's also ok.