Respironics need higher minimum pressure due to slower response? I'm not so sure.

[Shin Ryoku]

I have plain vanilla OSA (no significant CSA), and my events tend to come in clusters.  For example, here is my diagnostic sleep study hypogram:




Here's me on CPAP 4:




Me on CPAP 7:




I am entirely self-treated with help from the forums and other internet references. I have no managing health care provider. Seeing as how my events come in clusters, I decided to try APAP. I went for the least expensive, current model, data capable, popular machine I could find, which was the PR DreamStation Auto.

As I started to use the DreamStation, the feedback on got on this and other forums was that since PR devices are slower to adjust pressure, and that I would need a higher minimum pressure accordingly. This seems to be a commonly held belief.

That feedback gave me a pang of regret that I may not have bought the ideal machine for me, so I went ahead and bought a ResMed AirSense 10 AutoSet for Her to see if that would work out better for me. I subsequently also bought an AirCurve 10 Vauto but that is besides the point of this thread.

For demonstration purposes, I decided to do four consecutive nights of testing using a lower than optimal for me minimum pressure of 10. APAP 14-20 works out pretty well for me on all brand machines, but I wanted to show what I'm seeing, which is that using the same lower than ideal min pressure worksout about the same for me with both brand algorithms. I ended up missing one night due to forgetting to put in the SD card, but these are the full data, no cherry picking.


PR DreamStation 10-20 C-Flex 3:




ResMed AirSense Autoset 10-20 EPR 2:




PR DreamStation 10-20 Flex Off:




ResMed AirSense Autoset 10-20 EPR Off:




In looking at my SleepyHead data, I can see why people say the ResMed devices raise pressure quicker. That seems to be true. But I am not seeing anything that suggests that, in terms of outcomes, the PR devices need a higher minimum pressure setting. The same basic findings are also present when I use lower than 10 cm H2O minimum pressures (data not shown). I think the often stated claim that PR machines need a higher minimum pressure is likely incorrect.


Lastly, on an unrelated point, my clusters of OAs have nothing to do with chin tucking. Just saying...

[Walla Walla]

I own a Dreamstation Auto BiPAP and a ResMed VAUTO. On the VAUTO I can run it 1cm lower than the Dreamstation. That's based on my own results. I don't think you can take a few nights from one person and say if both machines are the same or not. The basis for the belief that the ResMed's are better at reducing the AHI's are from hundreds if not thousands of charts that have been posted here over the years.

As far as the chin tucking goes. More often than not that's exactly whats causes the event groupings. $15 for a cervical collar to avoid having to use high pressures is a pretty good trade off. If it doesn't work than OK it only cost $15. But what do you do if your on a Bipap and you reach 25cm without relief? Wouldn't it be better to try the collar early and at least see if it works? It would save a lot of time and energy if it does.

[Shin Ryoku]

I own a Dreamstation Auto BiPAP and a ResMed VAUTO.  On the VAUTO I can run it 1cm lower than the Dreamstation. That's based on my own results. I don't think you can take a few nights from one person and say if both machines are the same or not.

I only posted a few nights representative of the data I have seen over the long run. My conclusions are not based on only the nights shown.

The basis for the belief that the ResMed's are better at reducing the AHI's are from hundreds if not thousands of charts that have been posted here over the years.

I remain skeptical. Seeing the pressure jump up in the ResMed charts is likely a strong source of bias. Not to mention that once people starting saying something, it influences what others see and conclude.

As far as the chin tucking goes. More often than not that's exactly whats causes the event groupings.

I am skeptical about this as well. I believe that a collar helps some people. But "more often than not" is, IMO, a stretch which is not founded by good evidence.

If it doesn't work than OK it only cost $15. But what do you do if your on a Bipap and you reach 25cm without relief? Wouldn't it be better to try the collar early and at least see if it works? It would save a lot of time and energy if it does.

This part I agree with. There is a chance that it can help, and it is cheap. For that reason I tried several different soft cervical collars of different heights and firmness. All were uncomfortable, and none helped my sleep apnea.

Again, I believe that a collar helps some people and that it may be worth the price to try. But the conclusion that most event groupings are due to chin tuckings? That seems to be unfounded.

[Sleeprider]

I think this is a very interesting experiment using your own data for verification of the premise that a Respironics machine requires a higher minimum pressure to achieve optimal results, than a Resmed machine. I would have to agree that based on this example, the results favor the conclusion that result differentials between Respironics or Resmed at a range of 10-20, with or without EPR/Flex are insignificant or favor Respironics.

I think this shows that for any individual, generalized results from a broader cohort are not necessarily applicable, and everyone seems to respond differently, and to cause a different response from the machines depending on what signals the machine receives from the patient. In your case, both machines seem to increase pressure at a similar rate, and interestingly, the Philips machine does not display characteristic higher pressure tests. The Philips machine also (uncharacteristically) maintains higher pressure over sustained periods. This is accompanied by relatively high flags for flow limitation, and relatively low hypopnea.

In your case, I think the conclusions are correct, but I don't agree that they can be generalized. With a minimum pressure of 10 cm, I also have a similar event rate between my Philips and Resmed machines. I think that the concern for Philips auto machines has been that they tend to not increase pressure above the minimum for certain individuals that don't adequately warn of hypopnea and OA through the cues used by the machines to adjust pressure. This is most often seen as repeated hypopnea and OA with no sustained increase of pressure. Both you and I provide the Philips Auto with the cues needed to raise and maintain pressure. Many people I have seen on the forums do not.

I should more often say that many people get great results on a Philips Auto. I did for many years. The difference between a Philips and Resmed machine for me is actually very little, and would take long-term data to discern any differences. With that said, it has been my experience over many anecdotal observations, that the people using Philips machines have a more difficult path to optimization. This ranges from a generally higher event rate, particularly hypopnea, to gross errors in respiratory statistics, for inhale and exhale time. (I think this may be related as Flex may be mis-timed in these individuals). The biggest problem is nearly always greatest with people that for some reason show an inverse I:E ratio.

I appreciate what you have done here, and I know this responds to my frequently expressed preference for Resmed. I did not arrive at that preference through my own experience, but by the non-scientific sense that the same problems were repeatedly seen with the Philips machines. These "impressions" are not something that can be generalized to any given individual, but is more of an informed hunch, that better results are easier with the Resmed machine. This is partly due to the more predictive pressure increase of pressure ahead of events, and the different approach to exhale pressure relief which more resembles bilevel therapy, than simple relief at the trigger or cycle transition. Over time, I allowed my opinion to become less moderate, and I simply recommended Resmed over Philips. That is probably not fair for the majority of cases, but it seems to be good enough for the majority, and better for the minority that have poor results with Philips.

That's my story and experience.

[Shin Ryoku]

I appreciate what you have done here, and I know this responds to my frequently expressed preference for Resmed.

Honestly, what I wrote was not a response to you at all. My early experiences of hearing reasons why ResMed might be better for me came from reading posts by Pugsy on another site. And more recently I've seen a few posts where ajack mentioned needing a higher minimum with Respironics due to their slower response.

I do appreciate all your comments above, and I understand that my results are not necessarily generalizable to others. At the same time, you and I are unlikely to be exceptional in that we get similar results from ResMed and Respironics automatic algorithms.

With that said, it has been my experience over many anecdotal observations, that the people using Philips machines have a more difficult path to optimization. This ranges from a generally higher event rate, particularly hypopnea, to gross errors in respiratory statistics, for inhale and exhale time.  (I think this may be related as Flex may be mis-timed in these individuals).

The path to optimization is a separate topic IMO. For example, when I started therapy, I noticed a higher false positive rate (events scored while I was awake) with the DreamStation than with the ResMed. If my goal was to have a low machine-determined "score", it may have been harder with the Respironics.

But in terms of actual efficacy - reacting to and preventing real events - I'm not convinced that there are significant advantages one way or the other.

Here's what I mean: Suppose you took 100 people with OSA who, having already gone through the path to optimization, were well optimized on their ResMed APAP machines at a given pressure range. If you switched them all to Respironics machines with the same exact pressure range that worked for them with ResMed, what do you think would happen? Would most have a higher "true" (manually scored) AHI on the Respironics? I doubt it.

[Spy Car]

Shin Ryoku, while a minimum pressure of 10 might not be optimal for you, it still strikes me as a pretty high minimum pressure.

What would these tests look like with a minimum of 6 or 7? Or even wide-open (4-20), the way most patients have their APAPs set up.

I'd expect that's where the reputed faster response of the ResMeds might be more likely to show itself.

Bill

[Sleeprider]

Probably a subject for a different thread, but I would love to suss out this issue of inverse I:E ratio that we see fairly often, usually associated with intractable obstructive or complex apnea. I see it occasionally on Resmed, but much more often on Respironics.

Shin, great experiment, and not much to disagree with here. I think if more effort was made to properly set minimum pressure on any machine, we would not have much discussion of which auto CPAP was better other than subjective comfort issues.

[Shin Ryoku]

What would these tests look like with a minimum of 6 or 7? Or even wide-open (4-20), the way most patients have their APAPs set up.

I'd expect that's where the reputed faster response of the ResMeds might be more likely to show itself.

I alluded to this in the first post:

The same basic findings are also present when I use lower than 10 cm H2O minimum pressures (data not shown).

I've used pressure ranges of 7-20 with comparable results on ResMed and Respironics. Lower than 7 is very unpleasant for me unless I use a full face mask. Meanwhile higher than 15 is very unpleasant for me with a full face mask. So I can't really do a 4-20 experiment without undue suffering.

[PaulaO2]

Walla Walla: I split your post to its own thread since it was not related to the OPs question. You can find it here:
http://www.apneaboard.com/forums/Thread-...ation-info

[Shin Ryoku]

The only head to head study I could find in a medical journal with some decent data in the abstract is this one: https://www.ncbi.nlm.nih.gov/pubmed/17541262

The use of the Respironics REMstar Auto (RR) was associated with a significantly lower AHI in comparison with the ResMed Spirit (RS) [mean (SD) 6.9 (11.6)/h vs. 9.4 (9.2)/h, p = 0.004]. This result was obtained at significantly lower pressure levels [P95 9.2 (2.3) cm H(2)O vs. 10.2 (1.5) cm H(2)O, p = 0.001].

Likely things have changed in the past 10 years since that was published.

There is also this study: https://www.ncbi.nlm.nih.gov/pubmed/23600534

RESULTS:
Bland Altman plots showed good agreement between the recommended median and maximal pressure levels obtained with the two devices. A significant improvement was observed in all the sleep parameters by both A-CPAP machines to a similar degree.

Would be nice to see some more head to head APAP comparisons in the medical literature.