Resmed S10 weird difference between machines

** Does anyone know if Resmed has changed either either the event definitions or treatment algorithms in the last few years? I have two machines of different age, the older one works better.

I'm a longtime CPAP user (20 years). I have complex sleep apnea, my obstructive events are well controlled, usually about 1 per hour. My central apneas and hypopneas dominate my total AHI which is typically between 2 and 3.5, averaging around 3. I probably should have an ASV machine, but I don't qualify with my current numbers. In the last two or three years I've had issues with daytime sleepiness and have spent considerable time optimizing things with the help of OSCAR.

Here is the weird thing. I've got 2 Resmed S10 machines. A relatively new one which I got in late 2023, and an old one that I got in 2015 or so (its got ~19,000 hours). I use the old one as a travel machine and recently discovered that it apparently works better than the new one. My average total AHIs are around 1 with the old machine vs 3 with the newer machine. The central apneas and hypopneas are much lower with the old machine. All of the settings are exactly the same, and both machines register the same pressure with my gauge. They are running different firmware though, one has a 2G modem so this would be expected. I think I am less sleepy during the day, but its a little hard to tell if this is from vacation or the machine. I'm alternating weeks to try to sort out if its actually working better, or if it's just defining central apneas differently. My central apnea score is very sensitive to even minor changes in either machine settings and the environment. It seems possible that they have changed the algorithm slightly. But I think its also possible that the events are scored differently, or perhaps the blower responds differently because of age.

I use the Resmed P30i pillows mask by the way. Machine settings are minimum 9.2, maximum 10.4, no EPR. I've spent a lot of time finding these settings. My AHI considerably lower than before I started fine tuning. I wear a collar which improved my obstructive apneas, they were worse when my chin was low. There was a minor improvement in my central apneas as well because the average pressure is a bit lower when the obstructive apneas are better controlled. Any level of EPR and the central apneas increase significantly. Any pressure over about 10.8 triggers a large increase in central apneas. Like I said, my central apneas are extremely sensitive to even small changes in pressure or the pressure waveform.

Anyone have any experience or thoughts on this? I don't think Resmed ever talks about firmware changes...

There was an algorithm change from the S8 model machines to the S9 models.
That's the only algorithm change that I am aware of.
Nothing after the S9 model release and certainly no change in the AirSense 10 model algorithm.

Are the settings the same? Check the software version. Is that the same? If your AHI is truly better, that's good. But if the newer machine is detecting events that the older machine misses, than that is better.

Check to ensure For Her mode or Soft Response is not enabled.

Settings are exactly the same. There is no For Her, its the regular S10 Auto standard model. Response is set to standard, soft increases the number of central events for me. The firmware isn't the same on the two machines, one is SX567-0401, the other is SX567-0306SX567-0306 which is why I'm wondering about algorithms, a lot can happen in that many years in terms of firmware tweaks.

I'm going to alternate between the machines a week at a time and log my sleepiness. I also have a sleep ring that independently logs oxygen events. I'll see if it correlates with the AHI on the machines. If it does and I feel better then its the treatment, if not its probably the old machine missing events.

What I'm wondering is if they made adjustments to the algorithm to try to improve comfort/compliance with patients who have obstructive apnea only. There could be simple changes in play here, the blower response rate during exhale for example, or softening the standard response. In my experience features aimed at comfort like EPR, soft response and ramp all lead to increased central events for me. It leads me to believe I should bite the bullet and just pay out of pocket for an ASV.

I've now determined that there is probably is some sort of algorithm change on the S10. My oxygen logs seem to confirm that I am having fewer events with the old firmware than the new so its not the event characterization. More importantly I feel better.

I made some crude pressure measurements to see if the machines behave differently from one another when actively breathing. My measurements are almost certainly not accurate in the absolute sense, but the machines definitely behave differently. The static pressure is the same for both machines. With the newer newer SX567-0401 firmware IPAP is higher than the static pressure, EPAP is about the same as static. At my pressure minimum (9.2) IPAP was a full 4 CM higher than static pressure (no EPR and pillows mask type). This pressure increase largely disappears at higher pressure (11 CM or so). With the older SX567-0306 the IPAP, EPAP and static pressures are the same. Enabling EPR reduces the EPAP below the static target for both machines (as expected).

I measured at 9.2 CM in auto mode, but used CPAP mode to test at 10.4, which is my maximum. The newer firmware has less difference between IPAP and EPAP at higher pressure than at lower pressure, but there is still some differential, and at lower pressure in CPAP mode it still shows the elevated IPAP.

If I use the full face setting the newer firmware is closer to constant, like the old firmware. I think it is possible that the mask compensation profiles have changed over time to reflect changes in the design of masks people are using. 7 years is a long time for any firmware, there are always changes.

There is some uncertainty with my measurements. My setup adds both volume and leakage to the system. I put a T at the entrance to the mask and a pressure gauge off the T. There is additional leakage through the gauge, and additional tube volume. But the point is that the machines behave differently.

The differences in response may explain why the older firmware is better for me. I already know EPR and higher pressure significantly increase the frequency of central apneas for me. The difference between the machines would explain why the older firmware works better for me. I may be able to get by if I use the full face mask selection with pillows. But at the end of the day I think I need to break down and spring for an ASV machine. It would probably work better for me and hopefully require less fussing around. In the 20 years I've been using a machine I've never had numbers as low as I do now, often less than 1. There is real difference in the way I feel when its that low. I've been trying to avoid getting an ASV machine because I will have to pay for the it out of pocket. My current numbers probably don't qualify me under Medicare. Explaining all of this to my pulmonologist to get a prescription isn't going to be much fun ether...it's bad enough that I change my own settings.