Machine Chasing Leaks?

[cpdaniel]

I see the machine increasing pressure based on snores or whatever is being interpreted as a snore. You would need to very closely examine the flow line to see if there might be some reduction in respiratory volume that is also working to increase pressure. I think the leaks were coincidental and insignificant.

Sleepyhead badly needs an update to fully show the IPAP and EPAP pressures on the Aircurve, and properly interpret pressure settings. For all I know there may even be a bug in the leak reporting. Again, a closeup of the flow line would be more revealing here than the empty events chart.

Here's a close-up of the start of the same pressure ramp-up depicted in the previous screenshot.

   

At 01:41:55 there's the first increase in pressure, but there doesn't appear to be any change in the flow waveform that triggered this increase. I didn't show it in the screenshot, but there's no apparent change in Minute Ventilation around this time - it's nominally flat.

A few seconds later at 01:42:37 there's a hiccup in my breathing that I'm guessing may be me reacting to the pressure increase, or it may just be that I rolled over right at that second. There's clearly some flow limitation following that - which interestingly doesn't register at all on the flow limitation graph.

For comparison, here's a section of undisturbed sleep during which there were no leaks, no snores, no flow limitations, with pressure at the minimum.

   

For the record, the flow limitation graph isn't always zero. Here's another interval during the same night that's typical of what I see if I roll over onto my back:

   

[Sleeprider]

Really interesting cpdanial! The closeups are self-explanatory. Pressure increases are due to decreases in inspiratory volume, and flow limitation, that are not flagged events. It's difficult to deconstruct the algorithms that change pressure on these machines, but those are very good screen-shots.

Keep in mind, JediMark did not specifically write Sleepyhead to deal with the Aircurve 10, but a fix is in the works. I suspect that there are event flags that do not show. For example, why would the Autoset flag RERA, but the more expensive and later Aircurve not? The pending update may resolve some of the "undefined events".