What the heck was THAT?!

[cathyf]

So last night's data had something that I've never seen before. A 10-12 second range where the flow rate curve bounced between zero and spiking negative.

https://www.dropbox.com/s/51zis6xjo9cu4r...n.png?dl=0

Here's the same 1-minute view with the window stretched out
https://www.dropbox.com/s/ecqgrflxkhsljq...m.png?dl=0

Zero leaks. No snore.

It starts with that UF 1, which is an obstructive event but only seven seconds so it doesn't count. (I think of them as "sub-events") But then what comes after is bizarre!

Ok, it obviously didn't last very long so it's not going to kill me, but I'm really trying to wrap my head around the question of just what that picture is illustrating. So there's the wide humps of zero, and then the very narrow very deep dives into negative (exhale) territory. Is this a stuttering exhale where it starts and stops but there's no inhaling at all?

So there's "sleep disordered breathing" and that is serious and ominous but that looks like "sleep wacky breathing" ! Just so totally weird!

[2SleepBetta]

"Stuttering exhale". Exactly what I thought. I've no idea of any cause in sleep, but I have done a few such bursts when awake with a sense of something lodged in my throat and my not wanting to inhale out of fear of aspirating whatever the cause might be.

[sheepless]

can you replicate something that looks similar by, say, clearing your throat or a slight cough?

[Dormeo]

Like others, I'd guess this was a series of little coughs.  And speaking of anomalies, for your amusement, I attach the flow rate graph for a spell of hiccups.

[Dormeo]

Like others, I'd guess this was a series of little coughs.  And speaking of anomalies, for your amusement, I attach the flow rate graph for a spell of hiccups.

[Geer1]

First of all this was neither sleep disordered breathing or sleep wacky breathing. You had an arousal at 4:05:55 and were awake after that point. UF1 was not an obstructive event, it was you holding breath probably while changing sleep position. Then at 4:06:10 you took another breath and then had what appears to be a little coughing fit like others are suggesting. I think sort of akin to the response your body has when you accidentally have water go down the wrong pipe, potentially because you swallowed or something.

Any time you see a spike on top of a breath (like 4:05:55) preceding strange breathing you can be ~98% sure it is related to an arousal and that you are not asleep at that time. If this was caused by sleep disordered breathing then the cause would have to be the breathing prior to 4:05:55 which does show a little bit of flow limitation but not horrible by your standards. It isn't clear if that would have been a RERA or just a spontaneous arousal.

[cathyf]

Thanks, everybody -- coughing would make total sense, and thinking about what the numbers say and what coughing feels like, I totally get that. I agree that it doesn't count as "sleep-disordered breathing" and my fitbit thought that I was awake from 4:05:30-4:06:30, too.

Here's another oddball, where my fitbit was convinced that I was asleep the whole time. Those jagged ups and downs sure don't look like either my normal breathing or any sleep-disordered breathing. The thing that I'm not sure of is that the machine was reporting a low level of leak through both of the really weird periods and it lines up really closely. Everybody else (including the stupid ResMed happy face, LOL) thinks that the machine can compensate for leaks below redline, and these leaks running under the weird flow rate patterns are well under 10.

https://www.dropbox.com/s/ljq78jbb1e6shh...s.png?dl=0

But the machine is clearly confused by what I'm doing, and thinks my respiration rate is really high, and the EPR is struggling to follow along. Is it possible that something about the oscillations in the leaks is causing the machine to drastically misinterpret what's going on? The whole deal where the machine's sensors can figure out what's a leak vs what's weirdness in the breathing pattern is something that I think would be some pretty sophisticated math. As a former math major who is married to a physics professor who actually understands computational fluid dynamics  I'm looking at the graphs and wondering if the algorithm has just failed in some way --- like the leaks were much bigger than the machine thought and it was registering "mask trombone" as breathing, and for a couple of minutes the machine was just losing entirely the ability to sense what was happening?

I'm also wondering if my original 2014 autoset with the 18,000 hours on it might have intermittent failing electronics? I know that there are lots of folks who have machines with way more hours on them that are seemingly just fine, but the maybe the manufacturers are right that machines don't really have that long a life?



(Sorry I did a hit-and-run post last week. Life got very busy for the holiday...)

[Geer1]

That is just an example of your asynchronous breathing. I believe these machines operate on one (maybe two) pressure sensors and that they calculate the flow rate and leak from those sensors and settings input. I think the low level leaks reported in these periods are just calculation errors because the algorithm doesn't know how to deal with those wildly fluctuating pressures caused by your asynchronous breathing.

Have you tried recording sleep yet to try and get this breathing on video? I really do think that will give you the answers you are looking for and think you might be surprised by what you see.