Apnea

[Sonnyboy]

Sonnyboy,

The pressure curve has a big gap in it. Can you show the Leak Graph as well as the pressure and flow rate graphs. The other graphs are not needed. Also can you zoom in further and show just the period between 4:55 and 5:15.

The fact that the pressure curve just cuts out completely, but there are hints of a flow rate curve points to the possibility that the machine is just having a very tough time tracking your breathing for some reason. A largish leak might explain why the machine is has apparently lost track of your breathing. Our machines are good, but they're not perfect and sometimes this happens. When it does, it's possible for the events being scored to be "false" events. But it's also possible for the events to be "real" events. It can be very tough to figure out whether the events are real when the machine is having problems tracking the breathing in the first place.

When I first saw this kind of a pattern in my own data, I took the data (and my machine) back to the DME and asked them what the heck was going on and whether my machine was defective. (They had already replaced one machine under warranty when I woke up to it rebooting itself one morning.) The RT at the DME called the pattern a "patient disconnect", meaning that the machine had lost track of my breathing for some reason. (What reason was never explained since the leak rate was not particularly high.)

In my own data, this kind of "missing" data doesn't happen very often. Sometimes there is a large leak. Sometimes there isn't. I'll occasionally have a patient disconnect for 2 or 3 nights in a row, and then I won't see another one for several months (as in 6-12 months at times). Because these "patient disconnects" are rare in my data, I don't tend to worry about them.

[Imgur](http://i.imgur.com/dM3dunD.png)

[robysue]

Do you have auto on/off turned ON by chance?

It looks to me like the machine lost track of your breathing around 5:00:45. Right before that, the leak is rather high, even though it is still technically under the Red Line. It could indeed be that the mask got partially displaced in such a way that the machine just could not figure out what was going on.

It looks like the machine turned itself OFF around 5:02:15 (when the gap in the pressure curve starts) and then it looks like the machine turned itself back ON around 5:03:45. Can you look at the list of "sessions" in the left sidebar in the daily data and see if there is a session that ends at around 5:02:15 and another that starts around 5:03:45?

Even after the machine has either turned itself back ON or has found some small evidence of breathing going on around 5:03:45, your AirSense is still having a lot of trouble tracking the breathing. And where there is some clear air movement going into and out of the lungs, there's also a relatively high leak. Where the machine scores the OAs (that I think are artifacts), there is no leak rate. My guess is that the mask is still partially dislodged, and that it's dislodged in such a way that at times the airflow to the mask is simply being pinched off pretty completely.

At any rate, it looks like you somehow manage to reseat the mask in your sleep around 5:08 since a normal breathing pattern along with a 0 leak rate both start back up around that time. My guess is that the OA scored at 5:11 is a real OA, but that none of the OAs's scored between 5:00 and 5:08 are real.

Now an obvious question to ask: Is it really possible to reseat the mask and fix the leak in your sleep? The answer it, yes. Or perhaps more precisely, you may have woken up for a minute or two right around 5:07, realized the mask was askew, pulled it back into place (and fixed leak in the process) and then immediately fell back asleep by 5:08. If the wake lasted less than 3 minutes, you most likely would not remember it in the morning---most people don't remember a wake unless they're awake for at least 5 minutes or so.

[Sonnyboy]

Do you have auto on/off turned ON by chance?
Mode: Autoset for Her
It looks to me like the machine lost track of your breathing around 5:00:45. Right before that, the leak is rather high, even though it is still technically under the Red Line. It could indeed be that the mask got partially displaced in such a way that the machine just could not figure out what was going on.

It looks like the machine turned itself OFF around 5:02:15 (when the gap in the pressure curve starts) and then it looks like the machine turned itself back ON around 5:03:45. Can you look at the list of "sessions" in the left sidebar in the daily data and see if there is a session that ends at around 5:02:15 and another that starts around 5:03:45?
End: 5:02:13
Start: 5:03:33
Even after the machine has either turned itself back ON or has found some small evidence of breathing going on around 5:03:45, your AirSense is still having a lot of trouble tracking the breathing. And where there is some clear air movement going into and out of the lungs, there's also a relatively high leak. Where the machine scores the OAs (that I think are artifacts), there is no leak rate. My guess is that the mask is still partially dislodged, and that it's dislodged in such a way that at times the airflow to the mask is simply being pinched off pretty completely.

At any rate, it looks like you somehow manage to reseat the mask in your sleep around 5:08 since a normal breathing pattern along with a 0 leak rate both start back up around that time. My guess is that the OA scored at 5:11 is a real OA, but that none of the OAs's scored between 5:00 and 5:08 are real.

Now an obvious question to ask: Is it really possible to reseat the mask and fix the leak in your sleep? The answer it, yes. Or perhaps more precisely, you may have woken up for a minute or two right around 5:07, realized the mask was askew, pulled it back into place (and fixed leak in the process) and then immediately fell back asleep by 5:08. If the wake lasted less than 3 minutes, you most likely would not remember it in the morning---most people don't remember a wake unless they're awake for at least 5 minutes or so.

Thank you

[palerider]

Do you have auto on/off turned ON by chance?
Mode: Autoset for Her

she means "do you have SmartStart turned on?"

[Sonnyboy]

Do you have auto on/off turned ON by chance?
Mode: Autoset for Her

she means "do you have SmartStart turned on?"

Oops

SmartStart is turned off.

Thank you

[palerider]

Do you have auto on/off turned ON by chance?
Mode: Autoset for Her

she means "do you have SmartStart turned on?"

Oops

SmartStart is turned off.

Thank you

I'm just here to translate respironics speak into resmed speak

[robysue]

Sonnyboy,

After PaleRider provided the necessary PR to Resmed translation, you said you have SmartStart turned OFF

And yet you also write:

It looks like the machine turned itself OFF around 5:02:15 (when the gap in the pressure curve starts) and then it looks like the machine turned itself back ON around 5:03:45. Can you look at the list of "sessions" in the left sidebar in the daily data and see if there is a session that ends at around 5:02:15 and another that starts around 5:03:45?
End: 5:02:13
Start: 5:03:33

Since SmartStart is turned OFF, the machine should not have turned itself off at 5:02:13 and back on at 5:03:33. So someone or something turned the machine off and back on.

Do you own a cat? Of course even if the cat turned the machine off, you must have turned it back on.

Any evidence of a power glitch that night? Did you wake up to blinking clocks anywhere in the house?

More likely, howver, you are the one who turned the machine OFF and then back ON. Do you remember waking up at all?

This next question may sound off the wall, but I'm going to ask it anyway: Did you take any Ambien or other prescription sleep medication that night?

If you were awake or partially awake enough to turn the machine off and back on, then those OAs around the gap in the pressure curve may be a combination of mis-scored wake breathing combined with a very bad mask fit that made it difficult for the machine to figure out what your breathing pattern was.

[Jay Aitchsee]

This is interesting to me because, as I mentioned above, I've had the same thing happen with nearly identical graphics. Flat flow curve, a few breaths and apneas with a stop and restart. I don't have auto on/off selected either, so I've wondered about that. I have an S9 with one of the latest firmware.

I do think it has something to do with the mask/airflow. At this point, I'm not sure which mask I was using when this has happened to me. Could have been either the P10 or the DreamWear.

Instead of a large leak, what if the airflow was being blocked, or nearly so, with alternating short unblockings? What would that look like?

[Sonnyboy]

This is interesting to me because, as I mentioned above, I've had the same thing happen with nearly identical graphics. Flat flow curve, a few breaths and apneas with a stop and restart. I don't have auto on/off selected either, so I've wondered about that. I have an S9 with one of the latest firmware.

I do think it has something to do with the mask/airflow. At this point, I'm not sure which mask I was using when this has happened to me. Could have been either the P10 or the DreamWear.

Instead of a large leak, what if the airflow was being blocked, or nearly so, with alternating short unblockings? What would that look like?

No ambien or prescription sleep medicine.
Two cats.

It happened again last night. Leaks rose to about 18 and leveled off, pressure decreased, machine turned off for about 5 minutes.
I don't remember turning the machine off and on. I do remember feeling like my nose was blocked and adjusting the mask.
I did tighten my mask recently so maybe I need to loosen the mask a bit.
Looks like Jay might be correct.

[Jay Aitchsee]

So, an experiment. After about 2 minutes of breathing, I blocked the pillow openings of my P10 leaving the vent open. About 30 sec later, I let some air escape (UF1) and after another 30 sec some more (14:39:20). Then I left the openings blocked until 14:43:40 when I started breathing through the pillows again. At 14:44:20 I let the air exhaust freely for about 2 minutes until I started breathing through the pillows again and then shut the machine off. (the machine did not record these last breaths)
So with the pillows blocked, it looks like the machine runs the FOT and if nothing happens for about 1 minute it stops recording (14:40:20). During this time the fan did not quit and the mask continued to vent.
Notice SleepyHead scored the UF1 but there was no scoring from the S9. However, if the timing of the letting the air escape had been different, it probably would have scored something.
Interestingly, the machine also stopped recording after about 1 minute of max leaks (shaded area) and didn't start again, although I did take a few breaths before shutting down. Again the blower continued to run even though it wasn't recording.
Also notice, the pressure line didn't restart until the max leak period started at about 14:44:20. And it restarted at 6.8, which was the auto setting max. (6.4-6.8 )
I included the mask pressure graph to make things a little easier to see.
Look familiar, Sonnyboy?


http://i1090.photobucket.com/albums/i37 ... jdcd9p.png

[robysue]

Jay,

Thanks for the experiment. I think we have our answer as to what's going on here.

One question:

So, an experiment. After about 2 minutes of breathing, I blocked the pillow openings of my P10 leaving the vent open. About 30 sec later, I let some air escape (UF1) and after another 30 sec some more (14:39:20). Then I left the openings blocked until 14:43:40 when I started breathing through the pillows again. At 14:44:20 I let the air exhaust freely for about 2 minutes until I started breathing through the pillows again and then shut the machine off. (the machine did not record these last breaths)
So with the pillows blocked, it looks like the machine runs the FOT and if nothing happens for about 1 minute it stops recording (14:40:20). During this time the fan did not quit and the mask continued to vent.
Notice SleepyHead scored the UF1 but there was no scoring from the S9.

As I understand what you wrote: When the machine stopped recording at 14:40:20, the Resmed APAP stopped recording everything--i.e. no pressure, no leak data, no nothing, but the blower did not go off. That correct? What about the session begin/end times data? Does it show the end of one session and the start of another? That's what Sonnyboy's data showed: One session ended where the machine quit recording the flow rate and pressure data and another started where the machine started recording the flow rate and pressure data after the gap.

That's a bit different than what the PR System One does under similar circumstances. When a PR System One machine is on, but loses track of the breathing, that part of the detailed data gets flagged as "Breathing Not Detected" and there's a gap in the flow rate data, but the pressure curve and the leak curve do not have gaps in them. Here's an example from my own data:


Those big dark gray spots are where my S1 could not track my breathing. But notice that the Pressure and Leak graphs continue throughout the time the S1 machine is on, but cannot track the breathing.

[Jay Aitchsee]

As I understand what you wrote: When the machine stopped recording at 14:40:20, the Resmed APAP stopped recording everything--i.e. no pressure, no leak data, no nothing, but the blower did not go off. That correct?

Does it show the end of one session and the start of another?

Yes, the machine stopped recording but kept running. Two sessions were recorded. The first ended at 14:40:23 and the second started at 14:43:39 and ended at 14:45:42 (actual session times as reported by SleepyHead. Times in the post were my estimates from the graphic). I do not have SmartStart enabled.

So I think we can conclude, if SmartStart is not enabled and there is an apparent shut down, the operator turned off the machine or the machine stopped recording due to a lack of detected breathing. It looks like the FOT running for 1 minute could be the trigger. But this still leaves a lot of room to wonder in individual cases. Did they, or did they not turn off the machine? Did the pillows collapse? If so, did the patient breathe through the mouth? Etc.
In the one case I think I can remember, the pillow type mask had slipped under my chin and the pillows were tight against my neck preventing airflow creating a situation similar to that above.
I think there is a pretty strong clue in looking at the pressure and the flow rate charts. Notice in the experiment and in Sonnyboy's chart, the flow 'flat lines' and then the pressure line stops about a minute after the last breath is detected. If the machine were turned off with normal flow, both lines should end at the same time.
Interestingly, in looking back over some old charts, it looks like the pressure line normally doesn't start until about a minute after breathing is detected after the machine is turned on.
EDIT: I've never used SmartStart, but this looks to me as if it is probably that technology at work. Without SmartStart enabled, it doesn't shut down the blower, but it stops recording some time after breathing is no longer detected.