How are Pulse Change events determined?

[StuartC]

I have been monitoring SpO2 for the last week using a Wellue O2ring.  Along with the charts of oxygen levels, pulse rate and movement, the Events tab in the Daily Sidebar lists "Pulse Change Events".  I am trying to understand the definition of these.  Specific questions are that I am trying to get answered are:

1. Do these come from the O2ring, or are they generated by OSCAR?
   
2. What is the definition of an event, i.e. how much of a change over what time?
   
3. If I have a pulse spike, does that generate two events (one up and one down)?  In other words, is the number of spikes theoretically half the number of events?
   
4. In the list of Pulse Change events, there is a number (8) in brackets after the time of the event.  What does this indicate?
   

Thanks.

[SevereApnea]

Hi StuartC

I used to track the PR changes hoping to find some correlation with machine settings etc.
I found no correlation: See Post #15 and so I stopped looking there.

You can change the filter for flagging the PR changes under File Preferences Oximetry and change your preferred Pulse Rate change and Duration.

From memory you will have to rebuild the data for this to be recalculated.

From what I see on the charts there may be two changes recorded: firstly an increase in PR and then followed by a decrease in PR, if both pulse rate changes take place within your preferred time frame.

If your pulse rate spikes up by say 15 bpm in a period of 10 seconds, you will get a flag, but if it decreases by 15 bpm over a period of greater than  your selected 10 seconds you may not get a second flag. I often saw PR changes in change of position and also at times of cessation of mask or mouth leaks, spikes in flow limitations, pressures spikes and in association with change in Insp times etc. So many variables.

Here is a chart where PR changed up and then down in the selected time frame.

[attachment=36875]

[StuartC]

Thanks for the reply.  Pointing me to the preferences page answers most of most of my questions.  Then what you said about generating events for either increase or decrease answers the other - and matches what I thought so it is nice to have that confirmed.

The reason I asked is because I have been getting between 175 and 375 (!) per night.  It looks like I should change the settings to reduce that to a less spectacular and more meaningful indicator.  Or just to turn them off.

I have noticed some correlation between the larger pulse spikes and a disturbance in the breathing pattern shown in the Flow Rate trace.  Sometimes it is pretty subtle and the timing between the two is sometimes a bit variable, but I think it is generally there.  These are not disturbances that are detected in any other manner so I have been debating whether they might be indicate something like RERAs, which I can see that I used to get if I look back at the DS1 data that I imported (after I returned the machine) but do not see on in the data imported from my new AS10.  I can say with certainty that the day after the night that I had 375 pulse change events was an absolute shocker.  So I am not certain of what I see and I recognize that I could be seeing things that are not really there because I "want to", but I have not yet entirely discarded the possibility either.

I have attached a couple of screenshots, not to seek diagnostic advice because that is a topic for a different part of the forum, but just to show what I am seeing.  Perhaps it is of interest to you or others.

[sawinglogz]

Apologies, OSCAR's calculation of PC and SD are pretty bad. They don't match any particular clinical measure, so take them with a very large grain of salt.

It's one of many improvements on our list, but we have a long list!

As you suggest, it's probably only helpful if you tweak the preferences to highlight a few big events, otherwise it's too noisy.

[StuartC]

As you suggest, it's probably only helpful if you tweak the preferences to highlight a few big events, otherwise it's too noisy.

Thank you.  "A-tweaking" I will go!

I think I would like to set things so that I do still count the major changes because I remain interested in observing whether there is any correlation between number of pulse changes and how I feel in the morning.  As a non-medical person, it feels somewhat logical that a racing pulse could cause a sleep arousal, whether it is linked to a breathing anomaly or not.  And if sleep arousals are the "enemy", then they are worth knowing about.  Of course, what to about it becomes a whole different story ...

[SevereApnea]

Hi again StuartC,

Good luck in your quest. From one your zoomed in charts from the 31st one thing that jumps out is the cessation of leaks associated with incr. PR.

That was something I also saw early on in my struggles with mask leaks. I have marked up one of my older charts as an example.

[attachment=36915]
Note the Flat areas in orange with low PR, these are deeper periods of sleep, these are what we want. Our muscles get rest when we sleep because they are not exercising, but the heart has to carry on beating. It gets its rest by slowing down during the night. That's what we want, long periods of rest for the heart muscle.

Notice I slept through the ridiculously large mouth leaks!

Then notice the arousals circled in black where the PR jumped up and the Resp Rate increased with wildly fluctuating breathing patterns (Minute Ventilations and Tidal volumes.) Even though there were no recorded OA, HA or RERA these are clearly disturbed sleep, or transitions from one sleep stage to another. Notice the rising PR just before waking up after 5:40 AM.

Then look at the green lines, whenever there was a termination of the large mouth leaks there was a spike in PR. 

Message to self: need to control the mouth leaks! they cause arousals, and disturb sleep, reflected in the marker of increased PR.
Just another piece of the puzzle.

You might find something equally useful by taking a bird's eye view of your charts, including Resp Rate, Tidal volume, Minute ventilation and Leaks for example.

[StuartC]

Thanks for that.  I have looked at several other charts and I agree that changes in leak rate appear to have correlating PR changes.  The opposite is not always true, but tackling the leak rate should make a difference.

I am seeing someone on Wednesday about different masks to trial.  I used tape last night with some success, though still some leaks with characteristics that I did not expect.  It is not something that the DS1 reported much until just a few months ago.  The AS10, which I have only had for a month, has consistently shown huge leaks.

[selfcoacher]

Think of pulse change events as arousals of unknown origin. Some might be respiratory related, some dog related. The respiratory ones you can evaluate through flow graph before the pulse change.

In the case of SevereApnea, a plausible hypothesis is that he gets flow limits, APAP reacts by increasing pressure, Leak rate increases as a result of higher pressure, the leak end event causes arousal which results in pulse change

[StuartC]

Think of pulse change events as arousals of unknown origin.  Some might be respiratory related, some dog related.

Thanks for the reply.  I had more-or-less reached that conclusion.  

Two things then become interesting to me.  1) The impact, if any, of the number of arousals indicated by pulse change (much higher than AHI). 2) That I seem to be the only one interested in these observations on my charts because the neither my doctor, nor the CPAP consultants, nor even the sleep tech who did my sleep study seem to pay any attention to them (did make my GP's eyes open for a moment though).

I have not tried to match pulse increases against increased flow limitation.  I am pretty good at seeing things on charts so it might not be showing up.  Also, my flow limits generally occur in clusters that correlate to low SpO2 along with some variable breathing patterns, which the sleep studies show to be happening mainly during REM.  With that much going on at the same time, working out a root cause is tough.  

One thing I can rule out in my case is response to pressure changes as I am on fixed pressure.  Oh, maybe one other thing, I switched to an Evora Full mask and leaks are largely a thing of the past.

[selfcoacher]

Apologies, OSCAR's calculation of PC and SD are pretty bad. They don't match any particular clinical measure, so take them with a very large grain of salt.

It's one of many improvements on our list, but we have a long list!

As you suggest, it's probably only helpful if you tweak the preferences to highlight a few big events, otherwise it's too noisy.

Clinical measures are interested in pulse rises within 8 seconds.  Pulse drops need to be ignored and removed for the metric to be more useful and representative of actual arousals.  Hopefully that is simple enough to implement.

Another thing they use in the literature is to use the highest pulse rate from previous 12s as a reference for the delta pulse rise.