First oximeter use -- what does it say?

[GuyScharf]

I hooked myself up to a CMS50F last night and imported the data into OSCAR this morning.  I'd appreciate any thoughts you might have on the results.

I know leak rate was high last night; usually it's much better than this.  More snores than usual and they caused pressure increases as expected; probably I was on my back at that time. Flow limit graph is typical for me.

I see that SpO2 values were especially variable during a snoring episode.  I'm guessing that would be common?

I know the oximeter graph shows starting after CPAP starting and ending after CPAP ending.  That's real--I started the Oximeter recording late and turned of off after I got up.  I have times in CMS50F, computer, and CPAP machine all set the same.  I am importing directly from the oximeter and not using SpO2 Assistant.

Thanks!

        

[Jeff8356]

Looks relatively average to me. Don't know if the snores effect the desat's or not. When mine shows desat's like that it's usually due to RERA's or Hypopnea's. No snoring on my end. Looks like you had some Hypopnea's around those times also.

May help to turn on the dotted lines for Median and Lower Threshold. I don't remember what OSCAR defaults to for lower threshold but the setting is in Preferences>Oximetry. 88% is usually the norm for LT. Yours looks like you dipped below the LT briefly (twice maybe?), not unusual. Anything above 90 is usually considered normal.

[Sleeprider]

Guy, the SpO2 data shows a deficiency with OSCAR where for most measurements we want to know the 90, 95 or 98 percentile, values exceeded 10, 5 or 2% of the time. For example, it is useful to know what our 95% pressure, flow limitation etc. When dealing with SpO2 our focus is exactly the opposite, and we want to know the values at the opposite end of the distribution. The 2-percentile or 5 percentile SpO2 is %the number we really want to focus on, and OSCAR gives us only the minimum and median. If I know that SpO2 is at or below 90%, 5% of the time or below 88% 2% of the time, that has medical significance, while the minimum is simply a meaningless spike or erratic. An even more valuable metric would be time at or below a particular value like 90% or 88%. These are far more common metrics for pulse oximetry than statistical percentiles. On the other hand, the upper percentile summary data for pulse rate is something useful. I bring this up because you are in a unique position to not only recognize this deficiency in meaningful summary data, but are also able to actually do something about it. I could see value in optionally including 2% and 5% value statistical data, especially where SpO2 data is being reported, but the most useful metric for SpO2 would be time in minutes at or below a selected value.

At 10.2 to 15.0 cmH2O pressure in Autoset mode with EPR 3, your results are good, with predominately hypopnea events. We can see an alignment of SC events with flow limitation and SD events correlate with hypopnea. The large leaks interfere with interpretation of much of the night. The cluster of hypopnea at about 00:45 starts with a spike in snoring and rapid increase in pressure. This appears to be a positional event that results in arousal and change in position. From 1:30 to 04:00 appears to be a period of very steady pulse rate and gradually increasing SpO2. This may be a deeper sleep stage. There is an increase in flow limitations and some RERA at 02:20 that does not reflect in the pulse-ox data, and the smooth sleep pattern resumes with an overall low flow limitation. In summary the few respiratory events you have don't appear to adversely affect your ability to maintain good oxygen saturation, and there are no spikes in pulse that we often see with obstructive apnea and arousal. That's what I see...fix the leaks.

[Gideon]

Oximetry right now takes on a more important role than it ever has before. In the face of COVID 92% has become, at least a suggested value to be screened at.
I'd also like to see horizontal lines drawn at 92, 90, and 88% on the chart and time at and below for these 3 values.
I would target a 1.1.2 release, sooner rather than later because of the importance of oxygen levels in COVID.

Let's develop solid reporting standards.

[GuyScharf]

Thank you!  I appreciate your analysis.  As I said, that first night of oximeter use had extremely bad leaks, possibly because somewhat more movement as a result of the sensation of having a strange device on my finger.  Next night's sleep was much better, both in terms of the overall picture and that I felt more rested in the morning.  And leak rate is more typical.  The major cause of leaks is when I move my head on the pillow and drag the mask off-center, causing a leak at the side.  Add that my skin is sometimes more oily than others, so the grip of the silicone cushion isn't as secure as at other times, and leak rates can be variable.

       

And here are is the oximetry section of the Statistics page for last night:

        

I got the CMS50F because I was curious about my SpO2 levels and whether it might be related to overall fatigue.  All of the discussions of oximetry had been going over my head as I didn't have any measurements of my own to relate to; now I do.