Apple Watch SpO2 during sleep.

ChicagoGranny wrote: ↑

Wed May 25, 2022 1:00 pm

Janknitz wrote: ↑

Wed May 25, 2022 12:23 pm

Great info, you two. Thanks.

If all exchanges in all threads by everyone in this forum were as informative and civil as that, Pugsy would have much fewer gray hairs to dye.

ChicagoGranny wrote: ↑

Wed May 25, 2022 1:00 pm

Well, thanks. That's good to know. Gramps and I have been wearing Apple watches for about two years with no alerts.

And THAT's good to know!

lazarus wrote: ↑

Tue May 24, 2022 3:43 pm

chunkyfrog wrote: ↑

Tue May 24, 2022 3:32 pm

. . . watch . . . phone . . . device . . . life . . .

Seems all these newfangled computer-interweb-thingies are what life is now.

The Emay arrived today and I've been playing with it a bit. One thing I realized is that it takes 15 seconds to get a single number reading on the Apple Watch using the app, and then the spontaneous readings happen at wide intervals (based on the autosleep graph it looks like every half hour). On the Emay recording pulse oximeter, it monitors continuously and the readings go up and down between 1-3% second by second (haven't looked at the data graphs yet to see how that is handled).

I don't know if the Apple Watch takes multiple readings in the 15 seconds and averages them, or if it simply takes 15 seconds to have enough data for a single reading. But it explains why the AW reading may be off by 2-3% of the pulse oximeter reading which is very much immediate, so there's really no way to understand the accuracy of the AW readings, although they might be useful for trends.

Interesting!

Any suggestions on how to keep it on my finger all night? You mentioned blue tape, is that painter's tape?

Janknitz wrote: ↑

Wed May 25, 2022 6:03 pm

suggestions on how to keep it on my finger all night? You mentioned blue tape, is that painter's tape?

Yes. I (1) use a line of tape up each side then I (2) circle the tip end of the pulse ox with tape to hold it shut and then I (3) circle the finger with tape to hold both lines in place there. That may be overkill, but I've seen nurses and techs do similarly with surgical tape for monitoring thrashers. And I've always been sort of a belt AND suspenders kind of guy.

Note that in the settings you can change what the parameters are of what you consider normal to be. It has no alarm (which I consider an important feature, since sleep won't be disturbed), but changing what you consider normal will change how the various charts and statistics will be presented to you the next morning.

Interesting night. I slept with the EMAY recording oximeter. Unfortunately I was playing so much with my Apple Watch blood oxygen app during the day that my Apple Watch battery pooped out at about 3:30 am, so I don’t have much AutoSleep data to compare to the finger reader.

But the good news is a great result on the finger oximeter. I had an 02 sat ranging from 90 to 97 with an average of 94 (better than many of my daytime readings ). 71% of the readings were above 94.

I’ll have to do it again with my watch fully charged to compare the report with the auto sleep app.

These results were more of what I expected. Before I delved into the AutoSleep app, my casual observations of the health app charts for random blood oxygen readings was that I rarely had low blood oxygen levels at night on CPAP. Maybe that first night with more data was an outlier to begin with. ‍♀️

By the way, the graphs and reports on this recording device are very detailed. It gives two “oxygen desaturation index” (ODI) scores, one for number of 3% drops and one for numbers of 4% drops. I had 1% for 4% ODI and 3% for 3% ODI. It would be interesting to import that data into Oscar to see if it correlated to my AHI. Interestingly I had a “high” AHI (for me) last night. It’s almost always 0.2 -0.3. Last night it was 0.5. Which correlates nicely with the two ODI scores if averaged.

In any case I’m thankful to see I’m not hanging out with a blood oxygen sat level in the 80’s most of the night, and now I have a better way to track it if I have concerns.

Many thanks for the update!

I don't usually suggest routine recording of overnight O2 for friends with good AHI data. For most without breathing comorbidities, O2 usually gets fixed at a lower PAP pressure than what fixes sleep, I understand (with exceptions of course).

That said, $40 may be a small price to pay for the quality-of-life improvement that can come with some peace of mind regarding a niggling concern.

lazarus wrote: ↑

Thu May 26, 2022 9:54 am

Many thanks for the update!

I don't usually suggest routine recording of overnight O2 for friends with good AHI data. For most without breathing comorbidities, O2 usually gets fixed at a lower PAP pressure than what fixes sleep, I understand (with exceptions of course).

That said, $40 may be a small price to pay for the quality-of-life improvement that can come with some peace of mind regarding a niggling concern.

90 is still on the low side and I spent some significant time there. I'm considering tweaking my minimum pressure a bit, but not without more data. Because my AHI was a little higher, it may have just been a one night thing. I do have a few comorbities I should pay more attention to.

I'm curious to maybe try a night without CPAP to see just how bad my apnea is now, but I won't do that to myself because I have apnea almost immediately after falling asleep without my CPAP, and it feels horrible (how I tolerated that before CPAP--I have very severe apnea--I'll never know) and it causes me to have reflux, too. So nope, not gonna play that game. As they say, "curiosity killed the cat". I think I used up all 9 lives and a few more before I started CPAP.

Our electricity went out at bedtime the other night, and fortunately I have a portable rechargeable battery operated generator so I set it up to still have my CPAP until the electricity came back on a few hours later. Cannot sleep without it--satisfying curiosity is not an option!

Setting pressure relief on 3 might help a smidgen, if it isn't there already, just to get some bilevelish-action going, to lessen work of breathing.

Janknitz wrote: ↑

Thu May 26, 2022 12:30 pm

90 is still on the low side and I spent some significant time there.

Seems the significant overnight numbers in general to most pracitioners in many circumstances are the mean and the time below 90, as I understand it:

During a routine OPO [overnight pulse oximetry], one can get an idea about the mean overnight saturation (mean SaO2) and lowest SaO2 during the entire night recording. In general healthy controls, the normal overnight mean oxygen saturation is 96%. A decreased value points to an underlying cardiopulmonary problem.

Another parameter which is clinically helpful and is widely reported in most studies is time spent by the patient under 90% oxygen saturation (T-90) during the duration of the study. An elevated T-90 during overnight sleep study represents the hypoxic burden during the study. It provides the clinician a clue of the presence of coexistence of pulmonary hypertension, heart failure, and hypoventilation symptoms.

The Uses of Overnight Pulse Oximetry
Shruti Singh, Sara Z. Khan, [...], and Arunabh Talwar
Lung India. 2020 Mar-Apr; 37(2): 151–157

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7065557/

lazarus wrote: ↑

Fri May 27, 2022 12:23 pm

Janknitz wrote: ↑

Thu May 26, 2022 12:30 pm

90 is still on the low side and I spent some significant time there.

Seems the significant overnight numbers in general to most pracitioners in many circumstances are the mean and the time below 90, as I understand it:

During a routine OPO [overnight pulse oximetry], one can get an idea about the mean overnight saturation (mean SaO2) and lowest SaO2 during the entire night recording. In general healthy controls, the normal overnight mean oxygen saturation is 96%. A decreased value points to an underlying cardiopulmonary problem.

Another parameter which is clinically helpful and is widely reported in most studies is time spent by the patient under 90% oxygen saturation (T-90) during the duration of the study. An elevated T-90 during overnight sleep study represents the hypoxic burden during the study. It provides the clinician a clue of the presence of coexistence of pulmonary hypertension, heart failure, and hypoventilation symptoms.

The Uses of Overnight Pulse Oximetry
Shruti Singh, Sara Z. Khan, [...], and Arunabh Talwar
Lung India. 2020 Mar-Apr; 37(2): 151–157

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7065557/

Yes, it's not "run to the doctor low" but I think it could be a bit higher. I'm going to measure a few more nights to see if that's an outlier.

I don't know much about Apple watch but last time I checked my Garmin heartrate it is accurate to the machine doctor and nurses use.

Jman91 wrote: ↑

Fri May 27, 2022 5:32 pm

I don't know much about Apple watch but last time I checked my Garmin heartrate it is accurate to the machine doctor and nurses use.

As I understand it, the Garmin doesn't measure SpO2.

My Garmin 245 Forerunner does.
It doesn't automatically detect sleep but I set a time.
It will track movements, heartrate, SPo2 and breaths/min.

Jman91 wrote: ↑

Sat May 28, 2022 6:30 pm

. . . Garmin 245 Forerunner does . . .

Well. Sort of. Maybe. Kinda. I guess:

This study aims to evaluate the accuracy of the Garmin Forerunner 245 . . . . Overall SpO2 was higher than criterion (P<0.001) with a >50% error rate (>80% above 4800 m), and a poor reliability with criterion. The Garmin device . . . failed to provide trustworthy SpO2 values.--https://pubmed.ncbi.nlm.nih.gov/34000752/

I reckon I'm with Granny on this one:

ChicagoGranny wrote: ↑

Sat May 28, 2022 6:54 am

As I understand it, the Garmin doesn't . . .

ChicagoGranny wrote: ↑

Wed May 25, 2022 6:57 am

Janknitz wrote: ↑

Tue May 24, 2022 8:15 pm

(but it has validated the atrial fibrillation detection via the heart rate monitor)

I like the watch, but Apple irritates me in the way they promote this. They mislead by claiming Apple watch will warn of AFIB. The only time the watch does an ECG is when the wearer stops, sits quietly, presses the crown for an ECG, and quietly holds a finger on the crown for 30 seconds. Intermittent fibrillations are common. There is a good chance that most of them happen when the user is not taking an ECG. Of course, the problem of intermittent fibrillation is also present with ECGs in a doctor's office.

Apple watch saved my brothers life, I don't remember all the technicalities he shared, he felt fine but his resting heart rate was 120, abnormal sinus rhytem. He's very stubborn, believed watch was malfunctioning, cause he felt fine. Family finally convinced him to visit ER, he finally went late that evening, diagnosed AFib.

My Fitbit charge 5 rarely detects sp02, so I purchased a finger oximeter, which works 50%