Recording pulse oximeters - reliable and predictable?

[gpk111]

I wonder about the level of reliability for recording pulse oximeters. Most O2 readings are taken in docs offices without recording and seem to be reliable. The problem comes in when you try to use them to correlate to overnight breathing events. Sleep study interpreters seem to accept the readings as gospel, a result of proven accuracy of expensive professional models ...or denial!

I bought a Contec CMS50F about a month ago and like it. There are occasional drop outs, but the software seems to filter some/most of them. I am assuming that all oximeter data has to be treated gingerly when worn overnight. I'd be interested in any comments about the drop out rate and reliabilty/predictability of recording oximeters in general, particularly the consumer models. I also wonder if the software algorithms to identify "events" are standardized at all across the manufacturers.

Look forward to your comments.

[elena88]

I wonder about the level of reliability for recording pulse oximeters. Most O2 readings are taken in docs offices without recording and seem to be reliable. The problem comes in when you try to use them to correlate to overnight breathing events. Sleep study interpreters seem to accept the readings as gospel, a result of proven accuracy of expensive professional models ...or denial!

I bought a Contec CMS50F about a month ago and like it. There are occasional drop outs, but the software seems to filter some/most of them. I am assuming that all oximeter data has to be treated gingerly when worn overnight. I'd be interested in any comments about the drop out rate and reliabilty/predictability of recording oximeters in general, particularly the consumer models. I also wonder if the software algorithms to identify "events" are standardized at all across the manufacturers.

Look forward to your comments.

I am still on a learning curve, as I just got mine a few days ago.. I know this is not an expensive professional models, but those dont work very well
if you stay up watching them thinking about how much they are costing you.. That is what happened to me when the hospital sent me home with
one.. results came back.. Normal.. cuz I never slept.. so this little thing is VERY nice for my situation..
and yes, I bought it to see if I could get out of wearing my effing apap.. and yes, I will do anything, anything to get out of this. I will put a tin hat on
my head and dance a jig under the moon naked if that would do it.. In any case, Im very happy with my little oximeter, and when I get over the operator
error learning curve, that will probably help the accuracy of it.. I dont expect perfection from it.. Its just another tool, but I think is pretty sweet.

[ozij]

In my own case, the SPO Medcial 7500 recordings of 4% oxygen drops are highly correlated with the the Resmed S8's registration of Hypopneas. The SPO Medical 7500 lets you drill down to very fine resolution in the charts, as does ResScan; -- SPO's Vitascore registers "movement" i.e loss of signal in the all charts at the bottom graphic. I and I can see the O2 drop accompanying my hypopneas, one by one. There are differences: My ODI is higher that my AHI -- but then, I seem to have flow limitations driving up the auto algorithms (different machines) higher than I can tolerate -- and those may be a source for the additional desats.

I had more hypopneas than apneas in my original sleep study -- am having another study soon, and will try to wear my own oxymeter as well. With an ODI and AHI above 10 most nights , I don't have the time or patience to count the events on the charts one by one for comparison, and sadly, neither Vitascore nor ResScan has an export option of the raw data. They both only export calculated data to pdf's....

[slaaplekker]

Being a scientist at heart and Clinical Laboratory Scientist by training I've had long experience with testing, calibrating and validating complicated laboratory equipment including blood gas machines. There are a lot of things that can go wrong with any piece of equipment which is why laboratories spend a lot of time validating their equipment and comparing it to know standards.

But since personal Oximeters are mainly screening tools and not diagnostic tools by themselves they are good for providing baseline for understanding what your O2 saturation levels are doing over a period of time but not necessarily reliable enough to use to diagnose or prescribe treatment. In other words they are good enough to get a pretty good picture of how you are doing as far as 02 satuaration is concerned while you are sleeping.

I bought one within a week of starting APAP therapy since I knew that the stats from the APAP machine were just one piece of the puzzle and that having an Oximeter would be useful information in filling in another piece of the puzzle. My Dr. and I agreed before I started therapy that it was necessary to have a follow up overnight Oximetery study after a few weeks on therapy to verify that APAP was working as expected to keep my 02 levels in the normal range. As soon as I had a few nights of data from my own Oximeter I talked to my Dr. and he agreed that we needed to get an official overnight Oximetery report so that he could order supplemental O2.

I ran my Oximeter and and the official one from the Pulmonologist one on each hand the entire time. The results from the reports are close enough that I have no concerns about my Oximeter's accuracy at least as a screening tool. Going back to my previous experience though without out way to peroidically and compare it to known standards I will rely on it go give me only general guidance as to how therapy is working since there is no way to know if the accuracy will drift over time as the device ages.

Most dianostic equipment is maintined to rigid standards and verified against know benchmarks on a well defined schedule and I know of no way to do that with a personal Oximetery.

Like our own personal Oximeters the ones used by the Dr's staff when you go for a check are simply used as a screening not meant for diagnostics - you got an abnormal result and they are going to do more testing to get to the bottom of it.
Having said that if you get abnormal results with a personal Oximeter then you need to follow up as soon as possible with a Dr. and verify those results.

[torontoCPAPguy]

I have had the opportunity to compare at length (i.e. multiple times) my Contec CMS50E with the 'laboratory/hospital grade' oximeter in rehab and other places and have always found that the CMS50E is generally identical to the hospital grade unit. It does seem to be a bit more susceptible to movement... especially sudden, jarring movements.... but I expected that and I don't generally expect to produce sudden jarring movements while in bed (excepting last night... it was wonderful, thank-you darling - hope my wife isn't reading this).

Interestingly, I always take my Contec with me to my MD appointments (every other week these days. Sic.) And the Contec CMS50e readings match up quite nicely to manual readings in the doc's office. Wondering when they are going to start using them there?

[LoQ]

neither Vitascore nor ResScan has an export option of the raw data. They both only export calculated data to pdf's....

I don't know what you mean by that. I have both of those programs. Where in either of them can you export to a pdf? The only way I know how to do that is to "print" the file, and then select a pdf printer. That only works if you have a pdf printer installed. I cannot see in either of them an option to make a pdf file directly. Where is that located?


You certainly can export the raw data from Vitascore. It is in edf format. I've toyed with asking gvz for help with importing that into resscan, but he seems a little busy at the moment.

As for ResScan raw data, I think it is found in files with extension rlk. My guess is that rlk is a proprietary file type in resscan. I do not have a card with mine, but gvz's answer of getting the files off of the card is better.

[LoQ]

Hey that Vitascore sleep package looks pretty slick. Maybe a better option would be the reverse -- use Vitascore to view your ResMed data instead of ResScan.

Wow, that's a really interesting idea.

I'm not sure how to do that. Do you have any thoughts on the rlk file type? Is that just edf renamed, or is it something else entirely?

I don't have a card, so I can only deal with files on my computer in the resscan folders.

[LoQ]

The RLK files are not EDF. They are the local ResScan database files in a proprietary format it appears. I've glanced into them a while back. It should be possible to decode the format and directly interface with them, but it would be rather time consuming (for me, maybe someone else would have better/faster luck at it). Also the problem with any proprietary format is that ResMed could change the format at any time ... then it would have to be re-decoded all over again.

So I'm back to probably trying to display my vitascore data inside resscan. I don't have the example files that you mentioned on your thread where you talked about doing this, so I don't know where to put the O2 edf files, what they must be named, how the files must be "patched up" to make them work, etc.

Anywhere online that I could get those examples? Is it possible for someone to bundle up one of the example sets in a zip file that I could download? (It doesn't have to be gvz if someone else is willing to do that.) drop.io is an easy place to upload a file without a lot of hassle and with a 100 MB limit per upload for free.

[ozij]

ozij,

... nor ResScan has an export option of the raw data. They both only export calculated data to pdf's...

A lot of people on this board are recently upset with ResMed because of the auction thing, but I for one still think their stuff is pretty neat. For example, not only is ResScan basically free, but they store all the data on the SD-Card in Open EDF format. If the ResScan app won't export the data in a way you like, you can download another free utility called EDFBrowser, and load the data yourself. There is a "Convert to ASCII" feature you could try to convert the data to plain-text, then import it to your favorite spreadsheet program or whatever... Yes you might have to jump through a couple hoops to get what you want, but it's better than nothing. The SD-Card has a few different EDF files, each containing different streams of data but whatever you see in ResScan is available from the SD-Card files. I am happy ResScan chose to make the data in an open format.

Thanks for the info, gvz. I'm still on an S8 - no EDFBrowser for me, as fas as I know. Ahhh... an S9.... who knows, maybe I'll have one sometime.

[meloukha]

.