Questions Re Pulse Ox Monitors

[Madalot]

I've read through many threads about the pulse ox monitors, specifically the CMS50E and I'd like to ask a question or two for those that have it --

Does this monitor seems to be fairly accurate, as far as you can tell? Are the reports from it pretty much like the ones you get from the DME? Does it list artifacts? How easy is it to clear data from night to night? Is the software "fairly" easy to install and use? It doesn't have to be super simple as I'm fairly computer literate, but I don't want something really difficult.

I want to purchase a monitor, but I'm also really hoping for one that I can trust the results to send reports to my doctor as well.

One of the reasons I haven't purchase a monitor yet is that I'm kind of waiting to figure out if I can get by with the CMS50E or if because my situation is so weird, I need to go ahead and spend $300 - $400 for a monitor more like my DME uses. I hate to spent the $130 only to find that I'm going to need to spend MORE in a few months anyway. I don't want to waste money.

[JohnBFisher]

... Does this monitor seems to be fairly accurate, as far as you can tell? ...

With no way to calibrate the unit, I think it is accurate. Good nights (low variability) seem to be when I don't have headaches the following day. Bad nights (high variability) seems to be when I have headaches the following day. And the worst headaches are when I have serious desaturations during the night. So, just based on that experience, it seems to track correctly.

... Are the reports from it pretty much like the ones you get from the DME? ...

I'm not sure what the reports from the DME look like. But I did show my reports to my neurologist, who was both interested and confirmed they saw similar readings during my sleep study.

As you might remember (for other readers ... I think Madalot does remember), I have a progressively degenerative neurological problem that seems to impact my brain stem function. And part of that appears to be problems with breathing. It changes from night to night. But I do notice that the CMS50E seems to track per my headaches the following day.

The reports the software produces are fairly simple. But they get the job done.

... Does it list artifacts? ...

Yes.

... How easy is it to clear data from night to night? ...

Very. Just record a new set of data.

... Is the software "fairly" easy to install and use? ...

It's a bit odd to use it. You need to tell it to create a new set, and it waits for you to connect the unit and trigger the upload. But once you get the hang of it, it is pretty easy. Documentation is almost non-existant.

... It doesn't have to be super simple as I'm fairly computer literate, but I don't want something really difficult. ...

Outside of the odd upload process, it is fairly easy to use.

... I want to purchase a monitor, but I'm also really hoping for one that I can trust the results to send reports to my doctor as well. ...

It's not perfect, but it's pretty easy to use. And it does produce good enough reports to provide a discussion with your doctor.

... I hate to spent the $130 only to find that I'm going to need to spend MORE in a few months anyway. ...

Know that feeling! I found it a good enough tool. You need to be careful about the placement of the monitor as it can cause some stinging / bruising of the cuticle. It can actually read through the finger nail and does well if it rests there. But if you slip your finger farther into the unit, it will clamp on the cuticle and bruise it a bit by the morning. Yeouch!



You might want to look at a wrist model (the CMS-50F). It's more expensive, but might be easier for constant use:



http://www.oximetersonline.com/cms50f-w ... met50.html

And that's just one of the suppliers. There are plenty others. I did purchase my CMS-50E from this supplier and had a good experience with them. Otherwise, I have no affiliation with them.

Hope that helps.

[RLAUREN]

I agree with everything that John says about the unit. I the doctors office I had a chance to wear his $800 unit and mine. When I was still or steady moving the numbers were the same. The doctors unit was better able to read with quick erratic movement, CMS not so. For sleeping the CMS is fine. For exercise, it is not.

[Jay K]

I have the wrist version, CMS50F, and the first one I received was very inaccurate sometimes giving readings of 92 percent at the same time that another more accurate oximeter gave readings around 98 percent (all while awake). At times the oximeter readings changed by perhaps 6 percent over the course of 10 seconds (while I was awake). Also, while sitting doing nothing, the heart rate sometimes changed by 20 or 30 bps within the course of 15 seconds.

I returned the oximeter to echostore, they jerked me around for awhile saying there was nothing wrong with the oximeter, and then after a delay finally sent me a replacement which they claimed was a different one but as it turns out still had the exact same problems.

In any event, I would avoid Echostore, which I understand from another post here has rather poor overall customer service ratings. And I'm skeptical about the CMS50F after receiving two defective ones.

[jules]

I bought the CMS 50E - faulty - got it replaced - second one died in less than 2 months - rarely kept a good reading before then - they were not manufactured same month. My views are this is a cheap toy and it is basically a POS. I am probably in the minority.

[Madalot]

Thank you everyone -- for the input on this. I still haven't purchased a monitor because I really want to wait for the "OK" from my doctor to purchase one and hope for some input.

I REALLY appreciate you taking the time to give me your experiences.

[gpk111]

Interesting discussion. I have the following observations about the CMS50-F, which I bought on eBay for $90US directly from Contec (China):

Ergonomically, it's actually pretty well done (if you discount the Chinglish instructions). Customer service is also good. They actually write back promptly and offered my money back or an exchange if I pay postage on way. Fair enough.

Here are the reliabilty holes I found:
1. I am a runner and track my pulse rate pretty carefully to entertain myself on the treadmill. I even calibrated the various machines in the gym! I found the unit useless for heart rate monitoring when doing even light exercise. The readings are all over the place. I do remember seeing something in one of the manuals that it has to be stationary. ..so why market it as suitable for exercising. Oh, well.

2. Much bigger problem: I tried overnight measurements without my APAP to see if I have any desats. It showed some, but the waveshape looked very suspicious, so I investigated further. I discovered that the unit produces a spreadshhet (CSV format) in addition to the proprietary data files (SPOR format). The spreadsheet logs the O2 and HR readings every second. Very useful. Inexplicably, it doesn't log the time. Time waster, but it's possibly to reconstruct the time on the spredshhet.

It appears that in every "event" (maybe 4 of 5 per night), the unit logged zeros for 5 or 10 seconds, followed by a slow return to a normal pulse and O2 level after 10- 20 seconds. When this is translated into a graph in their output reports, it could easily be interpreted as a desat triggered by an aepnea event. I now know better. I speculate that the finger sensor stopped registering due to a defect or sleep movements and doesn't log it as "lack of data," but
a. averages the zeros in as valid readings and
b. doesn't account for the fact that it takes a few seconds to get back to a steady state reading after the zeros.

The artifacts, events/hr and other authoritative statistics are useless and misleading if this isn't considered.

I'm very disappointed, since the unit seemed well made and the data seemed useable and predictable when it's collected under very controlled conditions during the day. Back to the drawing board, I guess. Any other thoughts?

[gpk111]

gvz,
Answer to your second post: You get a flat line for the periods not in use. I tried it over several hours. That's a proper display, in my opinion.

What I don't like is the treatment of zeros (subject of your first post). I'd much rather see a flat line if that's indeed what they are. I believe they somehow bridge the zeros and end up with weird waveshapes.

I speculate that the cheapo units (like these) don't want to advertise they don't sense properly sometimes. I guess we just strip out the events which contain zeros? Lot of work, that!

Good to know a replacement unit probably won't fix the problem. It'll save me the postage to China.

I'm thinking about sending a description of this to my responsive client rep over there, but I'm concerned the language is too large a barrier.

Nice chatting with you. Maybe some other gurus (Arizona Willie, Dave, etc) will pick up this thread. Would be interested in their views.

PS Not clear on how you disected the spoR files. Any way to convert them to Excel readable? Step by step would be helpful.

[Rogue Uvula]

I've used my CMS50E a couple of times with a few drops in O2 showing each time. The last time I wrapped a piece of blue painters tape around it and the O2 stayed steady all night. From this, I am thinking it may be pretty sensitive to motion or getting shifted around on the finger.

[dtsm]

1. I am a runner and track my pulse rate pretty carefully to entertain myself on the treadmill. I even calibrated the various machines in the gym! I found the unit useless for heart rate monitoring when doing even light exercise. The readings are all over the place. I do remember seeing something in one of the manuals that it has to be stationary. ..so why market it as suitable for exercising. Oh, well.

FWIW, in all the websites that sell the CMS contour models, I believe they say can be used with exercise, meaning can do spot checks while in a resting/stationary position. They also explicitly say should not be used during exercise.

[frh]

I bought a CMS 50D+ from Cooper Medical (he participates here) a couple of months ago. Taping it to my finger overnight was not comfortable for me.

I just had surgery and wore an oximeter for 4 days in the hospital without any discomfort. Pictures of the rubber finger thing on the CMS 50F look like it would be a lot more comfortable to wear than a little plastic brick.

It's hard for me to justify spending money to replace something I just bought, especially since my business has been so slow this year.

[pulseoxstore]

Hi everyone.
I am glad to hear some honest feedback on Contec's oximeters. I also work at a DME company and we perform our testing with Nonin Wrist Ox units and Respironics 920M Plus oximeters. I tried several studies on patient's with Contec units that I also tested with our Nonin wristox unit to see how they matched up before becoming a dealer of Contec units. To be completely honest, they are not built as well as Nonin units but they are not junk either. For the price, these units are very good quality. I think the accuracy is there but sometimes the signal does not read as well as other higher priced units. I have had several patients that we have had to redo tests on Nonin wrist ox or 920M Plus units because they did not get good readings. It is the nature of the beast with oximeters. You will get nights with crazy data with high end oximeters too.
To answer the question of what the reports look like compared to a DME here is a report from Virtuox (an independent company used to qualify patients for Medicare) and a report from Contec.

Contec

DME

[gpk111]

Sorry about the long string of data. The middle column is the O2 level, right column is the heart rate and the left column is the time in seconds ( I added that column).

Readings #6251 through #6261 are normal. Reading #6262 is discontinuous and goes through #6267. No signal. At count #6268, the sensor picks up a signal through #6283, but it's displayed at an incorrect level. At #6284, the signal is lost again until #6338. At that time, the unit takes until #6451 or so to seek "normal" levels. Since these data points are somehow averaged to create the wave pattern in the canned reports, one can't tell if the resulting waveshape represents an aepnea event or is a result of misreadings. It appears that if the string of zeros is long enough (eg if the sensor is removed from the finger or disconnected), the report correctly shows it as discontinuous.

In this case, the wave pattern reflects erroneous data, but there is no way to tell without additional inspection.

(sec) spO2 HR
6251 96 47
6252 96 47
6253 96 47
6254 96 47
6255 96 47
6256 96 47
6257 96 47
6258 96 47
6259 96 47
6260 96 47
6261 96 47
6262 0 0
6263 0 0
6264 0 0
6265 0 0
6266 0 0
6267 0 0
6268 89 88
6269 89 96
6270 89 96
6271 89 96
6272 89 96
6273 89 96
6274 89 96
6275 89 96
6276 89 96
6277 89 96
6278 89 96
6279 89 96
6280 89 96
6281 89 96
6282 89 96
6283 89 96
6284 0 0
6285 0 0
6286 0 0
6287 0 0
6288 0 0
6289 0 0
6290 0 0
6291 0 0
6292 0 0
6293 0 0
6294 0 0
6295 0 0
6296 0 0
6297 0 0
6298 0 0
6299 0 0
6300 0 0
6301 0 0
6302 0 0
6303 0 0
6304 0 0
6305 0 0
6306 0 0
6307 0 0
6308 0 0
6309 0 0
6310 0 0
6311 0 0
6312 0 0
6313 0 0
6314 0 0
6315 0 0
6316 0 0
6317 0 0
6318 0 0
6319 0 0
6320 0 0
6321 0 0
6322 0 0
6323 0 0
6324 0 0
6325 0 0
6326 0 0
6327 0 0
6328 0 0
6329 0 0
6330 0 0
6331 0 0
6332 0 0
6333 0 0
6334 0 0
6335 0 0
6336 0 0
6337 0 0
6338 89 91
6339 89 91
6340 89 91
6341 89 91
6342 89 91
6343 89 91
6344 89 91
6345 89 91
6346 89 91
6347 89 91
6348 89 91
6349 89 91
6350 89 91
6351 89 91
6352 89 91
6353 89 91
6354 89 91
6355 89 91
6356 89 91
6357 89 91
6358 89 91
6359 89 91
6360 89 91
6361 89 91
6362 89 91
6363 89 91
6364 89 91
6365 89 91
6366 89 91
6367 89 91
6368 89 91
6369 89 91
6370 89 91
6371 89 91
6372 89 91
6373 89 91
6374 89 91
6375 89 91
6376 89 91
6377 89 91
6378 89 91
6379 89 91
6380 89 91
6381 89 91
6382 89 91
6383 89 91
6384 89 91
6385 89 91
6386 89 91
6387 89 91
6388 89 91
6389 89 91
6390 89 91
6391 89 91
6392 89 91
6393 89 91
6394 89 78
6395 89 68
6396 89 68
6397 89 68
6398 89 61
6399 89 56
6400 89 56
6401 89 56
6402 89 50
6403 89 50
6404 89 50
6405 89 50
6406 89 50
6407 89 50
6408 89 46
6409 89 46
6410 89 46
6411 89 46
6412 89 46
6413 89 48
6414 89 48
6415 89 48
6416 89 48
6417 89 48
6418 89 48
6419 89 48
6420 89 48
6421 89 48
6422 89 48
6423 89 48
6424 89 48
6425 89 48
6426 89 48
6427 89 48
6428 89 48
6429 89 48
6430 89 50
6431 89 50
6432 89 50
6433 89 50
6434 89 50
6435 89 50
6436 89 50
6437 89 50
6438 89 50
6439 89 47
6440 89 47
6441 89 47
6442 89 47
6443 89 46
6444 89 46
6445 89 46
6446 89 46
6447 90 46
6448 90 46
6449 90 46
6450 92 46
6451 94 46
6452 94 46
6453 95 45
6454 95 45
6455 95 45
6456 96 44
6457 96 44
6458 96 44
6459 96 45
6460 96 45
6461 96 45
6462 96 45
6463 96 45
6464 96 45
6465 96 45
6466 96 45
6467 96 45
6468 95 45
~~~~~~~~~~~~~~~~~~~~~~~~

[Goofproof]

1. I am a runner and track my pulse rate pretty carefully to entertain myself on the treadmill. I even calibrated the various machines in the gym! I found the unit useless for heart rate monitoring when doing even light exercise. The readings are all over the place. I do remember seeing something in one of the manuals that it has to be stationary. ..so why market it as suitable for exercising. Oh, well.

FWIW, in all the websites that sell the CMS contour models, I believe they say can be used with exercise, meaning can do spot checks while in a resting/stationary position. They also explicitly say should not be used during exercise.

If it says in your cars owners manuel, that your car won't fit thru a one lane bridge, while going sideways, and you try it anyway. It doesn't make sense to complain because you keep hitting the sides of the bridge. It's not a defect in the car or it's owners manuel, it's probably the driver.

If you really want A pulse Ox that measures while running, break out the bucks! For the difference in cost I can stand still for a minute. Jim

For a one lane bridge, buy a shorter car.

As far as the CMS 50 D, maybe the size of the batteries make it harder to stay on, My CMS 50 E, stays on all night no problem, no tape, no hurt, and I toss and turn all night.

[gpk111]

[quote="gvz"] Lately I've been working on a utility that converts the .spoR data to the .EDF format that ResScan will digest. Initially, all those 0 values would translate as major downward spikes in the ResScan chart since ResScan was plotting exactly what was given to it (taken directly from the Contec data file). But it made for an ugly-looking graph -- all those invalid reading values / downward spikes / etc. My hack workaround was to either strip out the zeroes entirely (causing a small time-skew, nothing else can be done about that), or do replace all the 0's with "last known good" values, which avoids the time-problem, but is not 100% accurate either because it just substitutes the most recent valid reading for the invalid ones. I also set up an optional filter to exclude/replace reading values that are outside preset ranges, so if you know your body and your pulse generally stays within 75-100, and you get anything below 45, then either strip it out (causing a time skew of 1 second for everything afterwards), or replace it with the last-known-good pulse.

quote]

I'd definitely substitute the zeros with the last known good value. Ideally, you'd interpolate between two valid data points. The problem in trying to do this is the time it seems to take to seek a steady state level. In my data string #6338 to #6453 for example. It shows real data, but it's invalid. Sometimes it seems to find a steady state in a few seconds. in this example, it took a little longer: 15 seconds.