Opinion on Sleep Study

[dataq1]

The attached is from an Apnea Link Air home sleep study. (no mask, nasal cannula sensor, and pulse ox) .
The analysis concluded AHI =11.1, RI=13, ODI = 10.5. (using all 7 hours) that resulted in a "medically necessary" prescription for Cpap therapy. 

What is striking to me is how bifurcated the study was. It seems obvious that the status between 2300 and 0300 is markedly different than that from 0300 to 0600, suggestive of some outside influence. 

What arguments should I make to request a new sleep study?  Or should I?

   

[OpalRose]

Is that the entire report?

[dataq1]

The next several pages are almost unreadable, but I'll attach an example. (BTW, I did not stack these, this is what was sent from the sleep lab)

What these pages look like (to my untrained eye) is a listing of desaturation events, that the operator can select (by highlighting) so that the flowrate, snoring, effort, and saturation graphs (for that segment of time) are displayed to put the selected desaturation event in to context.

What is also notable, is that the there were no selections made (on any of the pages) that encompassed the time period of 0300 to 0600.

   

What I think I am alerted to is the marked difference between the two time periods. Almost as if there were two different patients.

[SarcasticDave94]

We'd likely need the analysis pages. I think your DSX500 DreamStation is an Auto CPAP so don't accept anything less than ResMed AirSense 10 (or 11) AutoSet.

[dataq1]

Dave, I do not understand your reply.

[Geer1]

The pulse data looks suspicious which would maybe make me question the oximetry data. The desaturation marks do appear to correlate with hypopnea marks supporting the saturation data may be accurate so perhaps the pulse data is just poorly calculated or reported for some reason (my guess is 23-3 pulse data is not correct due to artifacts or something). If anything I would just ask for clarification that the data is accurate considering how the pulse data seems incorrect and get doctors thoughts on if the data can be trusted or if the test should be redone.

My home sleep study had some incorrect heart rate data as well and it makes me wonder if that happens often enough they just ignore it. I ended up seeing a cardiologist about it (because my max heart rate was over 200 bpm supposedly) who said it must have just been bad data.

[dataq1]

Pulse readings faulty.... hadn't considered that. I believe that the pulse and Spo2 sensor is one and the same, would that call into question the saturation readings as well?  
So, just for fun, imagine the pulse and saturation graphs are ignored.  What you have left is a bunch of hypopneas (readable from the flowrate traces) and 8 Obstructive events, some snoring and a boatload of flow limitations. 

By the way, do you know what "FS" is? (Right below the Flow limitation tics)

[Geer1]

Google oximetry artifacts. They are actually pretty common, any sort of movement will affect the readings. I think pulse rate is more easily affected due to the counting nature. Your pulse rate for that first bit appears to bounce between 65 and 130 indicating it might have for some reason been counting double what it really was. My guess is your heart rate was actually in the 50-70 range most of that time, same as the end data.

Another supporter of that is that your heart rate isn't really going to bounce from 70 to 130 and back and forth for any reason. Just doesn't make sense.

The hypopnea data would be based off nasal canula and effort belt and it matching the desaturations makes me think the saturation reported by oximeter was close if not accurate. Overall I think the pulse rate data is just affected and I assume that is what doctor would tell you if you ask about it.

Not sure what FS stands for, I assume flow something. Don't think it is important in your case, the regular hypopnea, snoring and flow limitation as well as odd apnea all support an obstructive/restrictive breathing issue hence the recommendation to use CPAP.

[dataq1]

Doesn't the pulse-ox use one LED, and the device's firmware sorts out the absorbed wavelengths to register change in capillary blood volume  (to derive pulse) and another wavelength to suggest SpO2. It would seem to me that if the pulse readings were intermittently faulty, it would be chancy to assume that the SpO2 readings are "spot-on".

I also just noticed a small pulse segment prior to 2300 when the pulse went whacky, that seems pretty consistent with the pulse observed from 0315 to 0600, suggestive of an sensor  (or wiring) fault in the period of 2300 to 0300.

It also appears to me that the frequency of desaturations was significantly greater during the "whacky pulse period"  as compared with the stable pulse segment

So unless I there is some other assignable cause for the "whacky" pulse I'm going to suggest that the scoring should be done without the pulse or desaturation data, or better yet, repeat the study completely.

To my untrained eye (and allowing it's hard to tell with any precision) how many H events were assigned in one time segment versus the other, it looks like IF the  scoring done in  the pulse-stable region only , that the AHI would fall below the insurance threshold for therapy.  [That may not make the doctor or the DME happy]

BTW, I thought that the principle purpose of the strain belt was to distinguish central from obstructive events.

[Geer1]

You are digging into this way too deeply...

I guarantee you that your heart rate wasn't going 70, 130, 70, 130, 70, 130... Every single vertical line on that pulse rate graph is garbage, the only question is if your heart rate was at 130ish or 65ish and the reality is you wouldn't have been sleeping for hours if heart rate was 130 so something was causing the heart rate measurement to be off for some reason (I would be very surprised if that wasn't the case).

Does that mean the saturation values are incorrect during that period? Maybe, I don't know the ins and outs about how the specific oximeter and program to review the data works. Unlike the heart rate graphs your saturation graph looks quite reasonable and for the most part inline with the events.

The reality is that you had multiple sensors picking up flow rate, sound (snoring), oxygen saturation, effort belts and all agree and indicate a sleep disordered breathing case.

In the two data views they shared you can also see some interesting information. In the first example you can see how at around 2 AM your heart rate supposedly dropped from ~ 120 bpm to 60 bpm in 10 seconds while breathing stayed constant which is impossible. You can also see how the heart rate graph fluctuates but saturation graph does not indicating the issues with heart rate sensor or calculation does not appear to affect saturation reading. In the bottom example you see some flow limited breaths and fluctuating effort with it although with being a home study it isn't super clear if this was sleep or awake breathing.

The big question is if you are already on treatment why are you getting home sleep studies done and questioning if other ones should be completed? What has been your experience on treatment? Do you have symptoms you believe are related to sleep apnea/disturbed breathing? If so and this study qualifies you for insurance you would be crazy to turn it down for another study that could come back not qualifying for insurance just because it was a good night of sleep. Results are more important than accuracy so step back and think about what you are trying to accomplish.