Respironics / EncoreViewer Data Error ?

[Velbor]

On my latest night of detailed data analysis using the EncoreViewer, I have encountered (for the first time) a disturbing instance of apparently flawed information provided by the machine and/or the software.

The data suggests an average apnea duration of about 8 seconds. This should not be possible, since by definition the minimal apnea duration must be 10 seconds or more.

Here is the daily graph:



There appear to be seven obstructive apnea events recorded. In fact, there are eight apneas. A bit of arithmetic based on the “Daily Events per Hour” table will show that there were 3 events at pressure 6, 4 events at pressure 7, and one event at pressure 8:



Utilization of the information contained in the .xml file created by the Encore Viewer definitively shows eight recorded apneas:



The problem which concerns me arises from the final page of the EncoreViewer report, which contains the only presentation of apnea duration information (nightly only; not for each event). The total duration of apnea for the one night of therapy covered by the report is 1 minute 6 seconds:



Sixty six seconds of apnea, distributed over 8 apneas, yields an average duration of 8.25 seconds per apnea. This is impossible, since Respironics defines an apnea as being of minimum 10 seconds duration. Even if the apneas recorded at 0:24:30 and 0:25:00 are in fact one single apnea event (there is no way to differentiate from the data available a short apnea during each of two contiguous 30-second data sampling periods, from one long apnea overlapping the two time-samples), that leaves 7 apneas lasting for 66 seconds, or an average of 9.4 seconds per apnea, which still violates the definition.

Any explanations? Am I missing something? Velbor

[ww]

You should post your equipment in your profile in text.

Respironics uses some funny math to make many of their calculations and some of their averages are rounded pretty badly. On my data, I see quite a few apneas that are 6 seconds in length, so 10 is not the magic number they use. That considered, your leak is not well controlled and may indicate mouth breathing which distorts the recoded data. What mask are you using?

[Wulfman]

Any explanations? Am I missing something? Velbor

From my point of view, your minimum pressure is too low. Needs to be about 8 cm.

PS. Don't get hung up on Resprionics' "funny math". It'll just add to the frustrations. Work at getting the AHI lower and then there won't be as many events to worry about.


Den

[2flamingos]

How are you seeing the length of apneas? I have only had the software a short time, but have not found how to get that information. I have both Encore Viewer and Encore Pro.

[Wulfman]

How are you seeing the length of apneas? I have only had the software a short time, but have not found how to get that information. I have both Encore Viewer and Encore Pro.

You would need Encore Pro Analyzer (from James Skinner).

Den

[2flamingos]

I went to the web site and entered request for software and am waiting on reply (has been alot longer than 30 minutes as stated on the site). reentered request using different email address as instructed, and still nothing. Any other suggestions?

[Velbor]

Many thanks for your responses.

You should post your equipment in your profile in text.

I do so when it is relevant to the specific topic being discussed. In this case, I can't see that it matters.

Respironics uses some funny math to make many of their calculations and some of their averages are rounded pretty badly.

The differences in rounding (different protocols for the same data on different pages of the report) I have seen. If "funny math" can be identified, this is what I would like to know about.

On my data, I see quite a few apneas that are 6 seconds in length, so 10 is not the magic number they use.

How are you seeing the length of apneas? I have only had the software a short time, but have not found how to get that information. I have both Encore Viewer and Encore Pro.

You would need Encore Pro Analyzer (from James Skinner).

To the best of my knowledge, neither EncorePro nor EncoreViewer provide the durations of individual apneas; only the total time in apnea for each session (on the last page). Knowing this, and knowing the number of apneas per session, the average apnea length can be calculated. This is what I did, and the resulting figure below 10 is the problem. I have looked at James Skinner's Encore Pro Analyzer sample charts (http://james.istop.com/EncoreProAnalyzer/charts/) and find that he too only seems to be displaying and working with number of apneas and total time in apnea - I see no sample charts or data displays with individual apnea event durations. I also see in his sample chart of "Average Apnea Duration" that there is one entry showing about 7 seconds, and several others perhaps a bit below ten. This tells me, along with your comment, ww, that my data observation is not unique. But that does not make it any less problematic. Respironics clearly and in many places defines an apnea as an event meeting several criteria, one of which is that it be of at least ten seconds duration. This confirms that the problem seems to be both real and generalized (i.e. not unique to me). It is a resolution (or at least an explanation) of this problem that I am seeking. Can anyone provide input specific to this issue???

That considered, your leak is not well controlled and may indicate mouth breathing which distorts the recoded data. What mask are you using?

I think that the chart I posted shows leak to be VERY WELL controlled. What are your criteria for thinking otherwise? On what basis do you think that the data might be distorted? What leak criteria would need to be achieved for data distortion to be considered a factor? In any case, at this particular time, in this particular context, I am not concerned about leak.

From my point of view, your minimum pressure is too low. Needs to be about 8 cm. .... Work at getting the AHI lower and then there won't be as many events to worry about.

Den, I agree that the minimum should be higher. In this particular instance, I am not seeking to optimize my therapy, but rather to collect some baseline data for analysis, and comparison to some other data I have collected using these settings. Thanks for your concern.

Don't get hung up on Resprionics' "funny math". It'll just add to the frustrations.

Den, frustration I can handle. It is a fine motivator. Inconsistencies in data, though, annoy me no end. Right now, if there is "funny math" going on, that is what I need to know about. Can you provide specific examples??

Blessings, Velbor

[6PtStar]

I get reported Apneas of 6 and 9 seconds fairly regular. So much for the defination. I agree with Den, Respirionics need to buy their programers a wally World Calculator!!!

Jerry

[Velbor]

I get reported Apneas of 6 and 9 seconds fairly regular. So much for the definition.

Jerry, If this is true, then it is the most frightening notion I have encountered in the xPAP world. It would mean that not only are we comparing apples and oranges (to paraphrase another recent thread), but the fruits themselves are not even real. I can handle a hefty dose of cynicism and doubt, but this is driving me in the direction of concluding that many of the things we THINK we know about the machinery, believing the technical information which has been provided to us, is bogus. I'm not yet ready to accept that view of the world. I hope there are alternative explanations. Velbor

[6PtStar]

Here is a 2 month report from Pro Analyzer. Look at it and it shows 5 Apneas at 6 seconds and 5 at 9 seconds. Some were multiples but they are still reporting 6 and 9 second apneas. I dont think I ever got a 8 second one.



Jerry

[GumbyCT]

Is it possible the number means in excess or over the 10 sec. requirement? Meaning the clock does not start until 10 sec. have elapsed? After all, how could it be an apnea before 10 sec. have elapsed? Just a thought.

[6PtStar]

Is it possible the number means in excess or over the 10 sec. requirement? Meaning the clock does not start until 10 sec. have elapsed? After all, how could it be an apnea before 10 sec. have elapsed? Just a thought.

Someone mentioned this once before. I always hoped it was not that long. 10 seconds + 18 seconds reported equals 28 seconds. Sure sounds like a long time not to breathe.

Jerry

[Velbor]

it shows 5 Apneas at 6 seconds and 5 at 9 seconds. Some were multiples but they are still reporting 6 and 9 second apneas. I dont think I ever got a 8 second one.

Jerry,

I understand this chart a bit differently than you describe it. What I see is 5 nights with an average apnea duration of 6 seconds. And 5 nights with an average apnea duration of 9 seconds. This chart alone does not tell you the number of apneas on the nights in question (other Encore Pro Analyzer charts will provide that), or the durations of individual apnea episodes.

But that is neither here or there. Either way is trouble. If the AVERAGE of a series of numbers is under 10, then at least one of those numbers must be under 10.

And Respironics states unequivocally that "An apnea is indicated if there is an 80% reduction in airflow for 10 seconds compared to the average airflow over an extended period of several minutes or if there is no airflow detected for 10 seconds." Ten seconds. Ten seconds. How can Respironics machines/software be reporting "apneas" with under 10 seconds duration? Is the definition inaccurate? Or is their implementation of sensing / measuring / reporting the definition untrustworthy? Either way, if there is so major, so central, a fault with their system, what data am I able to believe?

I hope someone can restore my faith!! Velbor

Is it possible the number means in excess or over the 10 sec. requirement?

Just saw your post. Part of the difficulty is that this is a derived number which Respironics does NOT report. And it comes from two other pieces of data, one of which Respironics does NOT report directly. Average Apnea Duration is calculated from "Total Time in Apnea" which Respironics DOES report, divided by Number of Apneas, which Respironic does NOT report as such (but which can be calculated from the "Daily Events Per Hour" table, as well as from other sources.) So the real question is whether "Total Time in Apnea" means something other than what it appears to mean, or whether the count of numbers of OA events means something other than what it appears to mean. Don't think so.

Thanks for trying, Gumby. Velbor

[ozij]

I the problem isn't in wrong time per apnea, the the problem must be a result of whatever it is Respironics considers a "day".
Maybe a "Respironics day" in the "Average time in apnea per day" exist only when sessions are longer than 4 hours?
You had 3-4 apneas in that one hour session.

You have to find out where Respironics starts its day and where it ends it, and how it handles sessions.

O.

[6PtStar]

I did not look at the 9 second days but all the 6 second days (April 5,6,8 &12 and May16). The 6 second apnea was the only one. All had a OA of 0.1. Respirionics starts it's day at 12 noon and runs until the next 12 noon unless there is a segment of more than one hour.

Jerry