How accurate does a remstar 60 series calculate AHI?

[DozingOff]

Do these auto setting machines accurately calculate AHI? Just curious as I am trying to read my sleep study report from last month and it shows lower numbers without cpap than with it. (I never got to the titration part of the study because I took too long to fall asleep. test started at 10 and I had to show enough apnea before 1 am and I didn't fall asleep till around 12:30) I'm doing another study next week and I hope I'm not weird for saying this but I hope they find something. I sleep great since using the machine. (I used it for 2 weeks and am stopping till after the study) Just not looking foreword to restless nights... (My machine reported an AHI of 18 when I first used it and now is showing around the 2.1 range with a 90% pressure averaging 9.7) Thanks for all of the great information you guys have been providing here.

[Paul B]

I think they're probably quite accurate depending how they define AHI, which may vary from one manufacturer to another. It's more important to compare the variation of your AHI from one day to another to look for trends in your treatment.

[JDS74]

The calculation of AHI depends on three variables, OSA, Hypopnea, and CSA

Of these three, the OSA component is the most accurate, especially with the later model machines as they can measure whether of not the airway is open or closed. In my experience, the lab said no obstructive events while my home machine report no obstructive events.

Hypopneas are the next in accuracy since it is a measure of airflow reduction. Because there is no input from sensors about the amount of respiratory effort, the machine cannot distinguish between a partial airway collapse and the patient just not taking as deep a breath as is normal for them at that time. For some patients, the amount of air inhaled, breath-to-breath varies and this variation may be enough for the machine to flag it as an hypopnea. In my case, the sleep lab in my most recent study, flagged only a few hypopneas while my home machine routinely flags 6 to 8 per hour. My sleep doctor and I concur that the cause of this difference is my breathing rate variability. Most patients won't have this extreme variability in breaths per minute.

The central sleep apnea variable is the least accurate since the machine cannot distinguish between centrals that naturally occur during the transition between awake and asleep. These are ignored in a sleep lab study. The machine also doesn't have any input from the respiratory effort sensors that a sleep lab has nor the EEG data. Both of these are needed to distinguish true centrals. In my case, the lab said 0.8 AHI while my machine routinely reports 1.2 to 2.4 central events per hour. The difference is that I have a large number of arousals (18 per hour) and awakenings (perhaps 1 - 2 per hour) and sleep-wake centrals I experience are recorded at home and ignored (correctly) in the lab. That said, the CSA numbers are pretty good anyway.

Overall, the machine computed AHI is a good measure of how well your therapy is doing. It is likely just a little higher when computed on a home machine than the sleep lab would measure for the exact same sleep experience because some of the events the home machine will count will be discarded in the sleep lab.

[photonic]

Great explanation!

[archangle]

I think they're probably pretty good overall. Realize that the meaning isn't all that exact anyway. Even trained human techs may disagree when scoring apneas for a sleep test.

I don't think the "errors" the machines make are that significant in terms of therapy, but here's where I think "errors" can occur.

- They don't know whether you're awake or asleep. Your normal breathing while awake may show up as apneas even if nothing is wrong. When you have a sleep test, they ignore what happens when you're awake.

- They may call a central apnea an obstructive apnea, but this doesn't change the AHI number.

- The hypopnea measurement is based off of your breathing rate before the apnea. A few deep breaths may cause it to show a hypopnea where a human evaluator wouldn't score one.

- Depending on the standards used, you don't score a hypopnea unless there is an O2 desaturation or an EEG arousal. The CPAP machines can't see this, so they may score "false" hypopneas.

- An apnea needs to be 10 seconds or longer. Since your breathing isn't always a simple on-off function, there's some uncertainty in deciding when an apnea starts or stops. When you have an apnea that is borderline in terms of being long enough or complete enough, the machine may score it "wrong."

All in all, I think this is not a big difference in terms of treatment. Realize that even an in-lab test doesn't give a clear, exact, black and white AHI number. Look at the airflow waveforms in SleepyHead and see whether the apneas and hypopneas look "real and severe."

I think you should not obsess over the question of accuracy or use it as an excuse to ignore the results. In particular, don't obsess over a point or two.

[Protectionist]

The last time you used your Philips Respironics RemStar Auto with A-Flex...
was last night (on Saturday, 8 November 2014)
You had an AHI of 3.59, which is considered technically "treated"
You machine was on for 9 hours, 28 minutes and 46 seconds.
You had no major mask leaks (maximum was 20.00 L/min).
Your pressure was under 13cmH2O for 95% of the time.
Philips Respironics System One (60 Series)
RemStar Auto with A-FlexModel 561P - P072032002372
PAP Mode: APAP (Variable)
Min 6 Max 13 (cmH2O)
Date Sleep Wake Hours
8/11/2014 12:19:35 06:53:46 09:28:46
Cheyne Stokes Respiration 0.41%
Large Leak 0.00%
Clear Airway 0.32
Obstructive 1.16
Hypopnea 2.11
RERA 2.43
Flow Limitation 1.79
Vibratory Snore 1.37
Vibratory Snore (VS2) 3.27
Pressure Pulse 2.95

The above is when I set the prescribed pressure 6cm- 13cm. and can someone explain the term technically treated( sleepyhead s/ware)


the one below is when i did my own setting at 11cm- 15cm.

The last time you used your Philips Respironics RemStar Auto with A-Flex... was last night (on Wednesday, 5 November 2014) You had an AHI of 1.22, which is considered reasonably good You machine was on for 6 hours, 34 minutes and 21 seconds. You had no major mask leaks (maximum was 20.00 L/min). Your pressure was under 15cmH2O for 95% of the time.

Philips Respironics System One (60 Series)
RemStar Auto with A-FlexModel 561P - P072032002372
PAP Mode: APAP (Variable)
Min 11 Max 15 (cmH2O)
Date Sleep Wake Hours
5/11/2014 23:54:27 06:39:14 06:34:22
Cheyne Stokes Respiration 0.00%
Large Leak 0.00%
Clear Airway 0.15
Obstructive 0.76
Hypopnea 0.30
RERA 1.83
Flow Limitation 0.76
Vibratory Snore 1.06
Vibratory Snore (VS2) 1.98
Pressure Pulse 2.13

[Wulfman...]

Do these auto setting machines accurately calculate AHI? Just curious as I am trying to read my sleep study report from last month and it shows lower numbers without cpap than with it. (I never got to the titration part of the study because I took too long to fall asleep. test started at 10 and I had to show enough apnea before 1 am and I didn't fall asleep till around 12:30) I'm doing another study next week and I hope I'm not weird for saying this but I hope they find something. I sleep great since using the machine. (I used it for 2 weeks and am stopping till after the study) Just not looking foreword to restless nights... (My machine reported an AHI of 18 when I first used it and now is showing around the 2.1 range with a 90% pressure averaging 9.7) Thanks for all of the great information you guys have been providing here.

They're pretty accurate......when you're asleep. They use algorithms, sensors and "airflow" to attempt to determine any events. While there can be some anomalies, for the most part, they do a pretty good job. While there can be some variations from night to night, what's important is to be able to look at longer term data and trends. You need to remember that it's not a full-blown all-wired-up sleep study. Everything it senses is from your breathing (or lack thereof)......"airflow".


Den

.

[DozingOff]

Thanks again for all of the input. I guess the biggest thing I am worried about is having enough happen so they do the titration part of the study. I feel so much better using the machine.... Just don't want to "Self medicate" Just want enough to show up so they prescribe me a machine and set up the pressure.