Central sleep apnea periodic breathing

[mollete]

I think that you will find that the definition of CSR includes an apnea in the trough. On the other hand patterned breathing has a hypopnea in the same location. So no, I do not think that it is possible to CSR without apneas.

Well, the definition certainly says that, but the question is about algorithm accuracy.

in this example, it is not yet a complete apnea, the algorithm has labelled this CSR (correctly)(although it's really PB-- SH quirk), but...



all the apneas are called obstructive!!

[vsheline]

Is it possible to have a CSR-type pattern of breathing where the machines will not report it as some kind of apneaic event (ap, hyp, CA)?

Question posed in part because some machines, including mine, do not have access to SleepyHead.

The definition of an apnea, whether in CSR or not, is the same: at least 90% reduction in respiration for at least 10 seconds. The apneas in CSR will be counted in the AHI and also in the RDI (Respiration Disturbance Index) if the machine also reports RERA and RDI.

But if a period of no breathing lasts 9.5 seconds it might not be reported. And if what looks like CSR repeats every 29 seconds (instead of 30 to 120 seconds) it might not be flagged as CSR, but the individual apneas (if lasting at least 10 seconds) will still be reported.

The RDI is equal to the AHI (average per hour of the sum of apneas plus hypopneas) plus the average per hour of RERA events. RERA is a Respiratory Effort Related Arousal, which is an arousal caused by needing to exert too much effort in breathing, usually caused by Flow Limitation.

Flow Limitation is caused by a partial restriction of the airway which limits the Flow rate while we are inhaling, which can make inhalation uncomfortably hard, which may lead to an arousal.

The fully data-capable PRS1 (Philips Respironics System One) machines have been (estimating and) reporting RERA for several years, but most "fully data-capable" ResMed machines do not attemp to (estimate and) report RERA. The A10 AutoSet For Her was the first ResMed model to report RERA. Recently, in some regions, users have reported their standard (meaning not the For Her model) A10 AutoSet also reports RERA. Haven't heard yet whether any of the new AirCurve 10 bilevel machines have started reporting RERA.

RERA events are arousals which disturb sleep, but because the reduction in respiration was less than 50% (or because some other requirement for being classified as hypopnea was missing) these will not be counted as hypopneas and therefore will not be counted in the AHI. (By definition, an event cannot be counted both as a RERA and as an hypopnea.)

So, conceivably, a person who has lots of RERA but few apneas and hypopneas may have a great AHI but may be unable to sleep well and may feel always fatigued and mentally foggy, etc.

Higher pressure or bilevel PAP therapy can help prevent Flow Limitation and RERA. If the person is using fixed-pressure CPAP therapy, it may help to turn on EPR (if available) and increase the Pressure setting by a corresponding amount - for example increasing the Pressure setting by 1 or 2 if now using an EPR setting of 2 cmH2O but the Pressure setting had been based on a lab titration which did not use EPR during the titration. (This would maintain the pressure used during exhalation closer to what had been determined to be needed during the lab titration.)

[Sleeprider]

Sleepyhead software improperly labels periodic breathing as CSR. it's a simple error by the programmer that can be changed by the end user. Just go to preferences/Events and re-label that as PB. In this case, just because Sleepyhead says it is, doesn't make it so.

[vsheline]

all the apneas are called obstructive!!

May be an unusual case where the apneas are both central and obstructive.

The slow smooth rise and fall in the Flow Rate seems to indicate these are likely central in type, but the airway may also be collapsing during the central apneas.

I am assuming here that the OAs (obstructive apneas) are being flagged as such by the machine and that SleepyHead is merely plotting the machine-generated OA flags.

[PaytonA]

Does anyone know how the machines that do estimate RERAs? What do they flag as an arousal?

Best Regards,

PaytonA

[vsheline]

Does anyone know how the machines that do estimate RERAs? What do they flag as an arousal?

RobySue described (perhaps in her excellent tutorial on SleepyHead in the Important Threads section at the top of the forum) what criteria are used by PRS1 machines, and I think the setup manual (Clinician Guide) for the AirSense 10 machines describes the criteria used by ResMed.

If memory serves, it is something like a 10 second or longer period of increased Flow Limitation followed by a sudden increase in Flow or Tidal Volume signifying the deep recovery breaths which would follow an arousal. Something like that.

When reading the criteria used, I remember it seemed likely to me that a significant number of true RERA (true RERA meaning RERA as measured when monitoring brain waves and muscle effort to reliably detect arousal and effort) may be missed, and the number of RERAs reported by the machines would likely be somewhat lower than had actually occurred.

But I haven't seen reports from ResMed or Philips Respironics on the relative likelihood of their algorithms missing a true RERA or being fooled into reporting as RERA something which would not have meet the clinical standards for RERA.

[mollete]

all the apneas are called obstructive!!

May be an unusual case where the apneas are both central and obstructive.

Perhaps, but in this case I believe that this is an entirely central disturbance.

We are fortunate enough to have a side by side comparison of machine scored central and obstructive apneas:


In the central event, there is no flow. 2 Pressure Pulses (Respironics CA technology) are injected and the event is appropriately scored central.

In the second event the limitations of the algorithm are exposed. This is clearly central phenomenon. However, there is still enough flow to prevent the Pressure Pulse to initiate so the algorithm cannot score this as a Clear Airway Event. However, there is >80% flow reduction so the only option is Obstructive Apnea.

The correct interpretation should be since their is still some flow, it cannot be called a central apnea, but 80 - 99% flow reductions need to be put somewhere, so is therefore a central hypopnea.

[richb]

Thanks.

I was curious to see if there were respiratory events that were missed by the machine. Indeed, there is a bit of periodic breathing (put http: in front of that):

//i79.photobucket.com/albums/j147/morbius10/apboard_zpsz5s1y8ze.jpg

In both periodic salvos, the trigger appears to be a arousal, suggesting a CompSAS type of thing. Yet those cycles are long, suggesting circulatory delay may be underfoot.

I was away for a couple of days. I looked at the waveform in the above sample and to me it sure looks like examples of periodic breathing. The trigger looks to be some deep breaths (maybe caused by arousals) that reduce CO2 concentrations just enough to set off the periodic breathing. I think we could refer to the events as periodic breathing with and without apnea. It seems that the medical community only wants to refer to Cheynes Stokes breathing when it is associated with congestive heart failure or end of life conditions. It seems that some of us suffer from periodic breathing scored as HA and CA by the machines with little or no OA involved. By the way I tried straight CPAP at 9 no EPR a couple of nights ago. I got a lot of HAs that probably would have been CAs at higher pressure with EPR. Anywhy I don't think a CPAP machine can be adjusted (for me) to eliminate the periodic breathing. I am trying a replacement machine starting tonight just to rule out machine error. My guess is that I will eventually wind up on an ASV machine and or medication to eliminate my periodic breathing (CAs).

[PaytonA]

Sleepyhead software improperly labels periodic breathing as CSR.

Out of my curiosity, what makes you sure that Sleepyhead improperly labels periodic breathing as CSR.

Best Regards,

PaytonA

[PaytonA]

I think that you will find that the definition of CSR includes an apnea in the trough. On the other hand patterned breathing has a hypopnea in the same location. So no, I do not think that it is possible to CSR without apneas.

Well, the definition certainly says that, but the question is about algorithm accuracy.

in this example, it is not yet a complete apnea, the algorithm has labelled this CSR (correctly)(although it's really PB-- SH quirk), but...



all the apneas are called obstructive!!

Where is it said that the apnea in CSR must be central apnea?