CSR issues - should I worry?

Hi, all.

New to the board here, hello to everyone. I have read extensively on the board here, quietly lurking and learning for the last week or two. This is a phenomenal collection of expert knowledge and a great support community, as far as I can see. I have studied chart and waveform analysis, and have viewed all the info videos on CPAP in general and SleepyHead software, so I would say I am a fairly well-informed newbie at this point. I do have a couple of questions about my readings in particular, and some CSR events that have showed up that give me some concern.

I have an Airsense 10 Autoset with Dreamwear nasal mask, and I've had it for 9 months. Getting noticeable better rest than before CPAP, I was diagnosed with AHI of 30 something, and with CPAP it's usually 5 or less. I just discovered Sleepyhead recently and got into data analysis, etc. Based on discussions here, I have turned EPR off, but otherwise have stayed with my basic prescription 5-15 cm H20.

Since I started using SH software, 17 nights ago, I have had 4 significant (like 30 mins long) CSR events and 3 others that had CSR-like periodic breathing, but without Central Apneas in-between. I normally get 0 to 1 CA's per night, typically, and have almost all OA, or Hypopneas, normally. Unless I get the CSR events, which always seem to happen just after I fall asleep, if and when they occur.

I have some charts, here is one of the most severe CSR events, and the zoomed-in version of it, and then one of the not-quite-CSR events, just the zoomed-in version of that one, as the board only seems to allow me to attach 3 files max. (?)

After browsing the 'net and reading about CSR's, I assume I am a walking corpse at this point, just looking for a place to fall down! Should I be concerned about these events? Should I talk to my Sleep Dr. about them, and if so, are they important enough to schedule an appt. right away? Otherwise, my next visit is in about 6 months.

Finally, one other question, I can't seem to find the leak redline option in the prefs, I see there's an option to flag leaks on the CPAP tab, but nothing about a red line. Is that a deprecated feature, perahps?

Thanks in advance for any advice, suggestions or other input!



CSR Event Max:
CSR Event Zoomed:
Near CSR Zoomed:

Welcome to the forum.

For the red line mystery....go down to where the leak graph is and hover your mouse in the little box where it says "Leak Rate" and then right click the mouse....a little menu will pop up with some choices...scroll down to where it say "dotted lines" and when the cursor is over the dotted line selection another menu should pop up and let you put a check mark in the box for "leak rate upper threshold" and you should then have a red line.

We can only do 3 attachments per post. If you want to add more just do another post or you can use imgur.com to host the images and do as many as you want in a post.

For the "almost CSR"....just ignore it. We can all have those from time to time. It's just some Periodic Breathing that leans toward CSR in appearance. Periodic Breathing is nothing more than a waxing and waning of the air flow that lasts a certain amount of time. CSR is but one of many forms of PB...so not all PB is CSR.

For the real appearing CSR....and it appears (for the most part but there's a few breaths in there with flags that don't look so classic CSR) real to me also or close enough to real that I would stand up and take notice but I wouldn't panic. Obviously you aren't a walking zombie corpse and not all CSR means someone is having major heart disease issues like the internet will tend to scare people into thinking. Sometimes CSR is just a bit of unstable breathing with those centrals happening that we don't know why it happens.

30 minutes is a sizable chunk of the night though..if it were 5 minutes I would ignore it since it doesn't happen often.
So how long do these chunks seem to last when you do have them? You've seen 4 out of 17 nights...that's roughly 1/4 of the nights...what has been the duration of the other chunks of CSR that look real to you? Do they tend to be brief or lean towards longer duration? What has been the pressure at when they started? This one above you were at a little less than 13 when it started which may or may not be a factor since you had another time during the night where you got that high and no centrals popped up.

Do you take any medications that might suppress your respiration? Those meds are generally of the pain medication type but can also be anti anxiety meds.

In general I tell people that if something concerns them...at least call up their doctor and alert them to what you see and let the doctor decide if it is worth worrying about. If you were seeing 30 minutes of CSR every night...I would call him up and at least convey that concern to him or her. I have seen this before where people have random periods of CSR that appears real and they obviously aren't ill and don't have any history of heart disease and no one knows why it happens. I have even chatted with a sleep tech about it...he says no one knows why it happens ..they just know it happens sometimes. If it bugs you a lot...call him up and let him know what you are seeing and let him decide if he wants to see you sooner than 6 months.

Your pressure line moves around a lot....what does your flow limitation graph look like. That's the only graph missing in your screen shot that I would be interested in seeing...some of the others might also like to see the snore graph. Something is driving your pressures all over the place and I suspect either the FLs or snores. I have my doubts as to how active the snore graph is but I am thinking your FLs might be what is driving the pressure all over the place.

Here's a nothing to get excited about FL graph.


Here's a really ugly FL graph. Could you share with us a typical night (be sure and include the FL graph and the snore graph if it has much on it) for you...maybe a lower AHI night or is the 4 ish AHI one of your lowest nights already?
If it were my report (without the CSR) and my AHI was 3 to 4 ish without centrals in the mix....I would try a little more minimum (maybe 7) and if I saw the CSR always at around 13 cm I would limit the max to 11 or 12 and see if that stops the CSR without allowing too many OAs and hyponeas to pop up. This would/should smooth out some of that rather wide pressure range variations... I am thinking that the higher pressures are probably related to the FLs....that's why I want to see the FL graph...see just how busy it looks when you have the higher pressures.

Thank you for your quick and informative response! You are amazingly prescient, I have a couple of the issues you touched on that relate to my condition - heart issues, and a fairly strong opioid prescription, which is sadly necessary. I also went back and took snaps of my Flow, Snore and Leak graphs for more information.

You see, I am pretty screwed up health-wise. I had the Big C about 15 years ago, had Non-Hodgkin's Lymphoma, and did full courses of Chemo and Radiation. That left my lymphatic system in a severely damaged state, and I developed terrible Lymphedema in both legs, think Popeye for how bad they are, it borders on Elephantiasis. So, for this I have to sleep in a hospital bed with my legs elevated so they drain at night. This makes we wake up every 2-3 hours having to urinate. I can go maybe 4 at most, but since I am kind of strapped in, I use a urinal for the mid-sleep urinations (point being I don't get up and disconnect for the quick pee).

The other major issue I have is failed spinal fusion surgery about 13 years ago. I had 2 discs fused, and only 1 healed correctly. The other is a MAJOR source of pain and immobility for me, and due to the screws and rods in my back it can't be operated on or corrected. The surgeon who did it was a quack and was found guilty of Medicare/Insurance fraud and doing too many of that type of operations (like 500 a year!). He lost his license to practice medicine in Wisconsin and other states, AFAIK. For the record, that was a "Dr." Cully White, formerly of Milwaukee, WI, just in case he's defrauding others elsewhere these days! Anyway, due to that, I slept in a chair for 10 years and take fairly large doses of opioids to deal with the intense pain and allow me some kind of life. Sleeping in a chair for such a long time really aggravated my leg issues but was the only way I could deal with the pain for a long time.

Anyway, things are somewhat better since I got the hospital bed and got out of the chair. Since I started using CPAP, my sleep is much better than it was. I can only stand to be in bed on my back for about 6-7 hours at the most, after that the pain in my back prevents me from sleeping any longer, despite the major painkillers.

Finally, I do have cardiac issues as well. I had atherosclerosis and had 4 angioplasties and carry around 11 stents now. The cardiac artery that goes over the top of the heart is 100% blocked for me. Happily, my smart heart and body figured out an alternative blood flow source from nearby capillaries and due to alternative secondary blood flow, I did not have a major heart attack. At least not yet, knock wood. I am on a very good heart meds regimen, Alpha and Beta blockers, BPM, Statins, etc. and I have great BP (130/70 to 140/80 typically) when properly medicated. I just had a full annual checkup and EKG last week, and it all came back good, so that's relieving.

I hope that gives you some insight into my situation medically, I hope this is all not TMI.

Thanks again!

Nov 29 (CSR event night) Snore/Leak/Flow graphs:
Dec 6 (Not-Quite-CSR event night) Snore/Leak/Flow graphs:

BTW the duration of the CSR events was typically about a half hour. The shortest one was 15 minutes, but the others were about 25-28 minutes, typically.

P.S. The flow limit was much lower it seemed before I turned EPR off. It was clear from it there was little to no limits, normally and you could see clear indication of limiting activity during REM sleep about 1-1/2 hours after sleep onset. It clearly showed spikes in leakage and flow limits around the time of REM onset every day. Then, I usually wake up to relieve myself about 2-3 hours in, and then go back to sleep for another few hours until someone calls or texts me and wakes me up.

Now that EPR is off (the later non-CSR graphs) it seems there is a lot of low-level flow limitations, background noise on the chart. You can sill see a Leak spike around REM time (about 8:00 AM) but the "noise" of the Flow Limit graph is so high, you can't see the associated spikes at that time, they are "lost in the noise". I have only tried turning off EPR for 3 days now, and haven't decided whether I like it or not, and how it affects my AHI and other stats yet.

Oh, I found the redline, thank you very much! I was looking in totally the wrong place, I expected it to be in the Prefs dialogs, the tutorials here aren't quite clear on that, since they were discussing other Prefs items at the time.

One more quickie question - Before I got SH and started recording my stats on an SD card, the machine was phoning home it's results by Cellular Modem, and I was looking at the Overview stats, and got some puzzling info - the AHI's recorded for 9 months before I started recording daily results were almost always under 2.0, usually way less, like 0.87 and 0.98, etc. However, once I started recording daily results, my AHI is typically 3-6 or so. I am attaching one more screenshot to show this. Just kind of puzzling that somehow recording the daily stats is somehow causing the reported AHI to be much greater, consistently. Kinda weird, or is it just me?

Thanks again!

Thank you for the additional details. It helps a lot.
You've got lots going on for sure and a very logical explanation for the sometimes central clusters looking like CSR.
I do think more likely related to the pain meds than the heart. If it were the heart I would expect to see it a lot more often.
Maybe the randomness is related to dosage and/or time of dose.
Pain meds can suppress our respiration...cause us to breathe more shallowly. When this happens we don't have a "normal" exchange of gases with our little respiration and we tend to breathe more shallow but maybe a little faster and this causes the body to "wash out" or "blow off" too much carbon dioxide.
It's actually carbon dioxide that is the trigger for the brain to send the "breathe" signal to the body. People tend to think it is low oxygen but it's not the case. We actually need higher carbon dioxide levels and when we wash out too much carbon dioxide with the unstable breathing the blood levels don't get high enough to trigger the brain's breathe signal to the body.
So you don't make an effort to breathe...and a central apnea happens because you simply didn't try.

You might keep a diary or log to see if you can correlate time of meds or dosage as impacting the breathing....you don't always have it happen so something must be a little different somewhere.

You probably have had this going on the past nine months but didn't know it because you didn't have a SD card.
Without the SD card you won't have the detailed graphs but you would have the AHI summary data available from those pre SD card nights but I think you would need ResScan to see it and I don't know if it is worth it to install just to see that data.
If you want ResScan...it's Windows only software...send me a PM and I will help you find it.
About all it will tell us is the AHI and the breakdown...you could see if some nights you had more centrals than other nights. To be honest I don't know how important that is at this point.

I would call up the doctor and at least convey to his nurse what you are seeing. Most likely he won't do anything but want to watch things because it's not happening every night....but at least alert him and let him decide if he needs to see you sooner than your next appointment. Better to be extra cautious IMHO.

I don't think it is bad enough at this point to need a different machine but again....that's something a doctor really needs to look at.
He might want you to wear an overnight pulse oximeter to see if these clusters are causing any significant desats.
5 minutes is unlikely to do much but 30 minutes might. Problem with that is you aren't having them every night and 1 or 2 nights is about all DMEs will loan out pulse oximeters for because of battery life issues. You might wear it 2 nights and nothing happen and still not know if oxygen levels are being impacted or not..

So the primary suspect.....the pain meds and the unstable breathing it sometimes causes.
Obviously you can't just stop the meds but maybe you could alter time of dosage and see if that impacts things a bit or even maybe take half a dose if possible.
It would remain my primary suspect until ruled out.

I don't think there is any need to panic or anything like that. I know this looks scary to you but I have seen much, much worse in terms of centrals and CSR just from pain meds than what little you are having.

Mr. Transistor wrote: ↑

Sun Dec 09, 2018 5:23 pm

the AHI's recorded for 9 months before I started recording daily results were almost always under 2.0, usually way less, like 0.87 and 0.98, etc. However, once I started recording daily results, my AHI is typically 3-6 or so. I am attaching one more screenshot to show this. Just kind of puzzling that somehow recording the daily stats is somehow causing the reported AHI to be much greater, consistently. Kinda weird, or is it just me?

Hmmm...that is weird and it's not because you just started using the SD card.
What else changed when the higher AHI started?
This is really where I am so sad we don't have those old graphs to refer back to......
Oh...one other thing...the AHI before the SD card use...and using SleepyHead...might not be totally spot on accurate.
SleepyHead needs some detailed files to get the AHI spot on correct and when the SD card wasn't in the machine you don't have those files.
SleepyHead is making an educated guess...and it won't be 100% accurate.
That's why I mentioned ResScan....ResScan is accurate even without the missing detailed files.
Now it's not hugely off...1 maybe 2 AHI difference...but there is a difference when SH doesn't have those files.

I doubt EPR is a factor in the centrals. You could add it back in and see what happens. When EPR causes centrals its very rare thing to happen.
I also doubt that the trigger is a higher pressure.

Pugsy wrote: ↑

Sun Dec 09, 2018 5:48 pm

Mr. Transistor wrote: ↑

Sun Dec 09, 2018 5:23 pm

the AHI's recorded for 9 months before I started recording daily results were almost always under 2.0, usually way less, like 0.87 and 0.98, etc. However, once I started recording daily results, my AHI is typically 3-6 or so. I am attaching one more screenshot to show this. Just kind of puzzling that somehow recording the daily stats is somehow causing the reported AHI to be much greater, consistently. Kinda weird, or is it just me?

Hmmm...that is weird and it's not because you just started using the SD card.
What else changed when the higher AHI started?
This is really where I am so sad we don't have those old graphs to refer back to......
Oh...one other thing...the AHI before the SD card use...and using SleepyHead...might not be totally spot on accurate.
SleepyHead needs some detailed files to get the AHI spot on correct and when the SD card wasn't in the machine you don't have those files.
SleepyHead is making an educated guess...and it won't be 100% accurate.
That's why I mentioned ResScan....ResScan is accurate even without the missing detailed files.

Sleepyhead does not calculate the AHI correctly from the summary info from the Autoset machine, it’s a bug nothing you can do about it, as Pugsy says Resscan can provide accurate stats for that time if you want. But in my mind your best plan is to look at reducing your AHI now nothing you can do about that old data

The discrepancy between ResScan summary AHI results and SleepyHead summary results when there is no SD card data files from the machine for SH to use is really a very small discrepancy. Talking maybe less than 1.0 difference to maybe 2.0 difference. It's not a huge difference and it's close enough for government work in most situations.
When I saw it...SleepyHead was almost always reporting slightly higher AHI than ResScan reported.
Like ResScan said 1.6 AHI and SH was saying 2.2 AHI.

Dang! Things are sure busy around here. It's been two or three days since my last post, and my topic has dropped off the 1st page already! Wow.

Anyway, I finally found ResScan 5.9 (thank you Pugsy for the hints!) and imported my data from the "before SD card" days, and yes, there is a distinct discrepancy between the numbers, and a LOT, not 1-2 AHI points. As I said SH is showing very small values like less than 1.0 values for most days, and usually under 2.0 overall with a few exception days.

I am attaching a graph from ResScan of my summary data from September 15 through current date (12/12/18) and you can see the AHI data is consistently around 4-6 or so, which is normal for me, AFAIK. If you look several posts up at the summary data from SH I posted, the data is quite different, starting on 11/21/18 when I started using an SD card to catch the detail data and started using SH. If you see the AHI's are VERY low (under 1.0 or 2) right up until that 1st SD card day on 11/21/18.

Looking at the ResScan data, there is almost no noticeable difference between days after 11/21/18, and those before the starting date of using the SD card. With SH, there is a huge spike in AHI numbers after the SD card 1st use date. Those are not false numbers, the AHI reported seems correct. What is not right is the SH-interpreted AHI numbers for days before the SD card was used, so somehow SH gets the AHI wrong for those days with only summary data recorded.

Oh, well, I guess it's not that critical for me, since it's in the past and there is no way to get back the detail data, but going forward the numbers appear to agree between ResScan and SH. That is good, since ResScan has a HORRIBLE UI, SH is MUCH better to use and for understanding the data in many ways, plus SH supplies quite a bit of more detail and information than ResScan does. ResScan is a clusterfun!

Thanks for all your help, I will keep an eye on the CSR events, which I have not had any for the last few days, thankfully. I am trying to match a cause to the symptom, trying to figure out if it's related to the time, my medication levels, the phase of the moon, or whatever. Perhaps this might give some insight to the SH folks if they decide to investigate this "bug" in SH behavior, I would be glad to share my data with them if it might help to diagnose and track this AHI issue, but I'm guessing it's not a great priority, plus as I said, it's only historical summary data anyhow, so it's not super-critical, just nice for tracking trends, etc.

Thanks again to all who commented!

Mr. T

You know the only differences I ever saw were minor but then my AHI is low anyway....maybe the discrepancy gets worse the higher the AHI. I never saw anytime where the difference was greater than 2...but then my AHI rarely hits 2 anyway.

Sleepy head will match ResScan when SH has the detailed data files and as you found out...must have SD card to store the detailed files.
To be honest...not worth redoing SH code just for that little discrepancy. Most people evaluating their data are going to have more than enough detailed data already.
You kinda fell through the cracks.
Besides...not like we can do anything about that old data anyway.
And while ResScan is a major PITA...it is available for verification of any questionable results in SH.

Now rumor control has it that there is a new version of SleepyHead out there in trial and some people (those who are part of the development team) have used it...I couldn't make it work for me and wasn't part of the development team because there's simply not enough hours in the day as it is...so I don't know if that issue was fixed or not in the new version.
And no...I have no idea when that new version will be finally released as a fully easy to install software and out of testing. Only one person working on it (doing actual code writing) and he has his own health issues that slow him down. This current version that we are using now is a bit buggy but then all software has some sort of bugs in it...even ResScan.
That's the main reason ResScan keeps updating the versions. Bug fix....bug fix...stuff like that...or when new machine models are released.

Ok, that sounds interesting. Once they get the new version ready to go, I'll try it and see if they addressed the issue. I understand there is no real push and SH is certainly very usable as-is, so I can wait until it's ready for Prime Time.

Thanks again for all the info and help!

Mr. T