Jeff Duntemann's Contrapositive Diary Rotating Header Image

sleep

A Year and Change on APAP

A year ago this past May, one of my doctors suggested that my lack of energy might be due to sleep apnea. Carol verified this; she has heard me stop breathing numerous times while lying beside me in bed. The doc prescribed an at-home “headband” sleep study, which at least verified his suspicion of apnea. The device (which was just that: a headband with electrodes) recorded an AHI of 33. Basically, I would stop breathing 33 times an hour. This seemed excessive and still does, for reasons I’ll explain a little later. But the next step was obvious: He handed me a prescription for an APAP machine. “APAP” is an adjustable pressure CPAP. The machine senses your breathing, and sends enough air through the hose to keep you breathing, no more.

I shopped around online, and got an NOS (new, old stock) ResMed S10 Auto. It was half the price of a new machine, even though it was still sealed in its original packaging. I bought a couple of different masks, and gave them all a good shot.

At first it made me nuts. I have never been a strong sleeper, and having this thing strapped to my face all night kept me awake. The full-face mask that most people use was a non-starter. I used a few other types of mask, and finally found that I could actually sleep a little using a “nasal pillows” mask, which has these two little soft silicone pads on a single strap that goes behind your head. The two pads each has a tube protruding from the middle, and those tubes go into your nostrils, while the soft pads keeps a good seal. Ok, a reasonably good seal. I still have problems with leaks around the edges of the pads, but that doesn’t negate the machine’s effectiveness.

It still kept me awake. So the doc put me on a new sleeping pill called Belsomra (Suvorexant) which, rather than sedating you, helps neutralize stimuli that prevent you from sleeping–like an APAP mask. And damn, it worked! I slept better than I had in a long time, with no interruptions but my two canonical bathroom breaks.

With the machine in operation, I was throwing just a few “events” every night. There are several kinds, and I don’t have the space to describe them all here. My personal favorite is hypopnea, which is shallow breathing, not airway obstruction. The doc said it doesn’t interrupt sleep. Some of the others I’m still not sure I understand, like Cheyne-Stokes Respiration; but that’s ok, as I think I’ve had it exactly once in thirteen months.

The S10 records everything it senses during the night on an SD card. You can pop the card out and read it any time. My correspondent TRX put me on to a free app called Sleepyhead, which takes the data from the card and throws up all kinds of graphs for Windows, Mac, and Linux. The damned thing literally graphs the shape of every single breath you take. You can see when you stop breathing, along with the following spike in pressure to open your airway again.

Ok. Now it gets a little odd, and a little disappointing. For the first six months I recorded AHIs of .5 to 7, which isn’t bad, especially compared to where I was starting from. The problem is, I still felt the profound lack of energy that’s been dogging me now for several years. I felt a little better, but I wonder if that was just the sleeping pill keeping me from reacting to dogs yipping in their sleep chasing archons in the akasha, or the drip system cranking up in the middle of the night. I stopped using the machine for a week. I didn’t feel any worse, nor better.

I’m still using it. I’ve worked out the optimal sleeping position through a year of trial-and-error: On my right side, with my head on a firm pillow, leaning back just a little to keep the mask from smooshing off my face due to pressure from the pillow. The data the S10 gives me showed me a few significant things: My AHI goes through the roof when I sleep on my back; something like 7-10. Oddly (and so far inexplicably) my AHI also goes up sharply when I sleep on my left side. Nobody can tell me why. So I sleep on my right side. I have to prop my right knee on a second pillow, but it works.

And now it gets more interesting still: I had been slowly putting on some weight for a year or two. So in February I cut my carb intake to as close to zero as I could manage, without starving myself. (Starving yourself doesn’t work. Really. What you might lose, you then gain back after the diet stops, and then some.) My weight went down from 163 to 148-150. It took a couple of months of this for me to notice, but eventually I saw it: As I lost weight, my AHI imploded. At the end of March I had my first perfect night: The S10 recorded no events at all, nothing. As spring continued, I saw my record improve even more: I started having perfect nights regularly, and then two or three (and once, four) in a row.

Carol mentioned something over breakfast one morning: I had lost weight in my face and my neck. I’ve never been seriously overweight, but I’ll be 67 in a week or so, and I’m trying to keep my A1C down to avoid Type 2. I was actually trying to eliminate visceral fat around my waist as much as possible. I didn’t even think I had fat in my face to lose.

I still haven’t regained my energy (which is one reason you don’t see as many Contra entries as you used to; I’m pouring most of what energy I have now into my fiction) and that problem remains unsolved. Maybe I’m just old. I don’t know. Coffee helps some. Beyond that, I’m out of things to try.

My only remaining theory is this: That headband sleep study was bogus. I suspect it was interpreting me jerking around in my sleep as apnea events–I’m an “active” sleeper and always have been. So although I did have sleep apnea, it wasn’t nearly as bad as the headband claimed. That said, I think the S10 has improved the quality of my sleep, which is beneficial in many ways beyond personal energy. This is why I continue to use it. My point here is that quality of sleep is not behind my energy deficit.

I’m still trying to figure that out. In the meantime, if you think you have apnea (spousal reports are good, and even a bad sleep study will give you some broad hints) I recommend two things:

1. Get yourself a recording APAP machine like the ResMed S10. Download Sleepyhead and watch your data, daily if possible. Development on the app has stopped, but it’s still available and works fine. You will learn a lot about how your sleeping position (and weight) affect your breathing.

2. Lose weight. This is good for lots of reasons (Type 2 being most important) but I’m pretty sure at this point that facial and neck fat are huge amplifiers for a tendency to apnea.

I’ve gone on long enough here for this busy morning, but if what I’ve experienced this past year will help my readers, it’s well worth it. Good luck. Cut carbs. Animal fat will not hurt you. (Certain vegetable oils will.) Sleep as much and as well as you can, even if it takes a machine to help you get there. Even (Gasp! The horror!) go to bed at 9 PM if that’s what it takes to get eight hours in before you have to go back to work or school. And pills; Belsomra is something entirely new in the human pharma cabinet. It is utterly unlike the nenzos or Z-drugs. Look into it if you have trouble sleeping.

More as I learn it. Let me know how you do, if you happen to be on this path as well.

Odd Lots

Hose Wars, Part 3: I Love It…But I Hate It

This is a series. Start here if you haven’t already.


Yes, I’m back. I didn’t pause the series because I was tired or busy. I was waiting because I wanted more data to analyze. So as of this morning I had four weeks in with the S10, and I decided to see what the trends were, and talk a little more about the experience itself.

In terms of what it was designed to do, the ResMed S10 Autoset is a complete win. If you recall from Part 1, my headband sleep study indicated an AHI of 36, meaning that over the time I was tested, I experienced and average of 36 events an hour. The events are of various species, some of which I still understand poorly. The biggie is obstructive apnea (basically, your soft tissues close your airway temporarily) which encompassed most of the events reported by Sleepyhead, assuming you include “Clear Airway” events with OAs. (I’m still trying to determine the precise difference between the two categories.) I’ve logged relatively little hypopnea (abnormally slow or shallow breathing) and almost no Cheyne-Stokes respiration. The machine is not capable of identifying central apnea events (which are basically an EEG issue) so I have no data on those.

And leaks. Lordy, do I have leaks. Still working on that. Fortunately, the S10 can tell what’s a leak and what’s some sort of breathing irregularity. It reports the leaks so I can try different things to minimize them. Useful, and some engineering is in process. Much of leak management is actually hose management, and the engineering lies in keeping the hose from pulling on the mask. I’ll describe what I end up with after I end up with it.

Now, results. For the first three nights, I tried the full-face mask I bought. It kept me awake, even with a Belsomra pill in me. I took a leftover clonazepam pill to knock me out a little more, and I managed to sleep. However, I have no intention of becoming dependent on a benzo just to sleep with a bigger mask. The USP of Belsomra is that it doesn’t disturb sleep architecture to the degree that benzos and the Z-drugs do. If I can’t do a mask on Belsomra, it’s unclear that I can do APAP at all.

So everything hinges on the “nasal pillow” mask I bought. It’s not exactly comfortable, but I’m able to sleep with it strapped to my face. It’s a ResMed AirFit P10, and has a very good reputation. I may try others as time allows.

Now, I can fall asleep with it…and sleep for about six hours. After six hours, the Belsomra is leaving my system, and there’s no longer enough to keep my orexin receptors neutralized. So come about 3:30 or 4, I can no longer fall back asleep. (I’ve been getting up twice a night for bathroom breaks for 25+ years, usually at 1:30 and 4.) Keeping the mask on if I’m not sleeping does nobody any good, so after my second bathroom break, I take the mask off and shut the machine down. This gives me 6-7 hours of treated sleep, plus another hour or two of untreated sleep. It’s not a perfect solution, but it may be the only solution I can manage. Even bad sleep is better than no sleep, and I’ll take whatever benefit from those last two hours that I can.

The improvement in my AHI has been spectacular. From a sleep study AHI of 36 I’ve gone down to an AHI of less than six on all 28 nights. And on only two nights did it go over 5. Most nights it’s less than 3. Last night, I had only four events across 5.53 hours with the mask on, for an AHI of 0.72. That’s not shabby. In fact, an AHI of less than one is considered no apnea at all. I don’t know why I have more events on some nights than others. That’s a subject of ongoing research.

There have been some weirdnesses. My prescription called for a pressure of 6-18 cm. (The S10 supposedly adjusts pressure to what it needs to clear an event.) What I found is that at least once a night, the pressure was up above 17, and I felt like I was being blown up like a balloon. I would wake up completely, and become so annoyed that I had a hard time falling asleep again. Not useful. So I set the machine to vary only between 6 and 13 cm. Now there are no excursions above 13, and from the graphs I can tell that I can sleep when it’s pumping in the vicinity of 12 cm. Median pressure is 7.7 cm. Given the reported AHIs, nothing of value was lost in the adjustment.

Now the bad news: APAP has taken all the pleasure out of sleeping. It’s a hard thing to describe. I’m aware of the mask as I try to fall asleep. It’s a constant irritation, and without the Belsomra I don’t think I would sleep at all. Relaxing completely is difficult. Maybe it’ll get better with more practice, but after 28 nights I’m thinking that whatever I’m experiencing now is what I’ll be experiencing for the rest of my life, which is nothing if not depressing. I’ve begun looking forward to the final two hours of the night as my reward for suffering through the first six hours.

I’m not sure what, if anything, can be done about this.

Now, one can’t argue with results. I don’t feel like a 10-year-old again, and I’m good with that. I wouldn’t mind feeling like a 20-year-old, but I’m not getting that either. The improvements are incremental but real: I’m getting more ideas, spending more time reading, and more time at the keyboard. I don’t feel a great deal more energetic, but something is getting the work done, and I can only credit that to better sleep.

I’m not sure there will be a Part 4 to this series, but when insights become available I’ll report here. So far…

…so good.

Hose Wars, Part 2: To Breathe, Perchance to Leak

This is a series. Start here if you haven’t already.


I’m not a good sleeper, and never have been. When my publishing company (now mostly forgotten) collapsed back in 2002, I developed severe insomnia. I was getting as little as three hours of sleep per night, often less, and sometimes none at all. After a couple of weeks of this, I started to hallucinate cute little cartoon devils doing calisthenics at the foot of my bed, along with other things I’m not sure I can describe. Sleep isn’t optional. I sometimes think we sleep in order to dream undisturbed, and that dreams are somehow where our humanity comes from. If we can’t sleep, eventually we start to dream while we’re awake.

My big fear in starting APAP therapy was that I couldn’t sleep with a mask on my face. Had I been a better sleeper, I’d probably have begun thereapy years earlier. I was given two masks: One covered my nose and mouth. This is called a “full-face” mask, even though it doesn’t cover your eyes. The other is harder to describe: It’s a little plastic thing on an elastic strap that inserts a couple of cushioned tubes into your nostrils. These are called “nasal pillow” masks, and they’re a great deal less intrusive than full-face masks.

The whole point of CPAP/APAP therapy is to push enough air into your nose to keep your airway open, and to open it if by some chance it closes. For this to work, you either need a full-face mask so that if your mouth opens it won’t matter, or with a nasal pillow mask you need some way to keep your mouth closed. There are chin straps of various sorts and other things lumped into a category called “headgear.” Yet more stuff to tie myself up in; no thanks. I did the obvious: I used that blue surgical tape you buy at Walgreen’s to tape my mouth shut.

It worked. It worked, at least, until the machine upped the pressure for some reason. The higher pressure blew the tape off one corner of my mouth, which became a massive air leak, one noisy enough to wake me up.

This is my problem in a nutshell: APAP is noisy and uncomfortable, and keeps me awake. The noise I’m getting used to, at least the fairly modest noise from the machine itself. Leaks are a separate issue. I sleep on my side, which means that both kinds of mask eventually contact my pillow. I can position myself carefully when going to sleep, and that generally works. But if I squirm around even a little while I’m asleep, my pillow nudges the mask to one side, making noise, or (with the full-face mask) spraying air into my eyes. That wakes me up in a hurry.

To keep me asleep despite masks and leaks and hoses flapping around, the doc gave me a prescrption for a sleeping pill called Belsomra (suvorexant.) It’s the first of a new class of insomnia treatments that target the orexin receptors in the brain, rather than the GABA receptors. Pills like Ambien (zolpidem) target GABA, and force you to sleep. If you take one and don’t hit the sack, you’ll start dreaming anyway, and say or do dumb things. The orexin receptors keep you awake. Interfere with their operation using an orexin antagonist like Belsomra, and the signals to stay awake go away. You drift off. I’ve taken Ambien, and it always felt like a whack to the back of my head. Boom! I’m out. Belsomra has a gentler touch, and from what I’ve read, it doesn’t affect sleep architecture (i.e., the different stages of sleep like REM) nearly as much as more preemptive pills like Ambien.

It’s expensive, but very fortunately, Medicare covers it. And so far, it’s done a pretty fair job keeping me asleep in spite of mask issues. As for mask issues, there’s a third sort of mask that I’m going to buy and try: A nose mask. This is like a smaller full-face mask that only covers your nose. It may not be any better than nasal pillows, but it’s cheap enough to do the experiment and be sure.

I’ve found that there’s a downside to blowing air up your nose. A couple downsides, actually, but there’s one big one, and that’s where I’ll start next time.

Odd Lots

Paradoxical Insomnia

For all the time I’ve been struggling with insomnia, I had never heard of “paradoxical insomnia” until Michael Covington recently called it to my attention.

Sleep is a weird business from top to bottom. I’ve encountered a lot of that weirdness, especially since 2001, when my company began to implode. I’ve never been a strong sleeper, but after that I began having nights when I might sleep for no more than an hour…

…or so it seemed at the time.

One of the weirdnesses of the post-Coriolis era is that, for as little as I thought I slept, I seemed to do reasonably well during the day. I certainly wasn’t at my best, but for the most part I wasn’t falling asleep in my chair. I wonder now if I experienced paradoxical insomnia, which is an unusual sleep disorder in which patients feel like they have slept little (or not at all) when in fact they slept adequately, if not normally. In paradoxical insomnia, a patient perceives time spent awake incorrectly. He or she might feel like it takes an hour or more to fall asleep when in fact it took only a few minutes. Early-morning awakenings during which hours seem to pass may again span only a few minutes. The condition is poorly understood. Researchers now think that patients are dreaming that they’re awake. This may seem bizarre to people who sleep normally, but let me tell you, I understand completely.

Here’s why: In my case, at least, the border between wakefulness and dreaming is rubbery. I’ve had some success with a technique I read in one book or another, which involves imagining some quiet activity that reflects daily life. Counting sheep may work for people who live and work with sheep. I’ve seen live sheep half a dozen times in my life, and you can have ’em. What works for me is imagining things like taking walks, sorting books on bookshelves, and having boring conversations with unexceptional people. Although I have “interesting” dreams about one shot in ten (along with the very occasional lulu) the vast majority of my dreams are very much like that: walking alone or sometimes with a nameless companion, or doing domestic things of little consequence, like taking towels out of the washing machine and putting them in the dryer.

I know that the technique works because a time or two I recall sliding from guided meditation into a dream without any change of scenery. I know that it was a dream because it stopped following the script. Here’s the best example: Jeff and a nameless and poorly imagined female companion are walking down a country road on a generic summer evening, talking about dumb stuff. No mosquitoes.

JEFF: Hey, there’s a sycamore tree! I like sycamore trees.

COMPANION: Me too.

JEFF: My grandfather planted one in the back yard when I was a kid. It had the biggest damned leaves.

COMPANION: I remember those.

JEFF: And seed balls. We used to throw them at each other.

COMPANION: That must have been fun.

JEFF: It was. We used to be able to burn the leaves in the street.

COMPANION (Turning): Jeff, what do you want most from God?

JEFF: Unconditional love.

Bzzzzt! In my directed meditations my imaginary companions do not ask me questions. So when my imaginary companions begin taking control, I know (in hindsight) that I’m dreaming.

In the grim days after Coriolis went under, I had plenty of experience lying awake much of the night and staring at the wall. At some point it became part of ordinary life, and thus completely unremarkable dream-fodder. I also seem to slide from conscious thought into dream states very smoothly. This is why dreaming about lying awake is no stretch at all, and may have continued long after I had gotten over the loss. It may continue to this day. Short of monitoring my own brain waves, I’m not sure how to tell.

But boy, it’s probably better than talking beavers.

Odd Lots

  • Here is the entire sky projected on a plane, and zoomable.  (LINK REMOVED–SEE COMMENTS.) That doesn’t do it anything like justice. Cruise the image a little and gasp. (Give the site time to refresh; it’s newly slashdotted.) Read the rest of the page too–it’s fascinating, and full of great photos. Chile looks a great deal like Mars in some places.
  • There is apparently a correlation between sleep loss and amyloid tangles in the brain, which are a key element in Alzheimer’s Disease. Causation is still a little unclear, but I find it significant that in our era of Anything But Sleep, the incidence of various dementias is exploding. Be in bed by 10:00PM and keep your brain. Now there’s a deal I can live with.
  • Wired has an interesting retrospective on tablet computing, which I found worthwhile mostly for the mention of a steampunk-era electromechanical handwriting encode-and-transfer device, which ferdam sounds like what Sherlock Holmes would use to IM Watson.
  • Here’s another worthwhile perspective on the Google Books Settlement.
  • A chap who calls himself the Jolly Pirate wrote to tell me that the Pirate Party is alive and well in the US (I was under the impression that it was a European thing) and some interesting links may be found on its site, many of which have nothing to do with piracy. Now, would an American instance of the Pirate Party lean left or right? (Or would it be port and starboard?) I’ll be damned if I can decide…
  • It’s been a bad season for big-time wine critics, who can’t seem to find a business model and keep getting busted in conflict-of-interest scandals. The Internet allows the crowdsourcing of critique of all sorts of things–why should wine be any different?
  • Pertinent to the above: What we need is the wine implementation of the “People who liked this also liked…” mechanism I see (and use) in the book world. I very much like Campus Oaks Old Vine Zinfandel, though the 2007 vintage now in stores is a pale shadow of 2005. What would be a wine similar to that? (If you know of such a system for wine, please share.)
  • There are candy Legos.
  • The charger for my Kodak pocket camera is a thin little slab with two 110V power plug pins that swivel out to plug into the wall, and then swivel back into hiding when they’re not needed. Why can’t they build that mechanism right into the back of an ebook reader? (I was without my Sony Reader for a couple of months after I lost the charger.)
  • After an unexplained absence of several weeks, Fort Carson’s cannon is back. (See my entry for September 15, 2009.) Maybe the cannon was broke and they had to send it back to the factory for repairs…

Odd Lots

  • Sleep seems to be key in allowing the brain to infer big-picture relationships from scattered facts–and by implication, memories. I’ve never been a strong sleeper, and I wonder if some of the “fails” in my older memories stem from inadequate deep sleep.
  • Also from Wired Science: People who don’t get enough sleep are touchy and angry. (Not like that’s news to anyone who’s prone to sleepless nights.) Could this explain the peculiar unpleasantness of the political blogosphere? I’ll vouch for the fact that peace, love, and tolerance of people who disagree with you is much easier on nine hours’ sleep than five.
  • If you’ll forgive the expression, here’s a long, beefy article on why everybody thinks animal fat is a bad thing (hint: it’s because a Right Man cherry-picked his data to confirm his hypothesis) and why, to the contrary, animal fat may be very, very good for us.
  • Maybe this will bring the issue of bad patents to the top of the queue. (Thanks to Pete Albrecht and numerous others for the link.)
  • From Jim Strickland comes a link suggesting that your family dog may be smarter than the average toddler. (Your dog, maybe. Mine, probably not.)
  • I was wrong about it coming from a Kum & Go store; while sorting my charge card receipts I discovered that Fat Dogs (see my entry for August 7, 2009) is a small chain of Conoco-franchised gas station/convenience stores limited to Western Nebraska. They had some water private-labeled by Sandhills Water, and put their wry corporate motto on the top of the label.
  • Lileks takes on the Sears Catalog for 1973. I remember that catalog. Fortunately, my clothes came from Goldblatts. And did real, human, breathing girls wear these things? Maybe the girls in Hoffman Estates did. In my neighborhood, well…no.
  • From the Words I Didn’t Know Until Yesterday Department: juiceboxer, a young, arrogant, high-status preppie-type. Their parents got their juice from bottles. And their grandparents mostly got it while it was still inside the fruit. Juice as a product is a relatively new thing; keeping juice from fermenting is difficult and one of the unappreciated miracles of our modern age.