As it happens, my doctor was right, and I was on my way to becoming a hosehead.
First someone from a home health agency came out and showed me how to use an oximeter, which clips onto your finger and records the oxygen saturation of your blood when you sleep. The normal range is 97% to 99%; mine went down at points to 85%. About a week later I went to a sleep clinic for an overnight study. They told me I had an AHI of 20.5, which means that I stopped breathing (apnea) or almost stopped breathing (hypopnea) 20 times an hour -- moderate sleep apnea. A week later I went back for another overnight, to titrate the machine (determine how much pressure I needed). I didn't sleep much either night -- too many wires and belts attached to too many parts of my body for real comfort. A week and a half after the titration study I finally got my CPAP machine -- CPAP for continuous positive airway pressure. The machine blows air into a mask which you wear on your face, and the increased pressure keeps your airways open. (And the mask is attached to the machine via a long hose -- thus "hosehead".)
The machine I got is actually an APAP -- adjustable positive airway pressure, able to vary its pressure automatically depending on what (it thinks) I need. But to start out they simply set it to a constant pressure of 9 cm (9 centimeters of water -- who comes up with these medical measurement units!?)
Thursday, AHI 14.6: I spent about 5 and a half hours on the machine, about four of those sleeping. Overall AHI was 14.6 -- not that much better. My machine records detailed data on an SD card, and there's an open-source program, SleepyHead, that will read it. Looking at the graphs, I saw that I had some totally clear periods alternating with periods of really frequent apneas -- once a minute or slightly more. And looking in even more detail, I saw that graphs looked an awful lot like the graphs of Cheyne-Stokes breathing in Wikipedia and other places on the web. Furthmore, the machine thought most of them were central apneas rather than obstructive apneas. An obstructive apnea is the kind I described above, where your airways close and you can't breath even though you're trying. Central apneas happen because for some reason your body decides not even to make the attempt. Cheyne-Stokes breathing is one kind of central apnea, and googling for "Cheyne-Stokes" is downright scary: it's strongly associated with heart failure and stroke. But it's also something that happens to some of us initially diagnosed with obstructive sleep apnea -- the CPAP or APAP machine clears out the obstructive events, which paradoxically leaves you vulnerable to a cyclic breathing pattern where the carbon dioxide in your blood goes so low that your body stops breathing; your blood CO2 finally rises enough that you breathe; you breathe so efficiently that you lower your blood CO2 level below the point that triggers breathing; you stop breathing; and so on.
Friday, AHI 11.3: Friday, as an experiment, I took Afrin to clear my nose of allergies. I had one obstructive apnea, in contrast to Thursday's twenty. (And discontinued the Afrin -- I'm definitely scared of the rebound that can happen if you use it for more than three days!). A bit under 4 hours on the machine, maybe three hours' sleep -- I kept waking up with a dry mouth. I turned the machine's humidifier up and up, and eventually decided it must have run out of water. But when I got up to refill it, it had a perfectly adequate amount of water. Some investigation on cpaptalk.com made me realize I'd learned to breath through my mouth in just two nights! (And short ones at that. ^_^) I hadn't done that in the sleep lab and had been hoping I was immune.
The cure for this is either a full-face mask (which I didn't have -- they're hard to make work with beards); or a chin strap to keep your mouth closed. I didn't have a chin strap either, and my bite is so terrible I hated the thought of something forcing my teeth together; but needs must, so I searched cpaptalk for "improvised chin strap", and lucked into SleepingUgly's description of a different mouth sealing method -- a self-adhesive Ace bandage wrapped horizontally around your head, like a blindfold for the mouth.
After that I took a late-morning nap, two hours on the machine. I wondered if the extra pressure was making me hyperventilate, causing the Cheyne-Stokes, so in the afternoon I took an hour's nap with the machine set to ramp up from 4 cm to 9 cm over 45 minutes -- no apneas at all during that period.
Saturday, AHI 16.0: I kept the 45-minute ramp-up. Again, no apneas during the ramp-up periods -- but lots of Cheyne-Stokes breathing after ramp-up, in periods of up to an hour. Google found a site that gave the secret to unlocking the clinician settings on my machine, and I set the pressure to float between 4cm and 9cm. Took an afternoon nap, a bit over an hour's sleep with one 3-minute episode of Cheyne-Stokes, AHI of 3.9 for the nap.
Sunday, AHI 7.3: Scattered sleep, about four and a half hours apnea-free (or almost free), another hour and a bit at 10 AHI, and twenty minutes of Cheyne-Stokes apnea-per-minute breathing.
Monday, AHI 2.3: No apneas to speak of! Yay! No sleep to speak of! Boo! -- Well, about four hours of sleep (out of six and a half on the machine). Really tired.
Tuesday, AHI 4.8: Another day relatively clear of apneas, but again only four hour's sleep.
Friday, on the advice of Pugsy from cpaptalk.com, I'd asked to have my sleep study mailed to me. When it came, I learned that in the titration study my apneas had already mostly been central. "Consider a repeat titration study if symptoms persist. ASV might need to be considered if central apneas are persistent with CPAP".
ASV stands for "adaptive servo-ventilation". The usual CPAP pressure is high enough to keep your airways open, which prevents obstructive but not central apneas. An ASV machine, when it detects Cheyne-Stokes breathing, ramps up the pressure enough to actually force you to breathe -- sort of like mouth-to-mouth resuscitation, only this is mask-to-nose.
Getting the study was reassuring in one way -- my interpretation of the readings agreed with the sleep center's. And it was infuriating: folks like me, who develop central apneas once the obstructive apneas are dealt with, often see the centrals go away in a month or two. But that's a month or two of a very rough ride. No one bothered to tell me that it would likely be difficult for some time, that the difficulties would quite likely go away, and that if they didn't there was a plan B. My geeky fascination with electronic toys showed me the central apneas; Google and the folks on cpaptalk.com told me what they meant and what to expect; but this kind of patient non-education is one reason why between 40 and 70 percent of people started on CPAP eventually stop.
Also, the 20.5 figure from the diagnostic study wasn't AHI, but RDI -- everything covered by AHI plus other respiratory-event-related arousals, semi-awakenings caused by breathing problems that aren't quite apneas or even hypopneas. The AHI was 14.1 -- so my first and third nights at home with CPAP were actually slightly worse than the sleep study night without CPAP.
Wednesday, AHI 4.7: First CPAP night with a semi-decent amount of sleep -- 9 hours in bed with 75 minutes up in the middle of the night, just under eight hours on the machine, probably a bit under 7 hours sleep. Felt half-way decent most of the day, very tired from 4pm to about 8pm, then annoyingly perky as I tried to get myself to bed. This is more or less the pattern I've been in for some years, and had been trying to reverse by better sleep habits when whatever happened in late April happened -- when the 4pm crash started coming at 10am.
Thursday, AHI 11.0: Oops! I tried setting my pressure lower, to avoid central apneas, and got lots of obstructive apneas instead. The 10am crash came back with a vengence. I was working at home Friday, so I set the pressure back up, took an afternoon nap (only one apnea!), and put in for some sick leave.
Friday, AHI 7.9: Disappointingly high AHI, eight hours of apparently solid sleep, but periods of Cheyne-Stokes breathing totaling an hour and a half scattered through it. I seem to be back at pre-April levels of tiredness -- a definite improvement over April and May, but there's still a long way to go. This is definitely frustrating ("What do we want? Sleep! When do we want it? Last night!"). Still, there's a part of me in geek heaven -- data! graphs! open-source programs! debugging my own breathing! (And even that part of me would gladly forgo the techno-euphoria in return for a nice boring reliable solution.)