sudafed causes big spikes in blood glucose, especially when you take it with carb heavy food. sigh. At least it’s coming down from the high of ~350.
This whole thing is exciting and frustrating at the same time. It’s like I’m relearning everything.
I’ve been splitting time between a number of projects lately.
Most of my time has been spent on Moonlight, of course. Things are progressing pretty nicely. We have real QA, so regressions don’t go unnoticed. We have a bunch of new (for us) features, and a bunch more coming.
A smaller amount of weekend/after hours time has been spent brushing off the Trac stuff for SCSharp, and clearing the 2 year old backlog of porn spam and unfixed bugs. I’m surprised just how little it’s bitrotted. Check it out.
My newest project, which was mostly motivated by my frustration generating local class-status pages for the work I’d been doing filling out the WPF apis is “gui-compare”, found in mono-svn in mono-tools/gui-compare.
Basically it’s a gtk# frontend for class-status, but it allows some pretty nice features. You can compare any dll to any other (to see, say, the differences between the 1.1 and 2.0 mscorlibs in mono). It also knows how to read the masterinfo files that are generated by the mono-api-info stuff. Lastly (and perhaps most importantly), it downloads and caches the correct masterinfo files for you. No more groveling around for them on the mono site. Just fire up gui-compare and e.g. Compare -> API 1.1 -> System, and it downloads the files it needs and performs the comparison. I also stole a neat feature from MoMA, and gui-compare will show you which methods throw NotImplementedException. Here’s what my current build looks like:
Another nice thing about gui-compare for use in fleshing out apis: You look at gui-compare, find the stuff you want to implement, stub it out in your code (or implement it, whatever), install your assembly, and then hit Ctrl-r in gui-compare to re-run. Quite a bit easier than having another terminal open in mono/web/web waiting for me to type ‘make’. File watcher support coming soon to make even this step unnecessary.
I sorta repeated my experiment again today. Much higher intensity. At no time was I in danger of going low, but it turns out bonking doesn’t necessarily mean going below 70 
First 30 minutes were relatively low gear, high cadence (avg cadence 99rpm, avg HR 156bpm, avg speed 17.1mph)
00:00 BG=147, started ride
00:15 BG=116
00:30 BG=89
Second 10 minutes I increased the incline on the bike, went up a few gears (avg cadence 90rpm, avg HR 163bpm, avg speed 19.6mph)
00:35 BG=81. Close to bonking here, legs getting really heavy, drank 7g CHO.
Last 5 minutes I cooled down (avg cadence 70rpm, avg HR 144bpm, avg speed 11mph)
00:45 BG=90
Felt fantastic again, although I’m clearly out of shape (figures, heh). Now I’m off to eat a ton.
Feeling mostly back to normal these days. Or at least back to the normal that existed for a while between my diagnosis and the start of the thymo trial.
I did a little experiment day before yesterday, of skipping my long acting insulin injection altogether. That afternoon, I got on my bike and rode on the trainer for about an hour. It was glorious. The past couple of attempts (once trying the ride at 14U lantus, the other at 10U) were total failures, with me eating large amounts of carbs almost constantly in an attempt to keep myself from going low. My heart rate never got about 150, and the rides were over at 45 minutes. Without any lantus (or any residual humalog from lunch) in my system, though, everything was fine. My pancreas is still producing some insulin, after all, and exercise makes it more efficient. I ate 46g CHO 30 minutes before I started the ride, with things looking like this:
00:00 BG = 152, start ride
00:15 BG = 136
00:30 BG = 107, drank half a cup of gatorade (7g CHO)
00:45 BG = 110, drank another half cup (7g CHO)
01:00 BG = 115, stop ride
02:00 BG = 106
That was really the first real exercise (my heart rate was up in the 160’s for a while at the 45 minute mark, which felt incredibly good) I’ve had in months. The other interesting thing is that I *think* given that I didn’t have a large amount of circulating insulin, I couldn’t have gone low. The pancreas releases another hormone when your BG runs low called glucagon, which signals the liver to release some of its stored glycogen. The mechanism by which this works is inhibited by insulin, though (I couldn’t tell you if the pancreatic cells which secrete glucagon are inhibited by the insulin, or if the insulin just forces the glucose right back into the liver after the glucagon has caused its release. The net effect is the same), which is why if you’re injecting insulin the real danger is no longer going too high, but going too low.
I think the next time I try this I’ll start out around 150-160 again, and see how my BG behaves over the hour without injesting any more carbs. I want to see if I can bottom out in the 70’s and not go any lower.
As an aside, my normal insulin/carb ratio when I’m covering meals is 1U/25g (I inject 1 unit for every 25g CHO in my food). The day of my ride I injected 10U humalog throughout a day, which is barely more than the 9 units I normally inject, but I also ate ~350g CHO. So when exercising even for 1 hour, I managed to drop my insulin/carb ratio to 1U/35g, which is nuts.
The only problem I had with the day was in the evening. After even that small amount of exercise, I was ravenous for the rest of the day. Trouble is, keeping track of all the 10g here, 15g there is hard, and it’s also hard to limit yourself to larger, discrete meals. I tore through my 100g burrito and the 60g of sorbet before bed, but still had a hard time keeping my BG in line with the rather constant snacking. I suppose I could have taken the long acting stuff after my ride, but I don’t like the idea of migrating the time I inject around, because of the 24 hour duration. More reason going to a pump will make life easier.
Anyway, more learning to be done on this front, as always
