Tuesday, February 08, 2005

Irish stew

"Crater Lake" was the phrase I noticed myself having trouble with first, but last night there was an even better one, because we had it for dinner. Irish stew. Try saying that quickly. I can say it OK if I give a big pause to readjust the mouth after "Irish." But if I try to say it quickly, I get "Irisch-too."

A friend writes: "I'm curious why your doctor is only doing 1 week on, then several weeks off. Why wouldn't they do a month or two straight? Perhaps you can explain the logic in your blog."

Well, I can't speak for the doc, but I will tell you my opinion, based on the as-yet unproven belief that the drug helps my ALS. Unproven in the sense that my experience is what they call anecdotal. And, since we've only done it once, it's not yet shown to be reproducible.

Part of my approach is the feeling I have, perhaps false, that I still have a lot of runway left. I mean, I still walk and talk and swallow. If I were much more advanced, or had gotten to this spot very quickly, I would want to do the drug every single day.

But since I do feel like I have runway left, I'm willing to go off the drug for a month, and see if the grip strength declines. And then I want to see what happens to the grip strength when I go back on it.

I feel that if the drug is helping and I stayed on it continuously, at a certain point I'd max out the benefit, and learn nothing. Also, I may possibly prompt my body to down-regulate the expression of the gene that creates glutamate transport protein. My doc reminded me that if you give someone regular testosterone injections, their testicles (if they have them) shrink.

I have this mental image of the motor neuron experiencing a buildup of old glutamate and gradually beginning to function more sluggishly, to send weak signals, slow signals, and misfire by sending signals when not needed (twitching). This could be total nonsense, and the motor neurons could just be dying one by one, in a binary fashion. But the fact that the live rodents, and this live human, could regain strength, argues against the binary model and for the buildup model. The evidence indicates that the nerves were somehow resuscitated, or revived, when the increased levels of GTP-1 cleared out the old glutamate.

Using this imagery, then, we can picture regular monthly infusions of ceftriaxone that stimulate GTP-1 and clean house for the motor neurons. Hopefully this might enhance their survival.

You can argue that my reaction to the ceftriaxone is a placebo effect. And that's fine, I would not debate you. But I know that for my purposes in this world, it's no illusion.

Yet I am such a tech hound that I really want to see that curve. I used to diagnose problems like this all the time when they happened in computers. I want to see if the grip strength stays level for a month, or declines, or even climbs, and if so, exactly how. I love debugging systems.

One of the things we often did to confirm that we understood the problem, and the solution, was to try to recreate the problem, and see if our solution fixed it. Let's say that hundreds of log files are being created every hour, and the system is getting bogged down wasting cycles full-text indexing all of the files. In this analogy, the log file buildup represents the glutamate buildup in my motor neurons. Imagine that we think the solution is to turn off the logging, but we can't do that, because the third-party application server won't allow it. So then we decide to turn off indexing of the log files, and that seems to fix it. Then what we should do, to find out if we have the real solution, is to turn back on the indexing and see if the performance hit recurs, then reapply our solution, and observe the effect.

Naturally, you do this on the backup system. You never want to mess with the production system if you can avoid it.

I always used to say that I could never be a doctor, because someone's life is in your hands, but when you operate on someone's computer, the worst that happens is that the computer dies.

Now I'm debugging a software problem on a live production server. And there is no backup.

I am awake at this awful hour because my daughter has been fussing for over an hour now. My lovely wife, because she wants to take care of me, is doing all the work. My thinking is that I really should set up that youth bed for the girl today, and that will reboot her into a better relationship to sleeping.
Weblog Commenting and Trackback by HaloScan.com