So why not just insulinate all our strokes and cardiac arrests? Certainly, your lab and others have accumulated evidence that insulin (and other growth factors) are salutary in the setting of brain ischemia. Moreover, insulin is cheap, available and familiar to physicians. What are we waiting for?

Here's the problem with insulin in brain ischemia: it doesn't work. (You may start to detect a pattern here.) At least not yet, and probably not by itself. There are some problems with deploying insulin in the clinical arena at this juncture.

1. The dose. In our studies on protein synthesis, Akt activation and cytochrome c release, we've found that huge doses of insulin are required--up to 20U/kg. That's about, oh, a zillion times the usual dose. Obviously, there is a huge potential for problems with hypoglycemia, hypokalemia, and other metabolic derangements with a dose this large.
2. Window of opportunity. Our default mode is to believe that treatment should begin earlier rather than later. Certainly, this is true when it comes to apoptosis. But when is the best time to turn on protein synthesis? As we noted earlier, there is some uncertainty about this.
3. Conflicting evidence. It's only fair to tell you that there is some evidence that insulin can actually worsen outcome in focal ischemic stroke. This evidence is limited and flawed, but cannot be ignored.
4. No clinical evidence. Until insulin is studied in humans, it's not ready for prime time.