much in the treatment of infectious diseases in people. Still, the clinical value of corticosteroids just did not appear to be present. When we began actually measuring the amount of the adrenal response in human volunteers in this building, who were being infected with various microorganisms, we were surprised that the adrenal response was minuscule compared to the immune response of surgical patients.

We did some of the same type of studies in animals that Dr. Wilmore just talked about. We found that a permissive effect of the steroids was necessary for a lot of responses by the animal and in adrenalectomized animals—the guy with the stress of just a small permissive dose would keep those animals going. I am still not convinced the large doses of steroids are of any protective value at all.

LEONARD KAPCALA: When our studies were performed, basically, what we were trying to do was to mimic what a maximal stress was, just showing that because, as you saw, the HPA axis of intact animals was completely protected so that these animals could tolerate very large doses.

I would just like to make one other comment. I know, in the medical literature right now, the pendulum has gone back and forth in terms of sepsis, and whether or not to give steroids. Right now it is in the camp that steroids have not been shown to be beneficial. I would just add one caveat though. I think that pharmacological steroids might be beneficial if they were given early enough because by the time you make the diagnosis, sepsis has advanced to a certain stage, and it is already too late, because animal models such as baboons, given sublethal or lethal doses of endotoxin can be protected. So I think the problem is that with humans, we just cannot really do that study. By the time we say this person is undergoing sepsis it is already too late, and the cytokine cascade has gone too far.

I would agree with your point about the importance of the permissive action of glucocorticoid in a certain situation such as the basal state. I think one reason we see relatively low cytokine levels in the basal, unstimulated state is because glucocorticoids are modulating some cytokine production and secretion. It has been shown that adrenalectomized animals will show an increased expression of cytokines just basally so that that is an important concept: that you do not necessarily need large amounts, but you do need larger amounts when the cytokine challenge becomes more significant.