Are endogenous glucocorticoid levels adequate in septic shock?