The use of steroids as an adjunct to antimicrobial therapy has been controversial for many decades. Recent reports of the use of steroids in supraphysiological rather than in "industrial" doses in patients with sustained circulatory instability has re-ignited the debate. Anecdotal reports of adrenal insufficiency in septic patients have suggested a relative cortisol deficiency in these patients with poor survival if not given supplementary steroids. The possibility that the hypothalamic pituitary adrenal axis is intimately involved in the pathogenesis of this entity has not previously been highlighted. This review looks at the relationship of sustained cytokine release and the possibility of altering the stress response with progressive loss of adrenocorticotrophic hormone release and subsequent diminution in adequate cortisol levels. The reliance on, and misinterpretation of, the short synacthen test in diagnosing the possibility of this condition is emphasized.