During sepsis or acute respiratory distress syndrome, the hypothalamic pituitary adrenal axis is rapidly activated through a systemic pathway, i.e. by circulating pro-inflammatory cytokines and through the vagus nerve. Subsequently, the adrenal glands release cortisol, a hormone which will likely counteract the inflammatory process and restore cardiovascular homeostasis. Both experimental models and studies in humans suggest that inadequate hypothalamic pituitary adrenal axis response to stress accounts, at least partly, for the genesis of shock and organ dysfunction in sepsis and acute respiratory distress syndrome. Relative adrenal insufficiency and peripheral glucocorticoid resistance syndrome are the two main features of the inappropriate hormonal response and provide the grounds for cortisol replacement in these diseases. In practice, a high dose of corticosteroids (i.e. one to four boluses of 30 mg/kg of methylprednisolone, or equivalent) had no effects on survival in severe sepsis or acute respiratory distress syndrome. There are at least seven randomised controlled trials reporting the benefits and risks of low dose corticosteroids (i.e. 200 to 300 mg daily of hydrocortisone or equivalent) given for a prolonged period in severe sepsis or in the late phase of acute respiratory distress syndrome. These trials showed consistently that, in these patients, the use of low dose of corticosteroids alleviated inflammation, restored cardiovascular homeostasis, reduced organ dysfunction, improved survival and was safe. Further studies are ongoing to better identify the target population. In the meantime, cortisol replacement (i.e. 200 to 300 mg daily of hydrocortisone or equivalent) should be considered as standard care for these patients.