The role of the hypothalamic-pituitary adrenal axis in the host response to infection is crucial. The initial inflammatory response to sepsis activates the endogenous release of cortisol, which in turn modulates the synthesis and release of both pro- and anti-inflammatory mediators to restrict inflammation in infected tissues. However, a number of factors, including vascular or ischemic damage, inflammation and apoptosis within the hypothalamic-pituitary adrenal axis, as well as use of drugs that alter cortisol metabolism, may cause adrenal insufficiency. One major problem ICU physicians are faced with is the diagnosis of sepsis-induced adrenal insufficiency at the bedside. A multidisciplinary international task force has recently recommended that sepsis induced adrenal insufficiency is best recognized by basal cortisol of less than 10 microg/dl or change in cortisol of less than 9 microg/dl after administration of corticotrophin. The diagnostic value of measuring salivary free cortisol in this setting remains to be investigated. While sepsis adrenal insufficiency is undoubtedly associated with a poor prognosis, the indication and practical modalities of corticosteroids therapy remained controversial. Based on the two largest randomised, placebo-controlled trials, many investigators, myself included, contend that septic shock patients with hypotension poorly responsive to fluid replacement and vasopressors should receive a seven day treatment with the combination of hydrocortisone at a dose of 200 mg per day and fludrocortisone at the dose of 50 microg per day.