Patients with adrenal insufficiency (AI) require adjustment of hydrocortisone (HC) dose to avoid life-threatening adrenal crisis during illness, surgery and trauma. However, current dose recommendations are based on empirical grounds only and choice of dose and administration modes vary considerably. We designed the PACS study to compare cortisol levels achieved by currently recommended HC stress doses to those in i) healthy controls (n=85, 2170 years), ii) military controls under combat stress conditions (n=105, 2040 years), iii) patients undergoing elective moderate or major surgery (n=22, age range 2160 years), and (iv) patients admitted after acute trauma (n=85, 3060 years). Ten patients with autoimmune primary AI (4064 years) underwent frequent serum sampling after 200 mg HC/24 h in four different administration modes: 50 mg orally every 6 h, 50 mg i.m. every 6 h, 50 mg i.v. every 6 h, or 200 mg HC per continuous i.v. infusion. Serum cortisol was measured by tandem mass spectrometry (Waters Xevo/Acquity uPLC). Cortisol levels during moderate (median 431, range 249570 nmol/l) and major elective surgery (611, 1651102 nmol/l) peaked between 2 and 4 h after anaesthesia induction. Cmax values for acute trauma patients were 433 (106685) nmol/l. Cmax after HC administration via any administration mode had median values ranging from 836 to 1440 nmol/l. However, nadir cortisol levels during intermittent bolus application of HC decreased to Cmin 277 (64398), 289 (148458), and 173 (118375) nmol/l 6 h after administration of oral, i.m. and i.v. HC bolus, respectively. By contrast, continuous infusion of HC yielded steady-state cortisol concentrations after 1 h with a median of 836 (range 6611073) nmol/l. HC dose cover during surgery, trauma and major illness in patients with AI should be provided by continuous i.v. infusion of 200 mg HC/24 h, following an initial HC bolus administered at admission or anaesthesia induction.