Table 2.

Characteristics of included randomised controlled trials

Study	Design	Patients	Interventions	Outcomes
Annane 200212 (France)	2 parallel groups, 19 centres	300 adults with vasopressor and ventilator dependent septic shock	Hydrocortisone (50 mg intravenous bolus every 6 h for 7 days + fludrocortisone 50 μg taken orally every 24 h for 7 days); respective placebos. Treatments had to be initiated within 8 h from shock onset	Primary: 28 day mortality in non-responders Secondary: 28 day mortality in responders and all patients; ICU mortality; hospital mortality; 1 year mortality; shock reversal; organ system failure free days; safety
Bollaert 19989 (France)	2 parallel groups, 2 centres	41 adults with vasopressor and ventilator dependent septic shock	Hydrocortisone (100 mg intravenous bolus every 8 h for 5 d then tapered over 6 d); placebo. Treatments had to be initiated after 48 h or more from shock onset	Primary: shock reversal Secondary: 28 day mortality; improvement in haemodynamics; safety
Bone 198730 (US)	2 parallel groups, 19 centres	382 adults with severe sepsis (n=234) or septic shock (n=148)	Methylprednisolone (30 mg/kg 20 min intravenous infusion, every 6 h for 24 h); placebo. Treatments had to be initiated 2 h from time entry criteria were met	Primary: 14 day development of shock for severe sepsis; shock reversal for septic shock; 14 day death and safety
Briegel 199910 (Germany)	2 parallel groups, 1 centre	40 adults with vasopressor and ventilator dependent septic shock	Hydrocortisone (100 mg 30 min intravenous infusion followed by 0.18 mg/kg/h continuous infusion until shock reversal and then tapered off); placebo. Treatments had to be initiated within 72 h from shock onset	Primary: shock reversal Secondary: 28 day mortality; improvement in haemodynamics; organ system failure; safety
Chawla 199911 (US)	2 parallel groups, 1 centre	44 adults with vasopressor dependent septic shock	Hydrocortisone (100 mg intravenous bolus every 8 h for 3 days then tapered over 4 days); placebo. Treatments had to be initiated after 72 h or more from shock onset	Primary: shock reversal Secondary: 28 day mortality; improvement in haemodynamics; safety
CSG 196320 (US)	2 parallel groups, 5 centres	194 adults and 135 children with vasopressor dependent septic shock	Hydrocortisone (intravenous infusion of 300 mg for 24 h then 250 mg for 24 h, followed by 200 mg orally on day 3, then tapered off in steps of 50 mg/day—that is, total duration of treatment 6 days); placebo	Primary: hospital mortality Secondary: safety
Keh 200314 (Germany)	Crossover design	40 adults with vasopressor dependent septic shock	Hydrocortisone (100 mg 30 min intravenous infusion followed by 10 mg/h continuous infusion for 3 days); placebo. All patients received hydrocortisone for 3 days preceded or followed by placebo for 3 days	Primary: immune response. Secondary: improvement in haemodynamics and organ system failure; safety
Klastersky23 1971 (Belgium)	2 parallel groups, 1 centre	85 adults with disseminated cancer and life threatening Infection	Betamethasone (1 mg/kg/day in 2 intravenous doses for 3 consecutive days); placebo	30 day mortality; rate of adverse events
Lucas 198427 (US)	2 parallel groups, 1 centre	48 adults with septic shock	Dexamethasone (2 mg/kg as a single intravenous bolus followed a maintenance infusion of 2 mg/kg/24 h for 2 days); standard treatment	Primary: 14 day mortality (unclear) Secondary: improvement in haemodynamics; improvement in pulmonary function; safety
Luce 198832	2 parallel groups, 1 centre	75 adults with septic shock	Methylprednisolone (30 mg/kg 15 min intravenous infusion, every 6 h for 24 h); placebo	Primary: prevention of acute respiratory distress syndrome Secondary: hospital mortality
Schumer 197624 (US)	3 parallel groups, 1 centre	172 adults with septic shock with positive blood cultures	Dexamethasone (3 mg/kg as a single intravenous bolus); methylprednisolone (30 mg/kg as a single intravenous bolus); placebo. Treatments might have been repeated once after 4 h and had to be initiated at time of diagnosis	Primary: 28 day mortality Secondary: complications rates
Slusher 199633 (US, Kenya, Nigeria)	2 Parallel groups, 2 centres	72 African children aged 1 to 16 years with severe sepsis or septic shock	Dexamethasone (0.20 mg/kg every 8 h for 2 days); placebo. Treatments might have been repeated once after 4 h if shock persisted and had to be initiated 5-10 min before first dose of antibiotic	Primary: hospital mortality (unclear) Secondary: haemodynamic stability at 48 h; complications
Sprung 198428 (US)	3 parallel groups, 2 centres	59 adults with vasopressor dependent septic shock	Dexamethasone (6 mg/kg as a single intravenous 10 to 15 min infusion); methylprednisolone (30 mg/kg as a single intravenous 10 to 15 min infusion); no treatment; placebo. Treatments might have been repeated once after 4 h if shock persisted and had to be initiated at time of diagnosis	Primary: hospital mortality; shock reversal Secondary: complications of septic shock; treatments' safety
VASSCSG 198731 (US)	2 parallel groups, 10 centres.	223 adults with severe sepsis or septic shock (n=100)	Methylprednisolone (30 mg/kg as a single intravenous 10-15 min infusion, followed by a constant infusion of 5 mg/kg/h for 9 h); placebo. Treatments had to be initiated within 2 h	Primary: 14 day mortality Secondary: complications
Wagner 195519 (US)*	2 parallel groups, 2 centres	113 adults with pneumococcal pneumonia; shock present in only 3	Hydrocortisone (orally 80 mg on admission followed by 60 mg 3 times on day 1, then 40 mg 4 times on day 2, 20 mg 4 times on day 3, 10 mg 4 times on day 4, and 10 mg twice on day 5); standard therapy (first 85 patients); placebo (last 28 patients)	Fever; pleuritic pains; patient's wellbeing
Yildiz 200213 (Turkey)	2 parallel groups, 1 centre	40 adults with sepsis (n=14), severe sepsis (n=17), and septic shock (n=9)	Prednisolone (2 intravenous bolus, 5 mg at 6 am and 2.5 mg at 18 pm for 10 days); placebo	Primary: 28 day mortality Secondary: complications

CSG=Cooperative Study Group; VASSCSG=Veterans Administration Systemic Sepsis Cooperative Study Group.

^*Quasi-randomisd.