• Secondary outcomes : [ Time Frame: Day 180 ] [ Designated as safety issue: Yes ]
• 90-day and 180-day mortality. [ Time Frame: Day 180 ] [ Designated as safety issue: Yes ]
• Duration of life-supporting treatments (i.e. vasopressors and mechanical ventilation) [ Time Frame: Day 180 ] [ Designated as safety issue: Yes ]
• Time to resolve multiple organ dysfunction, i.e. to obtain a SOFA score < 8 [ Time Frame: Day 180 ] [ Designated as safety issue: No ]
• Hospital length of stay. [ Time Frame: Day 180 ] [ Designated as safety issue: No ]
• Number of hypoglycaemic events (blood glucose < 4 mmol/l) during insulin infusion [ Time Frame: Day 180 ] [ Designated as safety issue: Yes ]
• Muscle weakness at discharge from intensive care unit, 90-day and 180-day [ Time Frame: Day 180 ] [ Designated as safety issue: Yes ]
• Post traumatic stress disorders [ Time Frame: Day 180 ] [ Designated as safety issue: Yes ]

• Secondary outcomes :
• 90-day and 180-day mortality.
• Duration of life-supporting treatments (i.e. vasopressors and mechanical ventilation)
• Time to resolve multiple organ dysfunction, i.e. to obtain a SOFA score < 8
• Hospital length of stay.
• Number of hypoglycaemic events (blood glucose < 4 mmol/l) during insulin infusion
• Muscle weakness at discharge from intensive care unit, 90-day and 180-day
• Post traumatic stress disorders

This study will compare, in adults with septic shock, the safety and efficacy of a combination of moderate doses of corticosteroids and intensive insulin therapy to that of moderate doses of corticosteroids. In addition, this study will compare the efficacy and safety of hydrocortisone alone versus hydrocortisone plus fludrocortisone

Objectives:

Comparison, in patients with septic shock, of efficacy and safety of the combination of moderate doses of corticosteroids and intensive insulin therapy to that of moderate doses of corticosteroids; and of efficacy and safety of hydrocortisone alone versus hydrocortisone plus fludrocortisone Methods

Study design :

This is a multicenter, prospective, randomised trial on parallel groups

Study treatments :

Experimental arm A:

A1=50 mg iv every 6 hours of hydrocortisone (hemisuccinate), 50µg through the nasogastric tube of 9 alpha fludrocortisone, for 7 days AND strict control of blood glucose levels with a target of 4,4 to 6 mmol/L using continuous iv infusion of insulin up to intensive care unit discharge.

A2=50 mg iv every 6 hours of hydrocortisone (hemisuccinate) for 7 days AND strict control of blood glucose levels with a target of 4,4 to 6 mmol/L using continuous iv infusion of insulin up to intensive care unit discharge.

Control arm B:

B1:50 mg iv every 6 hours of hydrocortisone (hemisuccinate), 50µg through the nasogastric tube of 9 alpha fludrocortisone, for 7 days.

B2:50 mg iv every 6 hours of hydrocortisone (hemisuccinate)for 7 days. Study Primary outcome : In-hospital mortality

Sample size calculation :

The expected in-hospital mortality rate in the control group is 50%. To detect an absolute reduction in in-hospital mortality rate of 12.5 %, that is 37.5% in the experimental arm versus 50% in the control arm, and considering risk alpha of 0,05 and a risk beta of 0,20, 254 patients per treatment arms are needed, for a total of 508 patients.

• Drug: recombinant human insulin
• Drug: hydrocortisone
• Drug: fludrocortisone
• Drug: Hydrocortisone

• Active Comparator: Hydrocortisone and convention glycemic control
• Experimental: Hydrocortisone and fludrocortisone and conventional glucose control
• Experimental: Hydrocortisone and intensive insulin therapy
• Experimental: hydrocortisone, fludrocortisone and intensive insulin therapy

Inclusion Criteria:

Patients admitted in intensive care units for septic shock and meeting all following criteria

• Proven infection
• Need for vasopressor to maintain systemic arterial tension above 90 mmHg
• Multiple organ dysfunction as defined by a SOFA score ³ 8.
• Need for treatment with moderate dose of corticosteroids

Exclusion Criteria:

One of the following :

• Pregnancy
• Less than 18 years old
• Moribund (i.e. expected to die on day of intensive care unit admission)