The purpose of this trial is to determine the influence of colloid versus crystalloid volume resuscitation and of intensive vs conventional insulin therapy on morbidity and mortality of patients with severe sepsis and septic shock.

Severe sepsis and septic shock have a high mortality. Research has concentrated on adjunctive sepsis therapies; the role of supportive measures is comparatively unclear. In Europe the use of colloids is widespread, but there is no evidence on the role of either crystalloid or colloid volume therapy in sepsis. Recently, a higher incidence of kidney failure in sepsis was reported after administration of colloids.

In critical illness, a significant reduction in mortality was recently achieved by strict glycemic control, however it has to be determined whether this is true and safe for patients with sepsis as well.

Inclusion Criteria:

• Patients fulfilling the criteria of "severe sepsis" or "septic shock" according to the Society of Critical Care Medicine/American College of Chest Physicians (SCCM/ACCP) definitions not longer than 24 hours before ICU admission or 12 hours after ICU admission

Exclusion Criteria:

• Age < 18 years
• Pregnancy
• Known allergy against hydroxyethyl starch
• Pre-treatment with > 1000ml hydroxyethyl starch within 24 hours before inclusion
• Pre-existing kidney failure requiring dialysis or serum creatinine value > 320 mmol/l (3,6 mg/dl)
• Intracerebral hemorrhage
• Severe head trauma with edema
• FiO2 at time of study inclusion > 0,7
• Heart failure (New York Heart Association [NYHA] IV)
• Enrolment in another interventional study
• Immune suppression (cytostatic chemotherapy, steroid therapy, AIDS)
• Do not resuscitate (DNR) order