The purpose of this study is to compare eritoran tetrasodium and placebo in patients with severe sepsis and to demonstrate a reduction of mortality from all causes.

INCLUSION CRITERIA:

• Age >= 18 years

• Confirmed early-onset severe sepsis, defined as:

  o---Objective evidence of infection likely to be caused by a bacterial or fungal pathogen

  o---Presence of at least 3 of 4 systemic inflammatory response syndrome (SIRS) criteria

  o---Sepsis-associated organ dysfunction

• Baseline Acute Physiology and Chronic Health Evaluation II (Apache II) Score of 21 to 37

• < 12 hours between onset of the first qualifying organ dysfunction and expected administration of study drug

• A commitment to full patient support

EXCLUSION CRITERIA:

• Pregnancy or breastfeeding
• Extensive (>20% Body Surface Area) third-degree burns
• Weight > 150 kg at admission
• Patients whose death from sepsis is considered imminent
• Patients not expected to survive for at least 2 months due to a pre-existing and uncorrectable medical condition, or those in a chronic vegetative state
• Patients with severe congestive heart failure
• Patients currently receiving immunosuppressive therapy such as cyclosporine, azathioprine, or cancer chemotherapy
• Patients with granulocyte counts < 1000/mm^3 unless the decreased count is believed to be due to sepsis
• Patients that required cardiopulmonary resuscitation in the 4 weeks prior to evaluation for enrollment
• Human immunodeficiency virus (HIV)-positive patients with CD4 count <= 50/mm^3 within 4 weeks of enrollment, or end-stage processes
• Patients with significant hepatic impairment, portal hypertension, or esophageal varices
• Patients who are expected to be treated with endotoxin-removal devices
• Patients with active cancer
• Patients receiving polymyxin B or colistin