Severe infection in the intensive care unit is common accounting for about 10% of admissions and has a death rate of approximately 40-50%. It is almost always associated with significant reductions in blood pressure. Administration of fluid often in large volumes is essential to normalize blood pressure and prevent failure of organs and death. Two common classes of fluid solutions are crystalloid fluids (salt based, normal saline) and colloid fluids (protein based, albumin). Due to its properties, the albumin fluid may remain in the vascular space better than the normal saline solution. Hence, there may be faster attainment of normal blood pressure as well as a reduction in failed organs and death. Preliminary clinical trial data suggests a potential for benefit with albumin in this setting but these findings require confirmation in a large clinical trial.

There are few data to explain how albumin may exert its protective effects and lead to better outcomes for patients with severe infections. We will conduct a clinical study that will examine potential biological mechanisms for albumin's protective effects in 50 patients across 6 Canadian academic hospitals. We will also examine our ability to successfully recruit patients into this trial.

This study will provide information that will help to understand the biological mechanisms of albumin in severe infection. The information gained will guide the investigative team for future fluid related mechanistic questions. The study will also provide essential information that will aid in the design and conduct of the future large clinical trial that will examine death as its primary outcome.

Background: Severe sepsis and septic shock are the most common causes of mortality among critically ill patients and account for approximately 10% of admissions to the intensive care unit (ICU) and 3% of all admissions to hospital. Despite decades of therapeutic investigation, mortality of septic shock has remained at approximately 40-50%. Fluid resuscitation is an integral component of early treatment, as several liters of fluid may be administered in the first hours of septic shock in an attempt to re-establish hemodynamic stability. The two main classes of fluid expand the plasma volume include crystalloids and colloids.

In Canada, the main crystalloid fluids used for resuscitation are normal saline and ringers lactate; the main colloid fluids used for resuscitation are hydroxyethyl starches (HES) and albumin. Until recently, no randomized controlled (RCT) or systematic review had confirmed benefit or harm with either colloidal or crystalloid fluids in this patient population. An RCT published in 2008 demonstrated that 10% pentastarch (HES fluid) causes acute renal failure defined by the requirement for renal replacement therapy. In contrast, a subgroup analysis from a RCT of severe sepsis suggests that 4% albumin fluid may reduce death. However, these hypothesis-generating findings require confirmation in the context of a large RCT.

There are limited observational mechanistic studies designed to understand the potential biological mechanisms for albumin's potential protective effects in sepsis. Moreover, mechanistic data in the clinical trial setting are lacking. The PRECISE pilot RCT (Five Percent Albumin versus Normal Saline as Fluid Resuscitation Strategies for the Management of Early Suspected Septic Shock) will help to understand the biological mechanisms of albumin in early septic shock and explore feasibility issues in preparation for a large RCT.

Study Hypotheses: We hypothesize that resuscitation with 5% albumin as compared to normal saline in early septic shock may exert some of its protective effects through the following biological mechanisms: 1) Limiting thrombin generation, 2) Supporting endogenous protein C and activated protein C (APC) generation, 3) Limiting the production of systemic pro-inflammatory cytokines and chemokines and supporting the production of cytokines and chemokines with anti-inflammatory potential, and 4) Reducing the ability of septic plasma to generate Neutrophil Extracellular Traps (NETs).

Study Design/Setting: A multi-centre randomized controlled trial conducted in 6 Canadian academic centres.

Participants: The study will enroll 50 adult patients who are identified with early suspected septic shock from the emergency department (ED) and the intensive care unit (ICU). Patients need to be hypotensive or require vasopressor agents, have at least 2 systemic inflammatory response syndrome criteria, and have no other form of shock (e.g., cardiogenic, obstructive, or hemorrhagic shock) and no previous ICU admission for severe sepsis or septic shock after receiving at least 1 litre of crystalloid and not more than 250 mls of colloid fluid within 8 hours of the first hypotensive event.

Interventions: Participants will be randomly assigned to 1 of 2 arms. Patients, families, clinicians, and research personnel will be blinded to the fluid administered (5% albumin or 0.9% normal saline). Fluid resuscitation will begin immediately after randomization and continue for the first 7 days or until discharge from the ICU or ED if this occurs sooner. Blood and urine will be collected at randomization and then at 6, 12, 24, 72 hrs and 7 days later. Study fluid will be administered as rapidly as possible as 500 ml fluid challenges. All other co-interventions of care will not be protocolized or mandated during the study period.

Main Outcome Measures: The primary outcome is a comparison of thrombin generation at 72 hours, and protein C and APC levels at 7 days. Secondary outcomes include a comparison of inflammatory cytokine and chemokine levels in the blood and urine. Other clinical outcomes will include mortality in ICU, hospital, and at 30 days, length of stay in ICU and hospital, development of organ failure, and need for organ support (defined by the number of days requiring mechanical ventilation and renal replacement therapy) and feasibility measured in terms of patient recruitment.

Trial Team: The trial will be led by 3 primary investigators (Lauralyn McIntyre, Alison Fox-Robichaud, and Alan Tinmouth), who will have responsibility for all aspects of the trial. They have expertise in hematology, transfusion medicine, basic science and clinical trials in critical care. Our team includes co-investigators from the ICU (Drs. Lauralyn McIntyre, Joe Pagliarello, Anand Kumar, Sean Bagshaw, Alison Fox-Robichaud, Alexis Turgeon), ED (Drs. Ian Stiell, Brian Rowe, Andrew Worster, Marcel Emond), and ICU/ED (Drs. Robert Green, David Easton) who will act as clinical champions at each participating site. Many co-investigators have extensive experience in the conduct of multi-centre RCTs (Drs. Paul Hebert, Deborah Cook, Dean Fergusson, John Marshall, Ian Stiell, Brian Rowe, Simon Finfer, Alan Tinmouth, Lauralyn McIntyre). The translational co-investigative team includes critical care physicians with expertise in sepsis pathophysiology (Drs. Alison Fox Robichaud, Paul Kubes, Patricia Liaw, Anand Kumar, John Marshall, Tao Rui, and Claudio Martin).

Significance of findings: Results from our PRECISE pilot RCT may provide a biologic rationale for a beneficial effect of 5% albumin in early septic shock resuscitation. If a biologic rationale is confirmed, the study will also provide essential information to determine the feasibility and aid the design of a definitive phase III trial of albumin in resuscitation of septic shock.

Inclusion Criteria:

Eligible patients will meet the following 2 inclusion criteria.

1. Refractory Hypotension: The patient has received at least 1 litre of normal saline or ringers lactate crystalloid fluid for hypotension (defined as a systolic blood pressure less than 90 mm Hg, or a systolic blood pressure more than 40 mm Hg below baseline, or a mean arterial blood pressure less than 65 mm Hg) within 8 hours of the first hypotensive event and has refractory hypotension defined by: a) Presence of hypotension OR b) Initiation of a vasopressor agent(s).
2. Two or more criteria for the systemic inflammatory response syndrome (SIRS): i) heart rate greater than 90 beats per minute; ii) respiratory rate greater than 20 breathes per minute, or a PaC02 less than 32 mm Hg, or mechanically ventilated; iii) temperature greater than 38 or less than 36 degrees Celsius; iv) or a white blood cell count greater than 12,000 x 109/L or less than 4,000 x 109/L, or more than 10% bands on the differential.

Exclusion Criteria:

1. Clinically apparent other forms of shock including cardiogenic, obstructive (massive pulmonary embolism, cardiac tamponade, tension pneumothorax) or hemorrhagic shock
2. Previous admission to ICU with severe sepsis or septic shock during the index hospitalization
3. More than 250 mls of colloid fluid (Pentaspan, Voluven, Hextend, or albumin) from the first episode of hypotension
4. A known previous severe reaction to albumin
5. Burns
6. Known cirrhosis
7. Traumatic brain injury
8. Religious objection to albumin use
9. Less than 18 years of age
10. Pregnant
11. Family or patient not committed to aggressive care
12. Currently enrolled in another related interventional study.