SAFE
PubMed • Full text • PDF
Clinical Question
Among ICU patients, does albumin compared to normal saline for fluid resuscitation improve clinical outcomes?
Bottom Line
Albumin and normal saline result in similar clinical outcomes when administered to ICU patients for volume resuscitation, except for those with traumatic brain injury who do worse with albumin.
Major Points
The medical literature has few rigorous studies of optimal fluid choices for intravascular fluid resuscitation. The 2004 Saline versus Albumin Fluid Evaluation (SAFE) trial randomized 6,997 ICU patients requiring IV fluids to either albumin or saline. There was no difference in all-cause mortality between the two groups. A subgroup analysis demonstrated that albumin was associated with poorer outcomes when administered to patients with traumatic brain injury. There was a non-significant trend towards mortality benefit with albumin use in patients with severe sepsis. A 2013 Cochrane review[1] found no benefit of albumin over crystalloids for fluid resuscitation in critically ill patients (all-cause mortality RR 1.01; 95% CI 0.93-1.10), leading the authors to conclude that the use of colloids is unjustified given their increased cost and lack of benefit.
The 2014 ALBIOS study further investigated the use of albumin administration to maintain patients with severe sepsis or septic shock to a serum albumin level ≥30g/L There was no difference in survival at 28 and 90 days, although those treated with albumin had more favorable SOFA subscores and received fewer vasopressors or inotropes.
Guidelines
Surviving Sepsis Campaign severe sepsis and septic shock (2016, adapted)[2]
- Recommend crystalloids as the fluid of choice for initial resuscitation and subsequent intravascular volume replacement in sepsis and septic shock (strong recommendation, moderate quality of evidence)
- Suggest using albumin in addition to crystalloids for initial resuscitation and subsequent intravascular volume replacement when patients with septic shock require substantial amounts of crystalloids (weak recommendation, low quality of evidence)
- Suggest using balanced crystalloids or saline for fluid resuscitation in sepsis and septic shock (weak recommendation, low quality of evidence)
Design
- Multicenter, double blind, parallel group, randomized, controlled trial
- N=6,997
- Albumin (n=3,497)
- Saline (n=3,500)
- Setting: 16 academic centers in Australia and New Zealand
- Enrollment: 2001-2003
- Analysis: Intention-to-treat
- Primary outcome: All-cause mortality at 28 days
Population
Inclusion Criteria
- ICU patients
- Age ≥18 years
- Requiring IV fluids as assessed by treating physician
Exclusion Criteria
- ICU admission following cardiac surgery, liver transplantation, or burn treatment
Baseline Characteristics
Comparisons are albumin vs. normal saline.
- Age in years: 58.6 vs. 58.5
- Females: 40.7% vs. 39.3%
- Surgical admission: 43.0% vs. 42.8%
- Subgroups:
- Trauma: 17.4% vs 17.2%
- Severe sepsis: 18.1% vs. 18.4%
- ARDS: 1.8% vs 1.9%
- Mean APACHE II score: 18.7 vs. 19.0
- Physiologic measurements:
- Heart rate: 91.4 vs 92.3
- MAP: 77.8 mmHg vs. 78.2 mmHg
- CVP: 9.0 vs. 8.6 mmHg
- Urine output: 89.7 vs. 95.0 ml/hr
- Serum albumin: 27.4 vs. 27.7 g/L
- SOFA score:
- 0: 57.2% vs. 55.1%
- 1: 31.4% vs. 33.5%
- 2: 9.8% vs. 9.6%
- 3: 1.5% vs 1.7%
- 4: 0.1% vs. 0.1%
- 5: <0.1% vs. 0%
- Mechanical ventilation: 63.8% vs. 64.8%
- Renal replacement therapy: 1.3% vs. 1.2%
- Albumin in prior 72h: 3.7% vs. 3.9%
Interventions
- Randomized to receive either 4% albumin or normal saline for fluid resuscitation
- Participants were stratified by institution and diagnosis of trauma
- Continuation of administration of fluid resuscitation until death, discharge, or for a total of 28 days
- Patients continued to receive other fluids (including maintenance, blood products, TPN, enteral feedings) as seen fit by clinicians
Outcomes
Comparisons are albumin vs. normal saline.
Primary Outcomes
- 28-day mortality
- 20.9% vs. 21.1% (RR 0.99; 95% CI 0.91-1.09; P=0.87)
Secondary Outcomes
- 28-day survival
- Alive in ICU: 3.2% vs. 2.5% (RR 1.27; 95% CI 0.96 to 1.68; P=0.09)
- Alive in hospital: 22.8% vs. 24.5% (RR 0.93; 95% CI 0.86-1.01; P=0.10)
- New organ failure
- No organ failure: 52.7% vs 53.3%
- One organ: 30.0% vs. 29.8%
- Two organs: 13.9% vs. 13.5%
- Three organs: 2.6% vs. 2.8%
- Four organs: 0.7% vs. 0.6%
- Five organs: 0.1% vs 0%
- (P=0.85 for no organ failures vs. any organ failures.)
- Duration of mechanical ventilation
- 4.5 vs. 4.3 days (absolute difference [AD] 0.24; 95% CI -0.06 to 0.54; P=0.44)
- Duration of renal replacement therapy
- 0.48 2.28 vs. 0.39 days (AD 0.09; 95% CI -0.0 to 0.19; P=0.41)
- Duration of stay
- In ICU: 6.5 vs. 6.2 days (AD 0.24; 95% CI -0.06 to 0.54; P=0.44)
- In hospital: 15.3 vs. 15.6 days (AD -0.24; 95% CI -0.70 to 0.21; P=0.30)
Subgroup Analysis
- 28-day mortality
- Trauma: 13.6% vs. 10.0% (RR 1.36; 95% CI 0.99 to 1.86; P=0.06)
- Severe sepsis: 30.7% vs 35.3% (RR 0.87; 95% CI 0.74 to 1.02; P=0.09)
- ARDS: 39.3% vs. 42.4% (RR 0.93; 0.61 to 1.41; P=0.72)
- TBI: 25% vs. 15% (RR 1.62; 95% CI 1.12-2.34; P=0.009)
Adverse Events
Not explicitly stated in article.
Criticisms
- Not powered sufficiently to detect small differences in the subgroups
- Death at 28 days may not be the most appropriate outcome measure for patients with brain injury
Funding
- Auckland District Health Board
- Middlemore Hospital
- Health Research Council of New Zealand
- Australian Commonwealth Department of Health and Medical Research Council
- Health Department of Western Australia
- New South Wales Health Department
- Northern Territory Health Services
- Queensland Health Services Department
- Royal Hobart Hospital
- South Australian Department of Human Services
- Victorian Department of Human Services
Further Reading
- ↑ Perel P, et al. "Colloids versus crystalloids for fluid resuscitation in critically ill patients." Cochrane Database of Systematic Reviews. 2013;2:CD000567.
- ↑ Rhodes A, et al. "Surviving Sepsis Campaign: International guidelines for management of sepsis and septic shock: 2016." Critical Care Medicine. 2017;45(3)1-67.