TRISS

Published

Holst L, et al. "Transfusion thresholds in Septic Shock". The New England Journal of Medicine. 2014. 371(15):1381-1391.
PubMed • Full text • PDF

Clinical Question

In patients with septic shock, how does a restrictive transfusion strategy (Hgb ≤ 7 g/dL) compare with a liberal transfusion strategy (Hgb ≤ 9 g/dL) in terms of 90-day mortality?

Bottom Line

Patients with septic shock who underwent transfusion at a Hgb threshold of 7 g/dL had similar mortality at 90 days but used 50% fewer units of blood compared with those who underwent transfusion at a Hgb threshold of 9 g/dL.

Major Points

Transfusion thresholds have been much contested. The 1999 TRICC trial demonstrated decreased rates of in-hospital mortality with a restrictive transfusion strategy as compared to a liberal transfusion strategy, establishing the paradigm for transfusion at 7 g/dL among euvolemic ICU patients. The study enrolled patients who were hemodynamically stable after initial fluid resuscitation, but excluded patients with septic shock. In patients with septic shock, the Surviving Sepsis Campaign recommends a high threshold of transfusion to maintain Hct >30% or Hgb >10 g/dL during the first 6 hours if hypoperfusion is present . However, this intervention has been questioned in light of data that more transfusions may be harmful to critically ill patients.

The 2014 Transfusion Requirements in Septic Shock (TRISS) trial randomized 998 ICU patients with septic shock and a Hgb <9 g/dL to a restrictive (Hgb ≤7 g/dL) or a liberal transfusion strategy (Hgb ≤ 9 g/dL). Patients with myocardial ischemia were excluded. Transfusions were administered as single units of prestorage leukoreduced RBCs. The primary outcome, mortality at 90 days, was similar between the restrictive and liberal strategy groups (43% vs. 45%; P=0.44). Secondary outcomes, including rate of ischemic events, use of vasopressor or inotropic therapy, need for mechanical ventilation, need for renal replacement therapy were also similar between the two groups. In addition, the restrictive strategy reduced blood transfusions by approximately 50% (1,545 vs. 3,088 units; P<0.001), about 1,500 fewer units or 500 L of blood.

The Surviving Sepsis guidelines currently advise a higher transfusion strategy until "hypoperfusion has resolved." Paul Hebert, the lead author of TRICC, argues for these guidelines to be re-written . TRISS supports the recommendation that a transfusion threshold of 7 g/dL should be adopted as the standard of care in patients with sepsis and septic shock.

Guidelines

Surviving Sepsis Campaign severe sepsis and septic shock (2016, adapted)

• RBC transfusion only when Hgb is <7 g/dL unless extenuating circumstances (e.g., MI, severe hypoxemia, hemorrhage; strong recommendation, moderate quality of evidence)

Design

• Multicenter, partially blinded, randomized, controlled trial
• N=998 ICU patients with septic shock
  • Restrictive strategy, Hgb ≤7g/dL (n=502)
  • Liberal strategy, Hgb ≤9g/dL (n=496)
• Setting: 32 ICUs in Denmark, Sweden, Norway, and Finland
• Enrollment: 2011-2013
• Analysis: Intention-to-treat
• Primary outcome: 90-day mortality

Population

Inclusion Criteria

• Age ≥18 years in the ICU
• Septic shock
• Hgb 9 g/dL or less

Exclusion Criteria

• Declined transfusion
• Previous adverse reaction to transfusion
• Received blood transfusion in ICU
• ACS
• Life-threatening bleeding
• Acute burn injury
• Withdrawal from active therapy
• Consent could not be obtained

Baseline Characteristics

• Median age: 67 years
• Male: 53.2%
• CV disease: 14.1%
• Admit to university hospital: 64.8%
• Emergent surgery during hospitalization: 40.9%
• Elective surgery during hospitalization: 11.3%
• Source of ICU admit:
  • ED: 16.9%
  • General ward: 52.6%
  • OR: 23.5%
• Source of sepsis
  • Lungs: 52.7%
  • Abdomen: 40.0%
  • Urinary tract: 12.0%
  • Soft tissue: 9.8%
• Positive culture from blood or sterile site: 34.9%
• Median interval from ICU admit to randomization: 21.5 hrs
• Median SAPS II: 51.5
• Median SOFA score: 10
• RRT: 12.1%
• Mechanical ventilation: 69.7%

Interventions

• Randomized to a group:
  • Restrictive transfusion - Threshold of Hgb ≤7g/dL
  • Liberal transfusion - Threshold of Hgb ≤9g/dL
• Single units of cross-matched, prestorage leukoreduced RBCs suspended in saline-adenine-glucose-mannitol solution were transfused when Hgb concentration decreased to assigned transfusion threshold
• Hgb reassessed within 3 hours after transfusion or prior to initiation of another transfusion
• Intervention period: entire ICU stay up to maximum 90 days after randomization
• If bleeding or ischemia that was potentially life-threatening, the treating physician determined appropriateness of additional transfusions
• Life support included use of vasopressor or inotropic therapy, mechanical ventilation, or RRT
• Ischemic event defined as cerebral ischemia, AMI, intestinal ischemia, or limb ischemia

Outcomes

Comparisons are restrictive vs. liberal transfusion strategy.

Primary Outcomes

Death by day 90

43% vs. 45% (RR 0.94; 95% CI 0.78-1.09; P=0.44)

Secondary Outcomes

Days alive without vasopressor or inotropic therapy

73% vs. 75% (P=0.93)

Days alive without mechanical ventilation

65% vs. 67% (P=0.49)

Days alive without RRT

85% vs. 83% (P=0.54)

Days alive and out of hospital

30% vs. 31% (P=0.89)

Number of transfusions

1545 vs. 3088 (P<0.001)

Transfused in the ICU

63.9% vs. 98.8% (P<0.001)

Subgroup Analysis

No difference in transfusion threshold on mortality at 90 days between: patients with and without chronic CV disease, patients ≤70 and >70, patients with SAPS II ≤53 and >53.

Adverse Events

Ischemic event

7.2% vs. 8.0% (P=0.64)

Use of life support at day 5

64.4% vs. 62.2% (P=0.47)

Use of life support at day 14

36.8% vs. 36.8% (P=0.95)

Use of life support at day 28

16.1% vs. 19.9% (P=0.14)

Criticisms

• Did not have systematic approach for evaluating rate of MI
• Not double blinded
• Unclear effects of leukoreduced blood as transfusion product

Funding

• Supported by grant from Danish Strategic Research Council and by Copenhagen University Hospital, Rigs-hospitalet, the Scandinavian Society of Anaesthesiology and Intensive Care Medicine (ACTA Foundation), and Ehrenreich's Foundation
• Multiple conflicts of interests from the authors