Background
Phase 3, multicenter, open-label RCT (TRISS trial). 998 adult patients with septic shock in 32 Scandinavian ICUs, requiring RBC transfusion for Hgb ≤9 g/dL within 24 hours of ICU admission for septic shock. Septic shock defined per Surviving Sepsis Campaign criteria. Prior TRICC trial had excluded patients with acute MI/unstable angina; TRISS specifically targeted septic shock to address this evidence gap.
Interventions and follow up
Arm A: Restrictive transfusion strategy — RBC transfusion triggered when Hgb <7.0 g/dL; maintain Hgb 7–9 g/dL
Arm B: Liberal transfusion strategy — RBC transfusion triggered when Hgb <9.0 g/dL; maintain Hgb 9–11 g/dL
Primary endpoint: 90-day all-cause mortality
mFollow up: 90 day
Arm B: Liberal transfusion strategy — RBC transfusion triggered when Hgb <9.0 g/dL; maintain Hgb 9–11 g/dL
Primary endpoint: 90-day all-cause mortality
mFollow up: 90 day
Results
90-day mortality: 43.0% vs 45.0%, RR 0.94 (95% CI 0.78–1.09), P=.44 — NS (noninferiority confirmed)
28-day mortality: 37.0% vs 41.1%, RR 0.90 (95% CI 0.76–1.04) — NS
Ischemic events: 5.0% vs 3.0%, P=.09 — slight trend more with restrictive, NS
RBC units transfused: 1.0 vs 4.0 units per patient — 75% reduction in transfusion exposure
Days alive without life support at 90 days: Similar between groups
28-day mortality: 37.0% vs 41.1%, RR 0.90 (95% CI 0.76–1.04) — NS
Ischemic events: 5.0% vs 3.0%, P=.09 — slight trend more with restrictive, NS
RBC units transfused: 1.0 vs 4.0 units per patient — 75% reduction in transfusion exposure
Days alive without life support at 90 days: Similar between groups
Adverse events
Main adverse events: Ischemic events (MI, stroke, limb ischemia, intestinal ischemia) trending higher in restrictive arm (5% vs 3%) but not statistically significant. Serious adverse reactions to transfusion similar. ARDS, acute kidney injury, and organ failure rates similar. Length of ICU stay and hospital stay similar.
Conclusions
A restrictive RBC transfusion strategy (Hgb trigger 7 g/dL) was noninferior to a liberal strategy (trigger 9 g/dL) in adults with septic shock, confirming the safety of the restrictive threshold even in this high-acuity population.
Key Limitations
Key Limitations: Open-label design (blinding of transfusion thresholds is impractical). Noninferiority margin was pre-specified at RR 1.20 for mortality — does not exclude modest harm. Small but non-significant increase in ischemic events with restrictive strategy warrants attention in patients with known cardiovascular disease or mesenteric ischemia risk. Enrolled in Scandinavian ICUs with high-quality supportive care — may not generalize to resource-limited settings. Liberal threshold of 9 g/dL is higher than many current "liberal" comparators in other trials.
Clinical Context
TRISS, combined with TRICC and TRICS-III, provides a complete evidence base supporting the Hgb 7 g/dL restrictive threshold across major ICU populations including septic shock. Surviving Sepsis Campaign guidelines recommend Hgb 7 g/dL as the target trigger in septic shock without active myocardial ischemia, hypoxemia, or acute hemorrhage. Higher thresholds (8–9 g/dL) may still be appropriate in patients with acute coronary syndrome or severe hypoxemia.
References
References: Holst LB et al, NEJM 2014 (TRISS primary)