Background
Phase 3, multicenter RCT (TRICC trial). 838 euvolemic anemic critically ill adult patients in Canadian ICUs with hemoglobin ≤9 g/dL within 72 hours of ICU admission. Patients with active acute MI, unstable angina, or chronic anemia excluded. This was the landmark trial establishing restrictive transfusion thresholds in the ICU.
Interventions and follow up
Arm A: Restrictive strategy — RBC transfusion triggered when Hgb <7.0 g/dL; target Hgb 7.0–9.0 g/dL
Arm B: Liberal strategy — RBC transfusion triggered when Hgb <10.0 g/dL; target Hgb 10.0–12.0 g/dL
Primary endpoint: 30-day all-cause mortality
mFollow up: 60 days in-hospital
Arm B: Liberal strategy — RBC transfusion triggered when Hgb <10.0 g/dL; target Hgb 10.0–12.0 g/dL
Primary endpoint: 30-day all-cause mortality
mFollow up: 60 days in-hospital
Results
30-day mortality: 18.7% vs 23.3%, P=.11 — NS (trend favoring restrictive)
Hospital mortality: 22.2% vs 28.1%, P=.05
Subgroup (APACHE ≤20 or age <55): Restrictive significantly superior (P=.02 and P=.03)
RBC units transfused: 2.6 vs 5.6 units per patient — 54% reduction
Cardiac complications: 21.0% vs 29.3% (pulmonary edema, MI) — significantly lower restrictive
Hospital mortality: 22.2% vs 28.1%, P=.05
Subgroup (APACHE ≤20 or age <55): Restrictive significantly superior (P=.02 and P=.03)
RBC units transfused: 2.6 vs 5.6 units per patient — 54% reduction
Cardiac complications: 21.0% vs 29.3% (pulmonary edema, MI) — significantly lower restrictive
Adverse events
Main adverse events: Pulmonary edema: 5.3% vs 10.7% (P=.006 — significantly lower with restrictive). New-onset MI: 0.7% vs 2.9% (P=.02). Multi-organ failure similar. Grade ≥3 bleeding and infectious complications not different between arms.
Conclusions
A restrictive RBC transfusion strategy (trigger Hgb <7 g/dL) was at least as safe as and likely superior to a liberal strategy in most critically ill patients, significantly reducing transfusion exposure and cardiac complications, and establishing Hgb 7 g/dL as the standard ICU transfusion threshold.
Key Limitations
Key Limitations: Primary endpoint (30-day mortality) was not statistically significant (P=.11), making the trial technically negative by its primary outcome, though the trend and subgroup data drove broad adoption of restrictive strategies. Patients with acute MI/unstable angina excluded — limiting generalizability to cardiac ICU patients. Single-country (Canadian) pragmatic design. Predates modern resuscitation practices and erythropoiesis-stimulating agents in ICU.
Clinical Context
TRICC established the restrictive transfusion threshold (Hgb 7 g/dL) as standard of care in non-cardiac critically ill patients, adopted in AABB, NICE, and CHEST guidelines. Subsequent trials (FOCUS, TRISS, TRICS-III) validated this threshold in cardiac surgery and orthopedic surgery populations. Liberal strategies are still considered for patients with active ACS, and there is ongoing debate in septic shock (TRISS showed non-inferiority of Hgb 7 in septic shock).
References
References: Hébert PC et al, NEJM 1999 (TRICC primary)