TL;DR: 🟢 Positive — liberal transfusion significantly improved neurologic outcome in acute brain injury (TBI/SAH/ICH combined; 62.6% vs 72.6% unfavorable, P=0.002), challenging the 25-year TRICC-derived restrictive default.
1. Publication
- Title: Restrictive vs Liberal Transfusion Strategy in Patients With Acute Brain Injury: The TRAIN Randomized Clinical Trial
- Acronym: TRAIN
- Year & Journal: JAMA, published October 9, 2024 (332(19):1623-1633)
- Citation: Taccone FS, Rynkowski Bittencourt C, Møller K, et al; TRAIN Study Group. JAMA. 2024;332(19):1623-1633. doi:10.1001/jama.2024.20424
2. Context & Rationale
Background: TRICC (1999) established restrictive transfusion as standard in general critical illness, but acute brain injury (TBI, SAH, ICH) patients may have unique oxygen-delivery vulnerabilities where restrictive thresholds could be harmful.
Research Question/Hypothesis: In acute brain injury patients (TBI, SAH, or ICH; GCS≤13) with anemia, does a liberal transfusion strategy improve neurological outcome at 180 days compared with restrictive?
3. Design & Methods
- Study Type: Prospective, multicenter, randomized interventional trial
- Population: 820 analyzed adults (18-80y) with TBI, SAH, or ICH, GCS≤13, expected ICU stay >72h, Hb≤9 g/dL within 10 days of injury; enrolled Sept 2017–Dec 2022
- Intervention: Liberal strategy — transfusion trigger Hb≤9 g/dL (n=397)
- Comparator: Restrictive strategy — transfusion trigger Hb≤7 g/dL (n=423)
- Statistical Power & Follow-Up: Primary: unfavorable neurologic outcome (GOS-E 1-5) at 180 days. Thresholds maintained up to 28 days or discharge/death.
4. Key Results
820 patients analyzed.
Outcome | Liberal | Restrictive | Effect Size | 95% CI | p-value | Notes |
Unfavorable neurologic outcome at 180d (primary) | 62.6% | 72.6% | Adjusted RR 0.86 | — | 0.002 | Significant benefit with liberal strategy |
Cerebral ischemic events | 8.8% | 13.5% | RR 0.65 | — | — | Fewer with liberal strategy |
ICU/hospital LOS, organ failure at 28d, mortality | No significant difference | No significant difference | — | — | — | No effect on these outcomes |
Transfusion burden | Median 2 units | Median 0 units | — | — | — | Expected separation |
Consistency: Findings were consistent across all subgroups of brain injury type and post hoc analyses.
5. Internal Validity Assessment
Large (820-patient), multicenter RCT with consistent subgroup findings. Overall: Strong — statistically significant, clinically important benefit (RR 0.86, P=0.002) with a coherent mechanistic companion finding (fewer cerebral ischemic events); consistency across subgroups and post hoc analyses strengthens confidence.
6. External Validity Assessment
Multicenter, international acute brain injury population (TBI, SAH, ICH combined) — broadly representative of severe acute brain injury requiring ICU care.
7. Strengths & Limitations
Strengths: Large, well-powered, statistically significant primary result; mechanistically coherent secondary finding (reduced cerebral ischemia); consistency across brain injury subtypes and post hoc analyses; findings consistent with the large HEMOTION trial in TBI specifically.
Limitations: Combines multiple brain injury subtypes (TBI, SAH, ICH) which may have differing pathophysiology; contrasts with the null SAHaRA result in aSAH specifically (this handbook), raising questions about whether the benefit is uniform across subtypes or driven by non-SAH patients.
8. Interpretation & Practice Impact
Supports a liberal (Hb≤9) rather than restrictive (Hb≤7) transfusion threshold in acute brain injury — directly challenging the TRICC-derived restrictive default that has dominated general critical care for 25 years, specifically for this population.
9. Controversies & Subsequent Evidence
The main findings are described as consistent with the large HEMOTION trial (742 adults with moderate/severe TBI and anemia). However, TRAIN's positive result contrasts with SAHaRA's null result specifically in aSAH (this handbook) — an active area of investigation, with a secondary analysis of TRAIN specifically examining spontaneous ICH patients (Stroke, 2026) to further dissect subtype-specific effects.
10. Summary & Executive Takeaway
Summary: TRAIN randomized 820 acute brain injury patients (TBI, SAH, ICH) with anemia to liberal (Hb≤9) or restrictive (Hb≤7) transfusion. Unfavorable neurologic outcome at 180 days was significantly lower with liberal transfusion (62.6% vs 72.6%, RR 0.86, P=0.002), with fewer cerebral ischemic events and no effect on mortality or LOS.
Overall Takeaway: TRAIN provides a significant, consistent, and mechanistically coherent challenge to the TRICC-derived restrictive transfusion default specifically in acute brain injury — supporting a liberal (Hb≤9) threshold, though the contrast with the null SAHaRA result in aSAH specifically suggests the benefit may not be uniform across all brain injury subtypes.
11. Bibliography
- Turgeon AF, et al. HEMOTION trial. N Engl J Med. 2024 (this handbook, Trauma category).
- English SW, et al. SAHaRA. N Engl J Med. 2025;392(11):1079-1088 (this handbook).
- Hébert PC, et al. TRICC trial. N Engl J Med. 1999;340:409-417.