Approaches to Guide Red Blood Cell Transfusion in Patients with Subarachnoid Hemorrhage: A Systematic Review and Meta-Analysis.

BackgroundInternational guidelines acknowledge the lack of subarachnoid hemorrhage (SAH) specific red blood cell transfusion thresholds. This study aimed to evaluate the efficacy and safety of red blood cell transfusion strategies in patients with SAH.MethodsWe systematically searched PubMed/MEDLINE, CENTRAL, EMBASE, and Google Scholar from inception to July 15, 2025. We included randomized controlled trials (RCTs) evaluating any transfusion strategy in SAH patients. Meta-analyses were conducted using random-effects models, and the certainty of evidence was assessed using the GRADE minimally contextualized approach. The protocol was registered in PROSPERO (CRD420251122925).ResultsThree RCTs comprising 966 participants were included. Two compared restrictive (hemoglobin threshold 7-8 g/dL) versus liberal (9-10 g/dL) transfusion strategies, and one compared a higher (≥11.5 g/dL) versus lower (≥10 g/dL) hemoglobin targets. A liberal transfusion strategy, compared with the restrictive approach, probably reduces unfavorable neurological outcomes (39.6% vs 45.5%; risk difference [RD]: -5 per 100; 95% CI: -10 to +1) and new cerebral ischemia (17.7% vs 22.9%; RD: -5 per 100; 95% CI: -11 to +2), while probably increasing hemoglobin levels (mean difference [MD]: +1.68 g/dL; 95% CI: +0.32 to +3.04), transfusion requirements (98.2% vs 37.7%; RD: +52 per 100; 95% CI: +24 to +92), and ICU length of stay (MD: +1.16 days; 95% CI: -0.11 to +2.42). It may also reduce the risk of cerebral vasospasm (30.0% vs 36.4%; RD: -5 per 100; 95% CI: -14 to +9), although the evidence is very uncertain. Evidence for higher versus lower hemoglobin targets (≥11.5 g/dL vs ≥10 g/dL) was very uncertain across all outcomes.ConclusionCompared with a restrictive strategy, a liberal transfusion approach in patients with SAH probably improves neurological outcomes and reduces cerebral ischemia. Evidence for higher hemoglobin targets remains very uncertain, underscoring the need for larger, high-quality trials to refine optimal transfusion thresholds.