Liberal Versus Restrictive Transfusion in Acute Brain Injury: A Systematic Review and Meta-Analysis

Abstract

Oxygen is critical for neurological function and survival, particularly in acute brain injury. Although transfusion at higher thresholds theoretically provides improved oxygen delivery to neurons, there is an associated risk of allogeneic reactions and increased utilization of limited blood resources. Conversely, although a lower threshold conserves resources, it may increase the risk of neuronal oxygen deprivation. The optimal transfusion strategy for patients with acute brain injury remains unclear. This systematic review and meta-analysis aimed to compare the efficacy and safety of liberal (hemoglobin threshold ≤10 g/dL) versus restrictive (≤8 g/dL) transfusion strategies in patients with acute brain injury (traumatic brain injury (TBI), subarachnoid hemorrhage, or intracranial hemorrhage), synthesizing evidence from randomized controlled trials (RCTs). We searched the PubMed, Excerpta Medica database (Embase), and the Cochrane Central Register of Controlled Trials (CENTRAL) databases to identify RCTs comparing restrictive and liberal transfusion strategies in patients with acute brain injury. Eligible trials reported outcomes including (1) unfavorable neurological outcomes defined as a score ≤5 on the Glasgow Outcome Scale Extended (GOSE) at six months, (2) mortality, (3) acute respiratory distress syndrome (ARDS), and (4) infections. Statistical heterogeneity was assessed using I² statistics. To account for statistical heterogeneity, a random-effects model was used to analyze all outcomes. We included six RCTs comprising 2,645 patients, of whom 1,303 (49.2%) were randomized to a liberal transfusion strategy. A reduction in unfavorable neurological endpoints (55.7% vs. 61.4%; risk ratios (RR) 0.92; 95% CI 0.84-1.01) was observed in the liberal group, although this difference was not statistically significant. In sensitivity analyses of unfavorable neurologic outcomes, statistical significance was achieved by removing a single paper (54.7% vs. 61.6%; RR 0.89; 95% CI 0.84-0.95) or by employing a fixed-effects model (RR 0.91; 95% CI 0.85-0.97). This meta-analysis suggests that there is no substantial difference in outcomes between a liberal and a restrictive transfusion strategy in patients with acute brain injury. While our primary analysis showed no statistically significant difference between strategies, sensitivity analyses suggested a potential benefit of liberal transfusion in reducing unfavorable neurologic outcomes. However, given the non-significant primary results and the importance of blood conservation, a restrictive strategy may be reasonable until further evidence emerges.

Keywords: acute brain injury; liberal transfusion; restrictive transfusion; subarachnoid hemorrhage; traumatic brain injury.

Figure 1. PRISMA flow diagram of study screening and selection

PRISMA: PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses; Embase: Excerpta Medica database

Figure 2. Pooled analysis of unfavorable neurological outcomes

A. There was a non-significant decrease in unfavorable neurological outcomes in the liberal group using the random-effects model. B. There was a significant decrease in unfavorable neurological outcomes in the liberal group when leaving out the study by Robertson et al. (2014). C. There was a significant decrease in unfavorable neurological outcomes in the liberal group using the fixed-effects model.

Figure 3. Pooled analysis of venous thromboembolism (VTE)

Figure 4. Pooled analysis of acute respiratory distress syndrome (ARDS)

Figure 5. Funnel plot analysis

SE: standard error; RR: risk ratios