Timing of Transfusion Strategy Initiation in Acute Brain Injury: A Subanalysis of the TRAIN Study

Recent randomized trials suggested that a liberal transfusion strategy (LTS) may be beneficial compared with a restrictive strategy (RTS) in patients with acute brain injury (ABI), although results remain inconsistent across studies. The aim of our study was to investigate whether the time of initiation of these transfusion strategies might influence neurological outcomes in patients with ABI.

This was a post hoc exploratory analysis of the international, multicenter TRAIN randomized clinical trial. Adult patients with traumatic brain injury, aneurysmal subarachnoid hemorrhage, or intracerebral hemorrhage were randomized to receive transfusion at a hemoglobin threshold of either below 9 g/dL (LTS) or below 7 g/dL (RTS). The exposure of interest was time from intensive care unit (ICU) admission to randomization, analyzed both categorically (i.e., as early, ≤ 3 days vs. late, > 3 days) or as a continuous variable. The primary outcome was the occurrence of unfavorable neurological outcome at 180 days (UO), defined as a Glasgow Outcome Scale-Extended score of 1-5. Associations were examined using multivariable logistic regression, propensity score matching, generalized additive models, and segmented regression. A formal interaction between transfusion strategy and timing was tested to assess effect modification.

Among the 820 included patients, baseline characteristics were well balanced between transfusion strategies within early and late randomization groups. LTS was consistently associated with a lower probability of UO compared with RTS, both in early- and late-randomization group, although point estimates suggested a numerically stronger association when randomization occurred earlier [odds ratio (OR) 0.51, 95% confidence interval (CI) 0.34-0.76]. LTS was associated with fewer cerebral ischemic events (OR 0.33, 95% CI 0.14-0.77), shorter mechanical ventilation and ICU stay in the early-randomization group, whereas in the late-randomization group LTS was associated with lower rates of sepsis (OR 0.33, 95% CI 0.13-0.86) and ARDS (OR 0.39, 95% CI 0.2-0.78]). Continuous modeling showed no clear relationship between time to randomization and UO. The interaction between transfusion strategy and timing was not statistically significant (p = 0.88). Exploratory exposure-response analyses suggested that greater cumulative exposure to hemoglobin levels above 9 g/dL during the ICU stay was associated with a lower probability of UO.

In patients with ABI, a LTS was associated with a lower probability of unfavorable long-term neurological outcome across different times to randomization. Timing to randomization may influence the impact of transfusion strategies and the occurrence of some adverse events, although prospective studies specifically designed to address this question are needed to confirm it.