Researchers in The New England Journal of Medicine have published findings from a randomized trial of blood transfusion strategy that explored a liberal versus restrictive approach. They sought to compare the impact of liberal versus restrictive red blood cell (RBC) transfusion strategy on the, “mortality and long-term functional and patient-centered outcomes of critically ill adult patients with moderate-to-severe traumatic brain injury (TBI).” The authors explained that the Hemoglobin Transfusion Threshold in Traumatic Brain Injury Optimization (HEMOTION) study featured 34 sites in Canada, the United Kingdom, France, and Brazil and included moderate or severe TBI patients 18 years-old or older with anemia defined as, “a hemoglobin level of ≤10 g per deciliter. [The study] excluded patients who received transfusion after ICU admission but before randomization and who had contraindications or objection to transfusion. Patients who received transfusion before ICU admission were not excluded.”
The authors of the study of liberal versus restrictive blood transfusion strategy explored and hypothesized that, “a liberal strategy (triggered by a hemoglobin level of ≤10 g per deciliter) would result in better outcomes than a restrictive strategy (triggered by a hemoglobin level of ≤7 g per deciliter).” They explained that the transfusion thresholds were chosen based on, “available evidence, expert opinion, and clinical equipoise and were similar to those considered acceptable by clinicians in an international survey. The liberal threshold was selected because maintaining hemoglobin levels above 10 g per deciliter may improve brain oxygenation. The restrictive threshold reflects the standard of care for critically ill patients. Patients received leukoreduced red cells, one unit at a time, when the specified hemoglobin threshold was met. Additional units were transfused when hemoglobin levels measured as part of routine care met the specified threshold. In both treatment groups, we aimed to transfuse red cells within 3 hours after the threshold was reached.”
The researchers also noted the blood transfusion strategy study’s primary outcome as, “unfavorable outcome (yes or no) at 6 months as assessed with the Glasgow Outcome Scale–Extended (GOS-E)” while secondary outcomes included, “mortality and scores on the Functional Independence Measure (FIM), the EuroQol visual analogue scale and EuroQol five-dimension, the Quality of Life after Brain Injury (Qolibri) scale, and the nine-item Patient Health Questionnaire (PHQ-9).”
The blood transfusion strategy study found that, “249 of 364 patients (68.4 percent) in the liberal-strategy group had an unfavorable outcome, as compared with 263 of 358 (73.5 percent) in the restrictive-strategy group (adjusted absolute difference, restrictive strategy vs. liberal strategy, 5.4 percentage points; 95 percent confidence interval [CI], −2.9 to 13.7). The overall relative risk of an unfavorable outcome in the liberal group as compared with the restrictive group was 0.93 (95 percent CI, 0.83 to 1.04), with findings consistent across groups of patients with the worst, intermediate, and best predicted prognoses and across prespecified subgroups; the results of sensitivity analyses were similar.” Additionally, the researchers explained that, “Mortality at 6 months was 26.8 percent in the liberal-strategy group and 26.3 percent in the restrictive-strategy group (hazard ratio for death, 1.01; 95 percent CI, 0.76 to 1.35)…Among patients who received a transfusion of red cells, 6 of 365 patients (1.6 percent) in the liberal-strategy group and 1 of 141 (0.7 percent) in the restrictive-strategy group had a reaction to the transfusion None of the reactions were severe.”
The authors concluded by stating that, “[o]ur trial was designed to assess the superiority of a liberal transfusion strategy at reducing unfavorable neurologic outcomes at 6 months. Although several patient-reported outcomes suggest potentially better results with a liberal strategy, firm conclusions may not be drawn. The trial was not designed to assess the noninferiority of a more restrictive transfusion strategy, so the possibility of harm with such a strategy cannot be excluded. In this international, randomized trial, a liberal transfusion strategy did not decrease the risk of an unfavorable neurologic outcome at 6 months as measured with the GOS-E in critically ill patients with traumatic brain injury.”
They acknowledged multiple limitations of the blood transfusion strategy study noting, “[b]y recruiting solely patients with anemia, we selected a population with more severe traumatic brain injury, which may explain the higher-than-expected baseline risk of an unfavorable outcome; We also observed imbalances between the groups at baseline, which included some prognostic variables of the TBI-IMPACT score that may have suggested a better prognosis at baseline in the liberal-strategy group; and it was not possible to mask the treatment assignments from the clinical team.”
Citation: Turgeon, A.F., Fergusson, D.A., Clayton, L., et al. “Liberal of restrictive transfusion strategy in patients with traumatic brain injury.” NEJM. 2024.