Liberal Transfusion Strategy No Better at Reducing Poor Neurologic Outcomes in Patients with TBI

A liberal transfusion strategy proved no better at reducing the risk of poor neurologic outcomes than a restrictive transfusion strategy in critically ill patients with severe traumatic brain injury (TBI) in a trial of more than 700 patients, according to the results of a randomized trial published recently in the New England Journal of Medicine.

Researchers found that 68.4% of patients who had a higher hemoglobin threshold for transfusion (liberal strategy) were considered to have poor neurologic outcomes at 6 months compared with 73.5% for those transfused at a lower hemoglobin threshold (restrictive strategy). The results come from the multicenter Hemoglobin Transfusion Threshold in Traumatic Brain Injury Optimization (HEMOTION).

While previous studies have examined restrictive versus liberal transfusion strategies for patients with TBI, those studies did not specifically examine neurologic functional outcomes in a critically ill population, the authors noted. Previous studies have also focused on mortality, and thus do not provide substantive guidance “for the care of patients with traumatic brain injury, for whom long-term neurologic function is the most important outcome.”

The patients included in this study were treated at trauma hospitals with specialized neurocritical care capacity. Patients had to have acute moderate or severe TBI based on a Glasgow Coma Scale (GCS) score of 3-12. Patients also had to have anemia, defined as hemoglobin (Hb) no more than 10 mg/dL. The intention-to-treat analysis included 369 patients randomized to a liberal transfusion strategy and 367 to a restrictive strategy.

The liberal transfusion strategy used a Hb threshold of 10 g/dL, while a threshold of 7 g/dL was used for the restrictive strategy. Patients received leukoreduced red blood cells (RBCs) one unit at a time when transfusion was triggered by the Hb thresholds. Median hemoglobin per ICU stay was 10.8 g/dL versus 8.8 g/dL for the liberal and restrictive strategies, respectively. Nearly all (98.9%) patients in the liberal strategy group received at least one unit compared with 38.4% of those in the restrictive strategy group.

The primary outcome was an unfavorable outcome (yes/no) at 6 months, as measured by the Glasgow Outcome Scale-Extended (GOS-E), which ranges from 1 (mortality) to 8 (full recovery to normal life). The researchers used a sliding dichotomy based on prognosis. For this, patients were categorized into three risk levels — worst, intermediate and best. Unfavorable neurologic outcomes were GOS-E scores less than or equal to 3 (worst), 4 (intermediate) and 5 (best).

Secondary outcomes included mortality and measures of function (motor and cognitive), quality of life and depression at 6 months. Mortality was comparable between the two groups at 26.8% (liberal strategy) versus 26.3% (restrictive strategy). The researchers noted that a liberal strategy was associated with better scores for some measures of motor function and quality of life, but that it was not possible to draw firm conclusions.