Generous Blood Transfusions Not Beneficial for Some With Burst Brain Aneurysms

For people with aneurysmal subarachnoid hemorrhage and anemia, there was no significant neurological benefit from getting red blood cell transfusions more liberally, the SAHARA randomized trial showed.

Between a liberal transfusion strategy (mandatory transfusion at a hemoglobin cutoff of ≤10 g/dL) and a restrictive strategy (optional transfusion at a hemoglobin level of ≤8 g/dL), an unfavorable neurological outcome was similarly likely at 12 months (33.5% vs 37.7%, RR 0.88, 95% CI 0.72-1.09).

Administered during the critical care period for people hospitalized for an aneurysmal subarachnoid hemorrhage, liberal and restrictive strategies were also associated with no different rates of adverse events and other secondary outcomes:

• Functional independence at 12 months (per Functional Independence Measure scale): 82.8 vs 79.8 points
• Quality of life (QoL; per EuroQol five-dimension, five-level utility index): 0.5 in both groups
• Health status or QoL (per visual analog scale): 52.1 vs 50 points
• Mortality at 12 months: 27.2% vs 27.1% (RR 0.99, 95% CI 0.77-1.28)

Researchers led by Shane English, MD, of Ottawa Hospital and University of Ottawa, Ontario, reported the SAHARA findings in the New England Journal of Medicine.

The best threshold for blood transfusions remains controversial in critically ill patients — existing guidelines only offering “vague recommendations to maintain hemoglobin levels within the normal range or to transfuse in the case of anemia,” according to English’s team.

While a liberal transfusion strategy might improve oxygen delivery and reduce brain tissue hypoxia, it has the potential downsides of increasing complications and mortality.

In this summer’s HEMOTION trial report, compared to an even stricter restrictive approach (transfusion at a hemoglobin level of ≤7 g/dL), a SAHARA-type liberal blood transfusion strategy failed to reduce the risk of an unfavorable neurologic outcome at 6 months when applied to critically ill patients with traumatic brain injury and anemia.

Soon afterward, however, the TRAIN trial results came out favoring a liberal strategy (hemoglobin cutoff

The SAHARA authors pointed out important differences among these studies, namely the choice of scale for neurologic function, patient populations included, and thresholds set to define liberal versus restrictive blood transfusion.

English and colleagues reported that in their trial, the liberal transfusion arm had gotten a median 2 units of red cells versus 0 units in the restrictive group. Median pretransfusion hemoglobin was 9.6 g/dL and 7.6 g/dL, respectively, and median time from hemoglobin measurement to transfusion was 6.1 hours vs 4.4 hours.

“Although the findings from our trial do not show a significant difference between the liberal and restrictive strategies, the possibility of smaller differences in benefit or harm that may be meaningful to patients and families is not excluded. Further meta-analyses of randomized trials may provide additional insight into appropriate practice,” the investigators wrote.

“In addition, refining patient-centered outcome measures to capture functional improvements is an important consideration for future studies in this population, especially when mortality benefit is unlikely,” they added.

The SAHARA authors had conducted an open-label trial with blinded outcome assessment. Seeking critically ill adults with acute aneurysmal subarachnoid hemorrhage and anemia (hemoglobin 10.0 g/dL or lower) in the first 10 days after admission, they enrolled 742 people at 23 centers in Canada, Australia, and the U.S. This figure includes the initial 60 people enrolled during the pilot phase of the study.

Study participants had a mean age of 59.4 years and were 82% women — to be expected, since younger women are disproportionately affected by subarachnoid hemorrhage from intracranial aneurysm rupture.

The ruptured aneurysms in SAHARA were largely in the anterior circulation (87.2%) and mostly secured using the endovascular approach (60.5%). Patients were randomized on median day 3 after hospitalization.

Study interventions lasted a median 17 days in the liberal group and 16 days in the restrictive group, according to the researchers.

The primary endpoint, an unfavorable neurological outcome, was defined as a modified Rankin scale score of 4 or higher on a scale from 0 to 6.

Duration of mechanical ventilation, length of stay in the intensive care unit, and length of hospital stay all came out similar between the two study groups.