Receiving a transfusion was associated with higher risk of ICU-acquired infection and severe hypoxemia among the study cohort of French ICU patients with sepsis.

FINDINGS:

Red blood cell transfusion did not appear to affect overall mortality in critically ill patients with sepsis, a study found. 

To estimate the effect of one or more transfusions within a day on three major outcomes (mortality, ICU-acquired infections, and severe hypoxemia) at day 30, researchers conducted modeling based on a database including patients with sepsis at admission, excluding patients with hemorrhagic shock at admission. Patients had been admitted to one of 23 participating French ICUs between April 1, 1998, and Sept. 1, 2014, for severe sepsis or septic shock. Results appeared online Sept. 12 in Critical Care Medicine.

RESULTS:

In the overall cohort of 6,016 septic patients, transfusion had no significant association with mortality rates (hazard ratio [HR], 1.07; 95% CI, 0.88 to 1.30; P=0.52). However, transfusion was associated with increases in ICU-acquired infections (HR, 2.77; 95% CI, 2.33 to 3.28; P<0.01) and severe hypoxemia (HR, 1.29; 95% CI, 1.14 to 1.47; P<0.01).

Patients were divided into subgroups by hematocrit levels during their ICU stay: low (26%; interquartile range [IQR], 24 to 28), medium (30.5%; IQR, 29 to 32), high (36%; IQR, 33 to 38), and very high (44%; IQR, 41 to 47). Because of the lack of transfusion for patients with high and very high hematocrit subgroups, the subgroup analyses could only be conducted in the low and medium hematocrit subgroups. Transfusion of at least one red blood cell concentrate was associated with a lower risk of death for the low hematocrit subgroup (HR, 0.72; 95% CI, 0.55 to 0.95; P=0.02) but not for the medium hematocrit subgroup (HR, 0.92; 95% CI, 0.65 to 1.28; P=0.62).

The study's finding of a protective effect on mortality from transfusions in the low hematocrit subgroup differs from similar research not focused on sepsis patients but supports the most recent research in critically ill sepsis patients, the authors noted.

“These results highlight that the lowering of the transfusion threshold might have overcome the transfusion risks and perhaps become too restrictive. Optimizing the transfusion's risk-benefit ratio seemed rather difficult and might not be achieved only thanks to a universal transfusion threshold,” they wrote.