Red blood cell transfusions worsen the outcomes even in critically ill patients undergoing a restrictive transfusion strategy

context and objective: anemia and blood transfusions are common in intensive care. this study aimed to evaluate epidemiology and outcomes among critically ill patients under a restrictive transfusion strategy. design and setting: prospective observational cohort study in an intensive care unit (icu) at a tertiary hospital. methods: all adults admitted to the icu over a one-year period who remained there for more than 72 hours were included, except those with acute coronary syndrome, ischemic stroke, acute hemorrhage, prior transfusion, pregnant women and jehovah's witnesses. the restrictive strategy consisted of transfusion indicated when hemoglobin levels were less than or equal to 7.0 g/dl. results: the study enrolled 167 patients; the acute physiology and chronic health evaluation ii (apache ii) score was 28.9 ± 6.5. the baseline hemoglobin level was 10.6 ± 2.2 g/dl and on day 28, it was 8.2 ± 1.3 g/dl (p < 0.001). transfusions were administered to 35% of the patients. in the transfusion group, 61.1% did not survive, versus 48.6% in the non-transfusion group (p = 0.03). transfusion was an independent risk factor for mortality (p = 0.011; odds ratio, or = 2.67; 95% confidence interval, ci = 1.25 to 5.69). icu stay and hospital stay were longer in the transfusion group: 20.0 (3.0-83.0) versus 8.0 (3.0-63.0) days (p < 0,001); and 24.0 (3.0-140.0) versus 14.0 (3.0-80.0) days (p = 0.002), respectively. conclusions: in critically ill patients, there was a reduction in hemoglobin with increasing length of icu stay. moreover, transfusion was associated with worse prognoses.