Does transfusion practice affect mortality in critically ill patients?

In 4,470 critically ill patients, we examined the impact of transfusion practice on mortality rates. As compared with survivors, patients who died in intensive care units (ICU) had lower hemoglobin values (95 +/- 26 versus 104 +/- 23 g/L, p < 0.0001) and were transfused red cells more frequently (42.6% versus 28.0%, p < 0.0001). In patients with cardiac disease, there was a trend toward an increased mortality when hemoglobin values were < 95 g/L (55% versus 42%, p = 0.09) as compared with anemic patients with other diagnoses. Patients with anemia, a high APACHE II score (> 20), and a cardiac diagnosis had a significantly lower mortality rate when given 1 to 3 or 4 to 6 units of allogeneic red cells (55% [no transfusions] versus 35% [1 to 3 units] or 32% [4 to 6 units], respectively, p = 0.01). Adjusted odds ratio (OR) predicting survival were 0.61 (95% CI; 0.37 to 1.00, p = 0.026) after 1 to 3 units and 0.49 (95% CI; 0.23 to 1.03, p = 0.03) after 4 to 6 units compared with nontransfused anemic patients. In the subgroup with cardiac disease, increasing hemoglobin values in anemic patients was associated with improved survival (OR = 0.80 for each 10 g/L increase, p = 0.012), We conclude that anemia increases the risk of death in critically ill patients with cardiac disease. Blood transfusions appear to decrease this risk.