3.3 Effect of red blood cell transfusion on outcomes
3.3.1 Effect of red blood cell transfusion
Mortality
In patients undergoing cardiac surgery, seven studies found that RBC transfusion was a significant predictor of short-term mortality.61–67 The odds of death increased with increasing units of blood transfused;62,66 however, a study of patients undergoing thoracic aortic surgery found no relationship.68 Three studies investigated longer term mortality63,69,70 and demonstrated that RBC transfusion was a significant predictor of 6-month,70 1-year63 and 5-year mortality.69
In patients undergoing noncardiac surgery, studies were less consistent. Five studies found that RBC transfusion was significantly associated with a higher risk of mortality;71–75 this association was dose dependent in two studies.71, 75 In contrast, three other studies found that RBC transfusion was not a significant predictor of mortality.76–78
Morbidity
In patients undergoing cardiac surgery, infection was the most common morbid outcome observed; it included wound infection, sepsis and pneumonia.61,64–66,79–84 All 10 studies found that RBC transfusion was a significant predictor of infection, and the odds of infection increased with increasing numbers of units of blood transfused. RBC transfusion was a significant predictor, in dose-dependent fashion, of cardiac,61,65,66,85,86 renal,61,66 respiratory61 and neurologic morbidities.61
In patients undergoing noncardiac surgery, the most common morbidity outcome investigated was also infection. Five studies found that RBC transfusion was a significant predictor for development of infection, including wound infections, sepsis and pneumonia.71,73,76,87,88 In patients undergoing major vascular surgery, RBC transfusion predicted development of venous thromboembolism.56 No studies reported cardiac, renal or neurological morbidity.