Should we reconsider triggers for red blood cell transfusion?

Very few randomized controlled trials on the benefits of red blood cell (RBC) transfusions in humans have been published. Consequently, most clinical practice guidelines remain based on expert opinion, animal studies and the limited human trials available. In the absence of definitive outcome studies, numerous theoretical arguments have been put forward either to support or to condone the classic transfusion threshold of 10 g/dL. However, the limited data available from randomized controlled trials suggest that a restrictive transfusion strategy (transfusion threshold between 7 and 8 g/dL) is associated with decreased transfusion requirements, that overall morbidity (including cardiac morbidity) and mortality, hemodynamic, pulmonary and oxygen transport variables are not different between restrictive and liberal transfusion strategies and, finally, that a restrictive transfusion strategy is not associated with increased adverse outcomes. In fact, a restrictive strategy may be associated with decreased adverse outcomes in younger and less sick critical care patients. The majority of existing guidelines conclude that transfusion is rarely indicated when the hemoglobin concentration is greater than 10 g/dL and is almost always indicated when it falls below a threshold of 6 g/dL in healthy, stable patients or more in older, sicker patients. In anesthetized patients, this threshold should be modulated by factors related to the dynamic nature of surgery, such as uncontrolled hemorrhage, coagulopathy, etc. Since transfusions are administered to correct inadequate oxygen delivery, whether global or regional, reliable monitors of tissue oxygenation will be required to study the benefits (or lack thereof) of RBC transfusions. The quest for a universal transfusion trigger, the holy grail of transfusion medicine, must be abandoned. All RBC transfusions must be tailored to the patient's needs, at the moment the need arises. In conclusion most published recommendations are appropriate but their conclusions are limited, as they are commensurate with existing knowledge. Reliable monitors to guide transfusion therapy and well conducted trials to determine optimal transfusion strategies are required.