The development of clinical practice guidelines for RBC transfusion has been challenged by a limited availability of high-quanlity evidence to support practice recommendations. There is general agreement that RBC transfusion is typically not indicated for hemoglobin (Hgb) levels of >10 g/dL and that transfusion of RBCs should be considered when Hb is <7 to 8 g/dL depending on patient characteristics. The decision to transfuse RBCs should be based on a clinical assessment of the patient that weighs the risks associated with transfusion aganist the anticipated benefit. As more studies addressing RBC transfusion become available, it becomes increasely clear that liberal transfusion strategies are not necessarily associated with superior outcomes and may expose patients to unnecessary risks.
The most recently published guidelines from the AABB (formerly the American Association of Blood Bank) are based on a systematic review of randomized, controlled trials evaluating transfusion thresholds. These guidelines recommend adhering to a restrictive transfusion stratety and consider transfusion when Hb is 7 to 8 g/dL in hospitalized, stable patients. This strong recommendation is based on high-quality evidence from clinical trials comparing outcomes in liberal versus restrictive transfusion strategies in this patient population. A restrictive transfusion strategy is also recommended for patients with preexisting cardiovascular disease. In this population, transfusion should be considered when Hb levels are <8 g/dL or for symptoms such as chest pain, orthostatic hypotension, tachycardia unresponsive to fluid resuscitation, or congestive heart failure. This weak recommendation is based on moderate-quality evidence due to limited clinical trial data directly addressing this population of patients. Additional clinical practice guidelines exist that specify Hb targets for critical care patients with conditions including sepsis, ischemic stroke, and acute coronary syndrome.
RBC transfusion is indicated in patients who are actively bleeding and should be based on clinical assessment of the patient in addition to laboratory testing. Much remains to be learned about the optimal resuscitation of the bleeding patient. However, a recent study examining transfusion in patients with active upper gastrointestinal bleeding showed superior outcomes in patients treated with a restrictive transfusion strategy (<7 g/dL).
The Physiologic Response to Anemia
The initial response to anemia is a shift in the oxygen dissociation curve to the right as modulated by an increase in production of 2,3-DPG in RBCs. This shift allows for the unloading of oxygen to the tissues at higher partial pressures of oxygen, ensuring adequate oxygen delivery despite the reduction in RBC mass.
As anemia progresses, the cardiac output will increase by an increase in the heart rate to preserve the delivery of oxygen in the setting of decreased oxygen content. As RBC mass is reduced in anemia, the viscosity of the blood decreases. This reduction in viscosity leads to an increase in regional blood flow at the tissue and organ level, driving up local perfusion area and pressures leading to increased oxygen extraction. While a change in viscosity may be the trigger for increased regional blood flow, there has been suggestion that local blood vessel dilatation may be mediated by the release of nitric oxide (NO) from the RBCs. In order for these mechanisms to work properly, the patient must be at or near a euvolemic state. In considering these regulatory mechanisms, it is important to understand that the transfusion of RBCs will incease viscosity by adding stored RBCs that may not have the same vasoactive capabilities of native RBCs. As such, a transfusion of RBCs may inhibit compensatory mechanisms for low oxygen states, without significiantly increasing oxygen delivery.
There is evidence that low levels of Hb can be tolerated in healthy subjects. Hematocrits of 10% to 20% have been achieved in experimental studies using normovolemic hemodilution without untoward effects. Weiskopf and colleagues studied patients who underwent isovolemic reduction of Hb to 7, 6, and 5 g/dL. No evidence of reduced oxygen delivery was detected at any of the tested values of Hb; however, there was a subtle reversible reduction in reaction time and impaired immediate and delayed memory observed at Hb below 6 g/dL. An important source of data regarding the impact of anemia on surgical outcome comes from studies of Jehovah's Witness patients. Carson has demonstrated that the risk of death in these pateints at Hb between 7 and 8 g/dL is low. However, the odds of death increase by 2.5 for each gram decrease in Hb below 8 g/dL. The mortality is very high at Hb levels below 5 g/dL. It should be noted that these data are from patients who refuse all RBC transfusions. There is time to intervene between a low Hb and resulting morbidity or mortality in most patients.