WE have all learned that 2,3-diphosphoglycerate (2,3-DPG) progressively decreases during storage of erythrocytes,1resulting in an increase of the affinity of hemoglobin for oxygen.2As a consequence, these erythrocytes seem to be less capable of releasing large amounts of oxygen to the tissue.3Therefore, it has been proposed to transfuse fresh rather than stored erythrocytes. For example, in a model of sepsis, only fresh (and not stored) erythrocytes could restore oxygen consumption.4Also, in septic patients with increased lactate levels, oxygen consumption did not increase after erythrocyte transfusions,5and gastric mucosal pH even decreased after the transfusion of erythrocytes older than 15 days.5Therefore, it is surprising that Weiskopf et al. 6in this issue of Anesthesiology found that transfusion of autologous erythrocytes stored for approximately 3 weeks is as efficacious as the transfusion of fresh (3–4 h) erythrocytes in reversing anemia induced cognitive dysfunction.
This finding is particularly surprising because the expected decrease in 2,3-DPG and the increase in the affinity of hemoglobin for oxygen was observed in the stored erythrocytes that reversed cognitive dysfunction.6Does this indicate that the retransfused stored erythrocytes had regained—at least partially—their 2,3-DPG levels at the time of testing? This is possible because Beutler and Wood7have shown that 1 h after transfusion, approximately 25–30% of prestorage 2,3-DPG was restored in donor erythrocytes. This does not necessarily mean that fresh and stored erythrocytes result in an equal increase of cerebral tissue oxygenation, which is the likely mechanism responsible for the reversal of the cognitive dysfunction after isovolemic hemodilution.8–10Perhaps the stored erythrocytes improved cerebral oxygenation sufficiently to reverse the cognitive dysfunction, whereas the fresh erythrocytes may have increased cerebral oxygenation to a greater extent. However, this did not translate into measurably better cognitive function. A more gradual increase of the hemoglobin concentration from 5 to 7 g/dl with intermittent neuropsychological testing might have resulted in a detectable difference in favor of fresh versus stored erythrocytes. In addition, it is important to note that the study subjects were healthy young volunteers. Storage changes of erythrocytes may have had less impact on tissue oxygenation in these individuals than in patients with underlying diseases such as sepsis or coronary artery disease. Last but not least, a recent study has challenged the common understanding that 2,3-DPG levels in stored erythrocytes are the key factor for oxygen off-loading capacity of transfused erythrocytes.11In this study, human erythrocytes stored for 2–3 weeks and containing almost no 2,3-DPG were equally efficacious in maintaining intestinal microvascular oxygen partial pressure as erythrocytes that were stored for 2–6 days. Only erythrocytes stored for 5–6 weeks were less efficacious.11Although undoubtedly there are differences between the oxygenation of the intestine and the brain, both studies indicate that 2,3-DPG levels of transfused erythrocytes may not play a major role with respect to their capacity for tissue oxygenation.
Apart from their contribution to the discussion of old versus fresh erythrocyte transfusions, Weiskopf et al. 8,9,12have opened the “window to the brain” with respect to monitoring the adequacy of cerebral oxygenation during acute anemia. Monitoring the effect of acute anemia on the cerebral function is essential because most experts would agree that erythrocyte transfusions are indicated to “treat or prevent imminent inadequate tissue oxygenation.”13Current monitoring assesses the heart for development of myocardial ischemia by electrocardiogram and transesophageal echocardiography. Also, the entire circulation can be evaluated by measuring mixed venous hemoglobin saturation in the presence of a pulmonary artery catheter and by calculating oxygen consumption using indirect calorimetry. With circulatory monitoring only, however, we have no direct access to the state of oxygenation and function of other organs. Monitoring the function of the brain in relation to the hemoglobin concentration is an important step toward physiologic transfusion triggers. However, cognitive testing with horizontal addition, digit symbol substitution, and memory tests6,8,9,12requires the cooperation of the patient and thus is unpractical during major operations or after trauma when patients normally are anesthetized. Anemia sensitive neurologic monitoring during general anesthesia is an area that requires further development,10,12,14–16 e.g. , analysis of evoked potentials aimed at central processing may in the future enable on-line monitoring of the adequacy of cerebral oxygenation.
The current study by Weiskopf et al. 6is therefore remarkable for several reasons. First, it is challenging the evolving way of thinking, that fresh erythrocytes are universally better than old erythrocytes. Second, it adds weight to the hypothesis that the 2,3-DPG level may not be the key factor determining the oxygen off-loading capacity of transfused erythrocytes. Finally, in conjunction with previous studies,8–10,12it suggests that on-line monitoring of the functional adequacy of cerebral oxygenation in relation to the hemoglobin concentration during acute anemia may be a valuable tool. Importantly, this study has established a very sensitive human model to monitor the early functional signs of oxygen supply–demand mismatch of the brain. This test can further be used to evaluate therapies that presumably increase oxygen delivery and tissue oxygenation such as hyperoxic ventilation9,10,17,18and artificial oxygen carriers.19
After developing the capacity to monitor the functional adequacy of the oxygenation of the heart, the circulation, and the brain during acute anemia, physiologic transfusion triggers will progressively replace arbitrary hemoglobin based transfusion triggers. This will render allogeneic erythrocyte transfusions more efficacious because physicians will be capable of using goal-directed erythrocyte transfusions.
* Department of Anesthesiology, University Hospital Lausanne, Lausanne, Switzerland. donat.spahn@chuv.ch