A Reductionistic Approach to Aged Blood

The changes that take place in blood components during storage have long been suspected as a culprit for part of the deleterious clinical effects attributed to blood transfusion. The case against “aged” blood can be made on two fronts: the clinical (often observational) evidence linking the advanced age of blood with adverse outcomes and the laboratory (experimental) evidence showing the changes in biology, chemistry, physics, and rheology as possible mechanisms of action. In this issue of ANESTHESIOLOGY, Vlaar et al. provide direct mechanistic evidence on the effect of storage of blood on lung injury and attempt to pinpoint the element(s) of blood that are adversely affected by storage and responsible for the observed harm using an animal model. The effective “shelf-life” of donated blood is affected by storage condition and preservative and additive solutions. For years, the US Food and Drug Administration has specified a 42-day expiration date for properly processed erythrocyte units stored in refrigerator, and any erythrocyte units stored for less than this period are deemed suitable for transfusion. However, the list of studies linking extended storage of blood (still within the 42-day permissible period) with unfavorable outcomes (e.g., morbidity and mortality) in various patient populations, including trauma, cardiac surgery, and critically ill patients, has steadily been growing (see Refs. 4–11 in the article of Vlaar et al.). Nonetheless, not all studies support this hypothesis, and a recent review of 24 studies concluded that the relationship between the age of transfused blood and the patients’ outcomes cannot be determined based on available clinical evidence, with the possible exception of massively transfused trauma patients suggestive of a possible dose-dependent effect in addition to the aged blood. Available studies are too heterogeneous to allow effective meta-analysis of the results, and the few available controlled trials often suffer from limited sample size and vaguely defined, overlapping arms. A number of large randomized trials have been designed to address these shortcomings. Notably, the Red Cell Storage Duration Study focusing on patients undergoing complex cardiac surgery,* the Age of Blood Evaluation trial in the resuscitation of critically ill patients,† and the Age of Red Blood Cells in Premature Infants Study‡ are currently ongoing. These trials will hopefully provide the definitive clinical evidence on the effect of age of blood on patients’ outcomes. Parallel to the ongoing clinical investigations, preclinical studies have provided invaluable insight into the physiology of the storage effect. Several reversible and irreversible changes that take place in blood during storage, such as decrease of adenosine triposphate, pH, and 2,3-diphosphoglycerate; modification of proteins and lipids; release of potassium, hemoglobin, and other intracellular components, and membrane and cytoskeletal alteration (collectively known as storage lesion), are well documented, and many others are being characterized. However, the significance of these changes in terms of causing clinically evident harm to the recipients of stored blood is uncertain, and further research is needed. To this end, the study by Vlaar et al. provides an interesting experimental insight into the link between blood storage and a relatively common and serious complication of blood: transfusion-related acute lung injury (TRALI). TRALI is the leading cause of transfusion-related death and is characterized by acute-onset lung injury within 6 h of receiving any allogeneic blood components, particularly plasma or plasma-containing products. Initial observations indicated that the pathogenesis of TRALI was antibodymediated, with donor’s antibodies against the recipient’s human leukocyte antigen class I and human neutrophil antigens or human leukocyte antigen class II antigens on monocytes, resulting in activation and sequestration of granulocytes in lung and endothelial damage. However, subsequent studies showed that these antibodies are neither necessary nor sufficient for TRALI to happen because in some cases, TRALI develops despite the absence of any detectable related antibodies and, moreover, many recipients of blood components containing high titers Accepted for publication February 17, 2010. The authors are not supported by, nor maintain any financial interest in, any commercial activity that may be associated with the topic of this article.