Blood transfusions are frequently required in the critically ill, with recent studies showing that about one-third of intensive care unit (ICU) patients receive a transfusion at some point during their ICU stay.(1,2) The rationale behind blood transfusion is the restoration of oxygen delivery, and transfusions are often given to ensure some reserve if further bleeding occurs. However, blood transfusions are not without risks, and in recent years, practitioners have shifted from a fairly liberal approach to a much more conservative one.

Risks of Blood Transfusions

Basically, three major types of transfusion risks exist:

1. Transmission of microorganisms, such as hepatitis B and C viruses and human immunodeficiency virus (HIV), has been a widely publicized risk of blood transfusion. Modern donor screening and laboratory testing of blood have reduced the risks of transfusion-transmitted infection, and the development of pathogen-inactivated red blood cells may further reduce the small risk that remains.(3)
3. Transfusion-related immunomodulation (TRIM) that may increase the risk of infections and may increase hospital mortality.(4)
4. Human errors (wrong type and crossmatch, incorrect patient identification, etc.) causing hemolytic reactions.

Problems Associated with Red Blood Cell Storage

For practical reasons, blood is often stored for up to 42 days before use. In the United States, some 38,000 units of blood are needed for transfusion every day (http://www.aabb.org). To have such amounts of fresh blood on tap, not to mention the problems of crossmatch and typing, would involve a complete overhaul of current blood banking policy. Demand and supply of blood are not constant, and the logistics required to ensure adequate availability, even if storage times were reduced to just a week, would be considerable.(5)

The significance of blood storage-induced lesions is, thus, a subject of considerable debate and has powered many intense discussions in the literature and at recent international meetings, including the Society of Critical Care Medicine’s (SCCM) 33rd Critical Care Congress in February 2004. What is clear is that red blood cells are altered with prolonged storage, and many of the “storage lesions” worsen with increased length of storage.6 What is not clear, however, is what impact these changes can have on the ability of the red blood cells to release oxygen, and what impact this may have on patient outcomes. Stored red blood cells (RBCs) certainly develop membrane damage with echinocytic change leading to increased membrane rigidity and reduced deformability,(7) making them potentially less able to penetrate the microcirculation to release oxygen where it is most needed. Stored RBCs also have reduced levels of 2,3-diphosphoglycerate (2,3-DPG), although again the potential effects of this on oxygen transport remain controversial.

Importantly, until relatively recently, all stored blood contained RBCs, white blood cells (WBCs), and platelets. White blood cells deteriorate more rapidly than RBCs, and as they do so, they release various bioactive agents, including histamine, procoagulants, and cytokines, which then accumulate in the waiting units of blood.(8) Stored blood also has higher levels of citrate, potassium ions, sodium ions, and glucose, and is more acidic. Finally, stored RBCs will have a shorter life span, and hence shorter potential oxygen carriage capacity, than freshly collected RBCs.

Taking all these factors at face value, it would seem to make sense to prefer fresh blood; however, the clinical consequences of transfusing stored versus fresh blood have not been ascertained. Marik and Sibbald(9) showed that in critically ill septic patients receiving red blood cells that had been stored for more than 15 days, the gastric intramucosal pH (pHi) consistently decreased following transfusion, suggesting a negative effect on splanchnic oxygenation. However, this study raised skepticism, in view of the vagaries surrounding the pHi measurements. Moreover, these results were not confirmed in a more recent study where transfusion of stored red blood cells had no clinically significant adverse effects on gastric tonometry or global indexes of tissue oxygenation in euvolemic, hemodynamically stable critically ill patients.(10)

Whatever the potential effects of storage on oxygen delivery, clinical studies have usually, although not always,(11,12) shown no difference in outcome related to the age of transfused RBCs.(13,14) Leal-Noval et al.(15) suggested an increase in nosocomial pneumonia with transfusions of blood stored for more than 28 days, and Zallen et al.(12) reported that stored RBCs given within the first six hours of resuscitation from trauma were associated with a higher incidence of multiple organ failure. The interpretation of the studies conducted in this area is fraught with difficulty due to variations in blood management strategies between hospitals and countries. Particularly with the increasingly widespread use of pre-storage leukoreduction, further study is needed to clarify the effects of use of stored blood on morbidity and mortality.

So, is leukoreduction — the filtering or apheresis technique whereby almost all the white blood cells are removed from the blood units — the answer to the problems of transfusion-related immunomodulation and to some of the problems of red blood cell storage? Universal leukoreduction has been introduced in approximately the last five years in many countries including the United Kingdom, Canada, France, and the United States. In many other countries, leukoreduction is widely, if not universally, used.

The potential benefits of leukodepleted blood are clear. Approximately 1% to 5% of patients develop febrile and other inflammatory reactions to RBC transfusion largely due to the WBC fraction, which would be virtually eliminated with leukoreduction. Human leukocyte antigens (HLA) alloimmunization and platelet refractoriness could also be reduced,(16, 17) and the transmission of WBC-associated pathogens such as cytomegalovirus would also be prevented. In a retrospective study on data gathered before and after the introduction of a leukoreduction program in Canada, Hebert et al. reported that patients transfused with leukoreduced blood had fewer febrile episodes, received fewer antibiotics, and had reduced mortality rates.(4) Hence leukoreduction, even though it is associated with substantial additional costs, may improve survival.

But again, studies have given conflicting results; although some have reported beneficial effects on various outcomes in various groups of patients,(4,18,19) others have found no benefits.(20-23) Interestingly, while an observational study on transfusions before widespread leukoreduction was introduced reported increased mortality in transfused patients,(1) a similar study where most blood given was leukodepleted showed no relation of transfusion with mortality.(2) Can these differences be ascribed to the benefits of leukoreduction? Leukoreduction almost certainly has beneficial effects in some patients, but whether the universal application of this process warrants the financial input is still unclear.(24)