Fresh thinking about fresh whole blood
J. Hess
DOI: https://doi.org/10.1111/j.1537-2995.2010.02854.x
2011-01-01
Transfusion
Abstract:T his issue of TRANSFUSION contains an article by Jobes and colleagues that investigates how long the platelets (PLTs) in cold-stored whole blood function in PLT function tests. However, they pose the question as “How long are the PLTs in fresh whole blood fresh?” It is a strangely worded question, and it does not allay concerns about the relevance of the in vitro data to clinical practice. Fresh, however defined, is not necessarily better than appropriately stored, and PLTs removed from the circulation cannot contribute to “integrated coagulation function.” Jobes and colleagues are from the Children’s Hospital of Philadelphia (CHOP) where for many years they have used “fresh whole blood” collected by the American Red Cross in their pediatric cardiac surgery program. They say they use this product because it provides superior hemostasis. The blood is collected from volunteers who meet all of the conventional donor criteria, is fully tested for infectious diseases, and is cooled to 6°C within 8 hours of collection as required by law. The whole blood is typically used within 2 to 5 days or made into leukoreduced red blood cells (RBCs) if not needed for the surgery or discarded. They report that in vitro PLT function is retained in cold-stored whole blood for at least 2 weeks. Others have found the same thing. But for 45 years, we have known that the PLTs in cold-stored whole blood do not stop the bleeding in acute leukemia; for 40 years, we have known that coldstored PLTs do not circulate; and for 8 years we have known why cold-stored PLTs are removed from the circulation. When the National Cancer Institute (NCI) began treating acute leukemia in the late 1950s, they rapidly discovered that bleeding and infection were the major consequences of myeloablative chemotherapy. The article by Gaydos and colleagues showing that bleeding was rare at PLT counts above 20 ¥ 10/L, and incidentally creating a transfusion trigger at the value that lasted for decades, came from this NCI group. PLTs from the cold-stored units of whole blood did not stop the bleeding in these patients. Treatment of an acute bleed required the rapid collection of several units of warm fresh whole blood and the separation of PLTs from the PLT-rich plasma removed from the tops of blood bottles. This was possible but logistically difficult in the days when the only infectious disease test was for syphilis. Tibor Greenwalt, chair of the National Academy of Science’s Committee on Blood Transfusion, and colleagues described this situation in a committee report in JAMA in 1966. Three years later, Murphy and Gardner showed that infused cold-stored PLTs did not persist in the circulation, and Murphy spent his career developing better methods of PLT separation, warm storage with agitation, and oxygen-permeable bags that enable current 5to 7-day PLT storage around the world. Finally, in 2003, Karin Hoffmeister and colleagues described the aggregation of glycoprotein Ib on the surface of cold-stored PLTs and their removal by galactose receptors on liver cells. So what is going on at CHOP? Clearly, they are exposing children to the added immunologic risks of nonleukoreduced whole blood for a potential coagulation benefit in which most of us do not believe. They justify this on the basis of a study their institution performed and published 18 years ago, which attempted to randomize infants and children undergoing open heart surgery to warm fresh whole blood, cold-stored fresh whole blood, and blood components. However, randomization did not work because the sickest and smallest children were frequently taken to surgery on short notice with available bank blood. Nevertheless, the study was evaluated as if it was randomized, and the relatively poorer outcomes in these sickest and smallest children made the use of banked blood in these children look like the least successful option. However, the study only achieved borderline significance and only that by failing to correct for multiple comparisons. Now that warm fresh whole blood is no longer available because of the need for more time-consuming infectious disease testing, they presumably view coldstored “fresh whole blood” as the next best available option. But even the cold-stored whole blood available today is stored longer because of the time required for nucleic acid testing than that used at the time of the study. Variations in the use of blood products in cardiac surgery are well-reported phenomena. Surgenor and colleagues described large differences in the use of blood in coronary artery bypass surgery in five US university hospitals in the late 1990s. Blood use in the most similar cohorts varied from 33% to 90% between the hospitals and was not related to patient mix or individual surgeon preference. The authors concluded, “We believe that the effects of the specific hospital on blood transfusion practice are the result of deeply ingrained institutional differences in training and hierarchical practices within the hospitals.” Andreasen and colleagues reported a similar difference of 30% to 62% transfusion rates among five hospitals performing bypass surgery in Denmark, and a TRANSFUSION 2011;51:5-7.