A colleague on the East Coast of the US reports that his hospital's main blood supplier currently provides only pre-storage leukoreduced blood. As a result, his hospital maintains an inventory of about 95% pre-storage leukoreduced units. Recently, his hospital's blood supplier offered to switch away from a universal leukoreduced blood product inventory and to provide a mix of leukoreduced and non-leukoreduced blood. The incentive to revert to a mix of products would be that the "direct costs" of leukoreduction would be taken off the price of the units that are provided to the hospital. The inquiring colleague wants to know if other blood bankers have been given similar offers, and if so, would they accept such an offer. He is somewhat reluctant to make the switch to a mix of leukoreduced and non-leukoreduced products, because he prefers a universal leukoreduced inventory, primarily to reduce the chance of a patient genuinely needing leukoreduced products accidentally receiving a unit of non-leukoreduced blood. For example, he worries about a patient such as a potential transplant patient who has not been identified to the Blood Bank "slipping through the cracks" and receiving non-leukoreduced blood products. He is also of the opinion that the rate of non-hemolytic febrile reactions is reduced when using leukoreduced blood products, as is the (remote) risk of "reverse TRALI". However, because cost is likely to be a consideration in the final decision, he is considering changing their inventory to about 20-25% leukoreduced and the rest non-leukoreduced. He comments that the randomized trial done at the Massachusetts General Hospital showed no real difference in giving leukoreduced or non-leukoreduced blood to patients lacking a clear indication for leukoreduced blood products.

The following responses have been received.

ADDENDA July 13, 2003

1. A medical director of a transfusion service in Texas prefers to have two inventories, one of leukocyte-reduced products and the other of non-leukoreduced products. In his practice, leukoreduced blood products are not indicated for 20-25% patients. He is of the opinion that the current literature does not make a strong case for universal leukoreduction (as reported in the Massachusetts General Hospital study). In his opinion universal leukoreduction and pathogen inactivation are driven by the industry.

ADDENDA July 14, 2003

2. A blood banker in Georgia reports that his institution has struggeled with the rising costs of blood products, and that they have had a lot of pressure from their administration to reduce expenditures in this area. His hospital has two major blood suppliers, and until just recently, one of them has only offered leukocyte-reduced products. His hospital's policy is to only use leukocyte-reduced products for patients who have a documented indication as determined by a physician.

3. A transfusion medicine physician in Detroit reports that they, too, have had an offer from their blood supplier of receiving a percentage of non-leukoreduced RBCs. Her hospital intends to take advantage of this, because they have never switched to an entirely leukoreduced RBC inventory. They are contracted with several blood suppliers, some of which continue to draw half or more of their RBCs as non-leukoreduced products, due to hospital customer demand in their respective areas. The Detroit physician reviewed published and local data available in the mid to late 1990s. Based on her review, her hospital has been unable to conclude that switching to a completely leukoreduced RBC inventory would be of significant advantage to patients outside the groups where clear medical indications exist for such products. Given the studies referred to by other discussants, she and her hospital still hold this opinion.

ADDENDA July 22, 2003

4. A transfusion medicine physician in New York believes those centers who have not adopted universal leukoreduction or who are going back to two inventories, have made, or are making decisions that in the long run will be regretted on the basis of patient care and cost. His hospital adopted leukoreduction for cardiac surgery patients in July 1998 and universal leukoreduction in July 2000. Their own analyses to date indicate that the decision led to reductions in morbidity, mortality and cost, as published by our own group (Amer. J. Clin. Path 118: 376, 2002) and in the much larger and more comprehensive studies recently published by the Canadians (JAMA 289:1950, 2003 and 289:1941, 2003). He comments that no one doubts that leukoreduction reduces the risks of febrile transfusion reactions, HLA alloimmunization and CMV transmission, all of which are worthwhile goals in all patients. Recent evidence from the responding colleague's own studies (Transfusion 43: 945, 2003) and a group in Finland suggests that leukoreduction reduces the risk of red cell alloimmunization as well, perhaps by about 30%. He adds that the single most compelling indication for leukoreduction in his opinion remains transfused surgical patients. Of the eight randomized trials examining the effect of leukoreduction on post-operative infection, which is the most common, costly and frequently fatal complication of surgery, six provide strong evidence for the efficacy of leukoreduction. He cautions that while it is premature to discuss his recent findings (data) in detail, his group is also in the process of re-evaluating the analyses in the literature that have argued that these existing studies are inconclusive or even negative. He believes that these analyses contain serious methodologic errors or have overstated the strength and validity of the evidence in the small number of negative trials. Based upon the evidence from their own studies and those in the literature, properly analyzed, the New Yorik physician is absolutely convinced that leukoreduction of transfusions reduces surgical morbidity and mortality, and saves money overall. Obviously it doesn't save money in the blood bank. He concludes that (verbatim) "Simply put, to assure patients with hematologic maligancies, hemoglobinopathies, and surgical transfusions that they receive only leukoreduced blood, it is necessary to inventory only leukoreduced red cells. In particular, those patients who receive larger amounts of blood and are most likely to benefit from leukoreduction in our studies, are those most likely to receive non-leukoreduced transfusions when the inventory is mixed. I would urge those who are planning to administer non-leukoreduced transfusions reconsider that policy based upon the data in the literature on short term benefits, and the long term benefits in reducing complications such as HLA alloimmunization, RBC alloimmunization and CMV infection that can come back to haunt us, and our patients, months or years to come."

ADDENDA July 24, 2003

5. On the subject of maintaining two inventories, one leukoreduced and the other not, in the opinion of Dr. Dennis Goldfinger of Cedars-Sinai Medical Center in Los Angeles (attribution used with permission), such an approach can be taken. However, colleagues will be interested to discover the rest of Dr. Goldfinger's opinion, in which he states (verbatim): "It is perfectly fine - just don't tell the patient what you are doing - that you are potentially alloimmunizing him or her, thereby making platelet transfusion more difficult in the event of future need; that you may be transmitting CMV infection (but who cares - certainly not the blood banker); that you may cause him or her to suffer a needless febrile transfusion reaction (some patients actually may enjoy the shaking chills followed by that nice warm feeling of fever) and finally that the transfusion may precipitate a TRALI reaction (only some of these are actually fatal). In fact, it seems that the first rule of transfusion medicine is don't ask the patient what he or she wants and never tell the patient that the cheap component that you select for transfusion may result in unnecessary harm - in other words, patient advocacy has no place in transfusion medicine and paternalism is alive and well in the world of blood banking."

ADDENDA July 25, 2003

6. A colleague who works at a blood center suggests that hospitals choosing to maintain a mixed inventory rather than a leukoreduced inventory may be overestimating the savings and underestimating the true cost of their decision. She is of the opinion that even if a few dollars are saved in the cost of each transfusion in the short run, this may be more than erased in overall costs to the patient (and the hospital), especially if the patient requires transfusion at another time. She also cautions that another factor should be considered when requesting a mixed inventory from local blood suppliers. Maintaining a duo-inventory for blood banks could result in more units outdating because the balance in supply of leukoreduced to non-reduced blood is difficult to maintain on a continual basis.

Hospitals in large urban areas who could purchase non-leukoreduced units from more than one blood supplier are at an advantage to those hospitals having only one blood supplier. Additionally, as blood centers experience an increase in outdates due directly to attempting to maintain a double inventory, they will eventually need to raise prices to cover the cost of the lost products. So then, where are the savings in the use of non-leukoreduced blood products in the end?

ADDENDA July 26, 2003

7. Editor's NOTE: e-Network colleagues might find the following two articles of interest, in response to the inquiring colleague's question:

• Blumberg, et al. WBC reduction of RBC transfusions is associated with a decreased incidence of RBC alloimmunization. Transfusion. 2003 Jul;43(7):945-52.
• Vamvakas EC. WBC-containing allogeneic blood transfusion and mortality: a meta-analysis of randomized controlled trials.Transfusion. 2003 Jul;43(7):963-73.

Please submit comments to the e-Network Forum.

Ira A. Shulman, MD
CBBS e-Network Forum Editor & Moderator