Febrile transfusion reactions and/or rigors with almost EVERY platelet transfusion

A blood bank physician reported that her hospital has been forced to use pooled platelets due to recent shortages of plateletpheresis products. She reports that the platelets used in these pools are manufactured by their community blood center from platelet-rich plasma derived from whole blood donations. The reason that she is sharing this information with the e-network is that her facility has been seeing febrile transfusion reactions and/or rigors with almost EVERY one of these transfusions despite different lots of leukocyte reduction filters being used. The blood banker also states that all of the patients experiencing these reactions have been oncology patients. The age of the platelets in each pool vary, but at least two of the platelet units in each pool are 4 days old, and each pool size is from 4-6 units. The pooled platelets are filtered by the inquiring blood banker's laboratory, using a POST-STORAGE FILTER system that is QC'ed regularly. She has no doubt that the reactions are cytokine- mediated, but she asks WHY NOW? HAVE THE FILTERS CHANGED? IS SOMETHING LEACHING? She wants to know if anyone else has seen this same problem or if any one has an idea as to what might be the cause.

The following replies were submitted in response to the above request.

1. 1One blood banker responded that in the June 23rd issue of the Lancet (The Lancet Volume 357, Issue 9273 23 June 2001 Pages 2023-2024) there was a letter discussing CD40 ligands and their implication in transfusion reactions. According to the authors, febrile responses are a common and sometimes life-threatening occurrence in patients receiving platelet transfusions. The cause is largely unknown. The authors reported that CD154 is present at 3000-7000 ng/L in the supernatant of nine preparations of human platelets used for transfusions and induces prostaglandin E2 synthesis in human fibroblasts. Since CD154 is a potent inducer of the cyclo-oxygenase-2 enzyme and prostaglandin E2, the main fever inducers, infusion of platelet preparations containing CD154 may incite febrile responses.
   
   
2. A blood banker from the United Kingdom suggested that if one intends to leukocyte-reduced pooled platelets, that the leukocyte reduction should be completed within 48 hours of the donation of blood from which platelets are derived. According to the responding blood banker, after 48 hours of room temperature storage, leukocytes release cytokines which result in febrile transfusion reactions. Furthermore, leukocyte depletion is ineffective once leukocytes begin to disintegrate.
   
   
3. A New York blood banker reported that his facility uses post-storage leukocyte reduced pooled whole blood-derived random donor platelets. Their reported reaction rate is about 1%. Perhaps another 5% or so of patients have very minor reactions that aren't being reported. If one is having nearly a 100% reaction rate (as reported by the inquiring blood banker), something is obviously wrong. According to the responding blood banker, the possibilities in descending order of likelihood are as follows: (1) The filters being used have a problem. Since the problem persists when different lots are used, if the reactions stop when using filters from a different manufacturer, that might shed light on the root cause. (2) Patients and staff have been intensely conditioned that apheresis platelets are a "superior" product and to not use random whole blood derived platelets. The responding blood banker claims having seen some patients who have been told at other medical centers that it is critical that they receive apheresis platelets. These patients are particularly anxious about this and often refuse the approach used at the New York blood banker's hospital. Anxiety over use of the pooled product might be inducing a high rate of reactions being called. (3) the transfusions are not ABO identical. It is the experience of the New York blood banker that ABO-mismatched platelets (even minor mismatches like O's to A's) have a higher rate of reactions, but not nearly 100%, as reported by the inquiring blood banker. The responding blood banker has no experience with mixing ABO types in the transfusion. If one is doing this as a matter of course, it is conceivable that routinely infusing immune complexes could cause a very high reaction rate, at least on theoretical grounds. One way of testing these hypotheses is to wash or otherwise plasma deplete the pooled platelets prior to their infusion. The responding blood banker would only do this if the reactions were exclusively mild febrile/inflammatory reactions. ABO-identical, washed platelet transfusions almost never cause febrile reactions in the responding blood banker's experience with thousands of transfusions over the years, even in patients with multiple previous reactions to ABO identical leukocyte reduced platelet transfusions.
   
   
4. Another blood banker suggested that if pooled platelets are not filtered prior to storage, then the accumulation of cytokines could be an explanation for transfusion reactions. He suggested using prestorage leukocyte reduced platelets (or platelets <48 hr old). He presumes that the blood bank laboratory evaluated the reactions and found no bacteria growing from the bag and the DAT was negative (since group O platelets given to a group A patient could cause symptoms). He thinks pools of prestorage leukocyte reduced platelets should be as effective as prestorage leukocyte reduced plateletpheresis.

ADDENDA Jan. 20, 2002

5. 5. A physician blood banker with extensive experience in blood product manufacturing commented that if febrile reactions are occurring with the transfusion of every pooled platelet transfusion, another possibility to consider is that there might be excessive time between pooling, filtration and transfusion. According to the responding physician, it is sometimes difficult for a hospital blood bank to complete the process of pooling platelets, filtering them and then having them transfused within the mandatory 4 hours, especially if the blood bank has not been doing this procedure routinely.

ADDENDA Jan. 23, 2002

6. A physician with experience on the AABB Standards Committee suggests that the e-network review the study by Kelley and Triulzi in Transfusion (2000) 40:439-442 that documents up to 11% FNHTRs with pooled platelets. The rate of reaction was directly proportional to how old the constituents of the pool were. The responding blood banker comments that one potential solution down the road is to pool and leukoreduce after completion of testing. The FDA might be receptive to this option once pathogen inactivation eliminates the risk of bacterial contamination, which is historically why they have not allowed pooling prestorage.

Submit comments to the e-Network Forum at enetworkforum@cbbsweb.org

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

W. Tait Stevens, MD
CBBS e-Network Forum Editor & Moderator

Elizabeth M. St. Lezin, MD
CBBS e-Network Forum Associate Editor & Moderator