A colleague at an academic medical center in Oregon wonders who at other institutions is responsible for determining what the ABO/Rh should be when selecting blood products for transfusions that are intended for ABO or Rh mismatched BMT patients? She also wonders what ABO/Rh selection rules are being used in the field, and if they are similar or different to the ones used at her institution. The inquiring colleague's local guideline (MS Word) is attached. Their medical director wants to use a computer program that can be accessed by doctors, nurses and blood bankers, so that when orders are written or processed for patients undergoing allogeneic BMTs, the likelihood of errors in donor ABO/Rh selection will be minimized.

ADDENDA Oct. 6, 2003

The following responses have been received.

1. A colleague in Virginia reports that on page 556 in the AABB Technical Manual (edition 14) there is a Table (#25-5) entitled Transfusion Support for Patients Undergoing ABO-Mismatched Allogeneic HPC Transplantation which contains information germane to this discussion. According to the Virginian's experience it is common practice to specify a 1st and 2nd choice for platelet selection for support of patients receiving incompatible transplants. If inventory forces the Virginia colleague's hospital to give plasma incompatible platelet products, they give products that have an anti-A/B titer <200 or that are washed.

2. Neil Blumberg, MD, Professor of Pathology & Laboratory Medicine and Director, Transfusion Medicine/Blood Bank University of Rochester Medical Center (attribution used with permission) reports that at his institution they have a very different approach than that proposed by the colleagues in Oregon. According to Dr. Blumberg, their practices are based upon:

• "Evidence from randomized trials that repetitive use of ABO-mismatched platelets leads to a higher incidence of refractoriness (about two-fold higher with the strategy of giving whatever is available regardless of ABO type)
  
  
• Evidence from cohort studies that creating large quantities of ABO immune complexes in patients by infusing ABO-incompatible antigen and/or antibody leads to a higher incidence of complications and earlier death in several clinical settings, including BMT."

Dr. Blumberg adds that in a recent unpublished comparison with literature data for patients with acute leukemia, use of ABO-unselected platelets is associated with a 15-20% rate of clinically significant hemorrhage. He comments: "Our rate in patients receiving only ABO-identical platelets is less than 5%. While not a randomized or even controlled study, a 3 to 4 fold difference is unlikely to be explained by confounding or bias. Thus our strategy is to avoid giving BMT patients any incompatible antigen or antibody, even if this means giving plasma-depleted (washed in our case) group O platelets and red cells instead of donor or recipient group identical. In addition, the concept that one should infuse FFP of the major mismatched type to 'neutralize' isoagglutinins is a policy that never had any significant clinical data to support it, as far as I know. Why would creating immune complexes actually benefit a patient? Even if one succeeded in minimizing hemolysis by this strategy (not a proven effect), one would be destroying platelets, activating white cells and doing all sorts of other things to the patient's immune system that are likely to be clinically unfavorable." He concludes stating: "One can avoid hemolysis in almost all cases by not transfusing cells or antibodies that are incompatible with the donor or recipient types. This appears a safer and more scientifically logical approach".

ADDENDA Oct. 10, 2003

3. A colleague at a large transplant center in South Australia strongly endorses Dr. Neil Blumberg's comments in posting #2 above. The Australian says: "We have followed an approach using a table from McCullough et al. which was to be found on page 540 in the 12th ed. of the AABB technical manual or in Vox Sanguinis 1994; 67(S3): 35-42. We are very particular when it comes to platelet selection for mismatched transplants and try to ensure that the ABO type required is provided by our local supplier. The protocols to be used should be set by the transfusion laboratory in consultation with the clinical unit concerned. We have also seen protocols which recommend using only group O red cells for any mismatch situation, with platelet and plasma support basically following McCullough's table. We also have about 50% of our platelet inventory as plasma-depleted buffy coat pools which reduces the amount of incompatible plasma that may need to be infused. Again Dr. Blumberg's approach of group O red cells and platelets [plasma-depleted] will avoid any incompatible antibody or antigen. Recently, the adverse effect of immune complex formation has also been a suggested hypothesis for reduced patient survival (Benjamin RJ et al, Transfusion, 1999, 39, 179-187)."

ADDENDA Dec. 15, 2003

4. The Editor wishes to draw colleagues' attention to a new AABB Slide Set entitled "Transfusion Considerations for ABO Incompatible Hematopoietic Progenitor Cell (HPC) Transplants", which is now available to AABB members.

ADDENDA June 13, 2004

5. A colleague in Southern California reports that recently a few patients have been seen at her facility with a history of a bone marrow transplant or and organ transplant. Two of these patients had a discrepancy between the forward and reverse ABO grouping results. After inquiring with the physicians caring for these patients, it was determined that both patients with ABO grouping discrepancies had undergone bone marrow transplantation at another facility. This information was then shared with their reference laboratory which then gave the hospital transfusion service recommendations as to what ABO group the blood products intended for transfusion should be. At the moment, the inquiring colleague's hospital does not have a policy for the selection of the appropriate ABO group of blood products intended for transfusion of bone marrow transplant patients. Instead, they follow the recommendations of their blood supplier and request leukocyte reduced irradiated blood products for these patients. The inquiring colleague would like to know how other facilities select products for transplant patients that have an ABO group discrepancy due to a bone marrow transplant.

6. At an academic center in Washington State, all patients who receive ABO mismatched hematopoietic stem cell transplants receive group O red blood cells regardless of ABO type beginning on the day of the transplant to reduce confusion. Platelets and plasma are selected in order to avoid any plasma incompatibility, to avoid antibodies against A or B antigens of donor or recipient and to avoid the need to volume reduce platelet components that have incompatible plasma. The responding transfusion medicine physician begins to monitor patient anti-A and anti-B titers and a percentage estimate of rbc's of recipient versus donor origin at 30 days post transplant. When titers are <4, monitoring of titers is performed weekly. When they have at least two consecutive ABO titers that are negative post-transplant and the patient is demonstrated to have a stable graft, without circulating recipient red cells (unless the recipient started as an O), the red blood cell and plasma ABO group selected is changed to match the ABO group of the hematopoietic stem cell donor. The responding physician adds that they send a transfusion summary along with the patient to be given to the physician caring for them that outlines the patient's transfusion history, including the need for matched platelet donors, any occurrence of reactions, the need for CMV negative or component modification because of reactions, recommendations for ABO group selection of blood products, and if they haven't been switched yet, when they would recommend the switch in ABO group. Patients are kept on irradiated components, but are removed from routinely receiving leukoreduced blood products after transplant, unless there is a compelling medical reason, such as repeated febrile transfusion reactions. In their experience, their patients are so immunosuppressed following transplant that they do not become alloimmunized or immune platelet refractory unless they were already refractory prior to transplant (in which case the leukoreduction seems to be of no benefit).

ADDENDA June 16, 2004

7. Dr. Neil Blumberg (attribution used with permission), adding to his previous response (#2, above), states that their approach now is to avoid giving any anti-A and/or anti-B antibody to recipients who have the corresponding antigen still present on red cells or elsewhere in the body, and to avoid infusing A and/or B antigen to anybody who has still has the corresponding antibody detectable. He admits that this is easier to say than to do. For example, he reports that they will give someone who has completely converted from group A to group O washed group O RBCs and platelets because they do not want to be giving substantial amounts of anti-A to an individual whose endothelial cells, hepatocytes and fibroblasts are covered in A antigen and who also have large amounts of circulating soluble A antigen. Dr. Blumberg admits that he does not have conclusive proof that this is necessary. However, he does have reason to think it might be a good idea. High molecular immune complexes probably compromise both immunologic and hemostatic function. While Dr. Blumberg agrees that most transfusion medicine colleagues might say this is an overly meticulous and obsessive approach, it also means that he does not have to deal with problems due to hemolysis caused by infused ABO antibodies or increased red cell and platelet consumption in the recipient due to the ABO system.

ADDENDA July 7, 2006

8. A physician at the University Health Network/Princess Margaret Hospital in Toronto, Ontario, Canada reports that as of June 2006, their policy with regards to transfusion support for patients undergoing ABO-mismatched allogeneic hematopoietic stem cell transplantation is to continue transfusing products that are compatible to both the recipient and donor indefinitely. They are, however, in the process of revising this policy such that once there is a clear forward and reverse group consistent with the donor (i.e. no mixed field reactions), transfusion support would be compatible only with the donor blood group. They acknowledge having found only one survey study looking at this topic showing heterogeneity in practice: Raimondi R et al. ABO-incompatible bone marrow transplantation: a GITMO survey of current practice in Italy and comparison with the literature. Bone Marrow Transplantation 2000;25:507-12. The inquiring physician wonders how other centers handle this issue. Are patients actively monitored for this change? Do any centers arbitrarily change to the donor group and if so, at what time post-transplant?

Please submit comments to the e-Network Forum.

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