A colleague in Ohio reports that many area hospitals transfuse platelets seemingly without regard to the recipients' ABO group, so that group A platelets are transfused to group O recipients, group O platelets are transfused to group A recipients, etc, etc. She was wondering what references support this practice, since she and her co-workers have had trouble finding evidence-based references on this matter. She wonders if colleagues have documented adverse patient outcomes following the transfusion of ABO incompatible platelets.

The following responses have been received.

1. Editor's note: Colleagues might find the following links to be germane to the discussion stemming from the above question:

• The practice of transfusing out-of-type platelets 
• Reducing supernatant plasma of pooled platelets before administration to ABO-incompatible adult recipients
• Washed and volume-reduced Plateletpheresis units
• Procedure for platelet donor ABO antibody titers (MS Word file)
• Procedure for selection of platelet products (MS Word file)
• High Titre Anti-A/B Testing of Donors within the National Blood Service (PDF file)

2. Dr. Neil Blumberg and Dr. Joanna M. Heal (attribution used with permission) offer the following opinions which will be appearing in print in 2004 in an upcoming issue of Blood Reviews, entitled "Optimizing Platelet Transfusion Therapy":

"Two randomized trials from the era before leukoreduction of platelet transfusions demonstrated that giving ABO identical platelets reduced the rate of HLA alloimmunization, and of platelet transfusion refractoriness in those patients receiving repeated transfusions. The rate of refractoriness was decreased five fold when ABO identical platelets were given instead of ABO major mismatched platelets (Carr, R et al Br J Haematol. 1990 Jul;75(3)) and two fold when given instead of randomly selected unmatched platelets (Heal, JM et al Eur J Haematol. 1993 Feb;50(2)). The deleterious effect of ABO mismatching is cumulative with increasing transfusion number, so patients likely to receive many transfusions will particularly benefit from receiving ABO identical platelets (Heal, JM et al Ann Hematol. 1993 Jun;66(6)). There is a possible effect of ABO matching of platelet transfusions on survival in acute leukemia (Heal, JM et al Am J Hematol. 1994 Feb;45(2)) and survival after cardiac surgery (Blumberg N and Heal, JM Transfusion. 2002 Nov;42(11)), but these studies were small or observational. In most trials of prophylactic platelet transfusions (see following section on transfusion threshold), 15-20% of patients with acute leukemia have major bleeding episodes. In our patients given only ABO identical platelet transfusions the prevalence of major bleeding has been less than 5%. More than 70% of patients had no clinically evident bleeding of any sort (unpublished data, n=43 patients). We are confident these results are not due to changed clinical practices such as use of antifibrinolytics. It seems possible that infusion of incompatible ABO antigen or antibody may contribute to bleeding through interference with inflammatory and/or hemostatic function.

There is good reason on theoretical grounds to believe that infusing large amounts of ABO incompatible antigen or isoagglutinin, and the ensuing creation of high molecular weight immune complexes of ABO antibody and soluble ABO antigen, might have deleterious effects on immunologic and even hemostatic functions (Heal, JM and Blumberg, N Transfusion. 1999 Nov-Dec;39). There is also the rare, but occasionally life threatening hemolytic anemia that can occur with infusion of incompatible plasma in platelet transfusions. In our own institution, our practice is to identify patients who will need repeated transfusions of platelets and administer only leukoreduced, ABO identical platelets to these patients. If ABO identical platelets are not available, we make an effort to provide plasma-depleted ABO antigen "compatible" platelets such as a washed group O platelet concentrate. In patients requiring only single platelet transfusions we attempt to give ABO identical, but when none are available in urgent situations, we try to give the least dangerous ABO mismatch that is on hand. For example, group A and B platelets are safer for a group AB recipient than group O platelets due to the absence of high titer anti-A,B. The only route to the ideal goal of ABO identical platelets for all patients is to either manufacture many more platelets than are needed (with increased outdating and cost) or extensions in the storage period of platelets so that larger inventories can be maintained."

ADDENDA Nov. 13, 2003

3. Mark Fung, MD PhD, Medical Director of Blood Bank/Transfusion Service at the Fletcher Allen Health Care/ University of Vermont reports that he recently completed a blood bank fellowship at the Institute for Transfusion Medicine in Pittsburgh, where he conducted an informal survey of transfusion services regarding their policies on the use of ABO-incompatible platelets. He reports having found the following:

• Institution A
  Volume reduction for all ABO-incompatible apheresis platelets
• Institution B
  Transfusion of up to 350 ml of incompatible plasma per day.  Transfusion of more than 350 ml per day required approval from pathology resident or attending on call.
• Institution C
  Transfusion of 300 ml of incompatible plasma per day. Transfusion of more than 300 ml per day required volume reduction of product.
• Institution D
  Transfusion of up to 500 ml of incompatible plasma per day.  Transfusion of more than 500 ml per day required volume reduction of product.
• Institution E
  Screening titer (1:50 dilution) performed on all potential ABO-incompatible apheresis platelet transfusions. If greater than 1:50 then labeled "high titer" and used only for compatible transfusions. Otherwise if less than 1:50 then labeled "low titer" and used.
• Institutions F and G
  No ABO restriction, only Rh phenotype compatibility and use of RhIgG to prevent Rh alloimmunization.

Dr. Fung concludes that there is no national standard regarding the use of ABO incompatible platelets and that he hopes to participate in helping to bring some standardization to this issue.

Please submit comments to the e-Network Forum.

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