A member wrote that his hospital blood bank does not have a sufficient blood inventory to always find a donor unit suitable for a patient who has an alloantibody. Therefore, his hospital blood bank must rely on the community donor center to 'find' RBC units that lack the appropriate alloantigen when an alloimmunized patient needs transfusion. The inquiring member's blood bank verifies the antigen typing results to assure that the units truly lack the antigen that corresponds to a patient's unexpected alloantibody. Apparently some time in the past, the community donor center made an error in red cell antigen typing a donor unit that was intended for a patient with alloantibodies. However, no regulation requires the confirmation of an RBC antigen typing (other than for ABO and D), and the inquiring member is now rethinking his position. Apparently this policy forces his hospital blood bank to keep a larger inventory of rare antisera than they would otherwise need, if only they would trust the antigen typing results of the community donor center. Furthermore, if a specific antiserum is not in stock, there is the potential for delay of transfusion until the specific antiserum is obtained or approval to vary from the policy is obtained from the Medical Director. The inquiring member asked the e-network what other members do regarding 'confirming' the results of antigen typings that have been done by their community donor center. In other words, do you trust their RBC antigen typing results?

To which the following responses were received:

1. Several members provided input previously when a similar question was raised about Kell typing.

2. A blood banker in Toronto states that the cost benefit of retyping units for RBC antigens that are not ABO or Rh that have been previously typed by a reputable blood center is not warranted. In cases of newly detected or known alloantibodies having potential clinical significance in possible blood transfusion recipients, the requirements are to perform an antiglobulin crossmatch on each of the selected donor units. Although erroneous antigen typings of donor units coming from reputable blood banks would be extremely rare, they can happen. However, the antiglobulin crossmatch will detect most of these errors thus, reducing risk to the patient to essentially zero. Therefore, the cost to maintain a substantial library of antisera in a small facility to retype these units is not cost-effective and provides very little extra protection to the blood recipient. It would be interesting to know how many times this blood transfusion service has found a discrepancy? Also, in the one instance that is mentioned, what happened to the recipient? Was there a transfusion reaction, and if so, how severe was it?

3. A blood banker in Los Angeles said that at his facility, they would not detect an error in RBC antigen typing if the alloantibody was no longer detectable in the patient's serum. This responding member read the anecdotes described previously for Kell (see #1 above) and commented that we always remember the exceptions and quickly forget the norm. He does not know how it is possible to draw a firm conclusion about a need to confirm every RBC antigen typing done by the donor center without studying the testing error rate, along with the rate of patient antibodies below detectable levels by AHG crossmatch. To those individuals who feel that their transfusion service is ultimately responsible, the responding member asks if they confirm the testing done at the donor center for HIV, HTLV, Hepatitis B, syphilis, CMV? Do they confirm donor selection? Do they confirm residual WBC counts on leukoreduced units? Do they irradiate irradiated products again? Do they confirm platelet counts on platelet units? Factor VIII and fibrinogen levels on Cryos? The point is, transfusion services are receiving licensed products from licensed laboratories that the transfusion services should have qualified as part of their supplier qualifications process in their quality program. Finally, the severity of the consequences of an ABO error do warrant confirming the accuracy of the blood product label.

4. A blood banker in Denver reported that her facility re-confirms antigen typings on units from the Blood Center because many of those units are selected based on computer searches of historical data from the donor. It seems to the responding member that a simple solution to this problem would be to establish a written policy between blood center and hospital that would authorize the blood center to perform the actual serologic testing of the units just prior to shipment, giving the hospital a written report of that testing along with the shipped units. This would not be so different from the way many hospitals utilize a central crossmatching lab in a blood center. The responding member's hospital occasionally accepts the crossmatch results from the reference lab at the blood center without actually repeating internally anything but the ABO/Rh confirmation. This would happen in the case of small children, for example, who have difficult work ups and for whom time to transfusion is of the essence and the drawing of additional samples for testing is out of the question.

5. A blood banker in Texas said they they had not been retyping the units that they ordered with special antigen typing from their reference lab. They generally have to order only for patients with multiple antibodies. Now that they have read these discussions and see the number of mistakes that have been discovered, they are considering changing their policy. They have most of the antisera available. They have never had a transfusion reaction with any of the units that they have received that were antigen typed, but now they don't feel as comfortable as they did before this discussion.

Editor's note: At the USC Kenneth Norris Cancer Hospital, we have come to the opposite conclusion of response #5. In our experience, the rarity of a donor center antigen typing errors, plus the likelihood that such an error would be discovered in an antiglobulin crossmatch suggest that it would be safe to trust the donor center antigen typing. We do realize that we might miss detecting an erroneous antigen typing if the patient no longer has the alloantibody detectable in their serum, however, this combination of events must be uncommon. If any e-Network Forum member has had such an experience, please share it.

6. A blood banker at a Health System (network of 8 hospitals, 1700 beds) re-confirms antigen typings done by the blood center, unless circumstances do not allow.

Please submit comments to the e-Network Forum.

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