A colleague in Argentina reports a case of an apparent transfusion reaction in a group A Rh positive (CDe/CDe) 85 year old male who had a history of Acute Lymphoblastic Leukemia (ALL) which was in remission since 1994. The patient experienced what appeared to be a hemolytic transfusion reaction in close temporal association with the transfusion of apparently crossmatch compatible donor RBCs; the reaction was manifest by the development of hemoglobinemia and icteric serum. Of interest, however, is that the patient was also being treated with Ciprofloxacine for about three days leading up to the onset of the transfusion-associated hemoglobinemia and icteric serum. The patient had been transfused without incident on several occasions in the past for his ALL. Following remission of his ALL he was no longer transfusion dependent.

Approximately a week prior to the onset of hemoglobinemia and icterus he required surgical repair of an aortic aneurysm. During the surgery he received several units of 'compatible' RBCs, at which time he had no demonstrable unexpected red cell alloantibodies by GEL-antiglobulin methodology. He tolerated those transfusions without incident. Seven days after his aortic aneurysm repair surgery a new blood sample revealed the presence of unexpected antibodies and a panel work up demonstrated anti-K and anti-c. A direct antiglobulin test was not done at the time. Because he needed additional RBC support, a c-negative donor RBC unit was selected for transfusion; the blood bank lab did not have anti-K typing serum. The patient was transfused with a crossmatch compatible (by GEL-AHG) c-negative donor RBC unit, following which he developed hemoglobinemia, icteric serum, and anuria within hours of the transfusion's completion. He required dialysis until his renal function recovered. A post-transfusion reaction blood sample showed a positive direct antiglobulin test with polyspecific IgG-C3d AHG, and an ether eluate showed only a warm panagglutinin. Neither anti-c nor anti-K were detected in the eluate. Both the anti-K and the anti-c were present in the patient's plasma and the autoantibody was only detectable in the eluate.These tests were performed 24 to 48 hours after the implicated transfusion.

The questions that the Argentina colleague wishes to be discussed include the following:

• What might be a recommended donor RBC selection strategy for this patient?
• If the Argentina colleague's hospital cannot find any compatible K-negative and c-negative RBC units, what would be the next best choice?
• This patient appears to have experienced a hemolytic event, but was the hemolysis more likely triggered by a drug that the patient was taking, by a warm autoantibody, or by an as yet undetermined red cell alloantibody reacting with transfused donor red cells?

Please share your thoughts and the scientific basis for them.

Please submit comments to the e-Network Forum.

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