Delayed hemolytic transfusion reaction and paroxysmal cold hemoglobinuria: an unusual association.

An 80-year-old female presented with melena and anemia due to bleeding from a benign gastric ulcer. Her blood group was O, D+. The serum contained anti-B and a weak anti-A (titer 2 at 18 degrees C). She was inadvertently transfused with approximately 3.5 units of group A red blood cells with no initial ill effects. One week later, the anti- A titer increased to 8 and the direct antiglobulin test (DAT) was weakly positive (IgG and C3d). The next day, intravascular hemolysis became evident. The DAT was still weakly positive and the serum contained a weak cold autoagglutinin, which did not correlate with the severity of the hemolysis. A Donath-Landsteiner test was performed and found to be strongly positive. The antibody showed P specificity, confirming a diagnosis of paroxysmal cold hemoglobinuria (PCH). Exchange transfusion was followed by rapid recovery even though the Donath-Landsteiner test remained positive for at least a month. The patient was well when last seen 11 months after presentation. It was thought that the original low titer of anti-A reflected compromised immune homeostasis in an elderly patient and that stimulation by incompatible blood in those circumstances resulted in a delayed hemolytic transfusion reaction that triggered, exacerbated, or was accompanied by an autoimmune response manifesting as PCH.