Introduction: A 4-month-old male infant received a small bowel transplant. He was readmitted at 9 months of age (m.o.a.) with marrow suppression and severe hemolytic anemia. Serologic evaluation demonstrated both cold and warm autoantibodies as well as two alloantibodies (anti-K and -E). T-activation was also present. The patient was unresponsive to Rituxan and IVIG, so therapeutic plasma exchange (TPE) was initiated at 12 m.o.a. A regimen of immunosuppression, IVIG and TPE was mildly successful in abating hemolysis. A third alloantibody was then identified (anti-Jka) at 13 m.o.a., and the patient continued to hemolyze. This case report describes the investigation of severe hemolytic anemia associated with small bowel transplantation.

Methods: Spreadsheets were generated with hematocrit plotted against these variables: Rituxan, IVIG, TPE, RBC transfusions, and serologic findings. Both venous and buccal samples were DNA-typed for RBC antigens. A venous sample was evaluated for chimerism at two different time points after a discrepancy was observed between buccal and venous DNA typing.

Results: The patient’s condition declined precipitously at 9 m.o.a., and stabilized at 10 and 11 m.o.a., then worsened again. Graphs of the hematocrit did not correlate with any one variable. Slight improvement was observed post-TPE, but hemolysis was observed during the procedures. Blood priming was required for all TPE. A heated circuit was devised to avoid extracorporeal cold agglutination. Samples sent for DNA analysis of the RBC antigens showed donor/recipient chimerism in the venous blood. Two of three “alloantibodies” (anti-E and Jka) were probably of donor graft origin against pt RBC. The third antigen was absent in both patient and donor DNA, and the antibody (anti-K) was likely formed in response to transfusion by either pt or donor lymphocytes. There is also a broadly reactive autoantibody. Genetic analysis showed 10% chimerism at 10 m.o.a., followed by 2% chimerism at 13 m.o.a. A decrease in chimerism was qualitatively associated with some decrease in hemolysis.

Conclusion: No one scenario explains all of this patient’s symptoms. The presence of chimerism suggests generation of autoantibodies and production of two alloantibodies by passenger lymphocytes. Alternately, the patient’s own autoimmune response, or a combination of both processes may play a causative role. Reduction of circulating cold and warm autoantibodies and alloantibodies formed by passenger lymphocytes plus priming with alloantigen-negative, least incompatible RBCs may account for the slight improvement seen with TPE.

Disclosure: No relevant conflicts of interest to declare.

Author notes

*

Corresponding author

Sign in via your Institution