Cytomegalovirus (CMV) infection still has a high mortality rate especially if one develops pneumonia.1,2 Although studies have shown the relationship between leukocytes in blood products and the risk of transfusion-acquired infection, we suggest that the risk of granulocyte transfusion–acquired CMV infection in patients with cancer appears to be low. Prevention of CMV infection requires screening the donors for CMV antibody or leukoreduction by filtration, which obviously cannot be done for granulocyte transfusion.3 4Furthermore, the prevalence of CMV antibody in our donor population is high, and using donors that are CMV seronegative would therefore remove 70% to 80% of available granulocyte donors.
We retrospectively reviewed 100 cases of cancer patients who received CMV–unscreened-granulocyte transfusions from January 1995 through April 2001 with the following diagnosis: acute myelogenous leukemia, 41; chronic lymphocytic leukemia, 10; chronic myelogenous leukemia, 8; multiple myeloma, 1; refractory anemia with excess blast, 1; acute promyelocytic leukemia, 1; acute lymphoblastic leukemia, 21; T-cell lymphoma, 11; MDS, 4; granulocytic sarcoma, 1; and B-cell lymphoma, 1. Our study showed that only 4 (4%) patients developed CMV infection. Three of these patients had concomitant fungal pneumonia, and 1 had bacterial pneumonia. All 4 patients were CMV seropositive prior to the granulocyte transfusions. Thus recommending serologic screening for cytomegalovirus in granulocyte donors seems unwarranted.
A previous study in our institution, involving 34 patients who received granulocyte colony-stimulating factor (G-CSF)–stimulated granulocyte transfusions for bacterial, fungal, or both infections clearly showed that the granulocyte transfusions were effective in 16 patients (47%).5 Therefore clinicians must individually weigh the risks and benefits obtained from granulocyte transfusions when confronted with cases of severe bacterial or fungal infections resistant to antimicrobial or antifungal agents alone.
We believe that CMV infection in these patients could be due to primary reactivation of latent CMV or perhaps secondary to infection with another CMV strain. Therefore, screening of potential granulocyte donors for CMV antibody may not provide additional benefit to fungal- and bacteria-infected severely granulocytopenic patients with cancer. A more detailed investigation of each individual case needs to be done.
