A considerable literature has accumulated over the past decade indicating that leukocytes present in allogeneic cellular blood components, intended for transfusion, are associated with adverse effects to the recipient. These include the development of febrile transfusion reactions, graft-versus-host disease, alloimmunization to leukocyte antigens, and the immunomodulatory effects that might influence the prognosis of patients with a malignancy. Moreover, it has become evident that such leukocytes may be the vector of infectious agents such as CMV, HTLV-I/II, and EBV as well as other viruses. An interesting development that has occurred coincidentally in transfusion medicine is that agencies responsible for the provision of blood products are being designated manufacturers of biologicals. The trend among manufacturers of biologicals is to try to produce pure products to provide for the specific therapeutic need of patients. Thus, with the realization that allogeneic leukocytes and their products may have adverse biologic activities, there is increasing pressure from various sources for the reduction of the leukocyte content in allogeneic blood components to minimize the occurrence of their adverse effects. Although the threshold leukocyte number below which these effects might no longer occur is still to be determined, a 2 to 3 log10 leukocyte reduction, provided by the currently available commercial leukocyte filters, has been shown to reduce the frequency of many of such reactions. On the other hand, the immunosuppressive effects produced by the infusion of allogeneic leukocytes might be beneficial for some patients, ie, for the maintenance of kidney allografts, in possibly reducing the relapse rate in patients with inflammatory bowel diseases, and in ameliorating recurrent spontaneous abortion. Moreover, therapeutic granulocyte transfusions may be of benefit in certain well- defined categories of patients infected with bacteria, yeast, and/or fungi. These include neonates with bacterial sepsis associated with bone marrow failure as well as severely neutropenic leukemic patients with an infection unresponsive to appropriate and specific antibiotic therapy. Many of the results obtained with the use of leukocyte depletion filters are tantalizing, but the actual clinical benefit of leukodepletion has not been established in most instances, because much of the available data are retrospective or from uncontrolled clinical trials. Moreover, issues of cost-effectiveness and quality control have not been considered adequately. Properly designed prospective clinical trials are essential to provide data with which to answer such questions and also to help define the optimal conditions required for the preparation of blood components ultimately destined for clinical use. Only with the availability of such data can sound decisions be made about the clinical value of leukodepletion.