There has been an increased interest in the use of therapeutic granulocyte transfusion in recent years because premedication of donors with granulocyte colony stimulating factors produces much higher doses of granulocytes for transfusion. Other factors which influence the outcome of transfusion include the types of infection being treated, the likelihood of recipient marrow recovery and recipient alloimmunization. This review provides a historical perspective on these issues.There has been a recent renewal of interest in the use of granulocyte transfusion, largely because of the observation that premedication of donors with granulocyte colony stimulating factor (G-CSF) permits the collection of large numbers of granulocytes resulting in higher post transfusion granulocyte increments, which are sometimes sustained for a few days post transfusion [1,2]. This 'resurrection' follows a period of more than 15 years in which granulocyte transfusions were used very sparingly. In the following, some of the reasons for this fluctuation over time will be addressed.
A brief history of the use of granulocyte transfusionGranulocyte transfusions were first utilized in the early 1960s in an era characterized by a high incidence of refractory infections with gram negative bacteria, due in part to the absence of effective antibiotics and the long periods of neutropenia related to the lack of effective therapies for acute leukemia [3]. In addition, technology was not available to efficiently separate granulocytes from the upper layer of red blood cells after centrifugation. In response to this problem, investigators at the National Cancer Institute studied the transfusion of leukocytes collected from patients with chronic myelogenous leukemia (CML), initially by simple centrifugation [4], and subsequently with the use of a continuous flow blood cell separator collaboratively developed by the NCI and the IBM Corporation [5].Two important principles were established by these early studies, both of which were somewhat ignored in subsequent decades. First, there was a direct relationship between the dose of leukocytes administered and the post transfusion increments with a suggestion that improved response rates were also associated with higher doses [4]. Secondly, increments were lower and the incidence of transfusion reactions was increased in recipients who were alloimmunized, as documented by the presence of leukoagglutinating antibodies. Because immature myeloid progenitors were obtained from CML donors, sustained levels of circulating granulocytes were commonly seen for a number of days after single transfusions owing to continued differentiation of the myeloid precursors. Interestingly, it was also shown that the leukocyte alkaline phosphatase levels were markedly increased in the granulocytes tested ex vivo after circulation in the infected host, whereas they were characteristically decreased in the CML donors [6].Although the CML transfusions were well tolerated and demonstrably of benefit, there was increasing reluctance to u...