Granulocyte Transfusions

Abstract

To the Editor: We wish to provide some additional information not covered in the review of granulocyte transfusions by Dr. Higby and Dr. Burnett.’ First, we and other investigators of granulocyte transfusions are still very preoccupied in research attempting to demonstrate whether granulocyte transfusions work. Most of our efforts have been stimulated by the vulnerability of previously published studies of granulocyte transfusion therapy to critical analysis2 and the fact that most infected neutropenic patients will respond to appropriate antimicrobial therapy alone.3’4 Criticisms against previous controlled studies of therapeutic granubocyte transfusions3 include: (1 ) the marked variability in the patients studied with respect to age, reversibility of underlying disease, the degree of neutropenia, and the types of infections and causative organisms; (2) the variation in the types and duration of antimicrobial therapy; and (3) the application of different criteria for evaluating response to therapy. Moreover, none of these studies included more than 20 patients in either the transfused or control group. The analysis of relatively few patients with very complex underlying illnesses and infections may only allow slight differences in prognostic factors between the transfused and control groups to bias the results. In most of these studies, the positive effect of transfused granulocytes was frequently short-lived, and a significant number of transfused patients died shortly after they were released from the study. For all of the above persons, a multicentered, controlled trial both therapeutic and prophylactic granulocyte transfusions sponsored by the National Heart, Lung and Blood Institute is now being carried out in order to accumulate the large number of patients needed to assure better comparability of transfused and control patients in terms of important prognostic factors and to provide data indicating which patients, if any, might benefit from granulocyte transfusions. This ongoing study should be completed soon. Second, the data on granulocyte procurement provided by Dr. Higby and Dr. Burnett may not be representative of most transfusion centers. Indeed, in a survey of 10 centers located in Southern California,’ we found that the mean cost of a granulocyte transfusion to the patient or his medical insurer is $468.00 (range, $280.00 -$600.00). The Haemonetics Model 30 blood processor is used almost exclusively by all the centers for collection of granulocytes by intermittent flow centrifugation. Only 4 of the centers stimulate the donors with both corticosteroids and hydroxyethyl starch, and a mean granulocyte yield per procedure of 1.3 x l0’#{176} (range,