Therapeutic granulocyte transfusions for documented infections. A controlled trial in ninety-five infectious granulocytopenic episodes.

Patients with granulocytopenia (granulocyte count less than 0.5 x 10(9)/L) and a documented infection were randomized to receive or not to receive daily granulocyte transfusions in addition to antimicrobial therapy. Thirty-four of 47 control patients responded to therapy compared to 30 of 48 transfused patients (type 2 error, pneumonia, or a soft tissue infection, respective response rates for the control and transfused patients were 11 of 11 and 11 of 16 (Yates' corrected chi-squared test, p = 0.12). Response rates for patients with gram-negative septicemia were lower but were influenced by recovery of bone marrow function. Eleven of 12 control patients and seven of seven transfused patients with recovery of marrow function survived the gram-negative septicemia. In contrast, 12 of 24 control patients and 12 of 25 transfused patients survived gram-negative septicemia and persistent granulocytopenia (type 2 error p = 0.13). Two thirds of all fatal infections were associated with an underlying disease refractory to medical therapy. Therapeutic granulocyte transfusions had no substantial benefit over optimal antimicrobial therapy alone in managing infected patients with granulocytopenia.