CIPROFLOXACIN VERSUS TRIMETHOPRIM-SULFAMETHOXAZOLE FOR PROPHYLAXIS OFBACTERIAL-INFECTIONS IN BONE-MARROW TRANSPLANT RECIPIENTS - A RANDOMIZED, CONTROLLED TRIAL

Purpose: To compare the efficacy and safety of ciprofloxacin (CIP) and trimethoprim/sulfamethoxazole (TMS) for the prevention of bacteriol i nfections in patients who received bone marrow transplantation (BMT) f or the treatment of solid and hematopoietic neoplasms. Patients and Me thods: Adult inpatients about to undergo BMT for lymphoma, leukemia, o r solid tumors were enrolled onto a prospective, randomized, double-bl inded, controlled trial that compared CIP (750 mg orally twice per day ) with TMS (160 mg trimethoprim and 800 mg sulfamethoxazole orally twi ce per day). Subjects were stratified before randomization according t o tumor and BMT type. Prophylaxis was begun within 96 hours of initiat ion of the BMT preparative regimen and continued until the onset of fe ver, signs or symptoms of infection, serious adverse effects, or recov ery of the absolute granulocyte count (AGC) to greater than or equal t o 400/mu L. Results: Seventy-five CIP recipients and 71 TMS recipients were assessable for efficacy. No difference was noted between the two groups in occurrence of fever during neutropenia, time to onset of fi rst fever, or overall infection rates. Ten bacteremias occurred in CIP recipients versus six in TMS recipients (P = .43). Ten episodes of Cl ostridium difficile enterocolitis occurred in TMS recipients versus no episodes in CIP recipients (P = .001). Four infections caused by gram -negative bacilli, including one bacteremia, occurred in TMS recipient s versus none in CIP recipients (P = .06). No differences were noted i n the incidence of rash or organ toxicity. TMS recipients had ranger d urations of granulocytopenia at AGC levels less than or equal to 500/m u L and less than or equal to 100/mu L than did CIP recipients (P = .0 8 for both comparisons). Mean peak and trough serum levels of CIP decr eased significantly between weeks 1 and 2 of prophylaxis. Conclusion: CIP and TMS were equally safe and effective in the prevention of bacte rial infections in BMT patients when the overall infection rate was us ed as the principal end point. TMS prophylaxis was associated with a h igher incidence of C difficile enterocolitis and infections caused by gram-negative bacilli, as well as a trend toward prolongation of granu locytopenia. (C) 1995 by American Society of Clinical Oncology.