Twenty-six oncology patients, 25 of whom received bone marrow transplants, were enrolled in a prospective, randomized, double-blinded, placebo-controlled trial assessing the efficacy of ciprofloxacin, 750 mg p.o. b.i.d., for preventing bacterial infections during prolonged neutropenia. Treatment was begun within 48 hr of initiation of chemotherapy and continued until the absolute granulocyte count recovered to greater than or equal to 500/microliters, or until the onset of fever (greater than or equal to 38.3 degrees C). Seven evaluable subjects received ciprofloxacin, and 11 received placebo. Risk factors for infection were comparable in both groups. Fever occurred in all study subjects, but onset was delayed in ciprofloxacin recipients (median = 6 days after the fall of the absolute granulocyte count to less than or equal to 500/microliters vs. 3 days for placebo recipients, P = 0.01). No clinically or microbiologically documented infections occurred in ciprofloxacin recipients vs. 10 infections in placebo recipients (5 bacteremias, 4 skin/soft tissue infections, 1 urinary tract infection, P = 0.0003). Ciprofloxacin recipients required fewer days of therapeutic antimicrobials (median: 28 antibiotic-days vs. 49, P0.02). The bioavailability of ciprofloxacin appeared comparable to that found in previously published studies of normal volunteers and patients not receiving chemotherapy. Adverse effects and colonization by ciprofloxacin-resistant microorganisms were monitored, but the sample sizes were too small to permit meaningful conclusions about these safety parameters. Ciprofloxacin appears to be effective for preventing bacterial infections in neutropenic patients. Additional trials are needed to establish the optimal dose of ciprofloxacin and to compare its safety and efficacy with those of currently used prophylactic regimens.