In 72 episodes of suspected or proven Hickman-catheter-associated infection occurring in 59 patients with various hematological disorders, patients were assigned to treatment with either vancomycin or teicoplanin in a randomized nonblinded prospective study. Of 60 episodes evaluable for response, 28 were treated with vancomycin and 32 were treated with teicoplanin. Sixteen infective episodes were microbiologically documented in the vancomycin group, and twenty-one were microbiologically documented in the teicoplanin group. Microbiologically and clinically documented infections treated with vancomycin had an 80% response rate, compared with a 69% response rate for those treated with teicoplanin (P = 0.316). Adverse events occurred in nine (25%) of the episodes in the vancomycin group, compared with three (8%) in the teicoplanin group (P = 0.044). Teicoplanin may provide an effective alternative to vancomycin in the treatment of Hickman-catheter-associated infection in patients with hematological malignancies.