In recent years the clinical use of the antibiotic rifampin, or rifampicin, as it is called in Europe, has been undergoing a gradual broadening. Once reserved almost exclusively for tuberculosis, rifampin has been used increasingly for treatment of a wide array of infections due to gram-positive and gram-negative bacteria, Chlamydia, Legion-ella, and even some fungi. Unfortunately many of these new uses have evolved, without benefit of clinical trials, out of frustration in situations where other drugs have failed. This review attempts to bring together available basic scienti-fic and clinical information about the expanded role of rifampin for nontuberculous infections, to identify potential areas where the drug might make future contributions and particularly to identify areas where more investigation is needed. Rifampin has several unique features that set it apart from other antimicrobial agents. Its mechanism of action is based on its ability to bind to and inactivate the bacterial DNA-dependent RNA polymerase, a mode not shared by any other anti-infective agent in use today. This action usually produces a bactericidal effect at very low concen-trations against most gram-positive and many gram-negative organisms, including Mycobacterium tuberculosis and Mycobacterium leprae, as well as the various Chlamydia and Legionella species. However, slight modifications in the target enzyme alter rifampin binding and render organisms highly resistant to this antibacterial activity. These mutations occur with a frequency of 1l0-6-10-7 among naturally occurring cells and predictably emerge in vitro and in experimental infections, such as endocarditis, where large numbers of organisms are present. This rapid emergence of resistance has required combination therapy with rifampin and another antimicrobial agent for established infections. Because of its unique mechanism of action, resistance to rifam-pin is not associated with resistance to other anti-microbial agents, and thus cross-resistance with other antimicrobial agents is rarely observed. Rifampin's main attribute may be its unique pharmacokinetic properties. It is highly lipid soluble, a property that allows it to penetrate into various hidden recesses, including the central nervous system and abscess cavities. It also ap-pears to be one of few antimicrobial agents that is capable of killing bacteria within the protected environment of phagocytic cells. In addition, rifampin induces hepatic microsomal enzymes and accelerates its own metabolism and excretion. The expanded uses of rifampin have been slow to emerge over the 20 years since its initial de-velopment. Several factors probably have played a role in retarding the expanded use of rifampin. Early in the course of its development the drug was found to be highly active against M. tuberculosis. Its introduction markedly altered treatment of this disease and, when used in combina-tion with isoniazid, rifampin allowed short-course therapy for the first time. Its utilization in therapy for other bacterial infections was discouraged because of the fear that widespread utilization could generate resistance of Mycobacterium to rifampin. Until it emerged as the preferred agent of prophylaxis for contacts of patients with meningococcal meningitis and for the elimination of the meningococcal carrier state, rifampin was reserved as essentially a" one-bug" drug. Since profits from antituberculous therapy were high, there was no economic pressure to expand its use. In addition, early studies in vitro and in vivo had demonstrated that resistant strains emerged rapidly when the drug was used alone. Conservative clinicians were quick to predict that wide utilization of this drug in hospitals …