The focus of the management of Mycobacterium avium disease in human immunodeficiency virus (HIV)-infected persons has shifted from treatment to prevention with the identification of effective prophylactic agents such as azithromycin, clarithromycin, and rifabutin. It is believed that M. avium disease is preceded by an ill-defined period of M. avium colonization of the respiratory tract or, probably more commonly, the gastrointestinal tract. It is most likely that disease occurs when the level and/or complexity (number of strains with perhaps different levels of virulence) of M. avium colonization reaches a critical threshold in combination with the development of a critical immunodeficiency, which places a patient at risk for infection. Indeed, the relationship between HIV infection and M. avium infection is unclear, and the combined effect of the two infections may influence the course and severity of each. The mechanism(s) of action of prophylactic agents for M. avium are also ill-defined, and the occurrence of breakthrough disease with strains of M. avium that are susceptible to the inhibitory effects of macrolides or rifabutin suggests that the subinhibitory effects of these agents may be important for a prophylactic effect. Nevertheless, a significant percentage of breakthrough does occur with strains that are resistant to clarithromycin (58%) or azithromycin (11%), and the rapid and accurate detection of these strains is important for the successful management of M. avium disease. Resistance to azithromycin has been defined as a minimum inhibitory concentration (MIC) ≥256 μg/mL, while resistance to clarithromycin has been defined as a MIC ≥32 μg/mL. The in vitro susceptibility of M. avium to azithromycin and clarithromycin can be reliably determined using a radiometric (Bactec) broth macrodilution assay. In addition, a molecular assay has been described that is both highly sensitive and specific for detecting mutations that are the primary cause of macrolide resistance.