Resistance to fusidic acid is determined by a number of mechanisms. The best described are alterations in elongation factor G, which appear in natural mutants that are harboured at low rates in normal populations of staphylococci (10(6) to 10(8)). Altered drug permeability has also been described, and appears to be plasmid-borne. Binding by chloramphenicol acetyltransferase type I and efflux are other described mechanisms of resistance whose prevalence is unclear. A large number of studies have examined rates of fusidic acid resistance in staphylococci. Most show low levels of resistance. Studies where high levels of resistance have been seen are from areas of the hospital where cross infection is common. Rates of resistance have tended to be slightly higher in methicillin-resistant strains of Staphylococcus aureus. Studies on the evolution of resistance have shown no major trends to the emergence of resistance. In one case this is despite increasing use of both systemic and topical fusidic acid over more than 24 years. Selection for resistant variants during treatment was recognised early in vitro and in vivo. However, evidence suggests that it does not occur at high frequency in clinical practice. Nevertheless, accumulated experience is that fusidic acid in combination with other agents results in less resistance emergence.