Mycobacterium marinum is recognized as a cutaneous pathogen requiring antibiotic treatment. We compared the E-test with a reference agar dilution method for susceptibility testing of M. marinum to amikacin, ciprofloxacin, clarithromycin, doxycycline, rifampicin, trimethoprim-sulfamethoxazole and ethambutol. MICs obtained after 6 d showed agreement between the E-test and agar dilution within +/- 2 dilutions in 95% of all cases for amikacin, ciprofloxacin, doxycycline and rifampicin. Inhibitory concentrations of trimethoprim-sulfamethoxazole were difficult to define using the E-test because of gradually decreased growth in the presence of increasing concentrations. For clarithromycin, results were generally 1-3 dilution steps lower with the E-test and for ethambutol they were often > 3 dilution steps lower. These differences always appeared in the low MIC range and did not affect the categorization of the strains as susceptible to these 2 antimicrobial agents. All strains were interpreted as susceptible to all tested antibiotics, except for doxycycline, according to recommended breakpoints. Overall, our results suggest that the E-test can be considered an alternative for susceptibility testing of certain antibacterial agents against M. marinum.