RESUMO
BACKGROUND/PURPOSE: To understand the resistance patterns of rapidly growing mycobacteria (RGM) in Taiwan, antimicrobial resistance of clinical isolates was determined as part of the SMART (Surveillance from Multicenter Antimicrobial Resistance in Taiwan) program. METHODS: During the period from January 2002 to December 2003, clinical isolates were collected from eight hospitals located on the west side of Taiwan and one reference laboratory. Broth microdilution minimum inhibitory concentrations of 11 antimicrobial agents were determined for 312 clinical isolates of RGM, including the Mycobacterium fortuitum group (110 isolates), Mycobacterium abscessus group (168 isolates), and Mycobacterium chelonae group (34 isolates). RESULTS: Nearly all of the RGM were susceptible to amikacin and ofloxacin (= 90%) and resistant to doxycycline (less than 3% susceptible). Tobramycin showed similar in vitro activity against the M. fortuitum and M. chelonae (77%) groups, but was less active against the M. abscessus group (58%). Ciprofloxacin was active mainly against M. fortuitum (95%). Nearly all RGM were resistant to erythromycin and doxycycline. However, around half of the RGM isolates remained susceptible to minocycline (50-54%). Clarithromycin was active against the M. abscessus group (53% susceptible), with a high rate of resistance in the M. chelonae (38% susceptible) and M. fortuitum (15% susceptible) group. Cefoxitin was more active against the M. fortuitum group (65%) than the other two RGM (40-44%), and les than 40% of the RGM isolates remained susceptible to imipenem (21-38%). CONCLUSION: The resistance of RGM in Taiwan is not as high as previously reported (notably for tobramycin, ciprofloxacin and cefoxitin), but reduction in the susceptibility rates of clarithromycin and imipenem for the M. fortuitum and M. abscessus groups demonstrates the importance of in vitro susceptibility testing of clinically important isolates, as susceptibility may differ in different geographical areas, even regionally, and over time.