Tigecycline seems to be a promising agent for treatment of resistant Acinetobacter baumannii infections, however. approved criteria for susceptibility testing are still lacking. The aim of this study was to evaluate tigecycline activity against multidrug resistant A. baumannii clinical isolates, as well as susceptibility testing methods and disk diffusion (DD) breakpoints. Reference broth microdilution (BMD), Etest and DD methods were used. MIC50 and MIC90 by the reference method were 1 and 8 mu g/ml, respectively. Applying the breakpoints for Enterobactericeae, 85.7% of the isolates were sensitive to tigecycline. The Etest resulted in lower susceptibility rates (63.7%). Essential agreement between Etest and BMD was 75.8%. 21.9% of the strains were susceptible by BMD and intermediate by Etest. Provisional DD breakpoints >= 19/<= 14 mm produced unacceptable minor errors while adjustment to >= 17/<= 13 mm yielded the best results. In conclusion, resistance was not rare against our MDR A. baumannii isolates although tigecycline had not been used previously. Breakpoints >= 17/<= 13 mm may be preferred for DD. We suggest caution with intermediate results by Etest.