[Sensitivity to colistin: evaluation of cut-off points available in disk diffusion test].

Infections produced by multidrug resistant organisms are one of the greatest problems in health centers. Often, only polymyxines show good activity "in vitro" against the carbapenem resistant gram-negative strains; but the National Committee for Clinical Laboratory Standards (NCCLS) documents do not currently provide interpretative criteria for testing the polymyxines. The antimicrobial activity of colistin, and the correlation between the agar dilution test and disk diffusion test were evaluated against 186 gram-negative strains isolated at the Hospital de Clínicas "José de San Martín" of Buenos Aires city. All susceptibility tests were performed according to the NCCLS recommendations. Were evaluated two breakpoints, NCCLS 1981 (< or = 8 mm and > or = 11 mm), and R. Jones 2001 (< or = 11 mm and > or = 14 mm). Discrepancies on interpretative category were found (0.5% minor; 2.2% major and 4.4% very major) with NCCLS 1981, and (18.9% minor; 3.8% major and 0.5% very major) with R. Jones 2001 criteria. Conclusions. In spite of the fact that the breakpoint used by R. Jones 2001 decreases the very major error but increases the minor error, according to our results we recommend the use of MIC methods to assist the therapeutic application of colistin; however resistance to colistin was not detected with zone diameters > or = 16 mm.