Introduction: We assessed agreement among 3 assays for determining susceptibility to imipenem (IMP) and meropenem (MPM) of Acinetobacter baumannii (Ab) heteroresistant to carbapenems (Hr-CP).
Methods: Thirty Ab clinical isolates belonging to 6 clones (REP-PCR) were studied, in which IMP and MPM MICs were between < or = 1 and > or = 8 mg/l by Wider system. MICs determined by Wider were compared with those obtained by microdilution (MD) and E-test. Errors in the clinical category were determined considering MD as the reference method.
Results: Twenty-five Ab were Hr-CP (growth of resistant colonies within the inhibition area of disks and E-test strips). Agreement for the MICs (+/- 1log(2)) in Hr-CP colonies was as follows: Wider vs. MD (96% IMP, 100% MPM), Wider vs. E-test (50% IMP, 64% MPM), MD vs. E-test (64% IMP, 60% MPM). Major errors were not observed. Minor errors and moderate errors for IMP included Wider vs. MD (40% and 0%), E-test vs. MD (40% and 12%), and disks vs. MD (36% and 8%). Agreement for the MICs considering colonies growing within the inhibition areas (E-test and disks) was Wider vs. E-test (8% IMP, 12% MPM), and MD vs. E-test (8% IMP and MPM). Major errors were not observed in this case either; minor errors and moderate errors for IMP were seen in E-test vs. MD (40% and 48%), and disks vs. MD (40% and 48%).
Conclusion: Susceptibility testing methods based on microdilution (Wider system and MD) are not useful to detect subpopulations of Ab Hr-CP.