A meta-analysis of in vitro antibiotic synergy against Acinetobacter baumannii

The aim of the work was to describe the different in vitro models for testing synergism of antibiotics and gather the results of antibiotic synergy against multidrug-resistant Acinetobacter baumannii (MDR-Ab). The different original articles were obtained from different web sites. In order to compare the results obtained by the different methods for synergy testing, the Pearson chi-square and the Fischer tests were used. Moreover, non-parametric chi-square test was used in order to compare the frequency distribution in each analysed manuscript. In the current meta-analysis 24 manuscripts, which encompassed 2016 tests of in vitro synergism of different antimicrobials against MDR-Ab, were revised. Checkerboard synergy testing was used in 11 studies, which encompasses 1086 tests (53.9%); time-kill assays were applied in 12 studies, which encompass 359 tests (17.8%); gradient diffusion methods were used in seven studies, encompassing 293 tests (14.5%). And, finally, time-kill plus checkerboard were applied in two studies, encompassing 278 tests (13.8%). By comparing these data, checkerboard and time-kill methods were significantly more used than gradient diffusion methods (p <0.005). Regarding synergy rates obtained on the basis of the applied method, checkerboard provided 227 tests (20.9%) with a synergistic effect; time-kill assays yielded 222 tests (61.8%) with a synergistic effect; gradient diffusion methods only provided 29 tests (9.9%) with a synergistic effect; and, finally, time-kill plus checkerboard yielded just 15 tests (5.4%) with a synergistic effect. When,comparing these percentages, synergy rates reported by time-kill methods were significantly higher than that obtained by checkerboard and gradient diffusion methods (p <0.005). On the basis of the revised data, the combinations of a bactericidal antibiotic plus Tigecycline, Vancomycin or Teicoplanin are not recommended. The best combinations of antibiotics are those which include bactericidal antibiotics such as Carbapenems, Fosfomycin, Amikacin, Polymyxins, Rifampicin and Ampicillin/Sulbactam. (C) 2015 Elsevier B.V. All rights reserved.