In Vitro Activity of Fusidic Acid-Containing Combinations against Carbapenem-Resistant Acinetobacter baumannii Clinical Strains

LETTER

Acinetobacter baumannii continues to be one of the most problematic Gram-negative pathogens in health care settings (1). Carbapenem-resistant A. baumannii infections are typically treated with combinations of agents that include colistin (CST). However, the increasing use of colistin has led to the development of colistin resistance (2), and the lack of treatment options continues to present a critical unmet need. To address this gap, the activities of various combinations of agents have been tested against A. baumannii. A series of studies have shown that some agents, such as vancomycin, which has activity only against Gram-positive bacteria, may exert activity against A. baumannii in combination with colistin (3, 4). More recently, Phee et al. reported robust synergy between fusidic acid (FD) and colistin against a panel of multidrug-resistant A. baumannii clinical strains, including some colistin-resistant strains (5).

To independently confirm this promising finding, we first conducted by previously described methods (4) checkerboard synergy testing and time-kill assays using carbapenem-resistant A. baumannii strains identified from three patients. The strains were 1494 and 1508 for patient 1, 2382 and 2384 for patient 4, and 2949 and 2949a for patient 10, as reported in our previous work (2). MICs of colistin were 1 to 2 μg/ml and 4 to >64 μg/ml for the susceptible and resistant strains, respectively. MICs of fusidic acid were 32 to >64 μg/ml. By checkerboard testing, fractional inhibitory concentration (FIC) indices of 0.05 to 0.25 and 0.06 to 0.19 were achieved for CST-susceptible and -resistant strains, respectively. For example, for CST-resistant strain 2949a, growth was inhibited with CST at 4 μg/ml and with FD at 8 μg/ml.

We then conducted time-kill assays using colistin, fusidic acid, doripenem, and vancomycin alone and in combination with colistin (see Fig. S1 in the supplemental material). The final concentrations used were 1 μg/ml for colistin (2 μg/ml for colistin-resistant strains), 8 μg/ml for fusidic acid, 8 μg/ml for doripenem, and 20 μg/ml for vancomycin (4). In 2 of the 3 colistin-susceptible strains, colistin with vancomycin and colistin with doripenem achieved total killing at 4 h, and colistin with fusidic acid did so at 24 h. For the remaining colistin-susceptible strain, colistin with doripenem and colistin with fusidic acid achieved total killing at 24 h, whereas colistin with vancomycin achieved a 3-log decrease in the number of CFU at 4 h, followed by regrowth at 24 h. For the colistin-resistant strains, total killing was achieved with all combinations at 24 h against one strain for which the colistin MIC was 4 μg/ml. With the remaining two colistin-resistant strains with higher colistin MICs (>64 μg/ml), colistin with doripenem yielded 1.5- to 2-log decreases in the number of CFU at 4 h, followed by regrowth, but colistin with vancomycin and colistin with fusidic acid did not achieve appreciable synergy.

These results confirm the presence of in vitro synergy between colistin and fusidic acid that is comparable to the synergy between colistin and vancomycin but also suggest that this synergy with fusidic acid may be strain dependent and applicable to strains for which the colistin MICs are relatively low, at least at concentrations that are pharmacokinetically relevant.