In Vitro Antibiotic Synergy in Extensively Drug-Resistant Acinetobacter baumannii: the Effect of Testing by Time-Kill, Checkerboard, and Etest Methods

Abstract

This study examined the in vitro effects of polymyxin B, tigecycline, and rifampin combinations on 16 isolates of extensively drug-resistant Acinetobacter baumannii, including four polymyxin-resistant strains. In vitro synergy was demonstrated in 19 (40%) of a possible 48 isolate-antibiotic combinations by time-kill methods, 8 (17%) by checkerboard methods, and only 1 (2%) by Etest methods. There was only slight agreement between Etest and checkerboard methods and no agreement between results obtained by other methods.

Treatment of extensively drug-resistant (XDR) Acinetobacter baumannii infections is challenging because of both the limited choice of antibiotic and the tendency of such infections to occur in critically ill hosts with limited physiologic reserves (17). Polymyxins demonstrate in vitro activity against A. baumannii, but resistance is also reported (8, 17). In the absence of feasible alternatives, unconventional antibiotics, such as minocycline, rifampin, and tigecycline, have been used, both singly and in combination (19). In vitro combination susceptibility testing poses a significant challenge for clinical microbiology laboratories, with a lack of standardization and a variety of in vitro testing methods. This study evaluated the effects of various antibiotic combinations against a panel of XDR-Acinetobacter baumannii and compared in vitro synergy testing results obtained from time-kill, checkerboard, and Etest methods.

(Some elements of this study were presented at the 20th European Congress of Clinical Microbiology and Infectious Diseases, 2010.)

Clinical isolates of antibiotic-resistant A. baumannii were collected from four hospitals in Singapore over a 2-year period. Species identification was confirmed by a multiplex PCR assay (2). MICs to ampicillin-sulbactam, ciprofloxacin, gentamicin, imipenem, meropenem, aztreonam, piperacillin-tazobactam, polymyxin B, tigecycline, ceftazidime, amikacin, and cefepime were obtained by broth microdilution (Sensititre, Trek Diagnostics, United Kingdom), while MICs for rifampin were obtained by broth macrodilution (3). Categorical susceptibility was based on Clinical and Laboratory Standards Institute breakpoints (4). XDR A. baumannii isolates were defined as isolates that were resistant to all the tested antimicrobials except for tigecycline, rifampin, and polymyxin B (5, 6), while pandrug-resistant (PDR) A. baumannii was resistant to all tested antimicrobial agents (5, 6). Clonal relatedness of study isolates was determined by multiplex PCR strain typing (11), with cluster analysis of banding data performed using the unweighted pair group method with arithmetic mean.

All strains were tested for the presence of in vitro synergy to polymyxin B-rifampin, polymyxin B-tigecycline, and tigecycline-rifampin combinations by three methods. These combinations were selected based on previously published data, which demonstrated a high likelihood of achieving in vitro synergy (10). Time-kill assays were performed following methods published by the Clinical and Laboratory Standards Institute (13), with colony enumeration performed at 0- and 24-h time points. The lower limit of detection was 400 CFU/ml. Antibiotic concentrations used for the time-kill assay were 2 mg/liter for each antibiotic, representing achievable serum concentrations for polymyxin B (9) and rifampin (7) and achievable tissue levels for tigecycline (18).

Etest and checkerboard synergy testing were performed according to published methods (12, 20). Fractional inhibitory concentrations (FIC) were calculated as follows: FIC of drug A = MIC of drug A in combination/MIC of drug A alone; FIC of drug B = MIC of drug B in combination/MIC of drug B alone. The FIC index (FICI) was defined as the FIC of drug A added to the FIC of drug B.

The definition of synergy for time-kill assays used was a ≥2-log₁₀ decrease in CFU/ml for the antibiotic combination compared with its more active constituent (13). For Etest and checkerboard assays, the FICI was interpreted as follows: synergy, FICI ≤ 0.5; antagonism, FICI > 4.0; and indifferent, 0.5 < FICI < 4.0 (14).

Thirty-six isolates of A. baumannii were received, of which 4 (11%) isolates were PDR A. baumannii, with the remaining categorized as XDR A. baumannii. Categorical susceptibility was not reported for rifampin (MIC range, 2 to ≥32 mg/liter) and tigecycline (MIC range, 2 to ≥32 mg/liter). Molecular typing data demonstrated limited clonal clustering of the 36 isolates (Fig. 1).

FIG. 1.

Clonal relatedness of all A. baumannii isolates received for testing. The 16 isolates tested by in vitro antibiotic combination testing are indicated by asterisks.

Sixteen isolates were selected for in vitro synergy testing, including all PDR A. baumannii isolates and 12 other isolates from different clonal subgroups. In vitro synergy by time-kill was detected with polymyxin B-rifampin (n = 9; 56%), polymyxin B-tigecycline (n = 7; 44%), and tigecycline-rifampin combinations (n = 3; 19%) by time-kill methods. Checkerboard methods reported in vitro synergy with polymyxin B-rifampin (n = 3; 19%), rifampin-tigecycline (n = 3; 19%), and polymyxin B-tigecycline combinations (n = 2; 12%). Etest methods demonstrated in vitro synergy only to the polymyxin B-rifampin combination (n = 1; 6%).

In vitro synergy was demonstrated in 19 (40%) of a possible 48 isolate/antibiotic combinations by time-kill methods, 8 (17%) by checkerboard methods, and only 1 (2%) by Etest methods. Only one pandrug-resistant isolate demonstrated in vitro synergy with the polymyxin B-rifampin combination by all three methods, while another two demonstrated in vitro synergy by both time-kill and checkerboard methods (Table 1). The same test outcome between the three methods was reported for 25 (52%) of 48 isolate/antibiotic combinations. The same test outcome by time-kill and Etest was reported for 29 (60%) of isolate/antibiotic combinations and for Etest and checkerboard testing in 40 (83%) of isolate/antibiotic combinations.

TABLE 1.

Rest values obtained by time-kill, Etest, and checkerboard in vitro testing^a

Isolate	MIC (μg/ml)			Test resultb
				PB-RIF			PB-TGC			TGC-RIF
	PB	RIF	TGC	Time-kill	Etest	Checkerboard	Time-kill	Etest	Checkerboard	Time-kill	Etest	Checkerboard
8	1	8	8	5.5	1.2	0.6	5.5	1.8	0.5	1.2	1.0	0.4
16	1	4	4	1.0	1.4	1.0	1.0	2.3	1.1	1.6	1.7	0.8
23	1	4	8	1.0	1.2	0.6	1.0	1.8	0.6	1.2	1.0	0.5
25	1	4	4	4.4	1.5	1.0	4.4	2	1.1	1.4	2	0.8
32	1	≥32	4	3.9	0.9	1.0	2.6	1.5	1.2	0.6	0.8	0.8
41	1	8	8	1.8	1.5	0.7	1.8	2	0.8	1.2	1.5	0.8
91	2	2	4	2.3	1.5	1.0	2.3	2	1.1	5.5	1.0	1
104	2	8	4	1.0	1.2	1.0	1.0	2	0.8	1.5	1.0	0.8
128	1	4	≥32	2.2	0.9	1.0	2.9	1.5	0.3	1.1	1.0	1.6
112	1	2	4	3.6	1.5	0.5	3.6	2	1.1	5.7	1.5	1.1
8879	2	2	2	2.5	1.8	0.6	2.5	2	1.1	4	1.5	0.6
3160	2	4	≥32	1.0	1	0.5	1.0	1.7	0.6	1.3	1.5	0.6
27640	≥16	≥32	≥32	2.8	0.2	0.02	1.0	0.6	1.0	1.1	0.7	0.4
18351	≥16	2	8	4.5	0.7	0.1	1.0	5	0.3	1.2	0.9	0.4
9447	≥16	≥32	4	1.0	2	0.1	1.0	1.8	0.5	1.0	2.0	1
11171	≥16	≥32	4	1.0	1.7	0.5	1.0	1.7	0.5	1.0	2.0	16

^aAbbreviations: PB, polymyxin B; RIF, rifampin; TGC, tigecycline.

^bValues for the time-kill column demonstrate the reduction in CFU/ml expressed in log₁₀, while values for Etest and checkerboard columns represent the lowest FIC index achieved in testing. Antibiotic-test-isolate combinations demonstrating in vitro synergy are shown in bold italic text, while those demonstrating antagonism are shown in bold text.

This study highlights the lack of agreement between different in vitro methods for determining antibiotic synergy in A. baumannii. The limited data available from this and other studies (1, 16) suggest that in vitro synergy in Acinetobacter spp. is most likely to be reported when tested by time-kill methods and least likely to be reported by Etest methods. Conversely, on a practical note, time-kill studies are too labor intensive for use in routine diagnostic laboratories and are unlikely to provide results in a clinically relevant time frame. To date, there are no published studies that compare clinical outcomes of combination antibiotic therapy for Acinetobacter spp. with the results of in vitro testing. Limited data are available comparing in vitro studies with animal models of infection. Four isolates of A. baumannii demonstrated in vitro synergy with colistin-rifampin combinations by time-kill methods, but when tested in a mouse pneumonia model, the imipenem-sulbactam combination provided the best survival (85.7% compared to 43.8%) (15). Based on the limited data available, it is not yet possible to conclude which in vitro testing method best correlates with clinical outcome, and it is clear that further research is acutely needed. In summary, this study demonstrates little agreement between different methods of measuring antibiotic synergy for PDR and XDR A. baumannii. In vitro synergy was best observed in polymyxin B-rifampin combinations when tested by time-kill. The optimal in vitro testing method that best corresponds with the clinical outcome of infections with XDR-Acinetobacter spp. remains to be ascertained.