Synergistic Activities of Colistin Combinations with Meropenem,Sulbactam,Minocycline,Disodium Fosfomycin,or Vancomycin Against Different Clones of Carbapenem-Resistant Acinetobacter baumannii Strains

Abstract

Aims:

Colistin became the primary treatment option for Acinetobacters that had developed a high rate of resistance to carbapenems which were the first-line therapy in the past, and now Acinetobacters become resistant to nearly all antibiotics. Because of the resistance potential to colistin and the concerns about toxicity, especially for high doses, colistin combination therapies are preferred nowadays. In this study, we aimed to investigate whether combinations of colistin with meropenem, sulbactam, fosfomycin, vancomycin, and minocycline are synergic or not and to determine minocycline susceptibility rate, which is not in use in our country.

Results:

For the studied 23 Acinetobacter strains, the highest synergy was between colistin and vancomycin, which was shown in 4 (17.4%) strains. The synergy of colistin with meropenem and fosfomycin was detected for 1 (4.3%) strain, the synergy of colistin with minocycline was detected for 2 (8.6%) strains, and no synergy was detected for colistin–sulbactam combination. All the strains were susceptible to minocycline.

Conclusion:

None of the antibiotic combinations was antagonistic. They had synergistic and additive interactions. Thus, these combinations can be used in clinical practices. The remarkable synergistic interaction of colistin–vancomycin combination and high susceptibility to minocycline highlight the need for more researches on these subjects.

Introduction

In recent years Acinetobacter baumannii has emerged as a significant nosocomial pathogen, especially in intensive care units (ICUs). The rates of the carbapenem-resistant or pan-resistant strains are increasing day by day.^1,2 Antibiotic choices are limited, and optimal treatment is unknown for patients infected with these bacteria. The Infectious Diseases Society of America (IDSA) reported Acinetobacter strains as challenging to treat pathogens (ESKAPE pathogens).³ Polymyxins, tigecycline, carbapenems, and sulbactam are the most preferred antimicrobials. Currently, colistin is the backbone of the treatment of A. baumannii infections. Because of the potential resistance to colistin, unclear optimal dosing, and concerns about toxicity, especially for high doses, colistin-based combination therapies are favored regimens today.^4,5

Since therapeutic choices are very limited or absent in cases with multidrug resistant (MDR) or pan-drug resistant A. baumannii infections, the investigations of new antibiotic regimens are needed. Synergistic, additive, or antagonistic effects of various combinations were tested in several in vitro studies. Checkerboard method is one of the frequently preferred synergy tests based on microdilution technique.⁶ In this study, synergistic in vitro activities of colistin combinations with meropenem, sulbactam, vancomycin, fosfomycin, and minocycline were tested by the checkerboard method against carbapenem-resistant A. baumannii strains belonging to different clones.

Materials and Methods

Isolates

This study was conducted in the Clinical Microbiology Laboratory of Ankara Numune Training and Research Hospital (Ankara, Turkey). Ankara Numune Training and Research Hospital is a 1,200-bed tertiary care hospital and referral center, which has both surgical and medical ICUs and receives patients not only from the capital city of Ankara with a population of nearly 5.5 million people but also from surrounding cities.

The isolates used in this antibiotic synergy study were selected from 124 carbapenem-resistant A. baumannii strains isolated from ICU surveillance cultures previously collected between May 15, 2012 and May 15, 2013. These 124 A. baumannii strains included isolates from clinical samples of the ICU patients (blood, tracheal aspirate, urine, wound samples, etc.), rectal swab samples, and environmental samples. Clonal analysis of these 124 A. baumannii strains was done previously with Repetitive element palindromic polymerase chain reaction (Rep-PCR) DiversiLab^® (bioMérieux, Marcy l'Etoile, France). The Pearson correlation coefficient was used to determine distance matrices and the unweighted pair group method with arithmetic averages to create dendrograms of the isolates. All isolates with Rep-PCR patterns with a similarity index of >95% were grouped within the same cluster. Seven clusters were detected, and some isolates could not be arranged.

The identification of bacterial strains was performed according to the conventional tests and using the VITEK II^® system (bioMérieux). Confirmation of the identification of A. baumannii strains was done with MALDI Biotyper (Microflex LT; Bruker Daltonics, Inc., Germany). Antibiotic susceptibility tests were carried out using VITEK II system (bioMérieux) with Vitek GN AST cards, which contains ampicillin–sulbactam, piperacillin, piperacillin–tazobactam, ceftazidime, cefoperazone–sulbactam, cefepime, imipenem, meropenem, amikacin, gentamicin, netilmicin, ciprofloxacin, levofloxacin, tetracycline, tigecycline, colistin, and trimethoprim–sulfamethoxazole. Ertapenem, imipenem, and meropenem minimum inhibitory concentrations (MICs) were also evaluated using E-test (bioMérieux).

The results were assessed according to the Clinical Laboratory Standards Institute (CLSI), except for tigecycline.⁷ For tigecycline, the European Committee on Antimicrobial Susceptibility Testing (EUCAST) breakpoints were used.⁸ All A. baumannii strains were resistant to beta-lactam antibiotics, including carbapenems. For antibiotics other than beta-lactams (amikacin, gentamicin, netilmicin, ciprofloxacin, levofloxacin, colistin, and trimethoprim–sulfamethoxazole), the susceptible strains were classified as susceptible and intermediate susceptible, and resistant strains were classified as resistant. The strains which had identical susceptibility pattern were accepted as an antibiotype. In this way, the strains were grouped into 23 antibiotypes. For the antibiotic synergy tests, only one strain was selected from each antibiotype group. In this way isolates from all the clones were also included. Two of these 23 strains did not belong to any clone.

Isolates were stored at −80°C until the study began. Isolates were thawed, and primary and secondary cultures were inoculated to 5% sheep blood agar before the start of the experiments. Table 1 shows the distribution of clones and clinical samples of the study isolates.

Table 1.

Distribution of Acinetobacter baumannii Clones and Clinical Specimens

	n	%
Clones
A	12	52.2
B	2	8.7
C	2	8.7
D	2	8.7
E	1	4.3
F	1	4.3
G	1	4.3
Unidentified	2	8.7
Total	23	100
Clinical specimens
Rectal swab	13	56.5
Tracheal aspirate	5	21.7
Blood	2	8.7
Wound	2	8.7
Sputum	1	4.3
Total	23	100

Synergy testing with checkerboard methods

Synergistic activities of five antibiotics against colistin were evaluated. Meropenem trihydrate (Sigma-Aldrich, St. Louis, MO), vancomycin hydrochloride (Sigma-Aldrich), disodium fosfomycin (Sigma-Aldrich), minocycline hydrochloride (Sigma-Aldrich), colistin sulfate (Sigma-Aldrich), and sulbactam (Sigma-Aldrich) MIC were used for synergy tests. First, to set up the MIC ranges, which were going to be used in checkerboard tests, two randomly selected isolates were assessed with the broth microdilution method. Disodium fosfomycin was diluted with 25 μg/mL glucose-6-phosphate-Mueller Hinton Broth (Sigma-Aldrich), and others were diluted with only Mueller Hinton Broth. After 18–24 hours incubation, MICs were recorded.

Panels of 96-well microtiter plates were prepared according to results obtained from MIC determination with broth microdilution. Dilution intervals were determined to be four to eight times higher and 1/8–1/16 below the MIC values obtained from the preliminary study. For colistin 32–0.03125 μg/mL, meropenem 64–1 μg/mL, vancomycin 256–4 μg/mL, disodium fosfomycin 256–4 μg/mL, sulbactam 16–0.25 μg/mL, and minocycline 4–0.0625 μg/mL concentration range were used, and checkerboard technique was applied as recommended by Clinical Microbiology Procedures Handbook.⁶

Inoculum was prepared for the final microorganism density 3–5 × 10⁵ CFU/mL from stored strains. Except for the sterility control well, all of the wells were inoculated. From control well, control plates for purity and counting (5% sheep blood agar plate) were prepared. All of the microplates and agar plates were incubated for 18–24 hours at 35°C in ambient air.

Escherichia coli ATCC 25922 and Staphylococcus aureus ATCC 25923 were used as quality control strains except for sulbactam. There was no suggestion of quality control strain for sulbactam.

MIC₅₀ and MIC₉₀ values of the strains were also calculated other than disodium fosfomycin and vancomycin where Acinetobacter strains were intrinsically resistant. The fractional inhibitory concentration (FIC) was calculated for each antibiotic in each combination. $F I C C T = \frac{M I C o f c o l i s t i n i n c o m b i n a t i o n}{M I C o f c o l i s t i n a l o n e}$

F I C x = \frac{M I C o f t h e o t h e r a n t i b i o t i c s i n c o m b i n a t i o n}{M I C o f t h e o t h e r a n t i b i o t i c s a l o n e}

The FIC index (FICI) was obtained by the addition of the FIC colistin (CT) and the FIC x. FICIs were interpreted as synergistic interaction if FICI ≤0.5, as additive interaction if 0.5 < FICI <1, as indifferent interaction if 1 ≤ FICI ≤4, and as antagonistic interaction if FICI >4.

Results

Of the 23 strains, 13 (56.5%) were isolated from rectal swabs, and the others were from clinical specimens (endotracheal aspirate, blood, wound, and sputum). All of the isolates were meropenem resistant, and clone A was the dominant clone (Table 1). According to the CLSI 2017, all the isolates were susceptible to minocycline, and 73.9% (17 of 23 isolates) were susceptible to colistin.⁷ MIC₅₀ and MIC₉₀ values of the antimicrobials are as demonstrated in Table 2.

Table 2.

The Minimum Inhibitory Concentration, the Minimum Inhibitory Concentration 50, and the Minimum Inhibitory Concentration 90 Values; and the Minimum Inhibitory Concentration Range for Sulbactam, Minocycline, Colistin, and Meropenem (n = 23)

	MIC₅₀ (μg/mL)	MIC₉₀ (μg/mL)	MIC range (μg/mL)
Sulbactam	16	16	4–16
Minocycline	1	4	0.0625–4
Colistin	2	4	0.5–8
Meropenem	32	32	16–64

MIC, minimum inhibitory concentration.

The results of synergy tests are shown in Table 3. The highest synergy was detected between colistin and vancomycin (17.4%). The synergy rates of other antibiotic combinations were determined for colistin–minocycline as 8.6%, for colistin–meropenem as 4.3%, and for colistin–fosfomycin as 4.3%. However, no synergy was detected between colistin and sulbactam. Except for colistin–sulbactam and colistin–fosfomycin combinations, synergistic and additive interactions were more common than the indifferent interactions. No antagonism was found between any combinations. FICI values in antibiotic combinations and the distribution of interaction types are demonstrated in Table 4.

Table 3.

The Synergy Test Results of Antimicrobial Combinations, n (%)

Combinations	Synergistic effect	Additive effect	Indifferent effect	Total
Colistin–meropenem	1 (4.3)	15 (65.2)	7 (30.4)	23 (100)
Colistin–sulbactam	0	8 (34.8)	15 (65.2)	23 (100)
Colistin–fosfomycin	1 (4.3)	10 (43.5)	12 (52.2)	23 (100)
Colistin–vancomycin	4 (17.4)	15 (65.2)	4 (17.4)	23 (100)
Colistin–minocycline	2 (8.6)	10 (43.5)	10 (43.5)	22^a (95.7)

n (%), row percentages.

Since the MIC of minocycline was <0.0625 μg/mL in a strain, the fractional inhibitory concentration index could not be calculated.

Table 4.

Distribution of the Fractional Inhibitory Concentration Index and Antimicrobial Interaction Type in Antibiotic Combinations

Strain number	Colistin–meropenem		Colistin–sulbactam		Colistin–fosfomycin		Colistin–vancomycin		Colistin–minocycline
Strain number	FICI	Int	FICI	Int	FICI	Int	FICI	Int	FICI	Int
1	0.63	A	1.01	I	1.00	I	0.62	A	0.75	A
2	0.75	A	1.00	I	1.01	I	0.53	A	0.51	A
3	0.75	A	1.01	I	0.75	A	0.56	A	0.75	A
4	0.75	A	0.75	A	0.62	A	0.28	S	1.00	I
5	0.62	A	1.01	I	0.62	A	0.51	A	0.62	A
6	0.75	A	0.75	A	1.00	I	0.56	A	—	U
7	1.00	I	0.51	A	0.75	A	0.50	S	1.01	I
8	1.00	I	1.00	I	0.51	A	1.01	I	0.75	A
9	0.75	A	1.01	I	0.53	A	0.56	A	1.00	I
10	1.00	I	1.01	I	1.01	I	0.62	A	0.75	A
11	1.00	I	1.01	I	0.51	A	0.56	A	0.75	A
12	1.00	I	0.75	A	1.01	I	0.62	A	0.56	A
13	1.01	I	0.75	A	1.00	I	1.02	I	1.03	I
14	1.01	I	0.75	A	0.75	A	1.00	I	0.50	S
15	0.53	A	1.01	I	0.51	A	0.75	A	1.01	I
16	0.75	A	1.01	I	1.00	I	0.56	A	1.00	I
17	0.75	A	1.00	I	1.01	I	0.75	A	1.00	I
18	0.75	A	1.01	I	0.51	A	1.00	I	1.01	I
19	0.56	A	0.62	A	1.00	I	0.31	S	1.00	I
20	0.56	A	1.01	I	1.01	I	0.56	A	0.75	A
21	0.75	A	1.00	I	1.01	I	0.62	A	1.00	I
22	0.75	A	1.00	I	0.28	S	0.75	A	0.31	S
23	0.19	S	0.52	A	1.01	I	0.19	S	0.51	A

A, additive effect; FICI, fractional inhibitory concentration index; I, indifferent effect; Int, interaction type; S, synergy; U, uninterpretable.

Discussion

Up to now, the optimal antibiotic regimen for the therapy of MDR A. baumannii infections is still unknown. However, in vitro studies can help clinical decisions about the most appropriate antimicrobial regimen. Colistin is an old drug that has good activity against A. baumannii isolates. Because of the concerns about the toxic effects, uncertainty regarding the optimal dosing regimens and the risk of emergence of colistin-resistant A. baumannii isolates during colistin monotherapy leads the preference of combination therapy. Various combination therapies have been used with colistin to improve clinical success in multidrug-resistant A. baumannii infections, and several studies examined the in vitro interactions between the drugs in the combinations.

Carbapenems are one of the antibiotic classes most commonly used in combination with colistin. In a meta-analysis, the in vitro synergy of the polymyxin–carbapenem combination was examined, and high synergy rates (77%) with low antagonism (1%) and less resistance development were reported.⁹ Checkerboard studies were published as generally lower synergy rates than time-kill studies. In checkerboard studies, the synergy rate of 32% and synergy or additive interaction rates of 71% were detected. Synergy rates were found to be higher with meropenem or doripenem than imipenem. However, in our study, while the synergistic effect was lower with 4.3%, the additive effect was higher, which was 65.2%.

Colistin–sulbactam is another combination investigated in vitro for A. baumannii strains. Different results were reported in various studies. While Percin et al. detected a synergistic effect in 5 of 10 strains by checkerboard assay in colistin-resistant A. baumannii strains, Çetinkol et al. found the antagonistic effect in all of 50 strains according to FIC index in carbapenem-resistant A. baumannii isolates.^10,11 In our study, synergy was not detected between colistin and sulbactam in any carbapenem-resistant strains, but the additive effect was found in 34.8% of them. Similarly, Santimaleeworagun et al. found no synergy for the colistin–sulbactam combination using the checkerboard method in clinical isolates of carbapenem-resistant A. baumannii strains producing OXA 23.¹² Thus, these different results make it difficult to comment about the efficacy of the colistin–sulbactam combination in carbapenem-resistant A. baumannii infections.

It has been suggested that fosfomycin has activity against Gram-positive bacteria and some Gram-negative bacteria, including extended-spectrum beta-lactamase (ESBL) producing microorganisms.^13,14 However, fosfomycin has no in vitro activity against A. baumannii strains. MIC₅₀ and MIC₉₀ values were reported as 256 mg/L and >512 mg/L for MDR A. baumannii isolates, respectively.^12,15 In the study of Santimaleeworagun et al., the synergistic effect of the colistin–fosfomycin combination was found in one of eight carbapenem-resistant A. baumannii strains (12.5%) by checkerboard method.¹² The combination of sulbactam and fosfomycin demonstrated the best synergistic effect in all of the antibiotic combinations.

In a clinical trial, clinical outcomes and mortality at 28 days were not statistically different for colistin versus colistin–fosfomycin combination regimens in patients with infections caused by carbapenem-resistant A. baumannii isolates.¹⁶ However, microbiological eradication rates were found to be significantly higher in the combination group. In our study, while the synergistic effect has been seen in only one carbapenem-resistant A. baumannii strain (4.3%), 10 strains (43.5%) presented the additive effect.

Minocycline, a derivative of tetracycline, has better activity against A. baumannii isolates compared with tetracycline or doxycycline.⁴ Moreover, it can enhance the activity of colistin against MDR A. baumannii isolates. Recent studies have indicated that minocycline is a promising drug for the treatment of MDR A. baumannii infections. In a surveillance trial between 2005 and 2011 in the United States, 84.1% of 2,900 A. baumannii isolates were found susceptible to minocycline. Of these strains, 883 were multidrug resistant, and most of them were susceptible to minocycline (72.1%) with MIC values as MIC₅₀ 2 mg/L and MIC₉₀ 8 mg/L.¹⁷

In our study, it was demonstrated that all of the carbapenem-resistant A. baumannii isolates were susceptible to minocycline. However, a few studies are investigating the synergistic effect of minocycline with colistin. In another research, activities of drug combinations were investigated against Extensively drug-resistant (XDR) A. baumannii isolates in a time-kill study.¹⁸ In that study, all of the isolates were found as resistant to meropenem but susceptible to colistin and minocycline. In addition, the combination of colistin with minocycline and the combination of meropenem with minocycline had a synergistic effect. Nevertheless, clinical experience with minocycline is very limited.

In this study, colistin–vancomycin combination presented the highest synergistic and additive rates with 17.4% and 65.2%, respectively. Vancomycin does not penetrate the outer membrane of Gram-negative bacteria. Thus, it is inactive against Gram-negative pathogens. However, in vitro studies suggested a strong synergy between colistin and vancomycin in colistin-resistant A. baumannii strains. Disruption of the integrity of the outer membrane by colistin may get better the penetration of vancomycin molecules through the outer membrane of A. baumannii strains and may enhance the activity of the drug against these pathogens.⁴

A study investigating the colistin–vancomycin synergy by the checkerboard method in epidemic multidrug-resistant A. baumannii strains reported that four of the six strains presented synergy between vancomycin and colistin.¹⁹ In the same study, electron microscopic examination also showed significant differences between colistin-exposed and colistin-unexposed cells. It was demonstrated that the surface of colistin-exposed cells emerged rougher with marked pits and showed increased topographic variability. In another study, the colistin–vancomycin combination had synergistic and rapid bactericidal activity in 9 of the 10 colistin-resistant A. baumannii clinical isolates.¹⁰ It was also demonstrated that colistin–vancomycin combination had synergistic interaction in colistin-resistant A. baumannii isolates, as well as in colistin-susceptible isolates.²⁰ Nevertheless, a higher risk of renal failure is an important disadvantage of this combination.

In conclusion, we have very limited therapeutic options for the treatment of MDR or XDR A. baumannii infections, and combination therapies are frequently preferred regimens. The combinations investigated in this study were found to be synergistic or additive, and none had antagonistic interaction. Especially the combination of colistin with newly introduced drugs such as fosfomycin and minocycline is promising. In addition, the highest synergistic effect of the colistin–vancomycin combination is impressive. As a result, this study suggested that these combinations can be suitable regimens in routine practice.

Footnotes

Disclosure Statement

No competing financial interests exist.

Funding Information

This study was conducted with the financial support of Infectious Diseases and Clinical Microbiology Speciality Society of Turkey (Number: 84, 07.07.2017).

References

Fishbain

, and Peleg

A.Y.

. 2010. Treatment of Acinetobacter infections. Clin. Infect. Dis. 51:79–84.

Lemos

E.V.

, de la Hoz

F.P

, Einarson

T.R

, McGhan

W.F

, Quevedo

, Castañeda

, and Kawai

. 2014. Carbapenem resistance and mortality in patients with Acinetobacter baumannii infection: systematic review and meta-analysis. Clin. Microbiol. Infect. 20:416–423.

Boucher

H.W.

, Talbot

G.H

, Bradley

J.S

, Edwards

J.E

, Gilbert

, Rice

L.B

, Scheld

, Spellberg

, and Bartlett

. 2009. Bad bugs, no drugs: no ESKAPE! An update from the Infectious Diseases Society of America. Clin. Infect. Dis. 48:1–12.

Viehman

J.A.

, Nguyen

M.H

, and Doi

. 2014. Treatment options for carbapenem-resistant and extensively drug-resistant Acinetobacter baumannii infections. Drugs, 74:1315–1333.

Maragakis

L.L.

, and Perl

T.M.

. 2008. Antimicrobial resistance: Acinetobacter baumannii: epidemiology, antimicrobial resistance, and treatment options. Clin. Infect. Dis. 46:1254–1263.

Moody

2010. Synergy testing: broth microdilution checkerboard and broth macrodilution methods. In L.S. Garcia (ed.), Clinical microbiology procedures handbook. American Society for Microbiology Press,, Washington, pp. 5.16.1–5.16.23.

CLSI. 2017. Performance Standards for Antimicrobial Susceptibility Testing. 27th ed. CLSI Supplement M100, Wayne, PA.

The European Committee on Antimicrobial Susceptibility Testing. 2017. Breakpoint tables for interpretation of MICs and zone diameters, Version 7.1. Available at www.eucast.org.

Zusman

, Avni

, Leibovici

, Adler

, Friberg

, Stergiopoulou

, Carmeli

, and Paul

. 2013. Systematic review and meta-analysis of in vitro synergy of polymyxins and carbapenems. Antimicrob. Agents Chemother. 57:5104–5111.

10.

Percin

, Akyol

, and Kalin

. 2014. In vitro synergism of combinations of colistin with selected antibiotics against colistin-resistant Acinetobacter baumannii. GMS Hyg. Infect. Control, 9:Doc14.

11.

Çetinkol

, Telli

, Altunçekiç Yıldırım

, and M.K. Çalgın. 2016. [Evaluation of the efficacy of colistin/sulbactam combination on carbapenem-resistant Acinetobacter baumannii strains]. Mikrobiyol. Bul. 50:460–465.

12.

Santimaleeworagun

, Wongpoowarak

, Chayakul

, Pattharachayakul

, Tansakul

, and Garey

K.W.

. 2011. In vitro activity of colistin or sulbactam in combination with fosfomycin or imipenem against clinical isolates of carbapenem-resistant Acinetobacter baumannii producing OXA-23 carbapenemases. Southeast Asian J. Trop. Med. Public Health, 42:890–900.

13.

Falagas

M.E.

, Giannopoulou

K.P

, Kokolakis

G.N

, and Rafailidis

P.I.

. 2008. Fosfomycin: use beyond urinary tract and gastrointestinal infections. Clin. Infect. Dis. 46:1069–1077.

14.

de Cueto

, Lopez

, Hernandez

J.R

, Morillo

, and Pascual

. 2006. In vitro activity of fosfomycin against extended-spectrum- lactamase- producing Escherichia coli and Klebsiella pneumoniae: comparison of susceptibility testing procedures. Antimicrob. Agents Chemother. 50:368–370.

15.

Falagas

M.E.

, Kanellopoulou

M.D

, Karageorgopoulos

D.E

, Dimopoulos

, Rafailidis

P.I

, Skarmoutsou

N.D

, and Papafrangas

E.A.

. 2008. Antimicrobial susceptibility of multidrug-resistant Gram negative bacteria to fosfomycin. Eur. J. Clin. Microbiol. Infect. Dis. 27:439–443.

16.

Sirijatuphat

, and Thamlikitkul

. 2014. Preliminary study of colistin versus colistin plus fosfomycin for treatment of carbapenem-resistant Acinetobacter baumannii infections. Antimicrob. Agents Chemother. 58:5598–5601.

17.

Denys

G.A.

, Callister

S.M

, and Dowzicky

M.J

. 2013. Antimicrobial susceptibility among gram-negative isolates collected in the between 2005 and 2011 as part of the Tigecycline Evaluation

USA

and Trial

Surveillance

(T.E.S.T.). Ann. Clin. Microbiol. Antimicrob. 12:24.

18.

Liang

, Liu

X.-F

, Huang

, Zhu

D.-M

, Li

, and Zhang

. 2011. Activities of colistin- and minocycline-based combinations against extensive drug resistant Acinetobacter baumannii isolates from intensive care unit patients. BMC Infect. Dis. 11:109.

19.

Gordon

N.C.

, Png

, and Wareham

D.W.

. 2010. Potent synergy and sustained bactericidal activity of a vancomycin-colistin combination versus multidrug-resistant strains of Acinetobacter baumannii. Antimicrob. Agents Chemother. 54:5316–5322.

20.

Vidaillac

, Benichou

, and Duval

R.E.

. 2012. In vitro synergy of colistin combinations against colistin-resistant Acinetobacter baumannii, Pseudomonas aeruginosa, and Klebsiella pneumoniae isolates. Antimicrob. Agents Chemother. 56:4856–4861.