Biomedical Research

Journal Banner

Determination of MIC distribution of colistin, fosfomycin, and tigecyclin antibiotics against carbapenem resistant enterobacteriaceae

Serap Süzük1*, Banu Kaşkatepe2 and Havva Avcıküçük3

1Public Health Institution of Turkey Microbiology Reference Laboratories Ankara, Turkey

2Department of Pharmaceutical Microbiology, Ankara University Faculty of Pharmacy, Ankara, Turkey

3Clinical Microbiology Lab, 29 Mayıs Statement Hospital, Ankara, Turkey

*Corresponding Author:
Serap Süzük
Public Health Institution of Turkey Microbiology Reference Laboratories
Adnan Saygun Caddesi, Sıhhiye
Ankara, Turkey

Accepted date: January 31, 2017

Visit for more related articles at Biomedical Research


Introduction: There are limited treatments options for carbapenem resistant Enterobacteriaceae (CRE) species and especially colistin, fosfomycin and tigecyclin are used for treatment. Therefore, in this study, we aimed for determining the rates of antibiotic susceptibility against colisting, fosfomycin and tigecyclin in CRE isolates with microdilution method.

Material and Method: In our study, a total of 63 isolates, which were isolated from blood, urine and tracheal aspirate samples. Bacteria identification of the isolates was made with MALDI-TOF MS (Bruker Biotyper; Germany) device. Imipenem, meropenem and ertapenem susceptibilities were determined with the gradient strip test (Liofilchem, Italy) method and genotypic properties were determined with the PCR method. Colistin (0.06-32 μg/ml), fosfomycin (0.06-32 μg/ml) and tigecyclin (0.25-256 μg/ml) M?K values were determined with the microdilution method.

Results: Of the isolates included in the study, 60 isolates (95.24%) were defined as Klebsiella pneumoniae, 2 (3.17%) as Escherichia coli and 1 (1.59%) as Enterobacter cloacae. 57 isolates (90.48%) had OXA-48, 4 (6.35%) had NDM and 2 (3.17%) had OXA-48 and NDM resistant genotypes. M?K50 values for colistin, fosfomycin and tigecyclin were 0.50, 4 and 0.25 M?K90 values were 1, 16 and 1, respectively. Two isolates (3.17%) had resistance to colistin, 23 (63.51%) to fosfomycin and 18 (28.57%) to tigecyclin.

Conclusion: Under the light of the findings of the study, we believe colistin, fosfomycin and tigecyclin can be alternatives for CRE treatment, however, local and national level surveillance studies are important and treatment protocols should be organized and dosage should be arranged based on this surveillance data.


Carbapenem resistant enterobacteriaceae, Colistin, Fosfomycin, Tigecyclin, MIC, Microdilution


Antimicrobial resistance is admitted to be one of the greatest threats to public health. It is a source of concern that the infections caused by enterobacteriaceae-producing carbapenemases are increasingly reported worldwide. They have been known as carbapenem resistant enterobacteriaceae (CRE), resistant to all or most β-lactam antibiotics added last line carbapenems [1]. Carbapenem antibiotics have been used to treat infections reasoned by multi drug resistant gram negative bacteria [2]. On the other hand, high frequency of antibiotic use, misuse of drugs, high rate of healthcareassociated infections, all cause multidrug resistant organisms and these bacteria have been spread out all around the world [3]. The two well-known infectious agents in carbapenem resistant enterobacteriaceae family are carbapenem resistant Klebsiella pneumonia and Escherichia coli [4].

Production of carbapenemases, which hydrolyze not only carbapenems but also all other beta lactam antibiotics, is mostly responsible for the resistance to carbapenamases. The most five common carbapenemases include imipenemase (IMP) type, veronica integronmetallo beta lactamases (VIM) type, Klebsiella pneumonia carbapenemase (KPC) type, New Delhi metallo beta lactamases-1 (NDM-1) and also oxacillinase 48 (OXA 48) which is the most common enzyme in Turkey [5]. Susceptibility to carbapenems is detected by performing phenotypic tests which are enough for antimicrobial susceptibility categorization; nevertheless, it is vital to clarify the detection of resistance mechanism for infection control and public health [6].

Infections caused by CRE have high mortality rates due to very limited available treatment options. High dose carbapenem, polymixin B, colistin, tigecycline, fosfomycin or aminoglyside can be called as preferable CRE treatment options [7]. The treatment of serious CRE infections should base on Minimal Inhibitor Concentration (MIC) of antibiotic susceptibility results [8]. In this study, we aimed to determine the susceptibility of colistin, tigecycline and fosfomycin in CRE.

Materials and Method

Setting and study design

This is a prospective study that included consecutive inpatients of any age and sex with any infection by any CRE in a period between August 1st 2015-April 30th 2016 in Medical Microbiology Laboratory of 29 May?s Statement Hospital, Ankara. In this period, a total of 63 carbapenem resistant enterobacteriaceae species were isolated from blood, urine and tracheal aspirate. Clinical samples were analyzed with standard microbiology laboratory techniques. Only one isolate per patient got involved.

Bacterial identification and phenotypic analysis of carbapenems

All bacterial isolates were identified by MALDI-TOF MS (Bruker Biotyper; Bruker Daltonics, Bremen, Germany). Antibiotic susceptibility tests and extended-spectrum β- lactamases (ESBL) of isolates were analyzed by the BD Phoenix automated system (BD Diagnostics, USA). Minimal inhibitory concentration of ertapenem, imipenem and meropenem were also tested by gradient strip test (Liofilchem, Roseto Degli Abruzzi, Italy). Susceptibility test results of all antibiotics were interpreted according to the EUCAST 2016 break point values [9].

Genotypic methods for detecting carbapanemase gene

A total of 63 Enterobacteriaceae isolates (blaOXA-48, blaVIM, blaIMP, blaNDM-1, blaKPC-2 carrying strains) that showed decreased sensitivity to at least one carbapenem (ertapenem, imipenem or meropenem), and carriage of carbapenemase gene confirmed by polymerase chain reaction (PCR), were included in the study [10-14].

Determination of MIC by microbroth dilution

The MICs of colistin, tygecyclin, fosfomycin were determined by the broth micro-dilution method with cation-adjusted Muller-Hinton broth (Oxoid, Code: CM0405, UK) according to Clinical Laboratory Standard Institute (CLSI) guidelines [15-17]. Colistin sulfate (Sigma-Aldrich, St. Louis, MO, USA), tygecycline (Sigma-Aldrich, St. Louis, MO, USA) and fosfomycin (Sigma-Aldrich, St. Louis, MO, USA) were tested over a range of dilutions (0.06-32μg/ml), (0.06-32μg/ml) and (0.25–256 μg/ml) respectively. One hundred microliters of graded concentrations of antibiotics, freshly prepared, were added 96-well U bottom microplates. Bacterial suspensions, which prepared from bacteria grown on non-selective culture media, were added microplates. The microplates were incubated for 24 h at 37°C in ambient air. The breakpoints of the European Committee on Antimicrobial Susceptibility Testing (EUCAST) were used for the results as a reference [9].

Escherichia coli ATCC 25922 strain was used as quality control.


Clinical samples for the isolates included in the study and distribution of isolated bacteria are provided in Table 1. 49.21% of the isolates are from urine sample. The most frequently isolated isolate is K. pneumoniae. 57 isolates (90.48%) had OXA-48, 4 (6.35%) had NDM and 2 (3.17%) had OXA-48 and NDM resistant genotypes. All of the isolates with NDM as well as OXA-48 and NDM combination were identified in K. pneumoniae isolates. M?K50 values for colistin, fosfomycin and tigecyclin were 0.50 μg/ml, 4 μg/ml and 0.25 μg/ml M?K90 values were 1 μg/ml, 16 μg/ml and 1 μg/ml, respectively. Two isolates (3.17%) had resistance to colistin, 23 (63.51%) to fosfomycin and 18 (28.57%) to tigecyclin. It was determined that M?K50 and M?K90 values of the isolates were resistant to all antibiotics (Table 2). Susceptibility of the isolates to antibiotics is provided in Table 3. Both of the colistin resistant isolates were K. pneumoniae.

Clinical samples K. pneumoniae E. coli E. cloace Total
Tracheal aspirates 5 1 0 6 (9.52)
Blood 24 1 1 26 (41.27)
Urine 31 0 0 31 (49.21)
Total 60 (95.24) 2 (3.17) 1 (1.59) 63 (%100)

Table 1. The source and distribution of isolates in clinical samples (n=63).

Antibiotic MIC range MIC50 (μg/ml) Interpretation* MIC90 (μg/ml) Interpretation*
Colistin <0.06-> 32 0.50 Susceptibility 1 Susceptibility
Fosfomycin < 0.25-> 256 4 Susceptibility 16 Susceptibility
Tigecyclin <0.06-> 32 0.25 Susceptibility 1 Susceptibility

Table 2. Determination and interpretation of the MIC values of the antibiotics tested for CRE (n=63).

Bacteria Colistin Fosfomycin Tigecycline
S (%) R (%) S (%) R (%) S (%) R (%)
K. pneumoniae 58 (96.67) 2 (3.33) 38 (63.33) 22 (36.67) 43 (71.67) 17 (28.33)
E. coli 2 (100.00) 0.00 2 (100.00) 0.00 2 (100.00) 0.00
Enterobacter cloace 1 (100.00) 0.00 0.00 1 (100.00) 0.00 1 (100.00)
Total 61 (96.83) 2 (3.17) 40 (63.49) 23 (36.51) 45 (71.43) 18 (28.57)

Table 3. Determination of the MIC values antibiotics tested for isolates (n=63).


CRE has been a major public health threat worldwide since the 1990s. CRE infections both cause deterioration of clinical picture and economic losses [18,19]. OXA-48 type carbapenemases that were isolated for the first time in Turkey, are seen commonly in our country. Bacteria with this enzyme have multiple drug resistance, and they have limited the use of antibiotics for the treatment [20]. In this present study, extended spectrumbeta-lactamase positivity was found in 80.95% of isolates along with carbapenemases enzyme positivity and multi drug resistance.

NDM-1 is a type of carbapenemases seen in Saudi Arabia, Kuwait, Amman, United Arab Emirates and Middle East countries [21]. In our country, NDM-1 and NDM-1OXA-48 associations began to be seen in recent years by the reason of developing and spread of medical tourism [22]. We determined 6.35% and 3.17% NDM-1 and NDM-1 and OXA-48 positivity respectively in this present study.

The situation is worrying for CRE infections because of the high mortality rates and treatment options of these infections are limited due to their multi-drug resistance. Therefore, clinicians have included old antibiotics, which are not preferred because of the toxic effects, to the treatment protocols [7]. Colistin, often preferred for treatment of multidrug resistant Gram negative bacteria, has usage restrictions due to its nephrotoxicity [23]. Tigecycline is another antibiotic that is used for CRE infections because of wide tissue distribution. However, FDA defined tigecycline monotherapy as "Boxed Warning" [24]. The other antibiotic that has antimicrobial activity against CRE agents is fosfomycin. Although fosfomycin is often reported as effective in acute uncomplicated urinary tract infection, its intravenous use is preferred for CRE infections [25]. MIC value of agents in CRE infections is important to adjust the dose for treatment protocols. In this study, it was aimed to determine antimicrobial susceptibility of colistin, tigecycline and fosfomycin antibiotics that are preferred for CRE infections, with the micro dilution method.

Colistin, tigecycline and fosfomycin are used as primary treatment options. KPC-producing K. pneumoniae treatment algorithms depend on the site of infection. Colistin is the first option especially in blood, lungs and gastrointestinal tract infections and also preferred for urinary tract infections as a possible therapeutic agent. Tigecycline is the first option in the gastrointestinal tract infections and potential therapeutic agents for the blood and lung infections. Fosfomycin, while a first option for urinary tract infections, is among the agents that may be preferred for other sites of infection. MIC levels of antibiotics are important in treatment algorithm because algorithm is determined based on these MIC levels [7].

A total of 96.83% out of 63 CRE isolates, included the study, were found susceptible to colistin, and followed by tigecycline (71.43%) and fosfomycin (63.49%). Colistin was found the most effective antibiotic for our isolates among the tested antibiotics. Only two K. pneumoniae isolates were identified as resistant to colistin in all isolates. One E. cloaca isolate included in the study was resistant to both fosfomycin and tigecycline. Two E. coli strains were found to be sensitive to all antibiotics tested. In a study involving seven Latin American countries, CRA species, isolated from the hospitals of these countries, have been assessed and were found susceptible 96% and 79% of the isolates to colistin and tigecycline respectively [26]. In South Indian region, antibiotic susceptibility of multidrug-resistant E. coli and K. pneumoniae isolates against various antibiotics were determined by the broth microdilution method. While colistin MIC50 and MIC90 values of E. coli isolates were determined as 0.25 μg/ml and 0.5 μg/ml, for K. pneumoniae they were found as 0.5 μg/ml and 1 μg/ml. Also fosfomycin MIC50 and MIC90 values were found 0.5 μg/ml and 4 μg/ml for E. coli, 8 μg/ml and 32 μg/ml for K. pneumoniae [27]. Similarly to Indian data, in our study, colistin MIC50 and MIC90 values were found 0.5 μg/ml and 1 μg/ml, respectively. By contrast, with MIC50 and MIC90 values differ greatly for fosfomycin.

In 2011, in a study conducted with 81 CRE isolates, antibiotic susceptibilities of colistin, fosfomisin and tigecycline were determined 92.6%, 60.5% and 46.9% respectively. Tigecycline resistant isolates have been reported to be between K. pneumoniae and enterobacter spp [28]. Likewise, tigecycline resistant isolates were detected in K. pneumoniae and E. cloace in our study.

In a surveillance study between 2010-2013 involving hospitals in Europe, it was indicated that CRE isolates were observed to be more often in Poland, Greece and Romania. Also 73.9% and 88.6% of the isolates were reported to be effective with colistin and tigecycline, respectively [29]. In a study evaluating the fecal samples of 100 hospitalized patients colonized with CRE in the intensive care unit, all of the isolates were found susceptible to colistin (MIC range 0,125-1 mg/l) and tigecycline (MIC range 0.06- 1.5 mg/l) [30]. However, the assessment of treatment success in the CRE infection may provide more positive results for the use of these antibiotics. We believe that consideration of the MIC values in the treatments will increase the chance of successful treatment. In our study, we determined that MIC values of isolates approached to the clinical limits. In this case, it should be borne in mind that normal doses of antibiotics may not be the treat levels in plasma.

In accordance with the data obtained from the study, we conclude that colistin, fosfomycin and tigecycline in CRE treatment may be a good option, but for this infection, local and national surveillance studies are important and regulation of treatment protocols and dose adjustment should be done based on surveillance data.