- Email: [email protected]
Colistin resistance among blood culture isolates at a tertiary care centre in Hungary
Accepted Manuscript Title: Colistin resistance among blood culture isolates at a tertiary care centre in Hungary Authors: Emese Juh´asz, Mikl´os Iv´an, Eszter Pint´er, J´ulia Pongr´acz, Katalin Krist´of PII: DOI: Reference:
S2213-7165(17)30147-9 http://dx.doi.org/10.1016/j.jgar.2017.08.002 JGAR 471
To appear in: Received date: Revised date: Accepted date:
8-6-2017 31-7-2017 2-8-2017
Please cite this article as: Emese Juh´asz, Mikl´os Iv´an, Eszter Pint´er, J´ulia Pongr´acz, Katalin Krist´of, Colistin resistance among blood culture isolates at a tertiary care centre in Hungary (2010), http://dx.doi.org/10.1016/j.jgar.2017.08.002 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Colistin resistance among blood culture isolates at a tertiary care centre in Hungary
Emese Juhász, Miklós Iván, Eszter Pintér, Júlia Pongrácz, Katalin Kristóf Diagnostic Laboratory of Clinical Microbiology, Institute of Laboratory Medicine, Semmelweis University
Corresponding author: Emese Juhász E-mail: [email protected] Telephone: 36 1 459 1500 / 62106 Address: Semmelweis Univesity, Institute of Laboratory Medicine, Nagyvárad tér 4, 14. emelet, 1089, Budapest, Hungary
Highlights
The first mcr-1 positive E. coli strain was found in Hungary.
The prevalence of colistin resistance among enterobacterial strains isolated from blood cultures was 0.6%, but colistin resistant subpopulations were found in 17% of isolates.
The prevalence of colistin resistance among P. aeruginosa and A. baumannii strains was 1.3% and 2.6%, respectively, but MDR strains with colistin resistant subpopulations were revealed.
All S. maltophilia isolates were resistant to colistin at MIC50 64 mg/l.
Abstract Objectives. The emergence of colistin resistance has been detected worldwide in recent years. While colistin susceptibility has been tested in carbapenem resistant Enterobacteriaceae, MDR Pseudomonas spp. and Acinetobacter spp. during routine laboratory practice, the overall rate of colistin resistance was unknown in our centre. The aim of this retrospective study was to reveal the prevalence of colistin resistance among clinically significant blood culture isolates in two different periods (2010-2011 and 2016) in our laboratory. Methods. Consecutive non-duplicate strains (n=779) were screened for colistin resistance with agar plates containing 4 mg/l colistin. Strains cultured on colistin plates were further examined: the MICs of the colistin tolerant subcultures and the original cultures were determined in parallel by the broth microdilution method. Screening for mcr-1 mediated colistin resistance was performed by PCR. Results. The rate of colistin resistance was 0.6%, 1.3% and 2.6% in Enterobacteriaceae, Pseudomonas spp. and Acinetobacter spp., respectively. Seven colistin resistant strains were found, among them an mcr-1 positive E. coli, isolated from the blood sample of a hematooncologic patient in 2011. Colistin resistant subpopulations were found in 17%, 27% and 20% of isolates, respectively, with low frequency. All S. maltophilia isolates were resistant to colistin. Conclusions. The low prevalence of colistin resistance was in concordance with European data. The prevalence of heteroresistance was significantly higher, but the clinical significance of the phenomenon is unclear. We have identified the first mcr-1 positive E. coli strain in Hungary. Mcr-1 has been in Hungary since 2011, but has not expanded yet.
Keywords: colistin; antibiotic resistance; antibiotic heteroresistance; mcr-1
Introduction Colistin is one of the last-resort antibiotics against multi-drug resistant (MDR) bacteria. The emergence of colistin resistance has been detected worldwide in recent years (1, 2). Different complex mechanisms lead to the loss or modification of lipopolysaccharide production in Gram negative bacteria, resulting in colistin resistance. Resistance can develop during colistin therapy. However, there is also colistin resistance without prior colistin exposure, partly due to cross-resistance between colistin and lysozyme or other human antimicrobial peptides (1). Nosocomial transmission of resistant bacteria is another way by which the prevalence of colistin resistance can increase. Plasmid-borne mcr-mediated resistance is the dominant risk factor for further spread of colistin resistance. Several cases of human colonisation or infection caused by mcr-1 positive bacteria have already been described in Europe (2). Colistin therapy is usually prescribed in hospital settings either against carbapenemresistant Enterobacteriaceae or against MDR Pseudomonas spp. or Acinetobacter spp. Many diagnostic laboratory investigates colistin susceptibility just of the above mentioned bacteria, but not that of highly susceptible isolates. However, colonies in the clear zone of inhibition of colistin disk or gradient diffusion test strip should raise the suspicion of colistin heteroresistance. Methodical difficulties regarding colistin susceptibility testing have had a negative impact on the accuracy of detected resistance rates. Before the guidance for colistin susceptibility testing was published by EUCAST in 2016, the gradient diffusion method was possibly the most commonly used method in diagnostic laboratories, as it was in ours (3). False susceptibility obtained by this method and the consequent very major errors in interpretation have been discussed in several publications.
The phenomenon of colistin ‘clonal heteroresistance’ described by Band et al. can remain undetected during routine diagnostic processes and can lead to treatment failure and even lethal infection (4). Therefore, the detection of colistin resistant sub-populations of MDR isolates may be necessary to prevent insufficient colistin therapy and to support physicians ’ in the decision to change colistin monotherapy to colistin containing combination therapy. The aim of this retrospective study was to reveal the rate of colistin resistance among clinically significant bacterial isolates in two different periods (2010-2011 and 2016). A further objective was to screen for mcr-1 among colistin resistant strains.
Materials and methods Only clinically significant, invasive strains isolated from blood cultures were involved in this study. Isolates were collected in Diagnostic Laboratory of Clinical Microbiology, Institute of Laboratory Medicine, Semmelweis University (Budapest, Hungary). They were collected in two periods: 2010-2011 and 2016 January-December. After the analysis of the total number and the species distribution of blood culture isolates in 2016, the same number of isolates from each species was tested from 2010-2011. Pantoea agglomerans and Aeromonas veronii isolates were exceptions, as these bacteria were not isolated from blood cultures in 2010-2011, and in case of Achromobacter xylosoxidans, only two isolates were cultured. The total number of consecutive, non-duplicate isolates was 779. Identification of isolates was performed by MALDI-TOF MS (Bruker Daltonics), AST was performed by the disc diffusion method according to EUCAST rules. Screening for colistin resistance was performed on Mueller-Hinton (MH) agar plates containing 4 mg/l colistin sulphate (Sigma Aldrich). From overnight broth cultures, 10 µl were inoculated on plates. Plates were incubated for 48 h at 37°C. Strains cultured on colistin containing plates
were further examined: the MICs of the colistin tolerant subcultures and the original culture were determined in parallel. The MIC was tested using the broth microdilution method in cation-adjusted MH broth as recommended by EUCAST. The final inoculum was 5x105 CFU/ml of bacteria in 96-well polystyrene plates. Control strains were P. aeruginosa ATCC 27853, E. coli ATCC 25922 and mcr-1 positive, colistin resistant E. coli OKICR1. The prevalence of resistant subpopulations was determined by growing the strains in MH broth overnight at 37°C, followed by 10-fold serial dilution in saline and plating on agar plates with or without colistin (4 mg/l). Colonies were counted after incubation at 37°C for 48 h. The frequency of colistin resistance was calculated by dividing the CFU/ml value of the resistant sub-population growing on the colistin plate by the CFU/ml value of the total population growing on the colistin-free plate (5). To screen for mcr-1 mediated colistin resistance, PCR was performed on resistant and heteroresistant isolates, as described previously (6). In the positive isolate, additional PCRs with primers C1-F CTGTGCCGTGTATGTTCAGC, C1-R GGTGCGGTCTTTGACTTTGT and C2-F TGACACTTATGGCACGGTCT, C2-R TTTCTTGGTATTTGGCGGTA were done. Targeting the extremities of mcr-1 gene, the previously described CLR5-F1 primer and ER-F GCATTCATCCGCTGATTTCT with ER-R TGACTGTGCTCAAGGGTCAG primers were used (7). PCR products were sequenced bi-directionally by the Sanger method. Plasmid extraction was performed using SV Miniprep Kit (Promega). An mcr-1 positive E. coli strain OKICR1 was used as positive control.
Results Among the 504 enterobacterial isolates tested, three were resistant to colistin (0.6%). All of them were isolated in 2010-11 (Table 1). The MIC values of colistin resistant E. coli,
Enterobacter cloacae and Klebsiella pneumoniae were 8 mg/l, 8 mg/l and >64 mg/l, respectively. They were ESBL producers; the K. pneumoniae and E. cloacae isolates were resistant to ertapenem, but sensitive to imipenem and meropenem, while the E. coli isolate was susceptible to all carbapenems. Colistin tolerant subpopulations were found in 17% of enterobacterial isolates. The prevalence of isolates with colistin resistant subpopulations was 63% (n=44) in 2010-2011 and 34% (n=24) in 2016 in Klebsiella spp. It was 20% (n=5) and 12% (n=3) in Enterobacter spp. and 2.7% (n=6) and 4% (n=4) in E. coli isolates, respectively. One Citrobacter freundii isolate showed the same features. In E. coli and Klebsiella spp. isolates, the frequency of colistin resistant subpopulations varied between 1.25x10 -6 % and 1.1x10-3 %, whereas in Enterobacter spp. isolates this was 6x10-6 % to 10.3%. Colistin resistant sub-populations of Klebsiella spp. showed the highest MIC values at >64 mg/l, whereas in E. coli MIC values were between 4-32 mg/l. None of the isolates with colistin heteroresistance were resistant to carbapenems. Among these isolates, 31% were ESBL producers and 53% had wild type antibiotic resistance. Two resistant P. aeruginosa (1.3%) with MIC 4 mg/l and two resistant A. baumannii (2.6%) strains with MIC 4 and 8 mg/l were revealed. Only one of the A. baumannii strains was MDR. Colistin resistant subpopulations were found in 27% of Pseudomonas spp. and 19.7% of Acinetobacter spp. isolates. Of these heteroresistant isolates, 37% were wild type and 28% had the MDR resistance phenotype. Isolates non-susceptible to at least one agent in ≥3 antimicrobial groups were defined as MDR (8). The frequency of colistin resistant subpopulations varied between 1.8x10-5 % and 3x10-2%. All S. maltophilia isolates were resistant to colistin with MIC50 64 mg/l, as were Achromobacter spp. isolates. Only one strain, an E. coli isolated from the blood sample of a hemato-oncology patient in 2011 was positive for mcr-1. The full mcr-1 cds, 1626 bp was sequenced. It was identical to
the reference sequence (accession No. KP347127). The plasmid extract of the strain was negative for mcr-1.
Discussion The rate of colistin resistance among Gram negative strains isolated from blood cultures in two different periods was examined in our study. A European multicentre study reported low colistin resistance prevalence values (range 0.2-10.9%) in Enterobacteriaceae and P. aeruginosa in 2011-2012 (9). Our data show resistance rates in the same range (0.6-4%). Seven colistin resistant strains were found among Enterobacteriaceae, Pseudomonas spp. and Acinetobacter spp., of which one was mcr-1 positive E. coli. At the time of isolation, only one of the resistant strains was tested against colistin, an MDR A. baumannii. By the gradient diffusion method, an MIC value of 1 mg/l was determined and it was interpreted as susceptible to colistin. As its MIC was 4 mg/l by the broth microdilution method, we can declare retrospectively that a major error occurred in 2011. Despite this, clinical data revealed that the patient was successfully treated with colistin. The rate of clonal heteroresistance was significantly higher than those for resistance. In Klebsiella spp. was found to be the highest (48%), which is in concordance with previous studies (10). The frequencies of colistin resistant subpopulations were very low. Enterobacter spp. isolates were the exception, with comprising 10% of resistant subpopulations. The large portion of resistant subpopulations of E. cloacae strains is in concordance with previous reports (5). By the commonly applied gradient or disc diffusion tests, only high-rate heteroresistance can be detected. As Enterobacter spp. isolates with colistin heteroresistance frequently have high-rate resistance, they are detected more easily in routine laboratory practice. On agar plates with 10 mg/l colistin, the survival rate of A. baumannii colonies was 64% according to Hong et al. (11). This is significantly higher than what we observed (17%) on 4
mg/l colistin plates. As the peak serum concentration of colistin is 3.5-6.5 mg/l and steady-state level is 0.68-5.14 mg/l, plates with 4 mg/l fit with in vivo conditions when strains isolated from bacteraemia are used for examination (2, 12). Further studies involving respiratory tract isolates should examine heteroresistance at higher colistin concentrations as well, considering its accumulation in lung tissue and the possibility of inhalation therapy. Despite the fact that a resistant minority of a bacterial population can influence the outcome of an infection and treatment, the clinical significance of colistin heteroresistance is ambiguous (4). Cases reporting the selection of colistin resistant derivatives of heteroresistant strains during colistin treatment have been published, but the critical frequency rate of the colistin resistant subpopulation sufficient to replicate and cause in vivo effects it is not yet clarified (13). Colistin susceptibility rates of S. maltophilia vary significantly in previous reports, mainly depending on testing methods and applied clinical break points (14). Without international guidelines on how to interpret susceptibility to colistin in non-fermenting Gram negative rods other than Pseudomonas spp. and Acinetobacter spp., the evaluation of results is limited. According to our results, colistin cannot be considered a therapeutic option as monotherapy against S. maltophilia or Achromobacter spp. Neither the rate of resistance, nor that of heteroresistance differed significantly in the two examined periods. Transposition of mcr-1 to the chromosome is a possible mechanism. A sequence type ST410 E. coli strain harbouring chromosomal blaCTX-M-15 and an ISApl1-mcr-1 cassette from 2013 was detected in Germany (15). The chromosomal location of mcr-1 may be an explanation as to why the plasmid extract of our isolate was negative, while that of the control strain was positive for mcr-1. However, mcr-1 usually coexists with other resistance genes such as ESBL genes on plasmids, which can reach sizes of 251 kb, and are thus more difficult to extract (2).
Colistin resistance was accompanied by a significant fitness cost in K. pneumoniae and A. baumannii (16, 17). As a limitation, it has to be considered that the low rate of resistance might be partly the consequence of the invasive feature of the studied strains. From an epidemiological standpoint, the monocentric nature of the study is a significant limitation. To the best of our knowledge, this is the first report of mcr-1 mediated colistin resistance in Hungary. Although mcr-1 was already in Hungary in 2011, it did not expand. The rate of colistin resistance is still low. Despite our findings, permanent surveillance for colistin resistance is recommended in order to reserve our antibiotic agent of last resort. The susceptibility to colistin of non-fermenting Gram negative rods other than Pseudomonas spp. or Acinetobacter spp. should be revised and interpretation should be guided by international committees. A global guideline for the accurate detection of colistin heteroresistance in daily clinical microbiology practice, as well as the interpretative criteria of the results and their clinical utility are lacking, but urgently required.
Acknowledgement The mcr-1 positive E. coli strain OKICR1 was kindly provided by Ákos Tóth (National Public Health Institute, Hungary).
Declarations Funding: No funding Competing Interests: None Ethical Approval: Not required.
References 1.Olaitan AO, Morand S, Rolain JM. Emergence of colistin-resistant bacteria in humans without colistin usage: a new worry and cause for vigilance. International journal of antimicrobial agents. 2016;47(1):13. 2.Poirel L, Jayol A, Nordmann P. Polymyxins: Antibacterial Activity, Susceptibility Testing, and Resistance Mechanisms Encoded by Plasmids or Chromosomes. Clinical microbiology reviews. 2017;30(2):557-96. 3.EUCAST. Guidance documents in susceptibility testing. Recommendations for colistin (polymyxin E) MIC testing. 2016. 4.Band VI, Crispell EK, Napier BA, Herrera CM, Tharp GK, Vavikolanu K, et al. Antibiotic failure mediated by a resistant subpopulation in Enterobacter cloacae. Nature microbiology. 2016;1(6):16053. 5.Napier BA, Band V, Burd EM, Weiss DS. Colistin heteroresistance in Enterobacter cloacae is associated with cross-resistance to the host antimicrobial lysozyme. Antimicrobial agents and chemotherapy. 2014;58(9):5594-7. 6.Liu YY, Wang Y, Walsh TR, Yi LX, Zhang R, Spencer J, et al. Emergence of plasmid-mediated colistin resistance mechanism MCR-1 in animals and human beings in China: a microbiological and molecular biological study. The Lancet Infectious diseases. 2016;16(2):161-8. 7.Cannatelli A, Giani T, Antonelli A, Principe L, Luzzaro F, Rossolini GM. First Detection of the mcr-1 Colistin Resistance Gene in Escherichia coli in Italy. Antimicrobial agents and chemotherapy. 2016;60(5):3257-8. 8.Magiorakos AP, Srinivasan A, Carey RB, Carmeli Y, Falagas ME, Giske CG, et al. Multidrug-resistant, extensively drug-resistant and pandrug-resistant bacteria: an international expert proposal for interim standard definitions for acquired resistance. Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases. 2012;18(3):26881. 9.Giske CG. Contemporary resistance trends and mechanisms for the old antibiotics colistin, temocillin, fosfomycin, mecillinam and nitrofurantoin. Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases. 2015;21(10):899-905. 10.Ah YM, Kim AJ, Lee JY. Colistin resistance in Klebsiella pneumoniae. International journal of antimicrobial agents. 2014;44(1):8-15. 11.Hong YK, Lee JY, Wi YM, Ko KS. High rate of colistin dependence in Acinetobacter baumannii. The Journal of antimicrobial chemotherapy. 2016;71(8):2346-8. 12.Gregoire N, Mimoz O, Megarbane B, Comets E, Chatelier D, Lasocki S, et al. New colistin population pharmacokinetic data in critically ill patients suggesting an alternative loading dose rationale. Antimicrobial agents and chemotherapy. 2014;58(12):7324-30. 13.El-Halfawy OM, Valvano MA. Antimicrobial heteroresistance: an emerging field in need of clarity. Clinical microbiology reviews. 2015;28(1):191-207. 14.Moskowitz SM, Garber E, Chen Y, Clock SA, Tabibi S, Miller AK, et al. Colistin susceptibility testing: evaluation of reliability for cystic fibrosis isolates of Pseudomonas aeruginosa and Stenotrophomonas maltophilia. The Journal of antimicrobial chemotherapy. 2010;65(7):1416-23. 15.Falgenhauer L, Waezsada SE, Gwozdzinski K, Ghosh H, Doijad S, Bunk B, et al. Chromosomal Locations of mcr-1 and bla CTX-M-15 in Fluoroquinolone-Resistant Escherichia coli ST410. Emerging infectious diseases. 2016;22(9):1689-91. 16.Beceiro A, Moreno A, Fernandez N, Vallejo JA, Aranda J, Adler B, et al. Biological cost of different mechanisms of colistin resistance and their impact on virulence in Acinetobacter baumannii. Antimicrobial agents and chemotherapy. 2014;58(1):518-26. 17.Choi MJ, Ko KS. Loss of hypermucoviscosity and increased fitness cost in colistin-resistant Klebsiella pneumoniae sequence type 23 strains. Antimicrobial agents and chemotherapy. 2015;59(11):6763-73.
Table 1. Distribution of colistin resistant and heteroresistant (S/R) isolates cultured from blood specimens
E. coli Klebsiella spp. Enterobacter spp. Citrobacter spp. Pantoea spp. Salmonella spp. Pseudomonas spp. Acinetobacter spp. Stenotrophomonas spp. Achromobacter spp. Aeromonas spp.
Number of isolates / period 146 70 25 8 2 1 76 38 20 3 1
2010-2011 R S/R 1 6 1 44 1 5 0 0 0 0 0 22 1 7 20 0 2 0 -
2016 R 0 0 0 0 0 0 2 1 20 3 0
S/R 4 24 3 1 0 0 19 8 0 0 0
S2213-7165(17)30147-9 http://dx.doi.org/10.1016/j.jgar.2017.08.002 JGAR 471
To appear in: Received date: Revised date: Accepted date:
8-6-2017 31-7-2017 2-8-2017
Please cite this article as: Emese Juh´asz, Mikl´os Iv´an, Eszter Pint´er, J´ulia Pongr´acz, Katalin Krist´of, Colistin resistance among blood culture isolates at a tertiary care centre in Hungary (2010), http://dx.doi.org/10.1016/j.jgar.2017.08.002 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Colistin resistance among blood culture isolates at a tertiary care centre in Hungary
Emese Juhász, Miklós Iván, Eszter Pintér, Júlia Pongrácz, Katalin Kristóf Diagnostic Laboratory of Clinical Microbiology, Institute of Laboratory Medicine, Semmelweis University
Corresponding author: Emese Juhász E-mail: [email protected] Telephone: 36 1 459 1500 / 62106 Address: Semmelweis Univesity, Institute of Laboratory Medicine, Nagyvárad tér 4, 14. emelet, 1089, Budapest, Hungary
Highlights
The first mcr-1 positive E. coli strain was found in Hungary.
The prevalence of colistin resistance among enterobacterial strains isolated from blood cultures was 0.6%, but colistin resistant subpopulations were found in 17% of isolates.
The prevalence of colistin resistance among P. aeruginosa and A. baumannii strains was 1.3% and 2.6%, respectively, but MDR strains with colistin resistant subpopulations were revealed.
All S. maltophilia isolates were resistant to colistin at MIC50 64 mg/l.
Abstract Objectives. The emergence of colistin resistance has been detected worldwide in recent years. While colistin susceptibility has been tested in carbapenem resistant Enterobacteriaceae, MDR Pseudomonas spp. and Acinetobacter spp. during routine laboratory practice, the overall rate of colistin resistance was unknown in our centre. The aim of this retrospective study was to reveal the prevalence of colistin resistance among clinically significant blood culture isolates in two different periods (2010-2011 and 2016) in our laboratory. Methods. Consecutive non-duplicate strains (n=779) were screened for colistin resistance with agar plates containing 4 mg/l colistin. Strains cultured on colistin plates were further examined: the MICs of the colistin tolerant subcultures and the original cultures were determined in parallel by the broth microdilution method. Screening for mcr-1 mediated colistin resistance was performed by PCR. Results. The rate of colistin resistance was 0.6%, 1.3% and 2.6% in Enterobacteriaceae, Pseudomonas spp. and Acinetobacter spp., respectively. Seven colistin resistant strains were found, among them an mcr-1 positive E. coli, isolated from the blood sample of a hematooncologic patient in 2011. Colistin resistant subpopulations were found in 17%, 27% and 20% of isolates, respectively, with low frequency. All S. maltophilia isolates were resistant to colistin. Conclusions. The low prevalence of colistin resistance was in concordance with European data. The prevalence of heteroresistance was significantly higher, but the clinical significance of the phenomenon is unclear. We have identified the first mcr-1 positive E. coli strain in Hungary. Mcr-1 has been in Hungary since 2011, but has not expanded yet.
Keywords: colistin; antibiotic resistance; antibiotic heteroresistance; mcr-1
Introduction Colistin is one of the last-resort antibiotics against multi-drug resistant (MDR) bacteria. The emergence of colistin resistance has been detected worldwide in recent years (1, 2). Different complex mechanisms lead to the loss or modification of lipopolysaccharide production in Gram negative bacteria, resulting in colistin resistance. Resistance can develop during colistin therapy. However, there is also colistin resistance without prior colistin exposure, partly due to cross-resistance between colistin and lysozyme or other human antimicrobial peptides (1). Nosocomial transmission of resistant bacteria is another way by which the prevalence of colistin resistance can increase. Plasmid-borne mcr-mediated resistance is the dominant risk factor for further spread of colistin resistance. Several cases of human colonisation or infection caused by mcr-1 positive bacteria have already been described in Europe (2). Colistin therapy is usually prescribed in hospital settings either against carbapenemresistant Enterobacteriaceae or against MDR Pseudomonas spp. or Acinetobacter spp. Many diagnostic laboratory investigates colistin susceptibility just of the above mentioned bacteria, but not that of highly susceptible isolates. However, colonies in the clear zone of inhibition of colistin disk or gradient diffusion test strip should raise the suspicion of colistin heteroresistance. Methodical difficulties regarding colistin susceptibility testing have had a negative impact on the accuracy of detected resistance rates. Before the guidance for colistin susceptibility testing was published by EUCAST in 2016, the gradient diffusion method was possibly the most commonly used method in diagnostic laboratories, as it was in ours (3). False susceptibility obtained by this method and the consequent very major errors in interpretation have been discussed in several publications.
The phenomenon of colistin ‘clonal heteroresistance’ described by Band et al. can remain undetected during routine diagnostic processes and can lead to treatment failure and even lethal infection (4). Therefore, the detection of colistin resistant sub-populations of MDR isolates may be necessary to prevent insufficient colistin therapy and to support physicians ’ in the decision to change colistin monotherapy to colistin containing combination therapy. The aim of this retrospective study was to reveal the rate of colistin resistance among clinically significant bacterial isolates in two different periods (2010-2011 and 2016). A further objective was to screen for mcr-1 among colistin resistant strains.
Materials and methods Only clinically significant, invasive strains isolated from blood cultures were involved in this study. Isolates were collected in Diagnostic Laboratory of Clinical Microbiology, Institute of Laboratory Medicine, Semmelweis University (Budapest, Hungary). They were collected in two periods: 2010-2011 and 2016 January-December. After the analysis of the total number and the species distribution of blood culture isolates in 2016, the same number of isolates from each species was tested from 2010-2011. Pantoea agglomerans and Aeromonas veronii isolates were exceptions, as these bacteria were not isolated from blood cultures in 2010-2011, and in case of Achromobacter xylosoxidans, only two isolates were cultured. The total number of consecutive, non-duplicate isolates was 779. Identification of isolates was performed by MALDI-TOF MS (Bruker Daltonics), AST was performed by the disc diffusion method according to EUCAST rules. Screening for colistin resistance was performed on Mueller-Hinton (MH) agar plates containing 4 mg/l colistin sulphate (Sigma Aldrich). From overnight broth cultures, 10 µl were inoculated on plates. Plates were incubated for 48 h at 37°C. Strains cultured on colistin containing plates
were further examined: the MICs of the colistin tolerant subcultures and the original culture were determined in parallel. The MIC was tested using the broth microdilution method in cation-adjusted MH broth as recommended by EUCAST. The final inoculum was 5x105 CFU/ml of bacteria in 96-well polystyrene plates. Control strains were P. aeruginosa ATCC 27853, E. coli ATCC 25922 and mcr-1 positive, colistin resistant E. coli OKICR1. The prevalence of resistant subpopulations was determined by growing the strains in MH broth overnight at 37°C, followed by 10-fold serial dilution in saline and plating on agar plates with or without colistin (4 mg/l). Colonies were counted after incubation at 37°C for 48 h. The frequency of colistin resistance was calculated by dividing the CFU/ml value of the resistant sub-population growing on the colistin plate by the CFU/ml value of the total population growing on the colistin-free plate (5). To screen for mcr-1 mediated colistin resistance, PCR was performed on resistant and heteroresistant isolates, as described previously (6). In the positive isolate, additional PCRs with primers C1-F CTGTGCCGTGTATGTTCAGC, C1-R GGTGCGGTCTTTGACTTTGT and C2-F TGACACTTATGGCACGGTCT, C2-R TTTCTTGGTATTTGGCGGTA were done. Targeting the extremities of mcr-1 gene, the previously described CLR5-F1 primer and ER-F GCATTCATCCGCTGATTTCT with ER-R TGACTGTGCTCAAGGGTCAG primers were used (7). PCR products were sequenced bi-directionally by the Sanger method. Plasmid extraction was performed using SV Miniprep Kit (Promega). An mcr-1 positive E. coli strain OKICR1 was used as positive control.
Results Among the 504 enterobacterial isolates tested, three were resistant to colistin (0.6%). All of them were isolated in 2010-11 (Table 1). The MIC values of colistin resistant E. coli,
Enterobacter cloacae and Klebsiella pneumoniae were 8 mg/l, 8 mg/l and >64 mg/l, respectively. They were ESBL producers; the K. pneumoniae and E. cloacae isolates were resistant to ertapenem, but sensitive to imipenem and meropenem, while the E. coli isolate was susceptible to all carbapenems. Colistin tolerant subpopulations were found in 17% of enterobacterial isolates. The prevalence of isolates with colistin resistant subpopulations was 63% (n=44) in 2010-2011 and 34% (n=24) in 2016 in Klebsiella spp. It was 20% (n=5) and 12% (n=3) in Enterobacter spp. and 2.7% (n=6) and 4% (n=4) in E. coli isolates, respectively. One Citrobacter freundii isolate showed the same features. In E. coli and Klebsiella spp. isolates, the frequency of colistin resistant subpopulations varied between 1.25x10 -6 % and 1.1x10-3 %, whereas in Enterobacter spp. isolates this was 6x10-6 % to 10.3%. Colistin resistant sub-populations of Klebsiella spp. showed the highest MIC values at >64 mg/l, whereas in E. coli MIC values were between 4-32 mg/l. None of the isolates with colistin heteroresistance were resistant to carbapenems. Among these isolates, 31% were ESBL producers and 53% had wild type antibiotic resistance. Two resistant P. aeruginosa (1.3%) with MIC 4 mg/l and two resistant A. baumannii (2.6%) strains with MIC 4 and 8 mg/l were revealed. Only one of the A. baumannii strains was MDR. Colistin resistant subpopulations were found in 27% of Pseudomonas spp. and 19.7% of Acinetobacter spp. isolates. Of these heteroresistant isolates, 37% were wild type and 28% had the MDR resistance phenotype. Isolates non-susceptible to at least one agent in ≥3 antimicrobial groups were defined as MDR (8). The frequency of colistin resistant subpopulations varied between 1.8x10-5 % and 3x10-2%. All S. maltophilia isolates were resistant to colistin with MIC50 64 mg/l, as were Achromobacter spp. isolates. Only one strain, an E. coli isolated from the blood sample of a hemato-oncology patient in 2011 was positive for mcr-1. The full mcr-1 cds, 1626 bp was sequenced. It was identical to
the reference sequence (accession No. KP347127). The plasmid extract of the strain was negative for mcr-1.
Discussion The rate of colistin resistance among Gram negative strains isolated from blood cultures in two different periods was examined in our study. A European multicentre study reported low colistin resistance prevalence values (range 0.2-10.9%) in Enterobacteriaceae and P. aeruginosa in 2011-2012 (9). Our data show resistance rates in the same range (0.6-4%). Seven colistin resistant strains were found among Enterobacteriaceae, Pseudomonas spp. and Acinetobacter spp., of which one was mcr-1 positive E. coli. At the time of isolation, only one of the resistant strains was tested against colistin, an MDR A. baumannii. By the gradient diffusion method, an MIC value of 1 mg/l was determined and it was interpreted as susceptible to colistin. As its MIC was 4 mg/l by the broth microdilution method, we can declare retrospectively that a major error occurred in 2011. Despite this, clinical data revealed that the patient was successfully treated with colistin. The rate of clonal heteroresistance was significantly higher than those for resistance. In Klebsiella spp. was found to be the highest (48%), which is in concordance with previous studies (10). The frequencies of colistin resistant subpopulations were very low. Enterobacter spp. isolates were the exception, with comprising 10% of resistant subpopulations. The large portion of resistant subpopulations of E. cloacae strains is in concordance with previous reports (5). By the commonly applied gradient or disc diffusion tests, only high-rate heteroresistance can be detected. As Enterobacter spp. isolates with colistin heteroresistance frequently have high-rate resistance, they are detected more easily in routine laboratory practice. On agar plates with 10 mg/l colistin, the survival rate of A. baumannii colonies was 64% according to Hong et al. (11). This is significantly higher than what we observed (17%) on 4
mg/l colistin plates. As the peak serum concentration of colistin is 3.5-6.5 mg/l and steady-state level is 0.68-5.14 mg/l, plates with 4 mg/l fit with in vivo conditions when strains isolated from bacteraemia are used for examination (2, 12). Further studies involving respiratory tract isolates should examine heteroresistance at higher colistin concentrations as well, considering its accumulation in lung tissue and the possibility of inhalation therapy. Despite the fact that a resistant minority of a bacterial population can influence the outcome of an infection and treatment, the clinical significance of colistin heteroresistance is ambiguous (4). Cases reporting the selection of colistin resistant derivatives of heteroresistant strains during colistin treatment have been published, but the critical frequency rate of the colistin resistant subpopulation sufficient to replicate and cause in vivo effects it is not yet clarified (13). Colistin susceptibility rates of S. maltophilia vary significantly in previous reports, mainly depending on testing methods and applied clinical break points (14). Without international guidelines on how to interpret susceptibility to colistin in non-fermenting Gram negative rods other than Pseudomonas spp. and Acinetobacter spp., the evaluation of results is limited. According to our results, colistin cannot be considered a therapeutic option as monotherapy against S. maltophilia or Achromobacter spp. Neither the rate of resistance, nor that of heteroresistance differed significantly in the two examined periods. Transposition of mcr-1 to the chromosome is a possible mechanism. A sequence type ST410 E. coli strain harbouring chromosomal blaCTX-M-15 and an ISApl1-mcr-1 cassette from 2013 was detected in Germany (15). The chromosomal location of mcr-1 may be an explanation as to why the plasmid extract of our isolate was negative, while that of the control strain was positive for mcr-1. However, mcr-1 usually coexists with other resistance genes such as ESBL genes on plasmids, which can reach sizes of 251 kb, and are thus more difficult to extract (2).
Colistin resistance was accompanied by a significant fitness cost in K. pneumoniae and A. baumannii (16, 17). As a limitation, it has to be considered that the low rate of resistance might be partly the consequence of the invasive feature of the studied strains. From an epidemiological standpoint, the monocentric nature of the study is a significant limitation. To the best of our knowledge, this is the first report of mcr-1 mediated colistin resistance in Hungary. Although mcr-1 was already in Hungary in 2011, it did not expand. The rate of colistin resistance is still low. Despite our findings, permanent surveillance for colistin resistance is recommended in order to reserve our antibiotic agent of last resort. The susceptibility to colistin of non-fermenting Gram negative rods other than Pseudomonas spp. or Acinetobacter spp. should be revised and interpretation should be guided by international committees. A global guideline for the accurate detection of colistin heteroresistance in daily clinical microbiology practice, as well as the interpretative criteria of the results and their clinical utility are lacking, but urgently required.
Acknowledgement The mcr-1 positive E. coli strain OKICR1 was kindly provided by Ákos Tóth (National Public Health Institute, Hungary).
Declarations Funding: No funding Competing Interests: None Ethical Approval: Not required.
References 1.Olaitan AO, Morand S, Rolain JM. Emergence of colistin-resistant bacteria in humans without colistin usage: a new worry and cause for vigilance. International journal of antimicrobial agents. 2016;47(1):13. 2.Poirel L, Jayol A, Nordmann P. Polymyxins: Antibacterial Activity, Susceptibility Testing, and Resistance Mechanisms Encoded by Plasmids or Chromosomes. Clinical microbiology reviews. 2017;30(2):557-96. 3.EUCAST. Guidance documents in susceptibility testing. Recommendations for colistin (polymyxin E) MIC testing. 2016. 4.Band VI, Crispell EK, Napier BA, Herrera CM, Tharp GK, Vavikolanu K, et al. Antibiotic failure mediated by a resistant subpopulation in Enterobacter cloacae. Nature microbiology. 2016;1(6):16053. 5.Napier BA, Band V, Burd EM, Weiss DS. Colistin heteroresistance in Enterobacter cloacae is associated with cross-resistance to the host antimicrobial lysozyme. Antimicrobial agents and chemotherapy. 2014;58(9):5594-7. 6.Liu YY, Wang Y, Walsh TR, Yi LX, Zhang R, Spencer J, et al. Emergence of plasmid-mediated colistin resistance mechanism MCR-1 in animals and human beings in China: a microbiological and molecular biological study. The Lancet Infectious diseases. 2016;16(2):161-8. 7.Cannatelli A, Giani T, Antonelli A, Principe L, Luzzaro F, Rossolini GM. First Detection of the mcr-1 Colistin Resistance Gene in Escherichia coli in Italy. Antimicrobial agents and chemotherapy. 2016;60(5):3257-8. 8.Magiorakos AP, Srinivasan A, Carey RB, Carmeli Y, Falagas ME, Giske CG, et al. Multidrug-resistant, extensively drug-resistant and pandrug-resistant bacteria: an international expert proposal for interim standard definitions for acquired resistance. Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases. 2012;18(3):26881. 9.Giske CG. Contemporary resistance trends and mechanisms for the old antibiotics colistin, temocillin, fosfomycin, mecillinam and nitrofurantoin. Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases. 2015;21(10):899-905. 10.Ah YM, Kim AJ, Lee JY. Colistin resistance in Klebsiella pneumoniae. International journal of antimicrobial agents. 2014;44(1):8-15. 11.Hong YK, Lee JY, Wi YM, Ko KS. High rate of colistin dependence in Acinetobacter baumannii. The Journal of antimicrobial chemotherapy. 2016;71(8):2346-8. 12.Gregoire N, Mimoz O, Megarbane B, Comets E, Chatelier D, Lasocki S, et al. New colistin population pharmacokinetic data in critically ill patients suggesting an alternative loading dose rationale. Antimicrobial agents and chemotherapy. 2014;58(12):7324-30. 13.El-Halfawy OM, Valvano MA. Antimicrobial heteroresistance: an emerging field in need of clarity. Clinical microbiology reviews. 2015;28(1):191-207. 14.Moskowitz SM, Garber E, Chen Y, Clock SA, Tabibi S, Miller AK, et al. Colistin susceptibility testing: evaluation of reliability for cystic fibrosis isolates of Pseudomonas aeruginosa and Stenotrophomonas maltophilia. The Journal of antimicrobial chemotherapy. 2010;65(7):1416-23. 15.Falgenhauer L, Waezsada SE, Gwozdzinski K, Ghosh H, Doijad S, Bunk B, et al. Chromosomal Locations of mcr-1 and bla CTX-M-15 in Fluoroquinolone-Resistant Escherichia coli ST410. Emerging infectious diseases. 2016;22(9):1689-91. 16.Beceiro A, Moreno A, Fernandez N, Vallejo JA, Aranda J, Adler B, et al. Biological cost of different mechanisms of colistin resistance and their impact on virulence in Acinetobacter baumannii. Antimicrobial agents and chemotherapy. 2014;58(1):518-26. 17.Choi MJ, Ko KS. Loss of hypermucoviscosity and increased fitness cost in colistin-resistant Klebsiella pneumoniae sequence type 23 strains. Antimicrobial agents and chemotherapy. 2015;59(11):6763-73.
Table 1. Distribution of colistin resistant and heteroresistant (S/R) isolates cultured from blood specimens
E. coli Klebsiella spp. Enterobacter spp. Citrobacter spp. Pantoea spp. Salmonella spp. Pseudomonas spp. Acinetobacter spp. Stenotrophomonas spp. Achromobacter spp. Aeromonas spp.
Number of isolates / period 146 70 25 8 2 1 76 38 20 3 1
2010-2011 R S/R 1 6 1 44 1 5 0 0 0 0 0 22 1 7 20 0 2 0 -
2016 R 0 0 0 0 0 0 2 1 20 3 0
S/R 4 24 3 1 0 0 19 8 0 0 0