Phenotypic Characterization of Non-Beta Lactam Multidrug Resistance in Extended-Spectrum Beta-Lactamase Producer and Non-Producer Enterobacteriaceae

Purpose: The aim of this study is to detect the correlations between ESBL-producing, and the pattern, and extensiveness of MDR of other common nonbetalactam antibiotics in Enterobacteriaceae strains. Methods: A total of 170 ESBL-producer strains and the same amount of ESBL-non-producer Enterobacteriaceae strains were collected as study and control groups from our Clinical Microbiology Laboratory. Antibiotic susceptibility tests and ESBL detection were conducted on both microorganism groups by routine bacteriologic methods. The antibiotic susceptibilities have been listed on a strain basis, and they were classified according to the extensiveness of resistances. Results: When comparing the ESBL-producer to the ESBLnonproducer strains, higher individual resistance rates were detected to fluoroquinolones, trimethoprim/ sulfamethoxazole, aminoglycosides, piperacillin/ tazobactam, nitrofurantoin and carbapenems in various percentages (69. 40% vs 23. 50%; 68.20% vs 30%; 34. 70% vs 5. 30%; 27. 20% vs 4.73%; 23.50% vs 1.90 and 18.90% vs 5.20% respectively). The three to six-drug MDRs in the same isolate has been found in 36 different strains of ESBL-p, although no strains were detected in the control group (21% vs 0). The three most prevalent triplet co-existences of MDR were sulfamethoxazole/trimethoprim+fluoroquinolones +aminoglycosides, and sulfamethoxazole/trimethoprim +fluoroquinolones+piperacillin/tazobactam, and sulfamethoxazole/trimethoprim+fluoroquinolones +carbapenem resistance combinations in both groups, but, their proportions were statistically higher in the study group (43 vs 4, 31 vs 3, and 22 vs 1 strains). Conclusions: ESBL and carbapenemase-producing Enterobacteriaceae control the other wide range of nonbeta-lactam antibiotic resistance in various combinations, according to our study.


Introduction
Extended-spectrum beta-lactamase-producing (ESBL-p) Enterobacteriaceae strains may cause emergent infections, both in the community and in hospital settings. These enzymes may inactivate a wide-range of beta-lactam group antibiotics, including nearly all cephalosporins [1][2][3].
The other major problem in the treatment of infectious diseases is the presence of MDR in the wide range of microorganisms, including Enterobacteriaceae. These group resistances also, contribute to the emergence and limit the treatment options of these microorganism infections.

Magiorakos et al. categorized a group of microorganisms
including Enterobacteriaceae according to their extensiveness of antibiotic resistance. In order to achieve this, they firstly classified all effective antibiotics into "categories" and they placed similar "antibiotic agents" into the same category. The authors classified bacterial resistance as: "Multidrug-resistant (MDR)"; "extensively drug-resistant (XDR)"; and" Pandrug resistant (PDR)". They classified 31 effective antimicrobial agents for Enterobacteriaceae within 17 categories, including all beta-lactam antibiotics, as well as non-beta-lactams, and ESBL-resistant antibiotics (ESBL-RA) such as carbapenems [4].
The other possible mechanism of wide range of resistance is multidrug efflux pumps. These subcellular organelles are the expeller a broad range of antibiotics and, even some heavy metals-drive the acquisition of additional resistance mechanisms by lowering intracellular antibiotic and other compounds concentration [8].
One problem that still exists is how many resistance genes may harbor and are placed in genetic structures to move together in enteric isolates in clinical settings [9][10][11][12]. In daily medical practices, detailed information on the transferable coresistance genes containing elements will be a practical guide to choose effective antibiotic treatment options.
The aim of this study is to detect any correlations between ESBL positivity and MDR amongst ESBL-RA in Enterobacteriaceae strains and to analyze the screening of the extensiveness and co-existence of these resistance.

Study-related definitions
Various definitions have been made to facilitate the explanation of this study, as follows: ESBL-RA: Abbreviation of ESBL-resistant antibiotics. It consists all non-beta-lactam antibiotics, carbapenems, and piperacillin/tazobactam that is resistant to conventional ESBLs.
Individual resistance: Single resistance to ESBL-RA in MDR enteric isolates

Extensiveness of MDR:
The number of resistances to ESBL-RA detected within the same isolate. They are defined as "onedrug MDR" to "seven-drug MDR", or more.
Co-existence of resistance: The resistance combinations to ESBL-RA within the same enteric isolates This prospective, case-control observational study was implemented between March 2016 and September 2017 in our 150-bed tertiary care teaching hospital in Nicosia, Northern Cyprus. During the study period, a total of 170 ESBLp Enterobacteriaceae strains were collected.

Inclusion and exclusion criteria
The repeated cultures were excluded. Members of the Enterobacteriaceae family other than E. coli and Klebsiella strains were also excluded, because of their proportion was significantly low.
Control group: ESBL-np strains have been selected among our clinical isolates as the control group. The repeated cultures were also excluded.
Clinical microbiology studies: Study and control group microorganisms have been isolated from clinical samples by routine microbiologic procedures. Isolated microorganisms were further examined with the BD PhoenixTM 100 Automated Microbiology System (Becton Dickson, USA) and Oxoid combination disk test methods. The inoculated PhoenixTM panels were placed into the PhoenixTM instrument for incubation and continuous reading. Aztreonam and extensive cephalosporin resistant members of Enterobacteriaceae were classified as "possibleESBL-p" strains, whereas sensitive members were classified as "ESBL-np".
ESBL-p strains dertected by double disk synergy methods described by Kaur et al. [14].
Beta-lactam resistance was not evaluated in this study because, the aim of this study is to detect the correlations between ESBL positivity and other MDR. These resistances have only been used to detect and control ESBL positivity.
The relationship between ESBL positivity and the extensiveness and the co-existence patterns of resistance amongst MDR to ESBL-RA in Enterobacteriaceae strains have been compared to the control group. ESBL-MDR mosaics have been created for this purpose.
Statistical analysis: Results were statistically analyzed by 'Difference between two proportions for independent samples", and the mean age of patients has been analyzed by t-test for independent samples.

Results
Demographic and clinical characteristics of patients' isolated ESBL-p and ESBL-np group microorganisms have been outlined in Table 1.
resistance [5]. Generally, they are carried on a series of plasmids such as OXA-48, and others. These genes can harbor some non-beta-lactam resistance genes, and they can move together [6,7]. This characterstics may explain these ESBL -non-beta-lactam co-resistances. The most prevalent harboring genes are fluroquinolone, trimethoprim/sulfamethoxazole, and aminoglycoside resistant genes to ESBL's and/or carbapenemases. Individual resistance to ESBL-RA in ESBL-p and ESBL-np Enterobacteriaceae strains has been shown in Table 2. The extensiveness of MDR's are outlined in Table 3. As shown in Table 3, there are no strains in the three-drug to six-drug resistances range in the ESBL-np group.
Co-existences of MDR and their combinations amongst ESBL-RA antibiotics have been outlined in Figure 1.

Discussion
The aims of this study are to define the phenotypic MDR patterns of ESBL-RA's in ESBL-p and ESBL-np Enterobacteriaceae strains and to show whether ESBL positivity is a dominant characteristic especially on resistance to non-beta-lactam antibiotics.
The ESBL-p and control groups were largely similar in demographic and clinical characteristic in the study, except for only two minor statistical differences. The mean age of the study group was older than the control group (56.22 vs 48.72). Additionally, there was a significant difference between the patient's sources, meaning that a higher percentage of ESBL-p This article is available from: http://infectious-diseases-and-treatment.imedpub.com/ samples were collected from inpatient clinics. However, on the contrary, ESBL-np strains were mainly isolated from outpatient sources.

Journal of Infectious Diseases and Treatment
The higher individual resistance prevalence to all ESBL-RA have been detected in the ESBL-p group. All isolates of the ESBL-np group were sensitive to colistin. On the contrary; only six (3.5%) colistin-resistant strains have been isolated in ESBL-p group.
Our ESBL-p strains showed 8.8% resistance to tigecycline, but it was not possible to collect susceptibility test results to this antibiotic for the ESBL-np group because of the institutionally restricted antibiotic prescribing policies. The least resistant ESBL-RA's detected were colistin, nitrofurantoin, and fosfomycin in ESBL-np group, respectively. Fosfomycin was also the least resistant antibacterial agents in the ESBL-p group. These results reveal excellent treatment options still exist, particularly for lower urinary tract infections. Similar resistant patterns have been published previously [12][13][14][15].
A total of nine individual types of resistance in different combinations have been detected in our ESBL-p group.
A statistical difference between ESBL-p and ESBL-np in the extensiveness of MDR has been also detected. A total of 36 (21%) ESBL-p strains showed four-to-six drug MDR to ESBL-RA, while we could not detect any resistance strains with similar characteristics in the control group. When the was gradually decreased from three-drug to one-drug resistance, the proportional difference between the -p and -np groups also decreased so that one drug resistance in ESBL-np was higher than ESBL-p.
Ninety-one strains of our ESBL-NP were totally sensitive to ESBL-RA while only 16 strains were in the same charachteristics in ESBL-p group (53.3% to 9.4% respectively: p=0.0015) Our study group of enteric isolates had a total of 33 different resistance combination coexistences, compared with 11 different combinations in the control groups [16].
It was found that the most prevalent triplet combinations of MDR in ESBL-p strains was trimethoprim/sulfamethoxazole +fluoroquinolones+aminoglycosides in combination, and this was the ten times more frequent than in the control group (43 vs 4 strains). The second most prevalent triplet resistance was trimethoprim/sulfamethoxazole+fluoroquinolones+ piperacillin /tazobactam and this combination was also ten times more frequent than in the control group (31 vs 3 strains). Moreover, in the third most frequent combination was trimethoprim/ sulfamethoxazole+fluoroquinolones and carbapenems (22 vs 1 strains).
A series of MDR-including carbapenem-resistant OXA-48 positive enterobacterial isolates were recovered from Turkey has been previously reported [17]. Most of the reported isolates were resistant to fluoroquinolones, trimethoprim/ sulfamethoxazole, and aminoglycosides in addition to carbapenems and the other beta-lactams. These strains were sensitive to other antibiotics not indicated above. It is highly possible that the most of our strains also comprise OXA plasmids as a part of its epidemics.
Our three most prevalent resistant pattern-containing strains may carry different accessory resistance characteristics such as colistin, fosfomycin, nitrofurantoin and tigecycline resistance.
One of the most remarkable co-existence of resistance is carbapenemase containing ESBL-RA resistance combinations in our enterobacterial isolates. This reminds us that our ESBL-p strains should be classified into two different groups according to their carbapenemase producing/non-producing characteristics. One-half of our ESBL-p total isolates contains carbapenemase-producing combinations (81 isolates: 47.6%), whereas the other half are ESBL-p without carbapenemase production.
In this study, we presented only phenotypic resistance characteristics rather than genetic analysis. However, it is highly possible that the dominant characteristics of these carbapenemases controls and manages the ESBL-RA resistance. Qeenan and Bush denominated carbapenemases as "versatile beta-lactamases" because of their broad activity in a wide range of beta-lactams [18,19]. It is apparent that this versatility is not only limited to beta-lactams, but also to nonbeta-lactams.
Both carbapenemases and other ESBL genes may harbor a set of non-beta-lactam resistance genes such as fluoroquinolones, aminoglycosides and/or trimethoprim/ sulfamethoxazole resistances [7,20,21].
On the other hand, in this study, we presented the first MDR results in Enterobacteriaceae strains for Cyprus. Country-wide susceptibility test results are required to understand the precise resistance epidemiology.
Investigations of the genotypic characteristics of our strains will a topic of further studies.

Conclusion
This study showed us that ESBL positivity controls and commands MDR to ESBL-RA in enterobacterial isolates. This control is stronger in longer co-resistance, while on the contrary, the shorter co-resistance is independent of ESBL positivity. When the ESBL-p strains are classified as carbapenemase and non-carbapenemase producing in origin, there is no significant coexistence difference between these subgroups. In daily medical practices, time-consuming laboratory test methods can be beneficial for choosing successful treatment options. Immediate ESBL test results are also necessary, and these results may not only direct us to avoid beta-lactams but also other options such as trimethoprim/sulfamethoxazole, fluoroquinolones, and aminoglycosides.