*Article* **Implementation of Antibiotic Stewardship in a University Hospital Setting**

**Milan Kolar <sup>1</sup> , Miroslava Htoutou Sedlakova 1,\*, Karel Urbanek <sup>2</sup> , Patrik Mlynarcik <sup>1</sup> , Magdalena Roderova <sup>1</sup> , Kristyna Hricova <sup>1</sup> , Kristyna Mezerova <sup>1</sup> , Pavla Kucova <sup>1</sup> , Jana Zapletalova <sup>3</sup> , Katerina Fiserova <sup>1</sup> and Pavel Kurfurst <sup>4</sup>**


**Abstract:** The article describes activities of an antibiotic center at a university hospital in the Czech Republic and presents the results of antibiotic stewardship program implementation over a period of 10 years. It provides data on the development of resistance of *Escherichia coli*, *Klebsiella pneumoniae*, *Pseudomonas aeruginosa* and *Staphylococcus aureus* to selected antibiotic agents as well as consumption data for various antibiotic classes. The genetic basis of resistance to beta-lactam antibiotics and its clonal spread were also assessed. The study showed significant correlations between aminoglycoside consumption and resistance of *Escherichia coli* and *Klebsiella pneumoniae* to gentamicin (r = 0.712, r = 0.869), fluoroquinolone consumption and resistance of *Klebsiella pneumoniae* to ciprofloxacin (r = 0.896), aminoglycoside consumption and resistance of *Pseudomonas aeruginosa* to amikacin (r = 0.716), as well as carbapenem consumption and resistance of *Pseudomonas aeruginosa* to meropenem (r = 0.855). Genotyping of ESBL- positive isolates of *Klebsiella pneumoniae* and *Escherichia coli* showed a predominance of CTX-M-type; in AmpC-positive strains, DHA, EBC and CIT enzymes prevailed. Of 19 meropenem-resistant strains of *Klebsiella pneumoniae*, two were identified as NDM-positive. Clonal spread of these strains was not detected. The results suggest that comprehensive antibiotic stewardship implementation in a healthcare facility may help to maintain the effectiveness of antibiotics against bacterial pathogens. Particularly beneficial is the work of clinical microbiologists who, among other things, approve administration of antibiotics to patients with bacterial infections and directly participate in their antibiotic therapy.

**Keywords:** antibiotic stewardship; resistance; consumption of antibiotics; clonal spread

#### **1. Introduction**

Antibiotic stewardship may be defined as a set of measures leading to rational antibiotic therapy based on the adequate selection of antibacterial agents, appropriate duration of their administration and a suitable route of administration [1–4]. The need for antibiotic stewardship implementation stems from the likely prospect of antibiotics losing their effectiveness and thus their ability to treat bacterial infections [5–7]. The increasing prevalence of bacteria resistant to antibacterial drugs, mainly those producing extended-spectrum beta-lactamases including metallo-beta-lactamases and carbapenemases opens the possibility of a new non-antibiotic era in which adequate antibiotics will be unavailable to

**Citation:** Kolar, M.; Htoutou Sedlakova, M.; Urbanek, K.; Mlynarcik, P.; Roderova, M.; Hricova, K.; Mezerova, K.; Kucova, P.; Zapletalova, J.; Fiserova, K.; et al. Implementation of Antibiotic Stewardship in a University Hospital Setting. *Antibiotics* **2021**, *10*, 93. https://doi.org/10.3390/ antibiotics10010093

Received: 30 December 2020 Accepted: 14 January 2021 Published: 19 January 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

treat infections caused by multidrug-resistant bacteria [8,9]. To prevent this, antibiotic stewardship programs have been developed as comprehensive systems comprising a range of activities that may be briefly characterized as follows:


It must be stressed, however, that the scope of antibiotic stewardship is much broader, involving numerous other activities that are also very important for adequate antibiotic therapy and preventing the spread of multidrug-resistant bacteria. These activities may be described as follows:


At the University Hospital Olomouc, Czech Republic, antibiotic stewardship is coordinated by the Antibiotic Center, a section of the Department of Microbiology. Based on analyses of the development of bacterial resistance and antibiotic consumption, including the overall costs of this group of drugs, recommendations for initial antibiotic therapy and prophylaxis are formulated and quarterly presented to the hospital management who subsequently approve these recommendations and make them valid.

The article describes efforts of the Antibiotic Center and presents outcomes of its activity over a period of 10 years (2010–2019).

#### **2. Materials and Methods**

#### *2.1. Characteristics of the Healthcare Facility*

The University Hospital Olomouc is one of the largest inpatient healthcare facilities in the Czech Republic, dating back to 1896. It is part of a network of nine teaching hospitals directly controlled by the Ministry of Health of the Czech Republic. Basic data on the facility are shown in Table 1.


**Table 1.** Basic information on the University Hospital Olomouc in 2019.

#### *2.2. Process of Approving Antibiotic Administration*

To better understand the study, it is reasonable to define the process of approving antibiotic administration at the University Hospital Olomouc. For a particular patient with a bacterial infection, the attending physician selects an antibiotic based on their own clinical reasoning and microbiological results (if available), while observing the hospital's guidelines for initial antibiotic therapy. Alternatively, an adequate antibiotic is recommended by a clinical microbiologist based on a consultation with the attending physician. If an antibiotic is selected to treat a particular bacterial infection, its administration must be approved by an Antibiotic Center member. The approval is granted electronically using the hospital information system. The clinical microbiologist (always holding a specialist qualification in medical microbiology) verifies the selection of the antibiotic focusing on all microbiological test results and, if adequate, approves its administration. The Antibiotic Center member has the right to disapprove administration of an antibiotic in case:


In case of disapproval, the reason and a more adequate antibiotic or recommendations from a consultation with the Antibiotic Center clinical microbiologist are entered into the hospital information system. This takes place daily between 7 a.m. and 4 p.m. Outside these hours, antibiotic therapy is selected in line with the hospital's guidelines and the antibiotic therapy is scrutinized on the following day.

#### *2.3. Assessing Antibiotic Consumption*

A computerized database of the hospital's Department of Pharmacology was used to obtain data on antibiotic consumption during the study period. The data were processed according to the 2020 ATC/DDD system and expressed as numbers of defined daily doses for various antibiotic classes [11]. Antibiotic consumption was analyzed for both the entire hospital and its Department of Anesthesiology and Intensive Care Medicine with 25 intensive care beds.

#### *2.4. Identification of Bacteria and Determination of Their Susceptibility/Resistance to Antibacterial Agents*

Bacterial pathogens (*Escherichia coli*, *Klebsiella pneumoniae*, *Pseudomonas aeruginosa*, and *Staphylococcus aureus*) were isolated from clinical samples (tracheal secretion, bronchoalveolar lavage fluid, sputum, blood, urine, pus, puncture samples, wound secretion, bile, cerebrospinal fluid) obtained from hospitalized patients with a suspected bacterial infection. For each patient, only the first isolate from particular clinical samples was included.

The identification of bacteria was performed by MALDI-TOF MS (Biotyper Microflex, Bruker Daltonics, Bremen, Germany) [12].

The susceptibility/resistance to antibiotics was tested using a broth microdilution method according to the EUCAST [10]. The following reference strains were used as quality control organisms: *Escherichia coli* ATCC 25922, *Pseudomonas aeruginosa* ATCC 27853 and *Staphylococcus aureus* ATCC 29213. All strains of *Staphylococcus aureus* were also tested for the resistance to methicillin using selective diagnostic chromogenic media (Colorex/TM/MRSA, TRIOS, Prague, Czech Republic) and an immunochromatographic assay for the detection of PBP2a (PBP2a SA Culture Colony Test, AlereTM, Abbott, Prague, Czech Republic). The production of beta-lactamases, such as ESBL and AmpC, was detected by phenotypic tests [13]. The production of carbapenemases was detected by the Carba NP test [14].

Additionally, methicillin-resistant *Staphylococcus aureus* (MRSA) strains isolated from the Department of Anesthesiology and Intensive Care Medicine patients were confirmed by the *mecA* gene detection [15]. The production of ESBL and AmpC beta-lactamases in *Escherichia coli* and *Klebsiella pneumoniae* was confirmed by PCR detection of the *bla* genes

only in pre-defined groups of strains/patients from above mentioned department (from tracheal aspirates in patients with hospital-acquired pneumonia, from stool in hospitalized patients etc.) [13]. The search for potential production of carbapenemases in the meropenem-resistant *Klebsiella pneumoniae* strains at this department was carried out by simplex PCR with primers targeting *bla*FRI, *bla*GES, *bla*GIM, *bla*IMI, *bla*IMP, *bla*KPC, *bla*NDM, *bla*VIM, *bla*OXA-23 and *bla*OXA-48. Detailed information on the primers is listed in Table 2. PCR assays were performed on Rotor-Gene TM 6000 (Corbett Research, Mortlake, Australia). PCR was run in a final volume of 25 µL using 100 ng of DNA template, 0.5 µM of forward and reverse primers, 200 µM of each dNTP, 2.5 mM of MgCl<sup>2</sup> and 1.25 U Combi Taq Polymerase (Top-Bio, Vestec, Czech Republic) in 1× Buffer (Top-Bio, Vestec, Czech Republic). The PCR conditions were as follows: initial denaturation at 94 ◦C for 3 min, followed by 30 cycles at 94 ◦C for 30 s, 72 ◦C for different times (45 s to 60 s) with a final extension at 72 ◦C for 10 min. PCR products were then separated on a 1% agarose gel containing SYBR Safe (Invitrogen) and visualized on a UV transilluminator. Bacterial isolates for genetic analysis were stored in cryotubes at −80 ◦C (Cryobank B, ITEST, Hradec Králové, Czech Republic).

**Table 2.** Primer sequences used to detect the carbapenemase genes by PCR.


<sup>a</sup> For degenerate primers: R = A or G; S = G or C; Y = C or T.

#### *2.5. Clonality*

The clonality of MRSA and meropenem-resistant isolates of *Klebsiella pneumoniae* detected at the Department of Anesthesiology and Intensive Care Medicine was assessed with pulsed-field gel electrophoresis (PFGE). Bacterial DNA extracted with a technique described by Husiˇcková et al. [19] was digested by the *Xba*I restriction endonuclease (New England Biolabs, Ipswitch, MA, USA) for 24 h at 37 ◦C in *Klebsiella pneumoniae* isolates and by the

*Sma*I restriction endonuclease (New England Biolabs, Ipswitch, MA, USA) for 24 h at 25 ◦C in *Staphylococcus aureus* strains. The obtained DNA fragments were separated by PFGE on 1.2% agarose gel for 24 h at 6 V/cm and pulse times of 2–35 s for both *Klebsiella pneumoniae* and *Staphylococcus aureus* strains. Subsequently, the gel was stained with ethidium bromide. The resulting restriction profiles were analyzed with the GelCompar II software (Applied Maths, Kortrijk, Belgium) using the Dice coefficient (1.2%) for comparing similarity and unweighted pair group method with arithmetic means for cluster analysis. The results were interpreted according to criteria described by Tenover et al. [20].

#### *2.6. Statistical Analysis*

Trends in the consumption of antibacterial agents, or antibiotic classes, bacterial resistance and their relationships were analyzed with Spearman's correlation. The data were processed with IBM SPSS Statistics 22 (Armonk, NY, USA).

#### **3. Results**

Tables 3–6 show the prevalence of *Escherichia coli*, *Klebsiella pneumoniae*, *Pseudomonas aeruginosa* and *Staphylococcus aureus* strains resistant to selected antibiotics over the 10-year period for the entire hospital. The results indicate an increase in resistance of *Escherichia coli* to piperacillin/tazobactam (r = 0.939), gentamicin (r = 0.826), ciprofloxacin (r = 0.816) and cefotaxime (r = 0.734). In *Klebsiella pneumoniae*, resistance to ciprofloxacin (r = 0.665) and cefotaxime increased (r = 0.644). *Pseudomonas aeruginosa* was shown to increase its resistance to colistin (r = 0.722) and amikacin (r = 0.691).

**Table 3.** Resistance of *Escherichia coli* to antibiotics at the University Hospital Olomouc in 2010–2019.


Legend: Resistance percentages (total number of isolates tested), AMS—ampicillin/sulbactam, PPT—piperacillin/tazobactam, CTX cefotaxime, MER—meropenem, GEN—gentamicin, AMI—amikacin, CIP—ciprofloxacin, COL—colistin, TIG—tigecycline.

**Table 4.** Resistance of *Klebsiella pneumoniae* to antibiotics at the University Hospital Olomouc in 2010–2019.


Legend: Resistance percentages (total number of isolates tested).


**Table 5.** Resistance of *Pseudomonas aeruginosa* to antibiotics at the University Hospital Olomouc in 2010–2019.

Legend: Resistance percentages (total number of isolates tested), CTZ—ceftazidime.

**Table 6.** Resistance of *Staphylococcus aureus* to antibiotics at the University Hospital Olomouc in 2010–2019.


Legend: Resistance percentages (total number of isolates tested), OXA—oxacillin, VAN—vancomycin.

Consumption of antibiotics or antibiotic classes at the University Hospital Olomouc is shown in Table 7. The data indicate increasing consumption of carbapenems (r = 0.964), tigecycline (r = 0.879), third- and fourth-generation cephalosporins (r = 0.867) and fluoroquinolones (r = 0.733). Conversely, consumption of penicillins combined with betalactamase inhibitors decreased (r = −0.745). Analysis of the relationship between antibiotic consumption and resistance in the entire hospital showed significant correlations between aminoglycoside consumption and resistance of *Escherichia coli* to gentamicin (r = 0.712), fluoroquinolone consumption and resistance of *Klebsiella pneumoniae* to ciprofloxacin (r = 0.896) and aminoglycoside consumption and resistance of *Pseudomonas aeruginosa* to amikacin (r = 0.716) (Figures 1–3).

**Table 7.** Antibiotic consumption in defined daily doses (DDDs) at the University Hospital Olomouc.


*Antibiotics* **2021**, *10*, x FOR PEER REVIEW 7 of 16

**Figure 1.** Correlation between aminoglycoside consumption (in numbers of defined daily doses) and resistance of *Escherichia coli* to gentamicin**. Figure 1.** Correlation between aminoglycoside consumption (in numbers of defined daily doses) and resistance of *Escherichia coli* to gentamicin. **Figure 1.** Correlation between aminoglycoside consumption (in numbers of defined daily doses) and resistance of *Escherichia coli* to gentamicin**.** 

and resistance of *Klebsiella pneumoniae* to ciprofloxacin**. Figure 2.** Correlation between fluoroquinolone consumption (in numbers of defined daily doses) and resistance of *Klebsiella pneumoniae* to ciprofloxacin**. Figure 2.** Correlation between fluoroquinolone consumption (in numbers of defined daily doses) and resistance of *Klebsiella pneumoniae* to ciprofloxacin.

**Antibiotic Class/Antibi-**

**Penicillins combined with beta-lactamase inhibitors** 

2010–2019.

*Antibiotics* **2021**, *10*, x FOR PEER REVIEW 8 of 16

**Figure 3.** Correlation between aminoglycoside consumption (in numbers of defined daily doses) and resistance of *Pseudomonas aeruginosa* to amikacin**. Figure 3.** Correlation between aminoglycoside consumption (in numbers of defined daily doses) and resistance of *Pseudomonas aeruginosa* to amikacin.

**Table 7.** Antibiotic consumption in defined daily doses (DDDs) at the University Hospital Olomouc**. otic 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019**  89,977 80,212 77,168 76,803 76,937 81,889 70,248 71,774 74,446 76,427 Tables 8–11 document resistance of particular bacterial species at the Department of Anesthesiology and Intensive Care Medicine over the study period. The results show increasing resistance of *Escherichia coli* to piperacillin/tazobactam (r = 0.845) and cefotaxime (r = 0.729), resistance of *Klebsiella pneumoniae* to cefotaxime (r = 0.778) and resistance of *Pseudomonas aeruginosa* to meropenem (r = 0.988).

**3rd and 4th generation Table 8.** Resistance of *Escherichia coli* to antibiotics at the Department of Anesthesiology and Intensive Care Medicine in 2010–2019.


creasing resistance of *Escherichia coli* to piperacillin/tazobactam (r = 0.845) and cefotaxime Legend: Resistance percentages (total number of isolates tested).

*Pseudomonas aeruginosa* to meropenem (r = 0.988).

**Table 8.** Resistance of *Escherichia coli* to antibiotics at the Department of Anesthesiology and Intensive Care Medicine in **2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 AMS** 40 (115) 48 (143) 49 (140) 28 (129) 45 (110) 42 (103) 37 (97) 45 (102) 44 (116) 34 (182) **PPT** 13 (116) 21 (141) 20 (138) 16 (125) 15 (107) 23 (100) 22 (98) 25 (102) 27 (117) 25 (182) At the Department of Anesthesiology and Intensive Care Medicine, consumption of tigecycline (r = 0.939), carbapenems (r = 0.879), third- and fourth-generation cephalosporins (r = 0.867) and glycopeptides (r = 0.636) increased (Table 12). There were significant correlations between carbapenem consumption and resistance of *Pseudomonas aeruginosa* to meropenem (r = 0.855) as well as between aminoglycoside consumption and resistance of *Klebsiella pneumoniae* to gentamicin (r = 0.869) (Figures 4 and 5).

(r = 0.729), resistance of *Klebsiella pneumoniae* to cefotaxime (r = 0.778) and resistance of

**CTX** 9 (116) 16 (141) 17 (138) 14 (125) 10 (107) 20 (100) 14 (97) 18 (102) 26 (117) 31 (182) **MER** 0 (116) 0 (141) 0 (138) 0 (125) 0 (107) 0 (100) 0 (98) 0 (102) 0 (117) 0 (182) **GEN** 6 (116) 20 (143) 16 (140) 13 (129) 12 (110) 11 (103) 12 (97) 13 (102) 18 (117) 23 (182) **AMI** 7 (116) 6 (141) 9 (138) 2 (125) 8 (105) 3 (100) 5 (98) 5 (102) 1 (117) 2 (182)


**Table 9.** Resistance of *Klebsiella pneumoniae* to antibiotics at the Department of Anesthesiology and Intensive Care Medicine in 2010–2019.

Legend: Resistance percentages (total number of isolates tested).

**Table 10.** Resistance of *Pseudomonas aeruginosa* to antibiotics at the Department of Anesthesiology and Intensive Care Medicine in 2010–2019.


Legend: Resistance percentages (total number of isolates tested).

**Table 11.** Resistance of *Staphylococcus aureus* to antibiotics at the Department of Anesthesiology and Intensive Care Medicine in 2010–2019.


Legend: Resistance percentages (total number of isolates tested).

**Table 12.** Antibiotic consumption in defined daily doses (DDD) at the Department of Anesthesiology and Intensive Care Medicine.


Medicine**.** 

Medicine**.** 

**Antibiotic Class/An-**

**Penicillins combined with beta-lactamase inhibitors** 

**Penicillins combined with beta-lactamase inhibitors** 

**Antibiotic Class/An-**

**3rd and 4th genera-**

**3rd and 4th genera-**

*Antibiotics* **2021**, *10*, x FOR PEER REVIEW 10 of 16

*Antibiotics* **2021**, *10*, x FOR PEER REVIEW 10 of 16

**Figure 4.** Correlation between aminoglycoside consumption (in numbers of defined daily doses) and resistance of *Klebsiella pneumoniae* to gentamicin. **Figure 4.** Correlation between aminoglycoside consumption (in numbers of defined daily doses) and resistance of *Klebsiella pneumoniae* to gentamicin. **Figure 4.** Correlation between aminoglycoside consumption (in numbers of defined daily doses) and resistance of *Klebsiella pneumoniae* to gentamicin.

resistance of *Pseudomonas aeruginosa* to meropenem. **Table 12.** Antibiotic consumption in defined daily doses (DDD) at the Department of Anesthesiology and Intensive Care **Figure 5.** Correlation between carbapenem consumption (in numbers of defined daily doses) and resistance of *Pseudomonas aeruginosa* to meropenem. **Figure 5.** Correlation between carbapenem consumption (in numbers of defined daily doses) and resistance of *Pseudomonas aeruginosa* to meropenem.

**tibiotic 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019**  1539 1463 1369 1376 1357 1288 1486 1519 1339 1473 **Table 12.** Antibiotic consumption in defined daily doses (DDD) at the Department of Anesthesiology and Intensive Care **tibiotic 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019**  1539 1463 1369 1376 1357 1288 1486 1519 1339 1473 Genotyping of ESBL- positive isolates of *Klebsiella pneumoniae* and *Escherichia coli* in particular patient groups (from tracheal aspirates in patients with hospital-acquired pneumonia, from stool in hospitalized patients etc.) at the Department of Anesthesiology and Intensive Care Medicine showed a predominance of CTX-M-type, namely of the CTX-M-15 and CTX-M-9 types (data not shown). In AmpC-positive strains, EBC and CIT enzymes prevailed in *Escherichia coli* and the DHA type in *Klebsiella pneumoniae* (data not shown).

**tion cephalosporins** 125 130 234 144 428 241 260 325 394 556 **Carbapenems** 618 739 505 946 1290 1427 1298 1280 1498 1822 **Aminoglycosides** 209 691 629 667 682 806 812 589 677 841 **Fluoroquinolones** 589 460 514 576 501 639 670 484 398 510 **tion cephalosporins** 125 130 234 144 428 241 260 325 394 556 **Carbapenems** 618 739 505 946 1290 1427 1298 1280 1498 1822 **Aminoglycosides** 209 691 629 667 682 806 812 589 677 841 Between 2010 and 2019, a total of 19 meropenem-resistant strains of *Klebsiella pneumoniae* were detected in patients staying at the Department of Anesthesiology and Intensive Care Medicine. Only 2 strains were NDM-positive (data not shown). However, no other carbapenemase genes were detected. The total number of isolated MRSA at the Department of Anesthesiology and Intensive Care Medicine was 45 strains. In case of meropenemresistant *Klebsiella pneumoniae* strains and MRSA, no significant clonal spread was noted.

**Colistin** 167 253 233 410 433 498 340 190 228 478

**Fluoroquinolones** 589 460 514 576 501 639 670 484 398 510

No identical clone was detected in meropenem-resistant *Klebsiella pneumoniae* isolates and only two pairs of identical MRSA strains were identified.

#### **4. Discussion**

Today's medicine is characterized by exponentially expanding knowledge in all specialties, resulting in considerable improvements of both diagnostic and therapeutic activities. Despite past achievements, however, there is one issue posing a serious therapeutic challenge. It is the role of bacterial infections that have continued to increase in recent years. One reason is rising resistance of bacteria to the effects of antibacterial drugs and the associated risk of treatment failure. Numerous studies have been published documenting higher mortality and shorter survival of patients with infections caused by multidrug-resistant bacteria compared to those due to susceptible strains of the same species [21–25]. The present study yielded interesting results when compared with the national and European resistance rates as reported by the European Antimicrobial Resistance Surveillance Network (EARS-Net). In 2019, the mean prevalence of MRSA in the Czech Republic and Europe was 13% and 15%, respectively; the University Hospital Olomouc rates ranged from 3% to 6% [26,27]. Similarly, very low prevalence was also noted for meropenem-resistant strains of *Klebsiella pneumoniae*. According to the ECDC's Annual Epidemiological Report for 2019, the mean prevalence of carbapenem-resistant strains of *Klebsiella pneumoniae* in Europe was 8%, with some European countries even reporting rates higher than 10% [26]. At the University Hospital Olomouc, however, the resistance of this species to meropenem did not exceed 1% or, in case of the Department of Anesthesiology and Intensive Care Medicine, 3%. Only two strains were found to produce NDM- carbapenemases. For meropenemresistant isolates without the carbapenemase gene, we assume that the resistance is due to mechanisms such as loss or mutation of porins with AmpC beta-lactamase or ESBL hyperproduction or overexpression of the efflux pumps.

There were considerable differences in resistance of *Klebsiella pneumoniae* to thirdgeneration cephalosporins in Europe (31%) and in the Czech Republic (50%) in 2019 [26,27]. The University Hospital Olomouc rate (43%) was below the mean rate for the entire country.

Resistance of *Escherichia coli* to cefotaxime and resistance of *Pseudomonas aeruginosa* to ceftazidime, aminoglycosides and fluoroquinolones at the University Hospital Olomouc do not greatly differ from the mean rates in Europe.

Of concern is the prevalence of *Pseudomonas aeruginosa* strains resistant to meropenem (34%), exceeding both the Czech (15%) and European (17%) mean rates [26,27]. However, carbapenems are mainly needed to treat infections caused by members of *Enterobacterales* producing ESBL and AmpC beta-lactamases; the resistance of these bacterial species to meropenem does not increase. Despite that, there will be efforts to reduce carbapenem consumption in the following years. It should be stated that carbapenems account for 6% of the overall antibiotic consumption at the University Hospital Olomouc (unpublished data).

With the exception of a higher prevalence of meropenem-resistant *Pseudomonas aeruginosa*, prevalence rates of other studied phenotypes are below the rates reported by the EU-CAST [26,27]. The main causes of the development and spread of bacterial resistance are the administration of antibiotics and their selection pressure [28–34]. Therefore, the restriction of certain antibacterial agents and relevant antibiotic classes aimed to limit their selection pressure is a possible solution to the problem [35]. However, selection pressure is a more complex issue. Apparently, consumption of certain antibiotics may only be reduced if the consumption of others increases. Moreover, antibiotic resistance is often multiple, meaning that selection pressure of a particular antibiotic agent results in increased resistance to other antibiotics, for example, resistance of ESBL-positive enterobacteria to cephalosporins and fluoroquinolones or resistance of MRSA to clindamycin [36,37]. Another important aspect influencing the selective pressure is antibiotic concentration, that is the correct dosage of antibiotics and their distribution in the body. Clinical microbiologists and physicians care about the accurate dosage in terms of pharmacodynamic/pharmacokinetic parameters to achieve satisfactory outcomes in patients. However, the question is how the selected dosage

and the final concentration of an antibiotic promotes the genesis of resistant mutants. The phenomenon of bacterial resistance represents a complex problem and the emergence of antibiotic-resistant mutants depends on different aspects such physiology, genetics, historical behavior of bacterial populations, antibiotic-bacterium dynamics and others [38,39].

Studies have shown that there may not be a direct relationship between the administration of selected antibiotics and bacterial resistance. Several studies failed to confirm correlations between bacterial resistance to particular antibiotic classes and their consumption [40–42]. Similarly, Htoutou Sedláková et al. reported decreasing consumption of third-generation cephalosporins and fluoroquinolones but increasing resistance of *Enterobacteriaceae* to these drugs [43]. This may be due to multiple mechanisms. Some authors claim that the relationship between antibiotic consumption and resistance disappears after a certain resistance threshold is exceeded, since mobile genetic elements (in particular plasmids and transposons) circulate in bacterial populations and a decrease in antibiotic selection pressure does not influence this phenomenon any more [44]. It is documented that transfer rates of ESBL-plasmids are highest in the absence of the antibiotic [45]. Another explanation could be the collateral effect of antibiotics, which means that not only subinhibitory concentrations of an antibiotic could stimulate the emergence and the dissemination of its corresponding resistant gene, but that collateral stimulation by other antibiotics is also possible. For example, the mobile genetic element carrying the gene for tetracycline resistance is able to exhibit a 1000-fold increase of its transfer frequency when exposed to subinhibitory concentrations of tetracyclines, but also macrolides, lincosamides and streptogramins [46].

Our findings suggest that the increasing bacterial resistance is mainly determined by the selection pressure of antibiotics. Neither significant horizontal clonal spread of multidrug-resistant bacteria nor increasing bacterial resistance to a particular antibiotic whose consumption decreases have been observed.

As part of antibiotic resistance surveillance, the Antibiotic Center not only controls the appropriate administration of antibiotics, that is the adequate indication and dosage in a particular patient, but also regularly monitors important bacterial resistance phenotypes and genotypes, in particular MRSA, vancomycin-resistant enterococci, ESBLand AmpC-positive *Enterobacterales*, Gram-negative bacteria resistant to carbapenems, fluoroquinolones and others, as well as their clonal spread. For technical reasons, such surveillance is not performed in the entire hospital, but is mostly limited to selected departments and pre-defined patient groups and time periods. This approach to antibiotic stewardship has been reflected in numerous studies carried out at our department [47–50]. Based on their outcomes, certain conclusions have been drawn and relevant measures have been implemented such as evidence-based recommendations for consultant microbiologists and attending physicians concerning an adequate selection of antibiotic agents, guidelines for initial antibiotic therapy including antibiotic prophylaxis, restriction of certain antibiotic classes or improvement of hygiene and epidemiological measures.

The present study showed a significant relationship between aminoglycoside consumption and resistance of *Escherichia coli* and *Klebsiella pneumoniae* to gentamicin, results consistent with those in our 2014 study [43]. Moreover, there were correlations between fluoroquinolone consumption and resistance of *Klebsiella pneumoniae* to ciprofloxacin and between aminoglycoside consumption and resistance of *Pseudomonas aeruginosa* to amikacin, consistent with findings published by other authors [34,51]. Another reason for increasing bacterial resistance may be the horizontal or clonal spread of genetically identical strains of particular species among patients. In this case, the selection pressure of antibiotics may be of less importance and external environmental factors may play a role, for example, those related to healthcare staff. Examples include a study by Hricová et al. on vancomycin-resistant enterococci in patients with hematological malignancies at the University Hospital Olomouc reporting 67% clonality of isolated strains or outbreaks of epidemic MRSA clones in various parts of the world [48,52–54]. The present study, however, did not show a significant clonal spread of MRSA and meropenem-resistant strains

of *Klebsiella pneumoniae* isolated from Department of Anesthesiology and Intensive Care Medicine patients, highlighting the role of horizontal resistance gene transfer in the spread of antibiotic resistance. Further, there is no doubt that the use of antibiotics contributes to the development of resistance by acquiring resistance genes and maintenance of chromosomal resistance-associated mutations [38]. However, determining the exact effect of antibiotic use on the development of resistance is problematic. Moreover, it is increasingly claimed that the emergence, maintenance and spread of resistance traits are also influenced by social, economic and genetic factors.

#### **5. Conclusions**

The presented data suggest low rates of bacterial resistance at the University Hospital Olomouc, with the only exception being an increased prevalence of meropenem-resistant strains of *Pseudomonas aeruginosa*. This confirms the importance of antibiotic stewardship and surveillance of antimicrobial resistance, including the use of molecular biology methods, for maintaining the effectiveness of antibiotics and limiting the spread of multidrugresistant bacterial pathogens. Data on the prevalence of bacterial resistance and the results of molecular genetic analysis of multidrug-resistant strains must form the basis for practical antibiotic stewardship. These should include a definition of optimal regimens for initial antibiotic therapy and assessment of the sources and routes of spread of multidrug-resistant bacteria so that adequate hygiene and epidemiological measures may be introduced. It is apparent that besides obtaining data for the entire hospital, hospital departments need to be individually assessed and adequate antibiotic stewardship measures must be implemented based on the results.

**Author Contributions:** Conceptualization, M.K.; Data curation, M.H.S.; Formal analysis, J.Z.; Funding acquisition, M.K.; Investigation, M.H.S., K.U., P.M., M.R., K.H., K.M., P.K. (Pavla Kucova) and K.F.; Methodology, P.M.; Project administration, M.K.; Resources, M.H.S., K.U., P.K. (Pavla Kucova) and K.F.; Supervision, M.K.; Validation, M.R.; Visualization, M.K., M.H.S., K.U., P.M., M.R., K.H., K.M., P.K. (Pavla Kucova), K.F. and P.K. (Pavel Kurfurst); Writing—original draft, M.K., M.H.S., P.M., M.R., J.Z., and P.K. (Pavel Kurfurst); Writing—review & editing, M.K., M.H.S., K.U., P.M., M.R., K.H., K.M., P.K. (Pavla Kucova), J.Z., K.F. and P.K. (Pavel Kurfurst) All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Czech Health Research Council (project no. NV18-05- 00340), project IGA\_LF\_2020\_021, Junior Grant of UP in Olomouc JG\_2019\_005 and by MH CZ—DRO (FNOL, 00098892).

**Data Availability Statement:** Data sharing not applicable.

**Acknowledgments:** The authors thank med. Arne C. Rodloff (Facharzt für Mikrobiologie, Virologie und Infektionsepidemiologie, Krankenhaushygieniker, Germany) and Pavel Boštík (Faculty of Medicine, Charles University and University Hospital in Hradec Kralové) for critically reviewing the manuscript.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

