Trends in antimicrobial resistance and antibiotic use before and during the COVID-19 pandemic in a university research and practice hospital in Türkiye

  1. Merve Buyukcelik
  2. Fusun Zeynep Akcam
  3. Ersin Uskun
  4. Emel Sesli Cetin
  5. Onur Kaya
  6. Gul Ruhsar Yilmaz
  7. Onur Unal
  8. Esra Nurlu Temel

Reviewed by

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Introduction Antibacterial resistance and the increasing number of infections caused by multidrug-resistant bacteria threaten human health worldwide. The coronavirus disease 2019 (COVID-19) pandemic may have influenced antibacterial resistance patterns through changes in infection control practices and antibiotic prescribing. Aim This single-centre, retrospective study aimed to describe changes in bacterial distribution, antibacterial resistance and antibiotic consumption in a university research and practice hospital before and during the COVID-19 pandemic. Methods We analysed routine microbiology and pharmacy records from hospitalized patients between 01.04.2018 and 31.03.2022. The two years before 01.04.2020 were defined as the pre-pandemic period and the two years after 01.04.2020 as the pandemic period. Bacteria isolated from blood, urine and lower respiratory tract cultures, together with their antimicrobial susceptibility profiles, were compared between periods according to EUCAST criteria. Antibiotic consumption was calculated as defined daily doses (DDD) per 1000 inpatient-days for commonly used agents. No patient-level clinical data or ward/ICU stratification were available. Results A total of 7275 isolates from 47,729 culture samples were obtained in the pre-pandemic period and 5794 isolates from 47,210 samples during the pandemic. Coagulase-negative staphylococci remained the most frequently isolated species from blood cultures, Escherichia coli from urine cultures and Acinetobacter baumannii from lower respiratory tract cultures in both periods. Extended-spectrum β-lactamase (ESBL) rates and carbapenem resistance in E. coli and Klebsiella pneumoniae increased significantly during the pandemic, whereas teicoplanin and linezolid resistance in coagulase-negative staphylococci decreased. Carbapenem resistance in A. baumannii also decreased. Overall antibiotic consumption increased for most agents, particularly cephalosporins, carbapenems, aminoglycosides and fluoroquinolones, while vancomycin use decreased. Conclusion In this single-centre, retrospective analysis, the overall distribution of major bacterial species remained largely stable before and during the COVID-19 pandemic, whereas important changes were observed in antimicrobial resistance profiles and antibiotic consumption. The increase in ESBL and carbapenem resistance in Enterobacterales, together with higher use of broad-spectrum antibiotics, underlines the need for strengthened antimicrobial stewardship and continuous local surveillance.

Article activity feed

  1. Access Microbiology

    I am pleased to tell you that your article has now been accepted for publication in Access Microbiology. While there are a few grammatical and spelling errors within the manuscript, it is the hope that these would be polished during the publication phase.

  2. Access Microbiology

    Comments to Author

    I consider that the authors have addressed the questions and suggestions raised by the reviewers and the editorial team. Line 68. Typo for Introduction. Methods. Lines 96. The number of hospital beds wasn't included. Results. Lines 170-173, concerning the frequency of coagulase-negative staphylococci (CoNS) as potential skin contaminants and the interpretation of their role in infection, should be addressed with caution. This point would be more appropriately placed in the discussion section (it is already mentioned under the limitations) rather than in the results.

    Please confirm that no generative AI tools or large language models have been used to generate this peer review report or to assist with any part of the peer review process.

    I confirm no generative AI tools were used in preparation of this review.

    Please rate the manuscript for methodological rigour

    Very good

    Please rate the quality of the presentation and structure of the manuscript

    Very good

    To what extent are the conclusions supported by the data?

    Strongly support

    Do you have any concerns of possible image manipulation, plagiarism or any other unethical practices?

    No

    Is there a potential financial or other conflict of interest between yourself and the author(s)?

    No

    If this manuscript involves human and/or animal work, have the subjects been treated in an ethical manner and the authors complied with the appropriate guidelines?

    Yes

  3. Access Microbiology

    Comments to Author

    Dear Authors, Thank you for your careful revisions and detailed responses. You have adequately addressed the previous comments, particularly by clarifying the methodological limitations and ensuring that the findings are interpreted appropriately within the study design. The manuscript is now clearer and more balanced scientifically. I appreciate your efforts and wish you success with the publication. Best regards

    Please confirm that no generative AI tools or large language models have been used to generate this peer review report or to assist with any part of the peer review process.

    I confirm no generative AI tools were used in preparation of this review.

    Please rate the manuscript for methodological rigour

    Good

    Please rate the quality of the presentation and structure of the manuscript

    Very good

    To what extent are the conclusions supported by the data?

    Strongly support

    Do you have any concerns of possible image manipulation, plagiarism or any other unethical practices?

    No

    Is there a potential financial or other conflict of interest between yourself and the author(s)?

    No

    If this manuscript involves human and/or animal work, have the subjects been treated in an ethical manner and the authors complied with the appropriate guidelines?

    Yes

  4. Version published to 10.1099/acmi.0.001150.v3 on Access Microbiology
  6. Access Microbiology

    Comments to Author

    The study evaluates the impact of the COVID-19 pandemic on bacterial distribution, antimicrobial resistance, and antibiotic consumption in a university hospital in Türkiye. A retrospective analysis was conducted comparing two-year periods: before and during the pandemic. More than 13,000 isolates from blood, urine, and LRT cultures were included. During the pandemic, ESBL rates and resistance to 3rd-generation cephalosporins and carbapenems increased in E. coli and K. pneumoniae. Antibiotic consumption, measured as DDD per 1,000 patient-days, increased for most agents. These results highlight the opportunity for antimicrobial stewardship. The main strengths of the study are its large sample size, a symmetric study period (pre- and pandemic), and the integration of antibiotic consumption and resistance data. Some typos: Key words antimicrobial stewardship instead of antibiotic stewardship (as a mesh term) Figure page 2: Klebsiella instead of kekiella Samples instead of samles Abstract: Lines 37-39: To include an hypothesis in a retrospective study is not as useful as in other studies types. And this hypothesis is not repeated in the main text. I suggest to delete or to frame it as a plausible expected effect Line 44: … in same, in blood, urine and lower….. what is same? Line 50: coagulase-negative staphylococci could be in several moments a contaminant and not an infectious pathogen… was it investigated? Line 51: The first time a bacterium is described requires the complete name (genus and species). It applies to the rest of the document. Lines 52-54: 3rd generation cephalosporin is an indirect marker of ESBL. I suggest leaving ESBL. Line 57: Contrary to our wishes…it is not a conclusion from the study. Maybe it can be omitted Methods Include a sentence related to the hospital's main characteristics, for example: is it a general hospital? How many beds does it have? Lines 99-101 are repeated in 101-104 and in 106 Results Lines 130 to 132. It would be useful to separate the counts of blood, urine, and LRT cultures between the two periods in the text and delete table 1. K. pneumoniae instead of K. pneumonia Lines 143-146 could be summarized: The ESBL frequency of E. coli strains increased form 43.7% before the pandemic to 49.3% during the pandemic (p=0.021), and ESBL of K. pneumoniae increased from 58.9% to 67.9% during the same periods (p=0.004). Table 2 Having all denominators in each column makes the table difficult to read. They could instead be placed at the top: blood (3,648) in the pre-pandemic section and 2,970 in the pandemic section, with one row for urine including the corresponding numbers, and another for LRT. A table footnote should be added defining the abbreviation LRT. Table 3. Ampicillin resistance to CoNS and S. aureus is not useful Discussion Line196-197 MRSA and CoNS strains associated with mortality…. This outcome wasn't included in this study, and the next sentence is related to resistance; there isn't a link. The discussion is too long and includes a detailed description of the results. I recommend summarizing, and some results, such as the high resistance of MRSA and vancomycin to E. faecium, could be compared with other manuscripts.

    Please confirm that no generative AI tools or large language models have been used to generate this peer review report or to assist with any part of the peer review process.

    I confirm no generative AI tools were used in preparation of this review.

    Please rate the manuscript for methodological rigour

    Good

    Please rate the quality of the presentation and structure of the manuscript

    Good

    To what extent are the conclusions supported by the data?

    Strongly support

    Do you have any concerns of possible image manipulation, plagiarism or any other unethical practices?

    No

    Is there a potential financial or other conflict of interest between yourself and the author(s)?

    No

    If this manuscript involves human and/or animal work, have the subjects been treated in an ethical manner and the authors complied with the appropriate guidelines?

    Yes

  7. Access Microbiology

    Comments to Author

    Reviewer Report on: Trends in antimicrobial resistance and antibiotic use before and during the COVID-19 pandemic in a university research and practice hospital in Türkiye I would like to thank the Editor for the opportunity to review this revised manuscript, and I thank the authors for their careful and constructive response to the previous comments. The revisions made in this second version have clearly improved the clarity, methodological transparency, and overall scientific tone of the manuscript. In particular, the authors have taken care to better align the interpretation of their findings with the descriptive and ecological nature of the study, which is appreciated. 1. Methodological rigour, reproducibility, and availability of underlying/supporting data The study is methodologically appropriate for its stated aim as a single-centre, retrospective, descriptive surveillance analysis. The design, data sources, and analytical approach are clearly described, and the authors consistently frame the findings as ecological trends rather than causal relationships. Compared with earlier versions, Revision 2 demonstrates a substantial improvement in methodological transparency. Microbiological identification and antimicrobial susceptibility testing follow EUCAST recommendations and are adequately detailed. The decision to group "susceptible, increased exposure" isolates with susceptible categories is explicitly justified for trend analysis purposes. While this approach may influence absolute resistance proportions, it is reasonable given the descriptive objective of the study. Antibiotic consumption analysis using defined daily doses (DDD) per 1000 inpatient-days is appropriate and consistent with international standards for hospital-level surveillance. However, the absence of adjustment for patient case mix, hospital occupancy, admission rates, or disease severity during the pandemic limits interpretability. These constraints reflect data availability rather than methodological oversight. The study is reproducible at the institutional level, as similar analyses could be replicated in other hospitals with routine microbiology and pharmacy databases. Reproducibility at a more granular epidemiological level is limited by the lack of patient-level clinical data and ward/ICU stratification. Regarding data availability, the manuscript indicates that all relevant data are presented within the article, with supplementary material also provided. Minor clarification in the data availability statement would improve consistency and transparency. 2. Presentation of results The results are comprehensive and supported by detailed tables describing bacterial distribution, resistance patterns, and antibiotic consumption. The revised manuscript shows improved synthesis, with clearer emphasis on clinically relevant findings rather than exhaustive repetition of tabulated values. The explicit acknowledgement that the very high proportion of coagulase-negative staphylococci (CoNS) in blood cultures likely reflects a mixture of true infection and contamination is a significant strength and improves interpretive credibility. The addition of absolute and relative changes alongside p-values for key resistance outcomes (e.g. ESBL and carbapenem resistance in Enterobacterales) enhances clinical relevance. Nevertheless, the large volume of descriptive data and multiple statistical comparisons necessitate cautious interpretation, particularly where statistical significance is emphasized without full contextualization of effect size or underlying confounding factors. 3. Communication of key findings The manuscript is generally well structured and clearly written. The revised title appropriately reflects the descriptive nature of the study and avoids causal language that is not supported by the design. The abstract is clearer, more neutral, and better aligned with the data presented. The introduction now more clearly identifies the local evidence gap and justifies the relevance of hospital-level data from Türkiye. The discussion is more disciplined than in earlier versions, with clearer separation between data-supported observations and speculative explanations. Overall, the organization and style effectively communicate the key findings within the limits of the available data. 4. Literature analysis and discussion The discussion demonstrates good engagement with relevant international and regional literature on antimicrobial resistance and antibiotic use before and during the COVID-19 pandemic. Comparisons with other hospital-based studies are generally appropriate and informative. A notable improvement in Revision 2 is the explicit framing of potential explanatory mechanisms such as changes in infection prevention practices or antibiotic prescribing as hypotheses rather than inferred causes. This aligns well with the ecological design of the study and avoids overstating conclusions. Some repetition in literature comparisons remains, but this does not substantially detract from the overall quality of the discussion. 5. Substantive weaknesses of the study The main weaknesses of the manuscript are inherent to the study design and data sources: * Absence of patient-level clinical data (e.g. demographics, comorbidities, COVID-19 status, severity, outcomes), limiting adjustment for confounding. * Lack of ward- or ICU-level stratification, potentially masking heterogeneity within the hospital. * Inability to distinguish colonisation from true infection, particularly relevant given the high proportion of CoNS in blood cultures. * Treatment of repeated isolates obtained more than 10 days apart as separate episodes, which may overestimate resistance in patients with prolonged hospitalization. * Limited analytical depth, with reliance on chi-square testing across numerous comparisons and no formal time-series analysis. * Interpretation of antibiotic consumption data without contextual information on hospital activity, admissions, or case mix. These limitations restrict causal inference and detailed interpretation but do not undermine the descriptive value of the surveillance data. Overall assessment Manuscript represents a clear improvement in methodological transparency, interpretive caution, and scientific tone. The manuscript provides valuable descriptive information on antimicrobial resistance patterns and antibiotic consumption trends in a tertiary hospital during the COVID-19 pandemic. Importantly, the authors have explicitly identified and addressed the key methodological and interpretive weaknesses within a clearly structured Limitations section, appropriately framing the findings as descriptive ecological trends rather than causal evidence. Subject to minor editorial refinements, the manuscript is suitable for publication within the scope of the journal. Recommendation: Accept for publication

    Please rate the manuscript for methodological rigour

    Very good

    Please rate the quality of the presentation and structure of the manuscript

    Very good

    To what extent are the conclusions supported by the data?

    Partially support

    Do you have any concerns of possible image manipulation, plagiarism or any other unethical practices?

    No

    Is there a potential financial or other conflict of interest between yourself and the author(s)?

    No

    If this manuscript involves human and/or animal work, have the subjects been treated in an ethical manner and the authors complied with the appropriate guidelines?

    Yes

  8. Version published to 10.1099/acmi.0.001150.v2 on Access Microbiology
  9. Access Microbiology

    Thank you for submitting your manuscript entitled “The impact of the pandemic on antimicrobial resistance and antibiotic use in a University Research and Practice Hospital in Türkiye” to Access Microbiology. The study addresses a timely and clinically relevant question, and the dataset has the potential to contribute meaningfully to the growing literature on antimicrobial resistance trends during the COVID-19 pandemic. Before proceeding to external peer review, I consider that a Major Revision is required. This initial editorial assessment has identified several areas where clarification, refocusing and methodological restraint would substantially strengthen the manuscript and improve its interpretability. Addressing these points at this stage would both enhance scientific rigour and facilitate a more constructive peer-review process. Title and scope The title accurately reflects the study content and setting, and the focus on a defined pre-pandemic and pandemic comparison in a single tertiary hospital is appropriate for Access Microbiology. However, the phrasing implies a broad causal assessment of “impact”, which the retrospective, descriptive design cannot fully support. A more cautious framing that foregrounds observed trends rather than impact or effect would better align expectations with the data presented. Abstract The abstract is clearly structured and largely consistent with the main text, but the hypothesis statement is problematic. The assertion that antibacterial consumption and resistance “must have decreased” during the pandemic introduces an a priori assumption that is neither justified nor necessary, and it weakens the neutrality of the study design. The methods description is adequate but would benefit from a clearer indication that this is a single-centre, retrospective analysis without patient-level adjustment. The conclusions are broadly supported by the results, although they could be more restrained, particularly in avoiding implicit value judgements about institutional performance. Introduction The introduction provides a conventional overview of antimicrobial resistance and situates the study within the context of COVID-19–related disruptions. While the dual potential for reduced transmission and increased empirical antibiotic use is appropriately acknowledged, the narrative relies heavily on generalised statements and would benefit from a more focused engagement with prior hospital-level surveillance studies. The aim is clearly stated, but the underlying rationale would be strengthened by explicitly identifying the local evidence gap this study seeks to address, rather than reiterating the global AMR problem. Methods The methods section is generally clear and transparent, with defined study periods, specimen sources and microbiological workflows. The exclusion criteria are appropriate. That said, several methodological limitations require clearer signalling. The reliance on routine laboratory data without stratification by ward type, ICU status, COVID-19 diagnosis or patient demographics substantially limits interpretability. The aggregation of repeated isolates beyond a 10-day window may also inflate resistance estimates. While EUCAST standards are correctly referenced, the inclusion of “intermediate” results within the susceptible category should be justified more explicitly, given its potential impact on trend comparisons. The antibiotic consumption analysis using DDD per 1000 patient-days is appropriate, but the absence of adjustment for case mix or occupancy changes during the pandemic should be acknowledged upfront rather than deferred to the discussion. Results The results are comprehensive and clearly presented, with tables that contain substantial data. However, the volume of descriptive statistics risks obscuring key signals. Greater synthesis in the text, drawing attention to the most clinically relevant changes, would improve readability. The high proportion of CoNS in blood cultures raises questions about contamination versus true infection, which are not addressed. The reporting of statistically significant changes in resistance is technically correct, but effect sizes and clinical relevance are not discussed, leading to an overemphasis on p-values. In the antibiotic consumption data, absolute increases are shown, but without contextualisation to admission rates or case severity, limiting causal interpretation. Discussion The discussion demonstrates familiarity with the literature and situates the findings within a broad international context. However, it frequently attributes observed resistance changes directly to antibiotic consumption or infection control practices without sufficient evidence from the study design to support these links. Several mechanistic explanations, such as improved environmental cleaning reducing Acinetobacter resistance, are plausible but speculative and should be framed explicitly as hypotheses rather than inferred causes. Comparisons with other studies are extensive, sometimes repetitive, and could be more selective, focusing on settings and methodologies most comparable to the present study. The discussion would also benefit from a clearer separation between observations supported by the data and broader commentary on antimicrobial stewardship. Limitations Although limitations are acknowledged, they are treated briefly. Key constraints, particularly the lack of patient-level data, inability to distinguish colonisation from infection, and potential shifts in hospital activity during the pandemic, warrant more prominence and depth. Explicit discussion of how these limitations may bias the observed trends would strengthen the credibility of the conclusions. Conclusion The conclusion reiterates the main findings but adopts a normative tone that may exceed what the data can substantiate. Statements suggesting institutional failure to capitalise on the pandemic risk appearing judgmental and would benefit from more neutral phrasing. Emphasising the descriptive nature of the findings and their role in informing future stewardship or surveillance efforts would provide a more balanced ending.

  10. Version published to 10.1099/acmi.0.001150.v1 on Access Microbiology