Research | Open Access | Volume 9 (4): Article  187 | Published: 11 Nov 2026

Characterization of multidrug resistance and antibiotic susceptibility among bacterial isolates from selected clinical specimens in a Kenyan referral hospital

PDF in Progress
Crossmark: Check for updates

Views: 41

   Menu, Tables and Figures

Navigate this article

Table 1: Distribution of Bacterial Isolates by Specimen Type

Table 2: Antimicrobial susceptibility results by specimen type, percentage of AST results (N=870)

Keywords

  • Antimicrobial resistance
  • Multidrug resistance
  • Antibiotic susceptibility testing
  • Bacterial isolate
  • Clinical specimens
  • Antimicrobial stewardship
  • Kenya

Faith Joan Maina1,&, Alfred Owino Odongo1, Jesse Gitaka2, Brian Mburu3, Simon Irungu4, Anthony Kipkoech4, Susan Githii5

1Department of Epidemiology and Biostatistics, Mount Kenya University, Kenya, 2School of Medicine, Mount Kenya University, Kenya, 3Department of Quality Assurance, Thika Level 5 Hospital Laboratory, Kenya, 4Center for Research in Infectious Diseases, Mount Kenya University, 5AMR Surveillance Focal Point, Kenya National Public Health Institute, Kenya

&Corresponding author: Faith Joan Maina, Department of Epidemiology and Biostatistics, Mount Kenya University, Kenya, Email: fjoannecarrie@gmail.com

Received: 22 Jul 2026, Accepted: 02 Nov 2026, Published: 11 Nov 2026

Domain: Antimicrobial Resistance

Keywords: Antimicrobial resistance, multidrug resistance, antibiotic susceptibility testing, bacterial isolate, clinical specimens, antimicrobial stewardship, Kenya

©Faith Joan Maina et al. Journal of Interventional Epidemiology and Public Health (ISSN: 2664-2824). This is an Open Access article distributed under the terms of the Creative Commons Attribution International 4.0 License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Cite this article: Faith Joan Maina et al., Characterization of multidrug resistance and antibiotic susceptibility among bacterial isolates from selected clinical specimens in a Kenyan referral hospital. Journal of Interventional Epidemiology and Public Health. 2026; 9(4):187. https://doi.org/10.37432/jieph-d-26-00265

Abstract

Introduction: Antimicrobial resistance (AMR) complicates the treatment of bacterial infections and remains a major public health concern. Facility-level susceptibility data are needed to describe local resistance patterns and provide baselines for surveillance, particularly in referral hospitals. This study determined antibiotic susceptibility patterns and the prevalence of multidrug-resistant (MDR) bacterial isolates from urine, pus, and body fluid specimens at Thika Level 5 Hospital.
Methods: A retrospective cross-sectional study was conducted using archived bacterial isolates collected between January 2021 and December 2023 at Thika Level 5 Hospital. A total of 250 isolates comprising Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa, and Staphylococcus aureus were analyzed from urine (n=91), pus (n=134), and body fluids (n=25). Isolates were revived and tested by Kirby-Bauer disk diffusion according to Clinical and Laboratory Standards Institute (CLSI) procedures. A total of 870 antimicrobial susceptibility test (AST) results were analyzed using WHONET 2023 and SPSS version 26. Descriptive statistics, chi-square testing, and binary logistic regression were used. MDR was defined as non-susceptibility to at least one agent in three or more antimicrobial classes.
Results: A total of 250 bacterial isolates were revived and tested, yielding 870 AST results, of which 308 (35.4%) were resistant. Resistance proportions by specimen were 43.5% for body fluids, 41.3% for urine, and 28.7% for pus. E. coli predominated in urine samples (60.4%), while S. aureus predominated in pus specimens (50.7%). Overall MDR prevalence was 23.6%, highest among urine isolates (36.3%), followed by body fluids (24.0%) and pus (14.9%), and the specimen type was significantly associated with MDR (χ²=13.65, p=0.001).
Conclusion: Antimicrobial resistance was observed among isolates tested at Thika Level 5 Hospital during 2021–2023, with multidrug resistance particularly prevalent among urinary isolates. These findings provide a facility-level baseline and support strengthened antimicrobial stewardship, enhanced diagnostic capacity, and the development of specimen-specific, locally informed treatment guidance.

Introduction

Antimicrobial resistance (AMR) is an expanding global threat to the treatment of bacterial infections. The World Health Organization (WHO) Global antibiotic resistance surveillance report 2025, based on more than 23 million bacteriologically confirmed cases reported by 104 countries in 2023, estimated that one in six laboratory-confirmed bacterial infections causing common infections worldwide was resistant to antibiotic treatment; resistance increased in more than 40% of monitored pathogen–antibiotic combinations between 2018 and 2023 [1].

Kenya has strengthened its national response through the National Action Plan for the Prevention and Containment of AMR 2023–2027, which emphasizes surveillance, diagnostics, antimicrobial stewardship, infection prevention and research [2]. A 2025 assessment of the preceding national plan identified persistent challenges including weak laboratory–clinical linkages, intermittent availability of laboratory supplies and incomplete transmission of facility-level AMR data [3]. More recent national and hospital antimicrobial-consumption data also show increasing use, with national consumption rising from 14.3 to 22.2 defined daily doses per 1,000 inhabitants per day between 2020 and 2024 [4].

Facility-level evidence is particularly important because resistance patterns vary by pathogen, specimen and healthcare setting. At Thika Level 5 Hospital, a 2024 study reported high antimicrobial resistance among clinical and environmental isolates and detected an extended-spectrum beta-lactamase phenotype in 65.9% (29/44) of enteric isolates, but the study had a different sampling frame and included environmental specimens and organisms outside the present study [5]. Recent Kenyan studies in other referral settings have likewise documented substantial resistance among wound isolates and multidrug-resistant or carbapenemase-producing Gram-negative carriage [6,7].

Despite these reports, facility-specific analyses that compare susceptibility patterns across urine, pus and body fluid specimens and simultaneously characterize MDR using patient-level variables remain limited. The present study uses archived isolates collected from 2021 to 2023; therefore, its value is primarily as a historical facility baseline against which subsequent surveillance can be compared, rather than as an estimate of current 2026 prevalence. This distinction is important when interpreting retrospective AMR findings.

The study therefore aimed to determine antibiotic susceptibility patterns and the prevalence of MDR among E. coli, K. pneumoniae, P. aeruginosa and S. aureus isolates recovered from urine, pus and body fluid specimens at Thika Level 5 Hospital, and to assess whether specimen type and selected patient characteristics were associated with MDR.

Methods

Study design and setting
This retrospective cross-sectional study analysed archived bacterial isolates and laboratory records from January 2021 to December 2023 at Thika Level 5 Hospital (TL5H). The study focused on susceptibility patterns and MDR across urine, pus and body fluid specimens. The study was conducted at TL5H, a referral and teaching hospital in Kiambu County, Kenya, serving an urban and peri-urban population and receiving patients from surrounding counties. The microbiology laboratory routinely processes clinical specimens, including urine, pus and body fluids, for culture, organism identification and antimicrobial sensitivity testing.

The laboratory follows standard microbiological procedures and contributes to diagnostic and surveillance activities for infectious diseases at the facility. The facility follows standard microbiological procedures and contributes significantly to diagnostic and surveillance efforts related to infectious diseases.

Study population
The study population comprised archived isolates obtained from specimens submitted to the TL5H microbiology laboratory between January 2021 and December 2023. The analysis included E. coli, K. pneumoniae, P. aeruginosa and S. aureus.

Inclusion criteria

  • Bacterial isolates from urine, pus and body fluid specimens
  • Isolates with complete laboratory records containing age, gender, patient location, organism identification and AST results.

Exclusion criteria

  • Isolates with incomplete or missing data
  • Unidentified or unspecified bacterial species
  • Duplicate isolates from the same patient episode

Sample size and sampling procedure
A total of 250 isolates were included: urine (n=91), pus (n=134) and body fluids (n=25). Isolates were selected from the archived laboratory repository using stratified systematic sampling, with specimen type as the stratification variable and proportional allocation across strata. Stratification by specimen type allowed resistance patterns to be compared across infection sites, and systematic sampling was applied within each stratum.

Data source and data collection
The primary data source was the TL5H routine microbiology archive. A structured extraction tool was used to capture age, gender, inpatient/outpatient location, specimen type, organism identification and AST results. Data were anonymized at extraction. Because the archive was a routine clinical record system rather than a population-based surveillance database, the study was limited to information recorded in the available laboratory records.

Laboratory procedures
Archived bacterial isolates were revived by subculture on nutrient agar using standard microbiological procedures and re-identified using conventional biochemical tests. Antibiotic susceptibility testing (AST) was performed using the Kirby-Bauer disk diffusion method on Mueller-Hinton agar in accordance with Clinical and Laboratory Standards Institute (CLSI M02 and M100 guidelines) [8,9].

Quality control
Quality control was performed using reference strains E. coli ATCC 25922, K. pneumoniae ATCC 700603, S. aureus ATCC 25923 and S. aureus ATCC BAA-977. Results were interpreted from inhibition-zone diameters together with phenotypic confirmatory testing where applicable. These strains served as controls for routine susceptibility testing (E. coli ATCC 25922, ESBL-negative), ESBL phenotypic confirmation (K. pneumoniae ATCC 700603, ESBL-positive), testing of methicillin-susceptible S. aureus (S. aureus ATCC 25923) and inducible clindamycin resistance testing (S. aureus ATCC BAA-977, ICR-positive). The final interpretation of test results occurred through zone diameter readings, with confirmatory phenotypic testing where applicable.

Data management and analysis
Data were entered, cleaned and validated in Microsoft Excel and then analysed using WHONET 2023 and SPSS version 26 [10]. Frequencies and percentages summarised isolate distribution and AST results. The chi-square test of independence assessed the association between specimen type and MDR. Binary logistic regression evaluated age, gender and inpatient/outpatient status as predictors of MDR, with adjusted odds ratios (AORs) and 95% confidence intervals reported. MDR was defined as acquired non-susceptibility to at least one agent in three or more antimicrobial categories [11]. Statistical significance was set at p<0.05. Resistance percentages were calculated from AST results, whereas MDR prevalence was calculated at the isolate level.

Ethical considerations
Ethical clearance was granted by the Mount Kenya University Ethical Review Committee (MKU/IERC/3880). Permission to access archived records was obtained from TL5H Management (CGK/TL5H/02/19/2024), and national research approval was granted by the National Commission for Science, Technology and Innovation (NACOSTI-838371). Data were anonymized; individual patient consent was waived for this retrospective record-based study.

Results

Characterization of bacterial isolates
A total of 250 isolates from three specimen types were analysed: urine (91, 36.4%), pus (134, 53.6%) and body fluids (25, 10.0%). The most prevalent organisms were Escherichia coli (84, 33.6%) and Staphylococcus aureus (80, 32.0%), followed by Klebsiella pneumoniae (45, 18.0%) and Pseudomonas aeruginosa (41, 16.4%, Table 1). Specimen-specific distribution showed clear variation, where E. coli predominated in urine (55/91, 60.4%), S. aureus predominated in pus specimens (68/134, 50.7%), whereas body fluid isolates were distributed across all four species.

Antimicrobial susceptibility patterns
A total of 870 AST results were analysed. Overall, 308 (35.4%) results were resistant. Resistance was highest among body fluid isolates (43.5%), followed by urine (41.3%) and pus (28.7%). The corresponding susceptible and intermediate proportions are shown in Table 2. These percentages describe AST results and should not be interpreted as clinical treatment-response rates.

Multidrug resistance
Overall MDR prevalence was 23.6%. Prevalence varied by specimen type, with 36.3% among urine isolates, 24.0% among body fluid isolates and 14.9% among pus isolates. Specimen type was significantly associated with MDR (χ²=13.65, p=0.001).

Discussion

This study characterised AST results and MDR among four bacterial species isolated from urine, pus and body fluid specimens at TL5H during 2021–2023. The results provide a facility-specific historical baseline rather than a current prevalence estimate. The study found resistance in 35.4% of AST results. This burden should be interpreted in the context of continuing global and Kenyan AMR pressures. WHO surveillance reported widespread resistance in 2023, while Kenyan monitoring has identified ongoing implementation and laboratory-system challenges [1,3]. A 2026 analysis also documented increasing national antibiotic consumption between 2020 and 2024 [4]. These external data are not directly comparable with the present hospital isolate dataset because of differences in populations, sampling and definitions, but they reinforce the relevance of continued facility-level surveillance.

E. coli predominated in urine samples, while S. aureus predominated in pus specimens. This pattern is consistent with the expected concentration of particular pathogens within infection sites and supports specimen-specific analysis of susceptibility data. The distribution also complements previous observations from TL5H and other Kenyan referral facilities, although the earlier TL5H study used a different sampling frame and included environmental specimens [5,6]. The findings emphasize that an overall facility antibiogram can obscure important differences between specimen categories. For this reason, the present analysis reports resistance and MDR separately by specimen type.

The relatively even distribution of the four organisms in body fluids should be interpreted cautiously because the body-fluid stratum was small (n=25), which limits precision. Overall MDR prevalence was 23.6%, with the highest prevalence in urine isolates (36.3%). The operational definition used here follows the international consensus definition of acquired non-susceptibility to at least one agent in three or more antimicrobial categories [11]. Recent Kenyan studies have also documented MDR and other resistance phenotypes in referral-hospital settings, including the detection of ESBL phenotypes at TL5H and carbapenemase-producing Gram-negative carriage in a Kenyan county referral hospital [5,7]. The differing study designs and populations mean that these findings should be used for contextual comparison rather than direct prevalence ranking.

The higher MDR prevalence among urinary isolates may reflect the influence of repeated healthcare or antimicrobial exposure, recurrent infection and other factors, but these mechanisms were not measured in the present dataset and therefore cannot be established as causes.

Pus isolates had a lower MDR prevalence than urine isolates (14.9%), but the retrospective design does not permit inference about differences in treatment exposure or clinical outcomes between specimen groups.

Resistance varied by specimen type, with the highest proportion of resistant AST results in body fluids (43.5%) and the lowest in pus (28.7%). Because the body-fluid sample was small, the estimate may be unstable. The finding nevertheless illustrates why specimen-specific surveillance can add information beyond an aggregate resistance measure.

Possible explanations for the higher resistance observed among body fluid isolates include prior antibiotic exposure, longer hospitalisation, and greater disease severity. However, none of these factors was measured in the present study, so they should be regarded as hypotheses rather than demonstrated determinants in this dataset.

The lower resistance proportion observed in pus should not be interpreted as evidence of better treatment outcomes because treatment response and prescribing data were not available. Instead, the result indicates a specimen-specific difference in laboratory susceptibility profiles.

The findings support continued facility-level AMR surveillance and specimen-specific analysis. They provide a historical reference point for later surveillance at TL5H and illustrate the value of linking laboratory susceptibility data with patient characteristics. However, the study did not include prescribing data, prior antibiotic exposure or treatment outcomes; consequently, it cannot establish that empirical therapy was ineffective or quantify treatment failure.

Study limitations
Several limitations should be considered. First, the retrospective single-facility design limits generalizability beyond TL5H. Second, the source was a routine laboratory archive; completeness and accuracy of the underlying records could not be independently quantified, and exclusion of records with missing key variables may have introduced selection bias. Third, revival and re-testing of archived isolates may introduce recovery or viability bias. Fourth, only four bacterial species and three specimen categories were included, and the body-fluid stratum was small, limiting precision. Fifth, the dataset did not contain comprehensive information on comorbidities, prior antibiotic exposure, clinical severity, antimicrobial prescribing, treatment outcomes or length of stay, limiting causal interpretation of patient-level associations and preventing assessment of clinical treatment effectiveness. Sixth, no molecular assays were performed, so resistance mechanisms could not be characterized. Finally, the analysis did not model annual temporal trends; therefore, the findings should be interpreted as a 2021–2023 baseline rather than evidence of current AMR prevalence.

Conclusion

This study found resistance in 35.4% of AST results and MDR in 23.6% of isolates from selected clinical specimens at TL5H during 2021–2023. Resistance and MDR varied by specimen type, with the highest MDR prevalence observed among urinary isolates and a statistically significant association between specimen type and MDR.

Increasing age was associated with higher odds of MDR in the adjusted analysis. Taken together, these findings provide a historical facility-level baseline for comparison with subsequent surveillance, while emphasizing the need for future studies that capture contemporary laboratory, prescribing and clinical outcome data.

The retrospective design does not support conclusions about current prevalence or the effectiveness of empirical treatment. Such questions require prospective or continuously updated surveillance linked to treatment and outcome data. Future studies should extend surveillance longitudinally, include a broader range of pathogens and multiple facilities, capture prior antimicrobial exposure, prescribing, comorbidities and treatment outcomes, and incorporate molecular or genomic methods to characterize resistance mechanisms.

What is already known about the topic

  • Antimicrobial resistance is a global health threat and resistant patterns vary by setting and specimen type. However, facility-level, specimen-specific surveillance data from Kenyan referral hospitals remain limited.

What this  study adds

  • This study provides a 2021–2023 facility-level baseline of antimicrobial resistance and multidrug resistance among urine, pus and body fluid isolates at a Kenyan referral hospital, showing that MDR prevalence was highest among urinary isolates and differed significantly by specimen type, which can inform specimen-specific surveillance and local antimicrobial stewardship.

Competing interest

The authors of this work declare no competing interests.

Data availability statement
The datasets generated and analyzed during this study are not publicly available due to hospital data governance policies but are available from the corresponding author (F.N.) on reasonable request and with permission from Thika Level 5 Hospital management.

Disclaimer
The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of Thika Level 5 Hospital or Mount Kenya University.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Acknowledgements

The authors sincerely thank the management and laboratory staff of Thika Level 5 Hospital for facilitating access to the archival data. We also acknowledge the Center for Research in Infectious Diseases at Mount Kenya University for their academic support.

Authors’ contributions

Conceptualization: Faith Joan Maina
Methodology: Faith Joan Maina, Alfred Owino Odongo
Data Curation: Faith Joan Maina
Formal Analysis: Faith Joan Maina
Investigation: Faith Joan Maina
Validation: Alfred Owino Odongo, Jesse Gitaka
Supervision: Alfred Owino Odongo, Alfred Owino Odongo, Jesse Gitaka
Writing – Original Draft: Faith Joan Maina
Writing – Review & Editing: Alfred Owino Odongo, Jesse Gitaka

Tables

Table 1: Distribution of Bacterial Isolates by Specimen Type
Specimen Type E. coli K. pneumoniae P. aeruginosa S. aureus Total
Urine 55 25 8 3 91
Pus 23 15 28 68 134
Body Fluids 6 5 5 9 25
Total 84 45 41 80 250
Table 2: Antimicrobial susceptibility results by specimen type, percentage of AST results (N=870)
Specimen TypeSusceptible (%)Intermediate (%)Resistant (%)
Body Fluids47.19.443.5
Urine47.910.741.3
Pus53.118.228.7
 

References

  1. World Health Organization. Global antibiotic resistance surveillance report 2025 [Internet]. Geneva (Switzerland): World Health Organization; 2025 Oct 13 [cited 2026 Nov 11]. 114 p. Available from: https://www.who.int/publications/i/item/9789240116337
  2. Kenya. Ministry of Health. National action plan for the prevention and containment of antimicrobial resistance 2023-2027: monitoring and evaluation framework [Internet]. Nairobi (Kenya): Ministry of Health; 2023 [cited 2026 Nov 11]. Available from: http://guidelines.health.go.ke:8000/media/ME_Framework_for_NAP_AMR_2023-2027_-_KENYA.pdf
  3. Mukoko J, Wesangula E, Gitonga N, Kusu N, Odhiambo C, Tanui E, Azegele A, Ndanyi R, Joshi MP, Hafner T, Konduri N. Kenya’s National Action Plan on antimicrobial resistance: measuring implementation progress. Front Trop Dis [Internet]. 2025 Mar 20 [cited 2026 Nov 11];6:1540713. Available from: https://www.frontiersin.org/journals/tropical-diseases/articles/10.3389/fitd.2025.1540713/full doi:10.3389/fitd.2025.1540713
  4. Mariita KM, Ombajo LA, Ngacha CM, Tongola B, Khamira V, Njogu R, Wanga K, Momanyi L, Andrew J, Otieno E, Ong’ayo MN, Karuri S, Ochola L, Obala N, Oluka MO, Tanui E, Kandie SC, Kibira S, Aywak D, Omar S. National and hospital level antimicrobial consumption patterns in Kenya. Antibiotics (Basel) [Internet]. 2026 Jun 9 [cited 2026 Nov 11];15(6):587. Available from: https://www.mdpi.com/2079-6382/15/6/587 doi:10.3390/antibiotics15060587
  5. Kimani R, Wakaba P, Kamita M, Mbogo D, Mutai W, Ayodo C, Suliman E, Kanoi BN, Gitaka J. Detection of multidrug-resistant organisms of concern including Stenotrophomonas maltophilia and Burkholderia cepacia at a referral hospital in Kenya. PLoS One [Internet]. 2024 Apr 16 [cited 2026 Nov 11];19(4):e0298873. Available from: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0298873 doi:10.1371/journal.pone.0298873
  6. Onsando S, Masanta W, Nyerere A, Njire M, Moriasi G. Diversity and antibiotic susceptibility profiles of bacterial isolates from wound infections in patients at the surgical unit of Kisii Teaching and Referral Hospital, Kenya. PLoS One [Internet]. 2025 Jun 17 [cited 2026 Nov 11];20(6):e0326048. Available from: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0326048 doi:10.1371/journal.pone.0326048
  7. Maina AK, Maingi J, Mutuma CK, Musyoki AM. Asymptomatic faecal carriage of multidrug-resistant and carbapenemase-producing Gram-negative bacteria among inpatients and outpatients in a Kenyan county referral hospital: a cross-sectional study. BMC Infect Dis [Internet]. 2026 Apr 11 [cited 2026 Nov 11];26(1):991. Available from: https://link.springer.com/article/10.1186/s12879-026-12896-5 doi:10.1186/s12879-026-12896-5
  8. Clinical and Laboratory Standards Institute. M100 Performance standards for antimicrobial susceptibility testing [Internet]. Wayne (PA): Clinical and Laboratory Standards Institute; 2026 Jan 26 [cited 2026 Nov 11]. Available from: https://clsi.org/shop/standards/m100/
  9. Clinical and Laboratory Standards Institute. M02 Performance standards for antimicrobial disk susceptibility tests [Internet]. Wayne (PA): Clinical and Laboratory Standards Institute; 2024 Mar 19 [cited 2026 Nov 11]. Available from: https://clsi.org/shop/standards/m02/
  10. WHO Collaborating Centre for Surveillance of Antimicrobial Resistance. WHONET [Internet]. Boston (MA): WHO Collaborating Centre for Surveillance of Antimicrobial Resistance; 2023 [cited 2026 Nov 11]. Available from: https://whonet.org/
  11. Magiorakos AP, Srinivasan A, Carey RB, Carmeli Y, Falagas ME, Giske CG, Harbarth S, Hindler JF, Kahlmeter G, Olsson-Liljequist B, Paterson DL, Rice LB, Stelling J, Struelens MJ, Vatopoulos A, Weber JT, Monnet DL. Multidrug-resistant, extensively drug-resistant and pandrug-resistant bacteria: an international expert proposal for interim standard definitions for acquired resistance. Clin Microbiol Infect [Internet]. 2011 May 7 [cited 2026 Nov 11];18(3):268-81. Available from: https://onlinelibrary.wiley.com/doi/full/10.1111/j.1469-0691.2011.03570.x doi:10.1111/j.1469-0691.2011.03570.x
  12. Mishra SK, Sharma K, Georrge JJ. Antimicrobial resistance: enzymes, proteins, and computational resources. Curr Pharm Des [Internet]. 2025 Aug [cited 2026 Nov 11];32(10):731-41. Available from: https://www.benthamdirect.com/content/journals/cpd/10.2174/0113816128415482250721112427 doi:10.2174/0113816128415482250721112427
  13. Kachhadiya DK, Mishra SK, Siddiquee NH, Roy S, Georrge JJ. Marine Bacillus safensis DJ1 metabolites disrupt P. aeruginosa virulence: an integrated experimental and network pharmacology and dynamics approach. Folia Microbiol (Praha) [Internet]. 2026 Mar 10 [cited 2026 Nov 11]. Available from: https://link.springer.com/article/10.1007/s12223-026-01453-x doi:10.1007/s12223-026-01453-x
Views: 41