Clinical Significance of Extended-Spectrum β-Lactamase (ESBL) and AmpC β-Lactamase (AmpC) Coproduction among Uropathogenic Escherichia coli from Diabetic and Nondiabetic Patients: Implications for Antimicrobial Resistance and Treatment Outcomes

Abstract

Background

Urinary tract infections (UTIs) caused by Escherichia coli remain a major global health burden, particularly among diabetic patients who are prone to recurrent and complicated infections.

Aim

This article determines the prevalence, antimicrobial resistance patterns, and clinical predictors of extended-spectrum β-lactamase (ESBL) and AmpC β-lactamase coproduction among uropathogenic E. coli isolates.

Methods

Cross-sectional study was conducted on 300 adult patients with culture-confirmed E. coli UTIs at Misurata Medical Center, including 150 people living with diabetes (PLWD) and 150 nondiabetic participants. Midstream urine specimens were collected and cultured on standard media. Antimicrobial susceptibility testing followed the Clinical and Laboratory Standards Institute 2024 guidelines, while ESBL and AmpC β-lactamases were phenotypically detected using the double-disk synergy and inhibitor-based disk tests. Data were analyzed using SPSS v27 and R v4.3.2. Categorical and continuous variables were compared using chi-square/Fisher's exact and Mann–Whitney U tests, respectively. Predictors of coproduction and its impact on treatment failure were identified via multivariate logistic regression; a p-value of < 0.05 was considered statistically significant.

Results

The total prevalence of ESBL, AmpC, and ESBL + AmpC coproduction was 58.3, 14.3, and 14.0%, respectively. While the prevalence of ESBL-only and AmpC-only phenotypes did not differ significantly between PLWD and nondiabetic patients, coproduction was significantly higher in PLWD (22%) compared with nondiabetic patients (6%; p < 0.001) and was associated with extensive multidrug resistance (93%). Diabetes mellitus (adjusted odds ratio [OR] = 3.64; p = 0.001) and prior antibiotic exposure (adjusted OR = 2.18; p = 0.041) were independent predictors of coproduction.

Conclusion

ESBL + AmpC coproduction in uropathogenic E. coli represents an emerging therapeutic and epidemiological concern, particularly among PLWD. Routine AmpC screening and rational antibiotic policies are urgently required to mitigate treatment failure and limit resistance.

Keywords

Escherichia coli - extended-spectrum β-lactamase (ESBL) - ampC β-lactamase - antimicrobial resistance - diabetes mellitus - urinary tract infection - Libya

Introduction

Urinary tract infections (UTIs) are among the most prevalent bacterial infections globally, accounting for significant morbidity, antibiotic consumption, and health care costs. Escherichia coli is the leading etiological agent of community-acquired UTIs (CA-UTIs).[1] Over the past two decades, the rapid evolution of β-lactamase–mediated resistance has severely undermined the clinical utility of β-lactam antibiotics, particularly the third-generation cephalosporins that form the backbone of empirical therapy.[2]

Extended-spectrum β-lactamases (ESBLs) and AmpC β-lactamases represent two major enzyme families responsible for hydrolyzing β-lactam antibiotics. While ESBLs confer resistance to oxyimino-cephalosporins and monobactams but remain inhibited by clavulanate, AmpC enzymes hydrolyze cephamycins and are poorly inhibited by β-lactamase inhibitors.[3] The coproduction of ESBL and AmpC enzymes within the same E. coli isolate creates a particularly worrisome phenotype that can evade both inhibitor combinations and extended-spectrum cephalosporins, resulting in multidrug-resistant (MDR) profiles and frequent therapeutic failure.[4] [5]

Although ESBL production has been extensively reported worldwide, the simultaneous presence of ESBL and AmpC β-lactamases has received far less attention, especially in developing regions. Recent surveillance data indicate that coproducers may represent up to 10 to 15% of E. coli isolates in some hospital settings and are associated with increased mortality, prolonged hospital stay, and higher rates of carbapenem consumption.[3] [6] Failure to detect AmpC activity in ESBL-positive isolates can therefore lead to inappropriate therapy and foster silent dissemination of MDR strains.[7]

This study specifically targeted CA-UTIs among adults presenting with acute urinary symptoms, excluding recent hospitalizations or indwelling devices to ensure a well-defined outpatient-based population.

Diabetes mellitus (DM) further complicates this scenario. People living with diabetes (PLWD) are predisposed to recurrent and complicated UTIs due to glycosuria, impaired neutrophil function, and reduced uroepithelial defense mechanisms.[8] Evidence suggests that E. coli isolates from PLWD exhibit higher resistance rates, a greater likelihood of ESBL production, and less favorable treatment outcomes compared with people without diabetes counterparts.[9] However, the interaction between host metabolic status (diabetes) and bacterial resistance mechanisms, specifically ESBL + AmpC coproduction, remains insufficiently explored, particularly across North African and Middle Eastern health care settings.

In light of these gaps, this study aimed to (1) determine the prevalence of ESBL + AmpC coproduction among uropathogenic E. coli isolates, (2) compare clinical and microbiological characteristics between PLWD and people without diabetes, (3) identify independent clinical predictors of coproduction, and (iv) assess its impact on therapeutic failure and hospital stay. By integrating clinical and microbiological parameters, this work aims to provide region-specific evidence supporting optimized empiric antibiotic policies and improved management of high-risk UTI patients.

Methods

Study Design and Setting

This analytical cross-sectional study was conducted over an 8-month period (March–October 2023). This was a single-center study conducted at Misurata Medical Center, a tertiary care referral hospital in Misurata, Libya, in collaboration with the Central Microbiology Laboratory. The study population included adult patients clinically diagnosed with UTI who yielded pure growth of E. coli on culture. Both inpatient and outpatient cases were enrolled; however, only CA-UTIs were included, and inpatients were eligible solely if they presented with CA-UTI criteria.

Sampling Method

A consecutive convenience sampling strategy was used, whereby all eligible patients presenting during the study period were enrolled to minimize selection bias.

Ethical Approval and Informed Consent

Ethical approval was obtained from the Ethics Committee of the Faculty of Pharmacy, Misurata University (Ref. 49/2023), which serves as the institutional review authority for student-led clinical research. Misurata Medical Center did not require a separate institutional review board submission at the time of data collection. All data were fully anonymized, and no patient identifiers were recorded.

Verbal informed consent was used in accordance with institutional regulations for minimal-risk observational research. Participants were informed about study purpose and confidentiality measures. The use of verbal consent is acknowledged as a limitation.

Inclusion and Exclusion Criteria

All included participants presented with clinical symptoms consistent with acute UTI (dysuria, urgency, frequency, suprapubic discomfort, or flank pain) and demonstrating significant bacteriuria (≥ 10⁵ CFU/mL) due to E. coli in urine culture. Both PLWD and people without diabetes were eligible. DM was confirmed by medical records, ongoing antidiabetic therapy, or glycated hemoglobin (HbA1c ≥ 6.5%).

Inpatients were included only when their infection met criteria for community acquisition (no hospitalization in the preceding 14 days). Patients admitted for management of UTI and meeting CA-UTI criteria were therefore eligible.

Patients were excluded if they met any of the following criteria:

• (1) Polymicrobial growth in urine culture.

• (2) Recent hospitalization within the preceding 14 days.

• (3) Indwelling catheterization or recent urinary instrumentation.

• (4) Incomplete clinical or microbiological data.

• (5) Prior inclusion (duplicate isolates).

• (6) Initiation of antibiotic therapy more than 48 hours before specimen collection.

These criteria ensured the inclusion of well-documented, monomicrobial E. coli infections and reduced confounding factors related to prior treatment or nosocomial exposure.

Sample Size and Patient Groups

A total of 300 nonduplicate urine isolates were analyzed, including 150 PLWD and 150 people without diabetes. This sample size was designed to achieve ≥ 80% statistical power to detect a 15% difference in coproduction prevalence between groups at a 95% confidence level.

Specimen Collection and Culture

Midstream clean-catch urine specimens were collected in sterile containers under aseptic conditions and transported to the laboratory within 1 hour of collection. Samples were cultured on MacConkey agar and CLED (Cystine-Lactose-Electrolyte-Deficient) agar (Oxoid, United Kingdom) and incubated aerobically at 37°C for 24 hours.

Significant bacteriuria was defined as ≥ 10⁵ CFU/mL of a single organism. Identification was based on standard biochemical reactions and confirmed by API 20E (bioMérieux, France) or VITEK 2 automated systems when available (instrument availability during routine laboratory workflow); otherwise, biochemical testing was used (IMViC, triple sugar iron, citrate, urease).

Antimicrobial Susceptibility Testing

Antimicrobial susceptibility testing was performed using the Kirby–Bauer disk diffusion method on Mueller–Hinton agar according to the Clinical and Laboratory Standards Institute (CLSI) 2024 performance standards.[10] The antibiotic panel included cefotaxime (30 µg), ceftazidime (30 µg), cefepime (30 µg), cefoxitin (30 µg), ciprofloxacin (5 µg), nitrofurantoin (300 µg), gentamicin (10 µg), amikacin (30 µg), piperacillin–tazobactam (100/10 µg), ertapenem (10 µg), imipenem (10 µg), and meropenem (10 µg). All antibiotic discs and media were obtained from Oxoid, United Kingdom.

Only E. coli American Type Culture Collection (ATCC) 25922 was available locally during the study period; the absence of an additional resistant-quality control strain (e.g., ATCC 35218) is acknowledged as a methodological limitation.

Isolates resistant to three or more antibiotic classes were categorized as MDR, this definition aligns with the international MDR criteria described by Magiorakos et al.[10]

Detection of ESBL Production

Screening for ESBL activity was performed according to the CLSI 2024 guidelines using reduced susceptibility to cefotaxime (30 µg) and/or ceftazidime (30 µg). Confirmatory testing employed the phenotypic confirmatory method (double-disk synergy test, DDST), where disks of cefotaxime (30 µg) and ceftazidime (30 µg) were placed 20 mm (center-to-center) from their respective clavulanate-containing disks (cefotaxime–clavulanate 30/10 µg and ceftazidime–clavulanate 30/10 µg).

An increase of ≥ 5 mm in inhibition zone diameter for either cephalosporin–clavulanate combination, compared with the corresponding cephalosporin alone, was interpreted as ESBL production.[11]

Cefotaxime and ceftazidime were selected for ESBL screening and confirmation because they are the primary oxyimino-cephalosporins recommended by the CLSI for detecting plasmid-mediated ESBL activity in Enterobacterales. Clavulanate-containing disks were included to differentiate ESBLs known to be inhibited by β-lactamase inhibitors from AmpC enzymes, which are not significantly inhibited.

Detection of AmpC β-Lactamase Production

Isolates showing resistance to cefoxitin (zone ≤ 18 mm) were further tested for AmpC production using the inhibitor-based AmpC disk test described by Coudron et al.[20] A sterile filter paper disk impregnated with phenylboronic acid (PBA, 400 µg) or cloxacillin (200 µg) was placed adjacent to a cefoxitin (30 µg) disk on Mueller–Hinton agar inoculated with the test organism.

An increase of ≥ 5 mm in the inhibition zone around the inhibitor-containing disk compared with cefoxitin alone indicated AmpC β-lactamase activity.

Cefoxitin resistance is a well-established phenotypic marker for screening AmpC-producing Enterobacterales, as AmpC enzymes hydrolyze cephamycins while ESBLs do not. The use of PBA or cloxacillin enhances detection by inhibiting AmpC enzymes, enabling differentiation from other cefoxitin-resistant mechanisms.

Isolates positive for both ESBL and AmpC phenotypic tests were classified as ESBL/AmpC coproducers.

Clinical Data Collection

Demographic and clinical data were obtained from medical records. Variables included age, sex, diabetic status, hospitalization status (inpatient/outpatient), recent antibiotic exposure (within 90 days), and clinical outcome.

Treatment success was defined as symptom resolution within 5 days without change of antibiotic regimen. Treatment failure was defined as persistence or worsening of symptoms requiring therapy modification or readmission. For hospitalized patients, length of stay (in days) was recorded.

Statistical Analysis

Data were analyzed using SPSS v27 (IBM, United States) and R v4.3.2. Continuous variables were expressed as medians (interquartile range [IQR]) and compared using the Mann–Whitney U test. Categorical variables were summarized as frequencies and compared by chi-square or Fisher's exact test.

Binary logistic regression identified independent predictors of ESBL + AmpC coproduction (age, gender, diabetes, prior antibiotics, hospitalization). Odds ratios (ORs) with 95% confidence intervals (CIs) were reported.

A multivariate model was constructed to assess the effect of coproduction on treatment failure, controlling for confounders. Model discrimination was evaluated by the receiver operating characteristic (ROC) curve using the pROC package in R, with area under the curve (AUC) reported. A p-value of < 0.05 was considered statistically significant.

Results

As illustrated in [Fig. 1], coproducing isolates were predominantly detected in PLWD, and their presence was associated with increased rates of treatment failure and prolonged hospitalization.

Fig. 1 Graphical summary illustrating coproduction of extended-spectrum β-lactamase (ESBL) and AmpC β-lactamases in Escherichia coli isolated from diabetic urinary tract infection patients, showing its association with increased resistance, treatment failure, and longer hospital stay.

Demographic and Clinical Characteristics

[Table 1] presents a total of 300 nonduplicate E. coli isolates, including 150 from PLWD and 150 from people without diabetes. The mean age was 53.4 ± 14.2 years (range: 18–86 years), with a female predominance (67.3%).

Table 1 Demographic and clinical characteristics of patients with E. coli UTI

Variable	PLWD (n = 150)	People without diabetes (n = 150)	p-Value
Mean age (±SD), y	56.2 ± 13.6	50.6 ± 14.8	0.008
Female sex, n (%)	104 (69.3)	98 (65.3)	0.47
Prior antibiotic use (< 3 months), n (%)	67 (44.7)	38 (25.3)	0.002
Inpatient cases, n (%)	61 (40.7)	43 (28.7)	0.04
Mean hospital stay (±SD), d	6.4 ± 2.8	4.3 ± 1.9	< 0.001
Treatment failure, n (%)	37 (24.7)	18 (12.0)	0.009

Abbreviations: PLWD, people living with diabetes; SD, standard deviation; UTI, urinary tract infection.

Fever and dysuria were the most common clinical symptoms in both groups. Prior antibiotic exposure within 3 months was significantly higher among PLWD (44.7%) than people without diabetes (25.3%, p = 0.002), and the mean hospital stay was longer (6.4 ± 2.8 vs. 4.3 ± 1.9 days, p < 0.001).

Prevalence of β-Lactamase Phenotypes

The distribution of β-lactamase phenotypes revealed a significantly higher prevalence of ESBL and AmpC coproduction among isolates from PLWD (22.0%) compared with nondiabetic patients (6.0%; p < 0.001), while nonproducers were significantly more common in the nondiabetic group (42.0% vs. 12.6% in PLWD; p < 0.001). However, the proportions of isolates producing ESBL only (48.7% vs. 40.0%; p = 0.14) or AmpC only (16.7% vs. 12.0%; p = 0.29) were not significantly different between the diabetic and nondiabetic groups, respectively ([Table 2]).

Table 2 Distribution of β-lactamase phenotypes among E. coli isolates

β-lactamase phenotype	PLWD (n = 150)	People without diabetes (n = 150)	Total (n = 300)	p-Value
ESBL only	73 (48.7%)	60 (40.0%)	133 (44.3%)	0.14
AmpC only	25 (16.7%)	18 (12.0%)	43 (14.3%)	0.29
ESBL + AmpC (coproducers)	33 (22.0%)	9 (6.0%)	42 (14.0%)	< 0.001
Nonproducers	19 (12.6%)	63 (42.0%)	82 (27.3%)	< 0.001

Abbreviations: ESBL, extended-spectrum β-lactamase; PLWD, people living with diabetes.

Antimicrobial Resistance Profiles

Antimicrobial susceptibility patterns of the E. coli isolates are presented in [Table 3]. Overall resistance was highest to third-generation cephalosporins and fluoroquinolones, whereas carbapenems and nitrofurantoin remained largely active.

Table 3 Antimicrobial resistance rates (%) among E. coli isolates according to β-lactamase phenotype

Antibiotic	ESBL only (n = 133)	AmpC only (n = 43)	ESBL + AmpC (n = 42)	Nonproducers (n = 82)	p-Value
Cefotaxime (CTX)	92.5	79.1	97.6	11.0	< 0.001
Ceftazidime (CAZ)	88.7	74.4	95.2	9.8	< 0.001
Cefepime (FEP)	84.2	65.1	92.9	6.1	< 0.001
Cefoxitin (FOX)	38.3	88.4	91.0	8.5	< 0.001
Piperacillin–tazobactam (TZP)	53.4	62.8	81.0	12.2	< 0.001
Ciprofloxacin (CIP)	77.4	69.8	88.1	18.3	< 0.001
Gentamicin (GEN)	62.4	58.1	79.0	20.7	< 0.001
Amikacin (AMK)	21.8	18.6	35.7	6.1	0.004
Nitrofurantoin (NIT)	12.0	16.3	9.5	5.0	0.18 nanoseconds
Ertapenem (ETP)	5.3	7.0	9.5	2.4	0.29 nanoseconds
Imipenem (IMP)	4.5	4.7	7.1	1.2	0.37 nanoseconds
Meropenem (MEM)	3.8	4.7	7.1	0	0.31 nanoseconds
Multidrug resistance (MDR)	76.7	68.4	92.8	19.5	< 0.001

Abbreviation: ESBL, extended-spectrum β-lactamase.

Coproducers exhibited the broadest resistance spectrum, with ≥ 85% resistance to cefotaxime, ceftazidime, cefepime, and ciprofloxacin, and 93% qualifying as MDR (p < 0.001).

Predictors of ESBL + AmpC Coproduction

Univariate analysis identified DM, prior antibiotic exposure, and inpatient status as significant predictors (p < 0.01).

In multivariable logistic regression, only diabetes (OR = 3.64, 95% CI 1.74–7.61, p = 0.001) and prior antibiotic use (OR = 2.18, 95% CI 1.03–4.62, p = 0.041) remained independent predictors as shown in [Table 4].

Table 4 Predictors of ESBL + AmpC coproduction among E. coli isolates

Variable	Unadjusted OR (95% CI)	p	Adjusted OR (95% CI)	p
Age > 60 years	1.34 (0.65–2.79)	0.42	1.21 (0.52–2.80)	0.64
Female sex	1.19 (0.58–2.43)	0.63	1.12 (0.50–2.51)	0.77
Diabetes mellitus	4.45 (2.12–9.33)	< 0.001	3.64 (1.74–7.61)	0.001
Prior antibiotic use	2.72 (1.39–5.34)	0.003	2.18 (1.03–4.62)	0.041
Inpatient status	1.98 (1.01–3.87)	0.046	1.53 (0.73–3.21)	0.26

Abbreviations: CI, confidence interval; ESBL, extended-spectrum β-lactamase; OR, odds ratio.

Impact on Treatment Outcomes

Coproducers were associated with a significantly higher treatment failure rate (35.7%) compared with noncoproducers (14.6%, p = 0.002). Median hospital stay was 6 days (IQR = 4–8) among patients infected with coproducer isolates versus 4 days (IQR = 3–6) in those with noncoproducers (p = 0.004, Mann–Whitney U). This difference is illustrated in [Fig. 2], which demonstrates a visibly longer hospitalization period in the coproducer group.

Fig. 2 Boxplot showing longer hospital stay among coproducer infections (median 6 vs. 4 days, p = 0.004).

The logistic regression model was designed to predict treatment failure, integrating microbiological and clinical variables (adjusted OR = 2.81, 95% CI 1.21–6.52, p = 0.016). The overall performance of the predictive model was robust, with an AUC = 0.82 (95% CI 0.75–0.88), confirming good discriminatory ability ([Fig. 3]).

Fig. 3 Receiver operating characteristic (ROC) curve of the multivariable model predicting treatment failure (area under the curve [AUC] = 0.82).

Discussion

This study provides new insights into the coexistence of ESBLs and AmpC β-lactamases among uropathogenic E. coli in PLWD and people without diabetes. The findings highlight both the clinical relevance and the public health implications of β-lactamase coproduction in a resource-limited setting such as Libya.

Demographic and Clinical Characteristics

The present study demonstrates that DM remains a major determinant of complicated UTIs caused by E. coli. The longer hospital stays and higher treatment failure rates among PLWD reflect a complex interplay between host metabolic dysregulation and bacterial virulence. Persistent hyperglycemia impairs neutrophil chemotaxis, oxidative burst, and phagocytic efficiency, while glucosuria provides a nutrient-rich environment that promotes bacterial adhesion and biofilm formation.[12] These pathophysiological mechanisms explain the clinical observation that PLWD often experience delayed recovery and frequent relapse.

Comparable evidence has been reported globally. In Ethiopia, researchers documented delayed clinical resolution and increased MDR among diabetic UTI patients,[9] while similar findings were observed in Tunisian hospitals.[13] A meta-analysis of 35 studies estimated that DM nearly doubles the odds of infection with resistant organisms.[14] Such findings underscore the need for early, targeted antimicrobial therapy in PLWD, particularly in North Africa, where metabolic disorders and unregulated antibiotic use coexist.

Prevalence of β-Lactamase Phenotypes

The observed coproduction rate of 14% places Libya within the regional cluster of high-burden settings. Comparable figures have been reported in neighboring countries,[15] [16] contrasting sharply with southern Europe (6–8%)[6] [16] and East Asia (5–6%).[2] The convergence across North African countries likely reflects similar antibiotic use patterns and widespread over-the-counter access to antibiotics. Hospitals in this region often act as amplification hubs for plasmid-mediated resistance, driven by empirical cephalosporin use and limited microbiological capacity.

Previous Libyan studies identified ESBL prevalence ranging from 30 to 45% but did not routinely screen for AmpC.[17] [18] Consequently, the true magnitude of dual β-lactamase production may have been underestimated. Our data therefore provide the first robust analytical evidence that coproduction is both measurable and clinically meaningful in Libyan UTI isolates.

Antimicrobial Resistance Profiles

The resistance configuration observed, dominant resistance to third-generation cephalosporins and fluoroquinolones, moderate resistance to β-lactam/β-lactamase inhibitor combinations, and preserved activity of carbapenems and nitrofurantoin, mirrors recent global antimicrobial resistance (AMR) surveillance trends.[19] Dual ESBL + AmpC production enables hydrolysis of nearly all extended-spectrum cephalosporins while simultaneously conferring inhibitor resistance, effectively neutralizing common empirical regimens such as ceftriaxone–clavulanate.[7] [20]

At the molecular level, coharboring of blaCTX-M and plasmid-encoded AmpC genes frequently occurs on IncF and IncI1 plasmids, often alongside qnrA/B/S determinants that mediate quinolone resistance.[21] [22] Such genetic linkage explains the simultaneous resistance to both β-lactams and fluoroquinolones observed in our coproducers. Recent sequencing data from Saudi Arabia revealed identical plasmid backbones among hospital E. coli strains,[23] suggesting regional dissemination through health care networks.

Clinically, the preserved sensitivity to carbapenems and nitrofurantoin (> 90%) echoes earlier Libyan findings[17] and indicates that these agents remain reliable last-line options. However, sporadic carbapenem resistance documented locally and in neighboring countries signals the potential emergence of blaNDM and OXA-48-like genes.[24] This trend warrants proactive molecular surveillance and strict enforcement of antibiotic stewardship.

Predictors of ESBL + AmpC Coproduction

Our logistic regression analysis identified DM and prior antibiotic exposure as independent risk factors for coproduction. These associations are consistent with multicentric evidence showing that chronic diseases and repeated antimicrobial exposure facilitate the selection and maintenance of resistant flora.[6] [23] Antibiotic pressure, particularly from third-generation cephalosporins and fluoroquinolones, enriches gut reservoirs of ESBL/AmpC-producing Enterobacterales.[4]

Hospitalization initially appeared significant in univariate analysis but lost significance after adjustment for antibiotic exposure, suggesting that the hospital environment amplifies risk primarily through antimicrobial selection pressure. This reinforces a key stewardship principle: exposure, rather than location, drives coproduction risk.

Impact on Treatment Outcomes

The association between coproduction, treatment failure, and prolonged hospitalization has important therapeutic and economic implications. Coproducers were linked to longer hospital stays and higher relapse rates, consistent with previous observations.[24] ESBL/AmpC coproducers were twice as likely to cause recurrent infection compared with single producers.[15]

The predictive model's AUC demonstrated strong discriminative performance, confirming the reliability of phenotypic data in forecasting poor outcomes. Such models could be integrated into infection-control dashboards to identify high-risk patients upon admission. Nevertheless, predictive analytics should complement rather than replace laboratory confirmation. Routine AmpC testing in diagnostic protocols remains essential, yet is still absent in most Libyan microbiology laboratories. Establishing this capacity would substantially enhance clinical decision-making and antimicrobial stewardship.

Phenotypic differentiation between ESBL, AmpC, and coproducers remains challenging. Concurrent AmpC expression may obscure ESBL activity, leading to false negatives in standard DDST.[11] Recent studies advocate using inhibitor-based assays with cefoxitin and boronic acid or cloxacillin disks to unmask coproducers.[7]

From a systems perspective, Libya's lack of integrated AMR data pipelines hinders national risk assessment. The World Health Organization Global Antimicrobial Resistance and Use Surveillance (WHO GLASS) framework provides a valuable template for harmonizing data on antimicrobial use (AMU) and AMR.[19] Implementing such a model locally would facilitate trend monitoring, benchmarking against regional averages, and adaptive updates to empirical therapy guidelines. Institutional antibiograms, stratified by ward and infection type, should be generated quarterly to maintain therapeutic precision.

Overall Interpretation

Collectively, these findings highlight that coproduction of ESBL and AmpC β-lactamases in uropathogenic E. coli is not uncommon and exerts measurable clinical consequences. Diabetic status and prior antibiotic exposure emerge as key risk factors. Routine inclusion of AmpC screening in diagnostic protocols and periodic updates to local antibiograms are essential to optimize empirical therapy, particularly in high-risk diabetic populations.

Coproduction represents a convergent evolutionary strategy whereby bacteria combine ESBL and AmpC enzymes to overcome both oxyimino-cephalosporins and β-lactamase inhibitors. Horizontal gene transfer via conjugative plasmids facilitates this adaptation.[25] In vitro evolution experiments demonstrate that the presence of blaCTX-M on mobile elements accelerates the acquisition of AmpC genes through recombination events.[22]

In PLWD, the selective advantage of such strains is magnified. Hyperglycemia impairs cytokine signaling (interleukin-8, tumor necrosis factor-α) and disrupts epithelial tight junctions, facilitating bacterial translocation.[12] Furthermore, subinhibitory antibiotic concentrations in chronically treated patients may induce SOS response (bacterial DNA damage - induced global stress response) responses that stimulate horizontal gene transfer. Thus, resistance evolution in diabetics is driven by both pharmacological and immunological factors.

Public Health and Stewardship Significance

The dual enzyme mechanism documented here poses a serious challenge to empirical therapy in resource-limited settings. Unchecked prescription of third-generation cephalosporins and self-medication are major drivers in Libya, where regulatory oversight remains limited. Empirical management of UTIs in PLWD should prioritize nitrofurantoin or carbapenems, reserving fluoroquinolones for documented susceptible isolates.

At the national level, establishing coordinated AMR surveillance within the Ministry of Health, integrated with regional WHO GLASS nodes, would enable real-time tracking of resistance phenotypes. Capacity-building initiatives should focus on equipping provincial laboratories with reliable AmpC detection and molecular screening kits. Public awareness campaigns addressing antibiotic misuse would complement these laboratory interventions.

From a regional standpoint, the North African AMR consortium could harmonize data sharing among Libya, Tunisia, Egypt, and Algeria, similar to the European Antimicrobial Resistance Surveillance Network model. Such cooperation would provide policymakers with early warnings of plasmid-mediated resistance trends and cross-border dissemination.

Strengths, Limitations, and Future Directions

The principal strength of this study lies in its comprehensive integration of clinical, microbiological, and analytical data, enabling direct correlation between phenotypic β-lactamase mechanisms and patient-level outcomes. To the best of our knowledge, this is the first study from Libya to quantify the prevalence of ESBL/AmpC coproduction among uropathogenic E. coli and to evaluate its clinical predictors using the CLSI-standardized phenotypic methods, thereby improving reproducibility and facilitating comparison with regional and international studies. The inclusion of both CA diabetic and nondiabetic UTI cases provides a clinically relevant overview of resistance patterns in a high-risk population.

Despite these strengths, several limitations should be acknowledged. First, the study was conducted in a single tertiary care center, which may limit generalizability to other regions of Libya. Second, molecular confirmation of ESBL and AmpC genes (e.g., polymerase chain reaction or whole-genome sequencing) was not performed due to resource constraints, and phenotypic methods alone may under- or overestimate coproduction in some isolates. Third, important clinical variables were not fully available in patient records, including glycemic control (HbA1c levels), diabetes severity, presence of comorbidities, prior antibiotic classes, and details of empirical or culture-guided treatment regimens. Follow-up cultures after treatment were also unavailable, preventing microbiological confirmation of clinical cure. Fourth, minimum inhibition concentration (MIC) values and resistant quality control strains (e.g., ATCC 35218) were not available during testing, which may influence phenotypic accuracy. Fifth, although we attempted to minimize bias through consecutive sampling, no matching strategy (e.g., propensity score matching) could be implemented to balance age or sex differences between diabetic and nondiabetic groups. Sixth, the lack of EUCAST (European Committee on Antimicrobial Susceptibility Testing) breakpoints and reliance solely on CLSI may limit comparability for some international readers. Seventh, verbal consent rather than written documentation was used due to institutional policy for minimal-risk research, and this is acknowledged as an ethical limitation.

Future studies should incorporate multicenter sampling to improve representativeness, complemented by molecular assays to determine the genetic determinants and transmission dynamics of ESBL and AmpC enzymes. The integration of AMU metrics with resistance profiles would help identify ecological and behavioral drivers of coproduction. Prospective cohort designs with serial sampling may clarify the temporal progression from ESBL to ESBL/AmpC coproduction, especially under antibiotic selective pressure. Additionally, including glycemic control parameters, treatment regimens, comorbidity indices, and follow-up culture data would strengthen clinical interpretations and support the development of optimized, evidence-based management strategies for UTIs particularly among PLWD.

Conclusion

In summary, this study demonstrates that coproduction of ESBL and AmpC β-lactamases among uropathogenic E. coli in Libya is frequent and clinically consequential.

The phenomenon is amplified in PLWD and those with prior antibiotic exposure, leading to higher MDR rates and worse therapeutic outcomes.

Addressing this challenge requires integrating AmpC testing into standard diagnostic routines, enforcing antibiotic stewardship, and establishing national AMR monitoring systems to safeguard the efficacy of last-resort agents.

Conflict of Interest

None declared.

Acknowledgments

The authors would like to thank the staff of Misurata Central Hospital for their cooperation during specimen collection and laboratory analysis.

• References

• 1 Bunduki GK, Heinz E, Phiri VS, Noah P, Feasey N, Musaya J. Virulence factors and antimicrobial resistance of uropathogenic Escherichia coli (UPEC) isolated from urinary tract infections: a systematic review and meta-analysis. BMC Infect Dis 2021; 21 (01) 753
• 2 Qin X, Li R, Zhang W, Liu J, Chen Z, Li J. Antimicrobial resistance of clinical bacterial isolates in China: a review from 2005 to 2022. JAC Antimicrob Resist 2024; 6 (02) dlae052
• 3 Choi JH, Kim YJ, Park YS, Seo KW, Jeon HY, Lim SK. Prevalence and characterization of extended-spectrum β-lactamase and AmpC β-lactamase producing Escherichia coli isolates from diseased dogs and cats in Korea. Antibiotics (Basel) 2023; 12 (04) 745
• 4 Salvia T, Dolma KG, Dhakal OP, Khandelwal B, Singh LS. Phenotypic detection of ESBL, AmpC, MBL, and their co-occurrence among MDR Enterobacteriaceae isolates. J Lab Physicians 2022; 14 (03) 329-335
• 5 Sun L, Meng N, Wang Z. et al. Genomic characterization of ESBL/AmpC-producing Escherichia coli in stray dogs sheltered in Yangzhou, China. Infect Drug Resist 2022; 15: 7741-7750
• 6 Das D, Vohra P, Mane P, Shaozae CK. Extended spectrum, AmpC & metallo β-lactamases producing Escherichia coli in urinary isolates: a prospective study in north India. Indian J Med Res 2025; 161 (02) 167-173
• 7 Nepal K, Pant ND, Neupane B. et al. Extended spectrum beta-lactamase and metallo beta-lactamase production among Escherichia coli and Klebsiella pneumoniae isolated from different clinical samples in a tertiary care hospital in Kathmandu, Nepal. Ann Clin Microbiol Antimicrob 2017; 16 (01) 62
• 8 Mama M, Manilal A, Gezmu T, Kidanewold A, Gosa F, Gebresilasie A. Prevalence and associated factors of urinary tract infections among diabetic patients in Arba Minch Hospital, Arba Minch province, South Ethiopia. Turk J Urol 2018; 45 (01) 56-62
• 9 Yenehun Worku G, Belete Alamneh Y, Erku Abegaz W. Prevalence of bacterial urinary tract infection and antimicrobial susceptibility patterns among diabetes mellitus patients attending Zewditu Memorial Hospital, Addis Ababa, Ethiopia. Infect Drug Resist 2021; 14: 1441-1454
• 10 Magiorakos A-P, Srinivasan A, Carey RB. et al. Multidrug-resistant, extensively drug-resistant and pandrug-resistant bacteria: an international expert proposal for interim standard definitions for acquired resistance. Clin Microbiol Infect 2012; 18 (03) 268-281
• 11 Clinical and Laboratory Standards Institute (CLSI). Performance Standards for Antimicrobial Susceptibility Testing: 34th Edition (M100). Wayne, PA: CLSI; 2024
• 12 Owusu E, Adjei H, Afutu E. Similarities in bacterial uropathogens and their antimicrobial susceptibility profile in diabetics and their non-diabetic caregivers at a national diabetes management and research centre, Accra-Ghana. Diseases 2022; 10 (04) 124
• 13 Ben Sallem R, Ben Slama K, Estepa V. et al. Prevalence and characterisation of extended-spectrum β-lactamase (ESBL)-producing Escherichia coli isolates in healthy volunteers in Tunisia. Eur J Clin Microbiol Infect Dis 2012; 31 (07) 1511-1516
• 14 Carrillo-Larco RM, Altez-Fernandez C, Acevedo-Ramos A. et al. Type 2 diabetes mellitus and antibiotic-resistant infections: a systematic review and meta-analysis. BMJ 2022; 376: e069780
• 15 Arumugam K, Karande GS, Patil SR. Prevalence of extended spectrum b-lactamase and AmpC b-lactamase among Escherichia coli and Klebsiella pneumoniae in urinary tract infections. J Pure Appl Microbiol 2025; 19 (03) 2237-2246
• 16 Hassen B, Abbassi MS, Benlabidi S. et al. Genetic characterization of ESBL-producing Escherichia coli and Klebsiella pneumoniae isolated from wastewater and river water in Tunisia: predominance of CTX-M-15 and high genetic diversity. Environ Sci Pollut Res Int 2020; 27 (35) 44368-44377
• 17 El-Faitouri A, El-Amin H, Mohamed M. Extended-spectrum β-lactamase production among E. coli isolates from urinary tract infections in Tripoli, Libya. J Infect Dev Ctries 2019; 13 (08) 742-748
• 18 Al-Kesh AH, Abdussalam S, El-Debani H. Antimicrobial resistance patterns of Escherichia coli causing urinary tract infections in Misurata, Libya. Libyan J Med Sci 2020; 4 (03) 115-122
• 19 World Health Organization (WHO). Global Antimicrobial Resistance and Use Surveillance (GLASS) Report 2025. Geneva: WHO; 2025
• 20 Coudron PE. Inhibitor-based methods for detection of plasmid-mediated AmpC β-lactamases in Klebsiella spp., Escherichia coli, and Proteus mirabilis. J Clin Microbiol 2005; 43 (08) 4163-4167
• 21 Mohammed A, Ahmed S, Ibrahim Y. High prevalence of blaCMY AmpC β-lactamase in ESBL co-producers among Escherichia coli and Klebsiella spp. clinical isolates. J Glob Antimicrob Resist 2020; 21: 205-211
• 22 Shahkolahi S, Shakibnia P, Shahbazi S. et al. Detection of ESBL and AmpC producing Klebsiella pneumoniae ST11 and ST147 from urinary tract infections in Iran. Acta Microbiol Immunol Hung 2022; 69 (04) 303-313
• 23 Alghamdi SAA, Mir SS, Alghamdi FS, Al Banghali MAMMA, Almalki SSR. Evaluation of extended-spectrum beta-lactamase resistance in uropathogenic Escherichia coli isolates from urinary tract infection patients in Al-Baha, Saudi Arabia. Microorganisms 2023; 11 (12) 2820
• 24 Abdallah H, Youssef M, El-Gendy A. Co-existence of extended-spectrum β-lactamases and AmpC β-lactamases among uropathogenic E. coli: emerging therapeutic challenge. Antibiotics (Basel) 2022; 11 (10) 1369
• 25 Aref M, Fathi J, Ghazanfari P, Malekzadegan Y, Jamali B. Characterization of plasmid-mediated quinolone resistant genes among uropathogenic Escherichia coli isolates in Bushehr, south of Iran. BMC Res Notes 2025; 18 (01) 404

Publication History

Received: 16 November 2025

Accepted: 19 November 2025

Article published online:
04 February 2026