Urine cultures phenotypically identified as Escherichia coli with colony counts ≥100,000 CFU/mL, processed between July and August 2022 at a tertiary-level hospital of the Honduran Social Security Institute (IHSS) in Tegucigalpa, were donated to this study. The isolates were provided along with anonymized sociodemographic data (sex, age, and hospitalization status). No additional clinical information, such as treatment history, underlying diseases, or urinary cathether use, was available. Only molecular characterization of the isolates was performed. No direct contact with patients occurred at any stage of the study.

Bacterial identification was performed using the BD Phoenix™ Identification System for Gram-negative bacilli, following the manufacturer’s instructions. Briefly, one to two colonies from a 24-h culture were suspended in Phoenix ID broth to a 0.5 McFarland turbidity standard. From this standardized suspension, 25 μL were transferred to Phoenix AST broth, which was supplemented with a drop of Phoenix AST indicator (containing methylene blue and resazurin) to detect bacterial growth. The mixture was then loaded into BD Phoenix™ NMIC-500 panels (catalog number BD 449023), which were sealed, registered, and processed using the BD Phoenix™ M50 system. Results were analyzed using Epicenter data management software (version 6.61A, BD Diagnostic Systems) after incubation for 6 to 16 h.

Antibiotics were grouped into 12 categories, according to the classification proposed by Magiorakos et al. (2012) (1) Aminoglycosides: Amikacin (AMK), Gentamicin (GM) (2) Sulfonamides: Trimethoprim/Sulfamethoxazole (TSX); (3) Nitrofurans: Nitrofurantoin (MAC); (4) Fluoroquinolones: Ciprofloxacin (CIP), Levofloxacin (LVX); (5) Penicillins with β-lactamase inhibitors: Amoxicillin/Clavulanic acid (AMC), Piperacillin/Tazobactam (TZP); (6) Penicillins: Ampicillin (AMP); (7) First-generation cephalosporins: Cefuroxime (CFX); (8) Second-generation cephalosporins: Cefazolin (CZ); (9) Third-generation cephalosporins: Ceftriaxone (CRO); (10) Fourth-generation cephalosporins: Cefepime (FEP); (11) Cephamycin: Cefoxitin (FOX); and (12) Carbapenems: Ertapenem (ETP), Imipenem (IMP), Meropenem (MEM). Based on the resistance profile of each isolate, strains were classified as follows: Non-multidrug-resistant (NMDR): Non-susceptible to two or fewer categories; Multidrug-resistant (MDR): Non-susceptible to three or more categories; extensively drug-resistant (XDR): Susceptible to only one of the tested categories; Pandrug-resistant (PDR): Non-susceptible to all tested categories (Magiorakos et al., 2012).

The prevalence of resistance for each antibiotic was graphically represented as a percentage. Additionally, a resistance index (RI) was calculated for each isolate using the formula previously reported by Karim et al. (2023),

Where “a” represents the number of antibiotics to which a given isolate was resistant, and “b” corresponds to the total number of antibiotics tested in this study. The resulting RI values were subsequently used for comparative analyses based on patient sex, age group, phylogroup classification, and other variables.

Phenotypically confirmed E. coli isolates were inoculated in Luria-Bertani (LB) broth (Liofilchem®) and incubated at 37 °C for 24 h. Genomic DNA was extracted using the Wizard® Genomic DNA Purification Kit (Promega, Madison, WI, USA), following the manufacturer’s protocol. DNA extracts were eluted in 100 μL of elution buffer and stored at −20 °C until further use.

Molecular identification of E. coli was performed by PCR amplification of the ybbW gene, which encodes an allantoin receptor reported as highly specific to this species (Walker et al., 2017). PCR reactions were conducted in a final volume of 25 μL containing 12.5 μL of 2 × PCR Master Mix (Promega Corp. Madison, WI, USA), 1 μL of each primer (10 μM), 8 μL of nuclease-free water, and 1.5 μL of DNA (40 ng/μL). Thermocycling conditions were as follows: initial denaturation at 95 °C for 4 min; 37 cycles of denaturation at 95 °C for 70 s, annealing at 60 °C for 70 s, and extension at 72 °C for 70 s; followed by a final extension step at 72 °C for 10 min. All PCR products were resolved on 2% agarose gels stained with ethidium bromide and visualized using the Gel Doc™ EZ Imaging System. Primer sequences used for amplification are listed in Supplementary Table 1.

Seventeen virulence genes associated with UPEC pathogenesis in the urinary tract were analyzed. These genes were functionally grouped into five categories: (1) Adhesion: fimH, papG-II, papC, sfaS/focC; (2) Motility: fliCD; (3) Toxin production: hlyA, cnf-1, vat, sat; (4) Immune evasion: agn43, traT, kpsM; and (5) Iron acquisition: feoB, fyuA, iucD, iha, iutA. Gene detection was performed via multiplex PCR following the protocol described by Ballesteros-Monrreal et al. (2021). PCR reactions were carried out in a final volume of 25 μL, containing 12.5 μL of 2 × PCR Master Mix (Promega Corp. Madison, WI, USA), 1 μL of each primer (10 μM), and 2 μL of genomic DNA. Positive controls included DNA from E. coli strains CFT073, ATCC 25922, and UPEC 39, as described in prior publications (Ballesteros-Monrreal et al., 2021). Primer sequences, amplicon sizes, and annealing temperatures (Tm, °C) are provided in Supplementary Table 1.

Phylogenetic group assignment was performed using Clermont et al. (2000, 2013, 2019) methods, based on PCR detection of the arpA, chuA, yjaA, trpA, ybgD, and cfaB genes, as well as the TspE4. C2 DNA fragment. PCR reactions were conducted following previously described protocols (Mendoza et al., 2023).

Clonality among all isolates was assessed through the analysis of consensus sequences of repetitive intergenic regions (ERIC), a technique widely validated and recommended for molecular typing of E. coli (Alsultan and Elhadi, 2022; López-Ramírez et al., 2018; Dishan et al., 2023; Ardakani and Ranjbar, 2016; Mahmoud et al., 2020). The methodology followed that described by Movahedi et al. (2021). PCR reactions were carried out in a final volume of 25 μL containing 12.5 μL of 2 × PCR Master Mix (Promega Corp., Madison, WI, USA), 8.5 μL of nuclease-free water, 2 μL of genomic DNA, and 1 μL of each ERIC primer. Thermocycling conditions were as follows: initial denaturation at 94 °C for 5 min; 40 cycles of denaturation at 94 °C for 1 min, annealing at 53 °C for 40 s, and extension at 72 °C for 1 min; followed by a final extension at 72 °C for 5 min. Band sizes, in relation to the molecular weight marker, as well as the similarity matrix for dendrogram construction, were generated using the GelJ v2.0 software.¹ Cluster analysis was performed using the unweighted pair group method with arithmetic mean (UPGMA) and the DICE similarity coefficient. To avoid overestimation of clonality, isolates were classified as “suspected clones” only when they shared ≥90% similarity in ERIC-PCR patterns and exhibited identical or highly similar resistance, virulence, and phylogroup profiles. This integrative approach ensured a more stringent and reliable interpretation of genetic relatedness. Clonal distribution was further analyzed in relation to hospital wards and patient origin (hospitalized vs. non-hospitalized).

A comprehensive statistical analysis approach was employed to ensure the reliability and interpretability of results. Descriptive statistics were initially calculated, including means, medians, and interquartile ranges, according to the distribution of the data. Assumptions of normality and homogeneity of variances were evaluated using the Shapiro–Wilk and Bartlett tests, respectively. Due to violations of these assumptions in some groups, the non-parametric Kruskal–Wallis test was used for overall group comparisons. When significant differences were observed, post hoc analyses were conducted using Dunn’s test with Bonferroni correction to determine which specific groups differed. For categorical variables such as sex, phylogenetic group, or gene presence/absence, Fisher’s exact test was used. Additionally, odds ratios (ORs) and relative risks (RRs) with 95% confidence intervals were calculated to assess the strength of associations between categorical variables. To explore potential associations between virulence factors and antimicrobial resistance, Pearson correlation analysis was performed. A p < 0.05 was considered statistically significant in all analyses.

All statistical analyses were performed using GraphPad Prism version 7.0 (GraphPad Software, San Diego, CA, USA)² and JASP v0.19.3.³ Graphical representations were generated using GraphPad Prism and RStudio with R version 4.5.0 (2025-04-11). Data visualization in R was conducted using the ggplot2, pheatmap, and magick packages (R Core Team, 2025). ERIC-PCR dendrograms were visualized and edited using the interactive iTOL environment, version 7.2 (Letunic and Bork, 2021).

Between July and August 2022, a total of 126 clinical isolates were identified as Escherichia coli through both biochemical and molecular methods. Of these, 99 isolates (78.6%) were obtained from female patients, whose ages ranged from 0 to 88 years, while the remaining 27 isolates (21.4%) were from male patients aged between 0 and 85 years.

Analysis of age group distribution revealed that the 13–50 years category was the most prevalent among both sexes, accounting for 41% of female and 59% of male patients. This was followed by patients older than 50 years, comprising 38% of females and 22% of males. However, no statistically significant association was found between age groups and patient sex (p > 0.05).

To evaluate the potential clonal relationship among clinical uropathogenic E. coli isolates, a clustering analysis was performed based on profiles obtained by ERIC-PCR. This analysis was organized at three levels: by hospital ward, among non-hospitalized patients, and among hospitalized patients. At the hospital ward level, possible clonal relationships were identified among isolates from eight clinical departments. The wards with the highest number of clusters were Emergency (5 clusters; 13 strains), Internal Medicine (7 clusters; 14 strains), Gynecology (2 clusters; 6 strains), General Medicine (2 clusters; 4 strains), Pathological Puerperium (1 cluster; 2 strains), Endocrinology (1 cluster; 2 strains), Pediatrics (1 cluster; 2 strains), Urology (2 clusters; 4 strains), and Outpatients (1 cluster; 2 strains) (Supplementary Figures 5A–P). Among hospitalized patients, 58 isolates were grouped into 20 clusters (Supplementary Figures O,P), while in non-hospitalized patients, 39 isolates were grouped into 16 clusters (Supplementary Figure 5P).

Despite the high similarity in banding patterns obtained through ERIC-PCR, we observed variations in resistance and virulence profiles, and even in phylogenetic grouping. In this context, it is important to consider that ERIC sequences do not differentiate specific virulence or antibiotic resistance genes, which can be acquired through horizontal gene transfer events. Similarly, phylogenetic groups are determined by specific genes that cannot be distinguished via ERIC-PCR. However, we identified some isolates with features suggestive of possible clonality: they showed identical ERIC profiles according to our cut-off criteria, along with high similarity in both virulence and antibiotic resistance profiles. These isolates were detected by hospital ward (Figure 8A), in the group of hospitalized patients (Figure 8B), and the outpatient group (Figure 8C). Figure 8 visually summarizes some of the clusters with the highest similarity, integrating genomic, phenotypic, and epidemiological profiles.

On the other hand, the analysis of virulence genes revealed significant differences between clinical isolates from male and female patients. Overall, strains from male patients exhibited a significantly higher virulence gene burden, with an average of 10.48 genes per strain, compared to 8.06 genes in female patients (p = 0.0029). This difference suggests that infectious episodes in male patients may be associated with strains possessing a greater pathogenic potential, which aligns with previous studies that have described a higher requirement for adaptability to colonize the male urinary tract, particularly in the presence of obstructive factors or urinary catheterization (Ruiz et al., 2002; AP et al., 2017).

Regarding age group distribution, results in females showed a higher mean number of genes in patients under 5 years old; however, this difference did not reach statistical significance. In males, the 6-to-12-year-old group stood out with a significantly higher genetic load than that observed in adolescents and adults (p < 0.05). This finding suggests that, in specific pediatric subgroups, pathogenesis may involve a higher number of virulence factors, which could translate into more aggressive clinical presentations or increased risk of recurrence. The elevated virulence gene load in isolates from pediatric patients has been previously reported by other authors (Hernández-Chiñas et al., 2019; Luna-Pineda et al., 2018), and although the sample size in our study limits definitive conclusions, these data suggest that the pediatric population, particularly school-aged boys, represents a risk group that warrants closer clinical and microbiological monitoring. Implementing follow-up strategies in this cohort could be key to detecting emerging virulence patterns, preventing complications, and timely adjusting therapeutic decisions.

Among the most prevalent genes, fimH (96%) and feoB (100%) were identified, both essential for initial adherence to the uroepithelium and iron acquisition, respectively, as well as for urinary tract colonization, particularly in nutrient-limited environments (Alam Parvez and Rahman, 2018). The high frequency of these genes is consistent with their central role in the pathophysiology of UPEC and has been widely reported (Ballesteros-Monrreal et al., 2021; Zhao et al., 2009; Yun et al., 2014; Tabasi et al., 2015; Sarshar et al., 2020). However, when stratified by sex, significant differences were observed in the prevalence of genes with more specialized functions. In males, there was a higher presence of genes related to renal-level adherence (papG-II, papC), toxins (hlyA, cnf-1), capsule formation (kpsM), and iron acquisition systems (fyuA, iucD, iha), all of which are associated with increased persistence and tissue damage in complicated infections (Ballesteros-Monrreal et al., 2021).

These results are consistent with previous reports: Kudinha et al. (2013) described a virulence gradient with higher prevalence of virulence factors in pyelonephritis compared to cystitis and healthy controls; Firoozeh et al. (2014) showed that pyelonephritis isolates harbor a greater diversity of virulence genes, particularly pap, traT, aer, hly, and PAIs; Hyun et al. (2022) identified sex-based differences with male isolates enriched in toxin genes such as hlyA and cnf1; and Hsiao et al. (2024) demonstrated that phylogroup B2 predominates in UTI and bacteremic cases, with genes such as papGII enriched in invasive infections (Hyun et al., 2022; Hsiao et al., 2024; Firoozeh et al., 2014; Kudinha et al., 2013). Collectively, these findings support our observation that isolates from male patients harbor more virulence determinants and are also associated with complicated clinical outcomes.

Regarding the virulence profiles of the isolates, high variability was observed, with a total of 92 distinct profiles: 14 were shared among multiple isolates from female patients, and only 2 among male isolates. Notably, isolates with shared profiles were not concentrated in a specific hospital ward or age group, suggesting a random and unfocused distribution of genetically similar strains within the hospital environment, with no clear evidence of clonal clustering within this dataset. Additionally, the low frequency of shared profiles is consistent with the high heterogeneity observed in the ERIC-PCR patterns, suggesting that most cases correspond to independent infectious events and a high diversity in infection sources. Nevertheless, ERIC-PCR results revealed two isolates that showed high similarity in their virulence characteristics, resistance, phylogenetic group, and hospital origin (Figure 8C). This finding underscores the importance of implementing continuous molecular surveillance to identify potential outbreaks, common transmission routes, or the emergence of successful clones with predominant virulence profiles.

Likewise, the high heterogeneity observed in virulence profiles may be related to the acquisition and maintenance of genes through horizontal transfer mechanisms, such as plasmids and pathogenicity islands (PAIs), reflecting the remarkable adaptability of uropathogenic isolates to diverse ecological niches and selective pressures. In our analysis, the significant co-occurrence of genes such as papG-II, papC, sfaD/focC, vat, fyuA, kpsM, and iucD is consistent with the findings of Ballesteros-Monrreal et al. (2021), who identified similar clusters in uropathogenic E. coli strains. More recently, Ballesteros-Monrreal et al. (2024) corroborated the presence of these PAIs through molecular characterization, suggesting that the observed gene clustering in our isolates likely reflects the functional and structural organization of these genetic elements rather than random distribution. This finding supports the hypothesis that the simultaneous presence of multiple virulence factors may be mediated by shared genomic structures. Moreover, considering that genes such as iutA and iucD have previously been associated with ColV-type plasmids (Johnson et al., 2006; Johnson and Nolan, 2009), it is reasonable to assume that some of our isolates may carry plasmids with horizontal dissemination capacity. This aspect is particularly relevant in hospital settings, where the circulation of strains harboring PAIs and plasmids contributes not only to pathogenic potential but also to the spread of antimicrobial resistance. Future studies should aim to confirm the presence and characterize the structure of such plasmids to better understand their role in the dissemination of resistance and virulence determinants. Ideally through whole-genome sequencing (WGS) to resolve their genetic context and evolutionary dynamics.

Regarding the characterization of the isolates based on antibiotic resistance, the analysis of the resistance index (RI) revealed significant differences between clinical isolates according to the patient’s sex. Isolates from male patients exhibited a significantly higher antimicrobial resistance burden, with an average RI of 0.46, compared to 0.27 in female-derived isolates (p < 0.0001). Additionally, 69% of all isolates were classified as MDR (including XDR), with the highest proportion of MDR found in male patients (88%; p = 0.009). This difference reinforces the previously observed trend in virulence gene burden, suggesting that strains isolated from male patients not only possess greater pathogenic potential but also exhibit increased antimicrobial resistance. However, one of the limitations of our study is the smaller male sample size (n = 27), which restricts the possibility of drawing definitive conclusions in this subgroup, although the observed trends are consistent with international evidence (Ruiz et al., 2002; AP et al., 2017; Khanal et al., 2024).

These results are consistent with findings by Khanal et al. (2024), who analyzed over 85,000 uropathogenic E. coli isolates in Australia and found a high prevalence of MDR strains in both hospital and community settings, with a significantly higher proportion in males compared to females (16.5% vs. 12.8% in hospitals; 6.4% vs. 5.2% in the community; p < 0.001). Although the Australian study also lacked individual clinical metadata, the authors attributed these differences to the fact that urinary tract infections in males are more frequently associated with conditions that complicate the clinical course, such as benign prostatic hyperplasia, urolithiasis, use of urinary catheters, history of urological instrumentation, or prior surgical interventions (Khanal et al., 2024). These conditions lead to urinary flow obstruction, increased risk of persistent colonization, and the need for prolonged or broad-spectrum antibiotic treatments, favoring the selection of resistant strains. Similar to our findings, Khanal et al. (2024) noted that, despite UTIs being more frequent in females, the strains isolated from males showed higher multidrug resistance rates (Khanal et al., 2024). This disparity suggests that UTIs in males may be caused by strains with more complex virulence and resistance profiles. This finding underscores the need to adapt empirical treatments not only to the type of infection but also to the epidemiological features associated with patient sex.

Among the 17 antibiotics tested, the highest prevalence of resistance in the 126 strains was observed for AMP (80%), TSX (67%), fluoroquinolones (CIP and LVX), and several second- to fourth-generation cephalosporins. When stratified by sex, isolates from male patients showed higher resistance rates for all evaluated antibiotics, with statistically significant differences observed for aminoglycosides, simple and combination penicillins, cephalosporins, carbapenems, nitrofurans, and fluoroquinolones. Similarly, the ESBL phenotype was more prevalent in this group (59%; p = 0.049). These results are particularly relevant when contrasted with currently recommended empirical treatment regimens in Honduras (Honduras, 2019; Ucros, 2009; Giovanni Erazo Trimarchi et al., 2015; Honduras, 2016). For outpatient pediatric patients, first-line treatment includes TSX and AMC, whereas in hospital settings, AMP, GM, AMK, CRO, or ampicillin-sulbactam (AMS) are commonly used (Johnson et al., 2006; Johnson and Nolan, 2009; Ucros, 2009). However, our findings reveal a high prevalence of resistance to TSX (60–100%), AMP (83–85%), and AMC (27–29%) in pediatric isolates of both sexes (with resistance reaching 66.7% in males and 40% in females), as well as to GM (33.3%) in males and CRO (33.3%) in females (Supplementary material 2B), suggesting potentially limited efficacy of these therapeutic options in the pediatric population. This situation is particularly concerning in children under 5 years of age, not only due to the frequent use of these antibiotics as first-line treatments but also due to the detection of a high burden of virulence genes. The coexistence of these factors may compromise therapeutic success and worsen clinical prognosis. Additionally, among pediatric female patients, ESBL- and CAR-producing isolates were observed, whereas none were detected in male children (Supplementary material 2B). This difference may be explained by the higher frequency of UTIs in females, which increases exposure to antibiotic treatments and consequently the selective pressure favoring the emergence of MDR strains.

In adult outpatients, commonly used antibiotics include AMC, TSX, CIP, LVX, and CRO; in hospitalized patients, fluoroquinolones, CRO, AMS, TZP, and carbapenems such as ertapenem (ETP) are used (Giovanni Erazo Trimarchi et al., 2015; Loscalzo et al., 2022). In our study, high resistance prevalence was observed among adult patients against these antimicrobials, particularly TSX (60–83.3%), CIP (39–93.75%), and CRO (31–83%) (Supplementary material 2B). This resistance pattern is not unique to our setting: studies conducted in the USA and Australia have reported elevated resistance rates to these drugs, especially TSX, more commonly observed in patients with recurrent UTIs, anatomical abnormalities of the urinary tract, or recent antibiotic treatment history, which has raised concerns regarding its use as a first-line empirical therapy (Khanal et al., 2024; Wesolek et al., 2022). These findings also align with a systematic review and meta-analysis conducted in Mexico, which reported high resistance to ampicillin (79%), third-generation cephalosporins (40–50%), quinolones and fluoroquinolones (45–61%), as well as TSX (61.9%), along with a considerable prevalence of ESBL-producing isolates (39.46%) (Ballesteros-Monrreal et al., 2023). This body of evidence underscores the urgent need to reassess empirical treatment guidelines for UTIs, considering both selective pressure and local resistance contexts, which may vary significantly depending on geographic region.

The high prevalence of resistance to these antibiotics may be linked to well-established genetic mechanisms. For TSX, resistance has been associated with dfrA and sul genes, commonly found in mobile integrons (Poey et al., 2024). In the case of CIP, resistance can result from chromosomal mutations in gyrA and parC, as well as plasmid-mediated mechanisms involving efflux pumps (qepA), target protection (qnrB, qnrA, qnrS), or enzymatic modification [aac(6′)-Ib-cr] (Malekzadegan et al., 2019). Regarding CRO, one of the most frequent mechanisms is the production of extended-spectrum β-lactamases (ESBLs), which confer resistance to multiple cephalosporins. Notably, many of these determinants can coexist within conjugative plasmids that harbor resistance genes for various antimicrobial classes (Salah et al., 2019; Juraschek et al., 2022), facilitating co-selection phenomena.

The co-occurrence of antimicrobial resistance phenotypes in our isolates is unlikely to be random, but rather reflects their organization within mobile genetic elements such as integrons and conjugative plasmids, which frequently harbor multiple resistance determinants within the same replicon. This genetic architecture facilitates horizontal gene transfer and promotes co-selection, whereby exposure to a single antibiotic may indirectly maintain resistance to several unrelated drugs. Similar patterns have been described in Enterobacteriaceae worldwide, underscoring the role of these mobile elements in the persistence and dissemination of multidrug resistance. Future studies using whole-genome sequencing (WGS) will be essential to characterize the specific plasmids and integrons circulating in Honduras and to resolve the genetic context underlying these associations.

Although the presence of these genes was not directly assessed, a strong correlation was observed between resistance to ceftriaxone and ciprofloxacin (r = 0.433; p < 0.001), suggesting possible co-transmission of these phenotypes. Additionally, significant associations were found between the ESBL phenotype and resistance to both cephalosporins and fluoroquinolones (r = 0.355; p < 0.001), potentially indicating the presence of mobile genetic elements that integrate and disseminate gene clusters associated with multidrug resistance in the studied population, posing a considerable therapeutic threat, particularly in settings where these antimicrobials are frequently used empirically.

This pattern of co-dissemination of resistance mechanisms has been previously documented in recent studies conducted in Mexico. In clinical isolates from women with UTIs, Ballesteros-Monrreal et al. (2024) reported a high prevalence of bla_CTX-M-2 and aac(6′)-Ib genes in isolates with simultaneous resistance phenotypes to β-lactams, fluoroquinolones, aminoglycosides, and ESBL production. Similarly, Marquez-Salazar et al. (2025) found that more than 95% of community-acquired E. coli isolates resistant to both β-lactams and fluoroquinolones carried ESBL and plasmid-mediated quinolone resistance (PMQR) genes—mainly aac(6′)-Ib-cr, qnrB, and qepA—pointing to the presence of conjugative plasmids carrying multiple resistance genes. The coexistence of these phenotypes may offer clues regarding potential resistance genotypes.

A limitation of our study is that we did not specifically detect these genetic determinants, highlighting the need for future investigations aimed at identifying genes associated with mobile genetic elements, as well as the circulating plasmids and integrons in E. coli strains from Honduras. Such research is essential to anticipate therapeutic efficacy crises and guide rational antibiotic use policies. Collectively, the findings of this study reveal a concerning landscape of antimicrobial resistance in uropathogenic E. coli strains in the Honduran context, necessitating a reevaluation of current therapeutic protocols and the strengthening of local surveillance. Integrating phenotypic resistance data with epidemiological and molecular information should be a priority for clinical decision-making and the formulation of antibiotic stewardship policies. In this regard, the use of personalized antibiograms, continuous training of medical personnel, and the incorporation of local data into clinical guidelines are key pillars for containing the spread of multidrug-resistant strains and preserving the effectiveness of available antimicrobials.

The phylogenetic characterization of the 126 clinical E. coli isolates revealed a heterogeneous distribution, with a predominance of phylogroup B2 (29%), followed by A and D (21% each)—lineages commonly associated with extraintestinal infections. We also identified phylogroups B1, Clade I, F, G, and C, along with 5% of non-typeable strains. Notably, phylogroup G, recently incorporated into the scheme proposed by Clermont et al. (2019), was detected. Although rare, its clinical relevance cannot be ruled out due to its shared characteristics with B2. The application of this updated scheme may explain the lower proportion of non-typeable strains compared to previous studies. Nevertheless, the persistence of unclassified isolates suggests the possible presence of phylogenetic lineages not yet described in Honduras.

Regarding sex-specific distribution, greater phylogenetic diversity was observed among female patients, with predominance of groups A and D, whereas phylogroup B2 was significantly more prevalent in male patients (59%), potentially associated with a higher burden of virulence and resistance genes, as well as anatomical and clinical factors specific to males. Although no significant differences were found among hospital wards, we identified a trend toward the concentration of B2 strains in departments such as Internal Medicine and Urology, possibly reflecting their association with more complex clinical presentations. Virulence index analysis showed significantly higher values in phylogroup B2 compared to A and B1, with associations to key genes related to adhesion, cytotoxicity, immune evasion, and iron acquisition, such as papG-II, papC, sat, hlyA, vat, cnf-1, kpsM, fyuA, iucD, and iha. Although phylogroup F exhibited a high median index, the low frequency of isolates precludes definitive conclusions but suggests an underestimated pathogenic potential that warrants further investigation.

Regarding antimicrobial resistance, no significant differences were observed in the overall index across phylogroups, although sex-stratified patterns emerged: in males, phylogroup A showed higher resistance to carbapenems; in females, B2 exhibited elevated resistance to gentamicin, D to ampicillin, and non-typeable strains to fluoroquinolones. These findings point to the existence of phylogroup-specific profiles possibly modulated by distinct selective pressures. Our findings align with those of Ballesteros-Monrreal et al. (2021), who also reported high phylogenetic diversity in clinical isolates from women, with 25% of strains being non-typeable. In that study, phylogroup B2 harbored the highest virulence gene load (iroN 91%, papC 64%, fyuA 100%), whereas B1 harbored the greatest antimicrobial resistance. Similarly, Hemati et al. (2024) identified B2 as the most prevalent group (32%), followed by E, C, and B1, and reported that over half of the isolates could not be phylogenetically assigned. Kara et al. (2024) also highlighted B2 as the dominant phylogroup (45.5%), followed by E (22.4%) and 12% non-typeable strains. Complementarily, Wang et al. (2023) characterized 844 clinical isolates and confirmed the predominance of phylogroup B2 (64.8%), with the highest average number of virulence genes (10.1 per isolate). Moreover, although phylogroups C, E, and F were less frequent, they displayed high multidrug resistance rates, supporting the importance of monitoring minority lineages that may be acquiring clinically relevant traits.

Collectively, our data identify phylogroup B2 as the predominant uropathogenic lineage in this cohort, supported by its association with multiple virulence genes and prominent role in resistance profiles. These findings highlight the need to strengthen molecular surveillance of uropathogenic E. coli in response to increasing genetic diversity and the potential emergence of clinically significant new lineages. This study presents one of the first comprehensive molecular and phenotypic characterizations of UPEC in Honduras. Our findings demonstrate a high prevalence of virulence factors, MDR, and ESBL production among isolates from a tertiary-care hospital, with a particular impact on pediatric and male patient populations. The observed resistance to commonly prescribed antibiotics, including ampicillin, trimethoprim-sulfamethoxazole, fluoroquinolones, and cephalosporins, raises significant concerns regarding the continued effectiveness of empiric treatment protocols recommended in national guidelines.

Phylogenetic analysis revealed substantial genetic diversity, with phylogroup B2 emerging as the predominant lineage, associated with increased virulence and resistance. The ERIC-PCR profiles, in combination with overlapping resistance and virulence patterns, suggest potential clonal relationships and highlight the urgent need for enhanced molecular surveillance strategies. In a context where data on urinary tract infections remain limited, our results provide critical evidence to guide clinical decision-making, support antimicrobial stewardship initiatives, and reinforce microbiological surveillance at the local level.

Another limitation of our study is the short sampling period (2 months), which restricts epidemiological generalizability. Future investigations should extend the collection period to capture seasonal variation and temporal dynamics in UPEC populations. Recognizing this limitation underscores the importance of complementing cross-sectional molecular studies with longitudinal surveillance to fully appreciate the epidemiological behavior of uropathogenic E. coli in our setting.

Finally, one of the future goals of our research is the complete sequencing of the genomes of isolates exhibiting co-occurrence of virulence and resistance genes. Whole-genome sequencing (WGS) represents a more robust approach compared to PCR-based detection, as it allows for the comprehensive reconstruction of the genomic context in which these determinants are embedded. Through WGS, it will be possible to confirm the presence of mobilizable genetic elements such as conjugative plasmids, transposons, and pathogenicity islands, and to elucidate their structural organization and evolutionary dynamics. This level of resolution is crucial to assess their potential for horizontal transfer within the hospital setting, where the circulation of highly adaptable clones represents a major epidemiological threat. In addition, characterizing these isolates in such depth provides a unique opportunity to move beyond descriptive surveillance and into translational applications. WGS-based analysis will enable the identification of the most relevant virulence factors associated with urinary tract infections in Honduras, including those that directly contribute to host colonization, immune evasion, or tissue damage. Understanding these genetic traits in a population-specific context may guide the design of novel preventive or therapeutic strategies, either by prioritizing the development of vaccines, antivirulence drugs, or adjuvant therapies that target key determinants of pathogenicity as has been observed in previously published studies focused on other relevant models (Tiwari et al., 2014; Khan et al., 2023; Sohail et al., 2025; Bunduki et al., 2021). Consequently, the genomic dissection of local UPEC populations does not only represent a step toward a more precise understanding of their epidemiology but also opens avenues for the identification of clinically actionable targets that could improve the management of UTIs in vulnerable groups.