Antimicrobial resistance (AMR) is a major global health challenge with significant consequences for public health and healthcare systems. In 2024, the Global Research on Antimicrobial Resistance (GRAM) project estimated that by 2050, there will be 1.91 million annual deaths globally attributable to AMR, with 6.63 million deaths attributable to AMR in sub-Saharan Africa. Without effective prevention measures, AMR will lead to a cumulative total of 39·1 million deaths attributable to AMR from 2025 to 2050 (Global Burden of Disease 2021 Antimicrobial Resistance Collaborators, 2024).

Reports indicate that the widespread fecal carriage of Extended-spectrum beta-lactamase producing Enterobacterales (ESBL-PE) is an emerging public health concern, contributing to infections in both community and hospital settings. Even in the absence of active infection, colonization with ESBL-PE remains a significant cause for concern (Godonou et al., 2022).

The high prevalence of fecal carriage of resistant pathogens facilitates the global transmission of resistance genes through travel, trade, and migration (Bokhary et al., 2021; Zamudio et al., 2024). In resource-limited settings with inadequate sanitation and hygiene infrastructure, the spread of resistance genes is exacerbated through contaminated food, water, and direct contact. This one health dynamic extends to animal populations via zoonotic spillover, cycling resistance genes back to humans through the environment or food chain (Aslam et al., 2021; Gebreyes et al., 2014). These mechanisms not only challenge effective treatment options but also contribute to increased morbidity and mortality due to limited therapeutic options (Iskandar et al., 2021).

The fecal carriage of ESBL-PE varies geographically, with prevalence ranging from 6 to 20% in Europe and North America, approximately 20% in the Eastern Mediterranean and Africa, and 24.5–27% in the Western Pacific and Southeast Asia in community and hospital settings (Bezabih et al., 2021; Pilmis et al., 2018; Islam et al., 2017).

In Africa, studies conducted in Chad and Rwanda have demonstrated a higher prevalence of ESBL-PE carriage in hospital settings than in the community (Ouchar Mahamat et al., 2019; Kurz et al., 2017). This pattern may be attributed to antibiotic use, antibiotic-induced gut dysbiosis, and reduced colonization resistance, which might promote ESBL-PE transmission through various routes, including person-to-person, contaminated food and water, and environmental exposure (Mai and Espinoza, 2023; Djuikoue et al., 2016).

In Eastern Africa specifically, ESBL-PE prevalence ranges from 6 to 17% in community settings and 38–83% in hospital environments (Storberg, 2014). A study conducted in East African hospitals has documented an overall pooled proportion of ESBL-PE at 42% with variation among countries: 61.7% in Uganda, 45.8% in Kenya, 38.8% in Tanzania and 30.9% in Ethiopia (Sonda et al., 2016).

Despite the growing concern about ESBL-PE in Tanzania, there is a notable lack of comprehensive studies on carriage of ESBL-PE across diverse age groups and clinical presentations (Silago et al., 2022; G et al., 2017; Ngowi et al., 2021; Mlugu et al., 2023). The available studies primarily focus on specific subpopulations or environments, leaving a fragmented understanding of the broader distribution of ESBL-PE. For instance, a study conducted among pregnant women reported a high prevalence of ESBL carriage rate of 64.3% (Mwandigha et al., 2020). Another study in urban informal settlements found that over 24% of ESBL-producing E. coli were isolated from private and shared latrines, indicating these facilities act as reservoirs for ESBL-PE transmission (Erb et al., 2018). Similarly, studies targeting hotel employees (Büdel et al., 2019) and healthcare settings (Manyahi et al., 2020) have provided valuable insights but are limited in scope. Most of these studies are urban-centric focus leaving rural communities underrepresented resulting in an incomplete picture of the epidemiology and dissemination of ESBL-PE in Tanzania. To address this knowledge gap, our study aims to evaluate the fecal carriage rate of and risk factors for multidrug-resistant Enterobacterales in patients presenting with symptoms of urinary tract infection at St. Francis Regional Referral Hospital (SFRRH) in Ifakara, a rural area in Tanzania. This research will contribute to the understanding of ESBL-PE epidemiology in the region and inform strategies for antimicrobial stewardship and infection control.

We enrolled 326 eligible patients meeting IDSA criteria for UTI. Out of these, 166 (50.9%) were outpatients, 189 (58.0%) were female, and the median age was 35.5 years [IQR: 25–52]. The collective prevalence of ESBL-PE in stool was 70.9% (231/326 patients). A total of 860 phenotypically distinct bacterial isolates were obtained from the 326 stool samples. Among these, 347 (40.3%) were ESBL-producing Enterobacterales. E. coli accounted for the majority of all ESBL-producing Enterobacterales 219 (62.8%) followed by K. pneumoniae 115 (33.0%), details in Table 1.

Additionally, bacterial growth in urine was 3.7% (12 out of 326 patients). The bacterial isolates detected in urine included: four (Zamudio et al., 2024) E. coli, four (Zamudio et al., 2024) P. aeruginosa, three (Bokhary et al., 2021) K. pneumoniae and one (1) P. mirabilis. ESBL was only detected in one E. coli urinary isolate but was not identical to the concomitant ESBL E. coli detected in stool.

This study found that more than two-third (70.9%) of patients presenting with UTI were ESBL-PE carriers. This high prevalence of carriage in rural Tanzania serves as an important indicator of antibiotic resistance in the area. Prevalence rates for ESBL-PE colonization vary substantially among different populations in Tanzania (Manyahi et al., 2020). Geographical distribution data revealed potential hotspots of ESBL-PE carriage rates across villages, with some showing higher prevalence. These findings highlight areas with potentially elevated transmission risk, which could guide future studies and inform targeted intervention strategies.

Previous studies have reported lower carriage rates in the urban area of Dar es Salaam, with rates of 32.6% in HIV patients and 34.3% in children (Manyahi et al., 2020; Tellevik et al., 2016). Additionally, a rate of 59.7% has been observed in adult patients (Kibwana et al., 2020). In contrast, other research has found much higher (91.5%) colonization rates among hotel employees on the island of Zanzibar (Büdel et al., 2019). Furthermore, most studies have linked risk factors such as sharing of toilets (Erb et al., 2018), poor working environments and hygienic practices (Mwanginde et al., 2021), use of third-generation cephalosporins (Manyahi et al., 2020), history of antibiotic use (Moremi et al., 2018), old age (Kibwana et al., 2020) with increased ESBL fecal carriage.

These findings highlight the diverse distribution of ESBL-PE colonization within Tanzania, suggesting potential differences between urban and rural settings, as well as among various demographic groups.

Notably, this study highlights that younger individuals and inpatients were associated with an increased risk of being ESBL carriers. Inpatients exhibited a higher ESBL carriage rate than outpatients, consistent with literature identifying prolonged hospitalizations and extensive antibiotic treatments as key predictors for ESBL carriage (Mai and Espinoza, 2023; Moremi et al., 2021).

Our study did not reveal significant associations between ESBL carriage and factors such as previous antibiotic use, or animal contact. However, patients might not have reported previous antibiotic use on the one hand, on the other hand the relatively small sample size limits our ability to draw definitive conclusions. Further research with larger cohorts may be necessary to fully elucidate the relationships between these variables and ESBL carriage in this population.

Despite the high prevalence of patients with a history of UTI (97.2%), our study did not find an association between this factor and ESBL carriage. This disparity can be attributed to most individuals delaying seeking medical treatment, which may result in recurrent UTIs. Furthermore, patients frequently self-medicate using over-the-counter antibiotics (Torres et al., 2021; Maldonado-Barragán et al., 2024) or use traditional remedies, such as herbal medicines. This practice often results in a common history of UTIs among those who do not seek medical attention. In addition, in rural settings, poor sanitation, limited access to clean water, and inadequate hygiene infrastructure increase the risk of UTIs and their recurrence (Mwambete and Msigwa, 2017). We understand that the 97.2% figure might differ from patient populations in other settings. However, this difference highlights the challenges faced in rural health care due to the factors outlined earlier.

ESBL-producing E. coli emerged as the dominant colonizing species, followed by K. pneumoniae, with the CTX-M-15 enzyme being the most prevalent ESBL. The ubiquity of CTX-M-15 genes in both E. coli (86.3%) and K. pneumoniae (92.2%) aligns with global trends as well as with studies from Tanzania (Bevan et al., 2017; Moremi et al., 2017; Onduru et al., 2021; Slown et al., 2022; Sonda et al., 2018). Highly mobile genetic elements are responsible for carrying the predominant CTX-M genes between bacterial strains. The scenario is further supported by the established association between CTX-M enzymes and E. coli IncF resident plasmids, which are particularly well-adapted to E. coli strains and can readily transfer between them (Woerther et al., 2013).

Carbapenemases were only detected in a minority of cases, such as OXA-181 in 3.7% of E. coli and KPC-2 in 1.7% of K. pneumoniae. This low prevalence of carbapenemases suggests that carbapenems are likely to remain an effective treatment option for the time being. However, continued vigilance is essential to uphold infection control measures and prevent the potential rise of carbapenem-resistant isolates in our setting. We observed a phylogenetic diversity among the ESBL-producing E. coli, with the majority belonging to phylogenetic group A, followed by B1 and B2. ST-131 was the most common sequence type in the latter phylogroup. This finding aligns with earlier studies that have shown considerable phylogenetic diversity within ESBL-producing E. coli populations (Bevan et al., 2017). We also observed a notable diversity in sequence types, serotypes, and fimH alleles.

The diverse phylogroups and sequence types observed among ESBL-PE isolates, suggest complex transmission dynamics. Our research underscores the need for further research into community transmission patterns in resource-limited settings.

Limitations of the study include reliance on short-read sequencing data and the lack of environmental sampling, which restricted the comprehensive investigation of plasmid population structures and environmental transmission factors.

In addition, we were not able to investigate the relationship between fecal ESBL carriage and urinary tract infections, as most urinary cultures remained negative or showed only insignificant growth. The negative urine cultures could be due to (a) some symptoms may have been caused by conditions other than a UTI, such as sexually transmitted infections (STIs) which cannot be diagnosed by urine cultures, (b) medical incompliance, patients might be consuming antibiotics but hide this information from the clinician these consumed antibiotic agents may inhibit in-vitro culture growth of bacteria in the urine in this study we did not perform any inhibitory test to detect the presence of any previous antibiotic therapy in our urine samples (c) there may have been improper collection, handling and processing of urine samples in our setting. Our results challenge the assumption that all patients presenting with UTI symptoms will have positive urine cultures but it underscores the importance of including differential diagnoses in patients presenting with UTI-like symptoms. In future, further studies are needed to investigate the causes of these symptoms and the role of non-UTI infections and antibiotic stewardship in our settings. Without correlating fecal ESBL carriage to urine culture results, we cannot definitively establish the gut as a reservoir for invasive UTIs in this population.

In this study, we observed a low rate of history of antibiotic use. This might be because certain patients were unaware/ cannot accurately recall when last they used antibiotics. Additionally, we did not capture self-treatment with non-prescribed antibiotics or the use of traditional remedies in our setting. We recognize the potential for recall bias, given the reliance on self-reported information.

We recognize other ESBL risk factors such as history of surgery, hospitalization in the last 6 months, urinary catheterization, history of ESBL colonization, travel to other countries, and number of people living in the same house are under-recorded. In this study, due to limited access to patient records, we could not have details for these risk factors. Additionally, some factors, such as international travel history, may have been less applicable in our study population compared to urban settings.

Our study included all patients who met UTI IDSA criteria, which introduces selection bias and explains why we had a high proportion of patients with a history of UTI. The model selection for our multivariable model was prone to bias given that we used a p-value cutoff from the bivariate associations to determine which variables to include, however, this was chosen due to limited data availability and unbalanced groupings for other categorical factors. Statistical significance should be taken with caution due to the observational nature of the study.

In conclusion, this study reveals a high fecal carriage rate of ESBL Enterobacterales, especially among younger individuals in Tanzania, underscoring the need to investigate transmission dynamics beyond hospital settings. Although only one ESBL-positive isolate was identified as a cause of UTI, the high fecal carriage rate indicates a significant potential for infections with resistant organisms. This suggests a broader reservoir of resistant pathogens, particularly in areas with limited infection control measures.

Our findings stress the importance of understanding local epidemiology and highlight the need for healthcare interventions in rural areas to ensure proper treatments, appropriate antibiotic use, and education on personal hygiene. These results have significant implications for empirical treatment strategies and infection prevention policies in similar contexts.