Antimicrobial resistance (AMR) is a critical global health threat, particularly in low-resource regions such as Sub-Saharan Africa, where surveillance data are scarce and the burden is highest.1–3 High rates of intestinal carriage of antimicrobial-resistant Enterobacterales and Enterococcus have been documented in pediatric populations.4–6 Carbapenemase-producing Enterobacterales (CPE), especially those harboring Klebsiella pneumoniae carbapenemase (KPC), New Delhi metallo-beta-lactamase (NDM), oxacillinase 48 (OXA-48), Verona integron-encoded metallo-beta-lactamase (VIM) and imipenemase (IMP) genes, and extended-spectrum beta-lactamase (ESBL)-producing strains, are increasingly reported worldwide.7–10 Vancomycin-resistant Enterococcus (VRE) is also an emerging concern and is often linked to mobile genetic elements and nosocomial transmission.11–13

Rural hospitals in Africa face additional challenges: limited access to antibiotics, high rates of malnutrition, delayed diagnostics, and poor hygiene.14,15 The Global Antimicrobial Resistance and Use Surveillance System (GLASS) initiative of the World Health Organization (WHO) calls for expanded AMR surveillance in such settings. The aim of this study, conducted between May and June 2024, was to describe the prevalence of AMR genes in stool samples from pediatric inpatients in rural Ethiopia in order to provide urgently needed data on intestinal carriage rates to inform local and global AMR containment strategies.

This study, conducted in May and June 2024, found high fecal carriage rates of AMR genes among pediatric inpatients in rural Ethiopia, with detection of ESBL (CTX-M) in 80% of patients, carbapenemases (NDM, VIM) in 37.5%, and vanA in 32.5%. These frequencies are comparable to those reported in other studies from Sub-Saharan Africa (high ESBL rates among infants with neonatal sepsis at Jimma Medical Center, Ethiopia,9; 17% prevalence of ESBL fecal carriage in children aged less than 5 years in Addis Ababa, Ethiopia,16; high prevalence of faecal carriage of ESBL-producing Enterobacteriaceae among children in Dar es Salaam, Tanzania,22 [Tellevik et al.]; high prevalence of CRKP in cases of sepsis diagnosed in Ethiopian children aged less than 5 years,17; 61% prevalence of CPE among children hospitalized in pediatric wards in Luanda, Angola,18; CTX-M prevalence of 70% to 90% among pediatric isolates in Kenya/Uganda and high carbapenemase carriage in rural Kenyan pediatric centers,6,19; and colonization by ESBL-producing bacteria in 48% of pediatric inpatients at a South African referral hospital20) and higher than those reported in European pediatric populations (5%–20%).21,23–25

Multiple factors were associated with AMR gene detection (environmental transmission, hygiene deficits, possible vertical pathways) and there was a low prevalence of prior antibiotic exposure (30.6%), indicating that environmental and community-related factors may contribute substantially more than direct selective pressure.14

The median hospital stay was longer in patients carrying AMR genes (8.0 vs 6.0 days; P = .287). This difference was not statistically significant, probably on account of the low statistical power (n = 40, post-hoc power <30%) and unmeasured confounders. Therefore, no clinical inference can be drawn.

The small sample size was a limitation but reflects the reality of research in rural African hospitals, where logistical and ethical barriers make large studies challenging. Similar studies with sample sizes of less than 50 have been published in the International Journal of Antimicrobial Agents and other leading journals, having explained the context clearly and justified the limitations. Despite these constraints, the study yields several important public health implications. Our findings, though not powered for multivariable analysis, provide critical baseline data for this neglected population.

The unexpected inverse association between severe chronic malnutrition and carbapenemase carriage may reflect an altered microbiota or reduced health care exposure in the most malnourished, but this requires confirmation in larger cohorts. Given the cross-sectional design and small sample size, the associations observed in this study should be interpreted as hypothesis-generating rather than causal and warrant confirmation in future longitudinal research.

The key limitations include: (1) lack of phenotypic correlation, as gene detection does not confirm resistance expression (which would require antibiogram with minimum inhibitory concentration [MIC]/culture); (2) no species identification (generic identification of Enterobacterales/Enterococcus)26,27; both intrinsic to molecular surveillance with PCR in rural hospitals lacking infrastructure. These aspects do not invalidate the descriptive genotypic baseline prevalence data.4,10 The high prevalence of VRE gene detection observed in our hospital aligns with Ethiopian pediatric reports.26–28

Implications: There is an urgent need for investment in diagnostic capacity, infection control, and antimicrobial stewardship in rural African hospitals.1,15 Surveillance in underrepresented settings is essential to inform global AMR strategies.2,3 Given the high prevalence of malnutrition (65%) and AMR carriage observed in this cohort, future studies should incorporate standardized neurodevelopmental assessment tools to better explore the combined impact of these factors on long-term developmental outcomes. This critical aspect could not be addressed in the present study due to resource and logistics constraints typical of rural African settings. Future research should prioritize larger multicenter cohort studies exploring transmission (vertical/environmental) as well as VRE trends in Africa vs Europe.13,29,30

These results suggest the need for a comprehensive public health approach tailored to the local context. Specifically, the development of local empirical antibiotherapy protocols may be warranted in light of the high observed prevalence of intestinal carriage (CTX-M in 80%, NDM in 37.5%, vanA in 32.5%). At Gambo Rural General Hospital, this might involve prioritizing beta-lactams not compromised by ESBL activity for severe infections, with early review for de-escalation when possible, and implementing strict criteria for limited and sporadic access to carbapenems and vancomycin, which are only available through occasional donations from foreign collaborators.5,8

It is also essential to implement antimicrobial stewardship programs (ASPs) adapted to resource-limited settings. These interventions should include training on rational antibiotic use, de-escalation criteria based on local data, and antibiotic consumption monitoring. Particularly, given the extremely limited availability of carbapenems at Gambo Rural General Hospital—dependent exclusively on irregular donations from international collaborators—ASP programs should prioritize the strategic reservation of these scarce resources for culture-confirmed cases with no alternative therapeutic options, maximizing their impact while preventing further resistance selection.

Complementing these efforts, realistic infection control measures (such as reinforcing hand hygiene, basic separation of colonized patients, and caregiver education) are feasible even in rural hospitals and can mitigate nosocomial transmission of resistance genes.11,13

High fecal carriage of AMR genes was observed among pediatric inpatients in rural Ethiopia (May–June 2024). These baseline data highlight the need for enhanced surveillance, tailored stewardship, and infection control in resource-limited environments.

Partially funded by Werfen Ltd., which provided molecular biology reagents. Also, this study was conducted in the framework of the research project “The search for new compounds with antimicrobial properties against multidrug-resistant pathogens or with ineffective treatments” [Fundación para la Investigación Biomédica del Hospital Universitario Príncipe de Asturias (FIBHUPA)].