A retrospective assessment of antimicrobial resistance patterns in WHO-access, watch, and reserve-classified antibiotics across two large hospitals in a resource-limited setting

Abstract

Antimicrobial resistance (AMR) remains a significant threat. To ensure efficient use of antibiotics, the WHO AWaRe Antibiotics classification framework classifies antibiotics into three groups: Access, Watch, and Reserve. Tracking antibiotic resistance across these groups remains a critical task. This retrospective study examined the prevalence of resistance within the AWaRe Antibiotics classification framework in two large hospitals in Abuja Nigeria. A total of 14,423 microbial culture test results were sourced and retrieved from the hospitals between January and December 2023, of which 3,987 (27.64%) showed microbial growth and with antimicrobial susceptibility results. Six different sample types were analyzed, with urine samples (36.22%) yielding the highest number of isolates. Escherichia coli (23.78%) was the most identified microbe. The overall AMR rate across the AWaRe classification was 41.96%. For the Access category, highest resistance was reported for Doxycycline at 100% resistance while the lowest resistance was reported for Amikacin at 24.63%. For the Watch category, the highest resistance was reported for Cefuroxime at 85.71% while the lowest resistance was reported for ticarcillin/clavulanic acid and Netilmicin, both with 0% resistance. For the Reserve category, the highest resistance was reported for Aztreonam at 67.96% while the lowest resistance was reported for Polymixin B at 9.71%. Cumulatively, antimicrobials in the Access category had a resistance prevalence rate of 44.29%; those in the Watch category had a resistance prevalence rate of 46.65%; while those in the Reserve category had a resistance prevalence rate of 34.95%. More than 70% of multi-drug resistance was observed in frequently used antibiotics such as Cefuroxime, Ceftazidime, and Meropenem. Additionally, over 60% resistance was found in antibiotics like Ertapenem, Aztreonam, and Teicoplanin even though they are not readily available in Nigeria. Every microbe reported in this study was resistant to more than one antimicrobial agent. The most resistant microbes were Staphylococcus aureus, Klebsiella spp., Streptococcus spp., and Escherichia coli. Across gender, positive isolates were found more in women than men across almost all the microbes identified. These results, based on real patient data, highlight the urgent need for an institutionalized AWaRe Antibiotics framework across healthcare facilities in addition to broader strategies of antimicrobial stewardship, pharmaceutical supply chain enhancement, and national AMR surveillance.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-37665-x.

Introduction

The threat of antimicrobial resistance (AMR), caused by overuse and misuse of antimicrobial agents, continues to rise, complicating infection treatment and control and endangering public health globally^1,2. The scope of AMR is categorized by severity, with multidrug-resistant (MDR) isolates defined as non-susceptible to at least one agent in three or more antimicrobial classes; extensively drug-resistant (XDR) isolates defined as non-susceptible to all but one or two antimicrobial classes; and pandrug-resistant (PDR) isolates defined as non-susceptible to all agents across all antimicrobial classes tested². In 2019, 4.9 million deaths were attributed to AMR³. This could reach 10 million deaths in the next two decades, majority of which will occur in low-resource settings like Nigeria, if there is no strong response to this public health threat³.

Many response strategies have been adopted to curb the rising burden of AMR, such as the World Health Organization (WHO) Critically Important Antimicrobials for Human Medicine (CIA list)⁴; the European Medicines Agency Antimicrobial Advice Ad hoc Expert Group (AMEG) Categorisation⁵; and the Access, Watch, and Reserve (AWaRe) by WHO⁶. While the WHO CIA list categorizes antibiotics based on their importance to human medicine to support risk assessment and management strategies, AMEG’s classification is based on the risk to public health from AMR due to the use of antimicrobials in animals^4,5 The AWaRe classification, however, is aimed at optimizing the use of antibiotics depending on their spectrum of activity, sensitivity, and efficacy against bacterial infections^6–8. The Access group includes antibiotics that are recommended as first- or second-line treatments for most infections. The Watch group comprises antibiotics that have a higher potential for developing resistance and should be used prudently. The Reserve group consists of last-resort antibiotics used to treat infections caused by multidrug-resistant bacteria^9,10.

The WHO’s 13th General Programme recommends country-level targets for antibiotic consumption, with at least 60% of total antibiotic consumption being from the Access group¹¹. This recommendation and the AWaRe classification are valuable guides for healthcare professionals, policymakers, and researchers to develop and implement antibiotic stewardship policies and programs to monitor antibiotic usage and set goals for combatting AMR¹². In addition to the AWaRe classification, WHO has also launched the Global Antimicrobial Resistance and Use Surveillance System (GLASS) where over 114 countries, including Nigeria, have committed to contributing data on AMR. This data is meant to strengthen antimicrobial usage and track resistant pathogens via laboratory sensitivity testing data to ensure early detection and control of these pathogens^13–15. Laboratory microbial culture and antimicrobial susceptibility testing results for biological samples such as blood, urine, and aspirate are crucial for determining the most effective antibiotics for treating bacterial infections^16,17. These tests such as disk diffusion, minimum inhibitory concentration methods, and genotypic and phenotypic characterization of bacterial resistance, help in determining pathogens that are sensitive and resistant to different antibiotics. This then guides clinicians in treatment decision making^16–19.

Strategies and frameworks such as AWaRe are crucial for low-resource settings like Nigeria where One Health approach of integrating human, animal, and environmental health surveillance to ensure stewardship across sectors is crucial, and where previous studies have documented AMR patterns in hospitals and community settings. For instance, a point-prevalence survey in a teaching hospital in the southern part of Nigeria found that 51% of bacteria assessed were resistant to three or more drug classes²⁰. Another study in northern Nigeria revealed a high prevalence (54.0%) of Gram-negative bacterial infections of which a significant proportion (88.9%) was as a result of multidrug-resistant bacterial pathogens²¹. Some other studies have reviewed and meta-analyzed resistance patterns to certain bacterial pathogens in Nigeria²². However, none of these studies classify resistance data within a holistic framework such as the WHO AWaRe categories. Most focus on individual antibiotics or specific pathogens, rather than structuring findings across the broader AWaRe framework. As a result, their insights are limited with regards to patterns of resistance that reflect WHO’s stewardship guidance or the relative risk posed by overuse of “Watch” or “Reserve” antibiotics. Therefore, this study aims to ascertain the prevalence of AMR within the framework of AWaRe antibiotic classification. Specifically, the study assessed the prevalence of resistance amongst available AWaRe antibiotics in two large tertiary hospitals in Abuja, Nigeria.

Methods

Study design and setting

A retrospective cross-sectional design was used to collect demographic and microbiological data on microbial culture and antimicrobial sensitivity laboratory testing results from the hospital information system of two large tertiary hospitals in Abuja, North Central Nigeria from January – December 2023. One hospital, Federal Medical Center Abuja, is in an urban setting, has a 351-bed capacity, and its laboratory is Level III WHO standard medical laboratory. The other hospital, University of Abuja Teaching Hospital Gwagwalada, is in a peri-urban setting and has a 520-bed capacity. Both hospitals were chosen as they are located within Nigeria’s Federal Capital Territory which has one of the highest levels of medical regulation and treatment capacity in Nigeria.

Study population and data collection

All patients, including adults and children, with requests who underwent culture and sensitivity testing in microbiology laboratory units of both hospitals, constituted the study sample from which those meeting the inclusion criteria (having a positive culture test, irrespective of age) were drawn. Duplicate isolates from the same patient were not included in the analysis. Sensitivity results outside the study period and culture results with no microbial growth within the period under study were excluded from the analysis. Information collected include specimen type, patient gender, types of microorganisms isolated, and antibiotic susceptibility and resistance result from the microbiology lab records. Sample processing and identification of isolates by the two hospitals follow standard operating procedures for microbial culture and sensitivity testing. Samples are cultured on microbiological media, incubated at 37 °C for 24–48 h, and identified through colony morphology, Gram staining, and biochemical tests. Antimicrobial susceptibility is assessed using the Kirby-Bauer disk diffusion method on Mueller-Hinton agar, measuring zones of inhibition to determine resistance or sensitivity to antibiotics.

Laboratory methods

Urine specimens were cultured by plating the sample on selective media (Cystine Lactose Electrolyte-Deficient [CLED] media, blood agar, and MacConkey agar) using a calibrated loop. Plates were incubated at 35–37 °C for 24–48 h. Colonies were then visually examined for preliminary identification, and final bacterial confirmation was performed using biochemical tests to assess metabolic properties.

Blood specimens were cultured by inoculating 125 µL of blood from positive culture bottles onto Mueller–Hinton agar plates, followed by incubation at 35 °C. Rapid Antimicrobial Susceptibility Testing (RAST) was performed using disk diffusion on agar plates. Plates were incubated for 4–8 h, after which biochemical and susceptibility tests were carried out to determine bacterial identification and resistance profiles.

Swab specimens were cultured by inoculating onto selective media such as blood agar or chocolate agar plates and incubating at 35 °C in 5% CO₂ for 18–24 h, according to European Committee on Antimicrobial Susceptibility Testing (EUCAST) guidelines. Biochemical identification and antimicrobial susceptibility testing were performed on isolated colonies under similar incubation conditions to determine pathogen identity and resistance profiles.

Stool specimens were cultured on MacConkey agar and blood agar and incubated at 36 °C for 18–24 h. Biochemical tests for bacterial identification followed culture and were typically performed at 35–37 °C to assess metabolic capabilities and confirm pathogen identity during susceptibility testing.

Aspirate specimens were cultured on blood agar and chocolate agar and incubated aerobically at 35–37 °C with 5–10% CO₂ for 24–48 h. Biochemical tests followed incubation for bacterial identification and included enzyme activity assays performed at similar temperatures.

Body fluid specimens were cultured in blood culture bottles incubated at 35 °C. After positivity, 150 µL was sub-cultured onto Mueller–Hinton agar and selective media such as blood, chocolate, and MacConkey agar. Biochemical tests followed incubation to identify organisms and assess antimicrobial susceptibility.

Species-level identification was only performed for Escherichia coli and Staphylococcus aureus due to laboratory limitations in testing other microbes at species level. Fungal cultures were not performed routinely across specimen types. However, we included results for Candida spp. because they were assessed when clinicians requested testing based on suspicions of fungal infections such as candidiasis and because they are also implicated in polymicrobial infections with other bacteria and could impact disease severity and drug resistance²³.

VITEK 2D version and antibiotics disc diffusion was used. The hospital laboratories used EUCAST 13.0 (2023) version as breakpoint. The names of each of the antibiotics used to test for susceptibility and resistance in this study, their concentrations and manufacturers are listed in the supplementary files.

Data analysis

The data on antimicrobial susceptibility (sensitivity/resistance) performed by the hospitals using standard microbiological methods were retrieved. Descriptive statistics using the Statistical Package for Social Sciences version 21 (IBM Corp., Armonk, NY, USA) were then applied to summarize and analyze the frequency and rates of resistance by generating summary statistics and visualizations for better understanding of the overall patterns.

Ethical considerations

Ethical approval was obtained from the National Health Research Ethics Committee of Federal Medical Center Abuja and University of Abuja Teaching Hospital with protocol numbers FMCABJ/HREC/2024/128 and UATH/HREC/PR/388 respectively. All methods used in this study were performed in accordance with the relevant guidelines and regulations of the ethics committees above and align with the ethical standards outlined by the Declaration of Helsinki and other international guidelines governing biomedical research involving human subjects. Based on these guidelines, retrospective studies are considered non-interventional and primarily observational. Additionally, the initial data was collected for clinical purposes and given the volume of data, tracking patients and obtaining individual informed consent is impractical. The ethics committees (National Health Research Ethics Committee of Federal Medical Center Abuja and University of Abuja Teaching Hospital) waived individual informed consent following ethical approval. However, data obtained were anonymized and aggregated to further maintain confidentiality and privacy in line with ethical principles.

Results

During the period under study (January – December 2023), a total of 14,423 microbial culture tests were retrieved from both hospitals of which 3,987 (27.64%) of tests showed microbial growth. The remaining samples that did not show microbial growth were excluded. Based on the data obtained, the specimen samples provided by patients were aspirate, blood, body fluid, stool, swab, and urine. From the specimen samples, urine samples had the most microbial isolates—1444 (36.22%) followed by swab with 1122 (28.14%). Stool samples showed the least microbial growth with 199 (4.99%). Table 1 shows the demographic distribution of the sample from which microbial culture test results were retrieved.

Table 1.

Demographic distribution of the sample.

Characteristic	Category	Count (n)	Percentage (%)
A. Culture test result	Negative cultures	10,436	72.36%
	Positive cultures	3987	27.64%
Total		14,423	100%
B. Age distribution of positive cultures	Children (0–17 years)	1885	47.28%
	Adults (> 17 years)	2102	52.72%
Total		3987	100.00%
C. Gender distribution of positive cultures	Male	1290	32.36%
	Female	2697	67.64%
Total		3987	100.00%
D. Specimen type distribution of positive cultures	Urine	1444	36.22%
	Swab	1122	28.14%
	Blood	492	12.34%
	Aspirate	377	9.46%
	Body fluid	353	8.85%
	Stool	199	4.99%
Total		3987	100.00%

Escherichia coli was the most isolated microbe (n = 948, 23.78%), followed by Staphylococcus aureus (n = 831, 20.84%), and Klebsiella spp. (n = 558, 14.00%). The least isolated microbes were Corynebacterium spp. (n = 4, 0.10%), Serratia spp. (n = 5, 0.13%), and Shigella spp. (n = 7, 0.18%).

Table 2 shows the distribution of microbial isolates across the sample population.

Table 2.

Distribution of positive isolates across gender and age, and prevalence of AMR in the sample population.

Pathogens recorded	Distribution across gender		Distribution across age groups		Total, n (%)	AMR Prevalence, n (%)
	Female, n (%)	Male, n (%)	Children, n (%)	Adult, n (%)		Resistance	Susceptible
Acinetobacter spp.	51 (1.28%)	30 (0.75%)	40 (1.00%)	41 (1.03%)	81 (2.03%)	334 (0.94%)	182 (0.51%)
Candida spp.	427 (10.71%)	27 (0.68%)	56 (1.40%)	398 (9.98%)	454 (11.39%)	514 (1.45%)	440 (1.24%)
Citrobacter spp.	65 (1.63%)	41 (1.03%)	35 (0.88%)	71 (1.78%)	106 (2.66%)	376 (1.06%)	287 (0.81%)
Corynebacterium spp.	1 (0.03%)	3 (0.08%)	3 (0.08%)	1 (0.03%)	4 (0.10%)	20 (0.06%)	15 (0.04%)
Enterobacter spp.	71 (1.78%)	43 (1.08%)	48 (1.20%)	66 (1.66%)	114 (2.86%)	487 (1.37%)	368 (1.04%)
Enterococcus spp.	125 (3.14%)	45 (1.13%)	95 (2.38%)	75 (1.88%)	170 (4.26%)	729 (2.05%)	1,009 (2.84%)
Escherichia coli	595 (14.92%)	353 (8.85%)	490 (12.29%)	458 (11.49%)	948 (23.78%)	4,341 (12.23%)	4,976 (14.02%)
Klebsiella spp.	383 (9.61%)	175 (4.39%)	288 (7.22%)	270 (6.77%)	558 (14.00%)	2,859 (8.06%)	3,021 (8.51%)
Proteus spp.	110 (2.76%)	48 (1.20%)	88 (2.21%)	70 (1.76%)	158 (3.96%)	689 (1.94%)	781 (2.20%)
Pseudomonas spp.	106 (2.66%)	75 (1.88%)	80 (2.01%)	101 (2.53%)	181 (4.54%)	941 (2.65%)	621 (1.75%)
Salmonella spp.	61 (1.53%)	57 (1.43%)	65 (1.63%)	53 (1.33%)	118 (2.96%)	387 (1.09%)	774 (2.18%)
Serratia spp.	3 (0.08%)	2 (0.05%)	0 (0.00%)	5 (0.13%)	5 (0.13%)	20 (0.06%)	23 (0.06%)
Shigella spp.	3 (0.08%)	4 (0.10%)	1 (0.03%)	6 (0.15%)	7 (0.18%)	22 (0.06%)	51 (0.14%)
Staphylococcus aureus	515 (12.92%)	316 (7.93%)	484 (12.14%)	347 (8.70%)	831 (20.84%)	3,642 (10.26%)	4,653 (13.11%)
Streptococcus spp.	181 (4.54%)	71 (1.78%)	112 (2.81%)	140 (3.51%)	252 (6.32%)	1,196 (3.37%)	1,723 (4.86%)
Total sample isolates (percentage)	2697 (67.64%)	1290 (32.36%)	1885 (47.28%)	2102 (52.72%)	3987 (100%)	16,557 (46.66%)	18,924 (53.34%)

Antimicrobial susceptibility/resistance profiles

Every microbe reported in this study was resistant to more than one antimicrobial agent. Across gender, positive isolates were found more in women than men across almost all the microbes identified. Results show that Corynebacterium spp., Enterococcus spp., Escherichia coli, Klebsiella spp., Proteus spp., Salmonella spp., and Staphylococcus aureus were more prevalent in children.

Figures 1, 2, 3 and 4 show the prevalence of AMR across antibiotics analyzed in the study in line with the AWaRe classification. In the Access category (Fig. 1), there was total resistance to Doxycycline. High levels of resistance were observed for other antibiotics in this class such as 69.64% for Trimethoprim/Sulfamethoxazole, 57.22% for Oxacillin, and 51.26% for the readily available Amoxicillin/Clavulanate combination. The least resistance in the Access category was seen in Amikacin (24.63%) and Gentamycin (31.73%).

Fig. 1.

Prevalence of AMR across Access antibiotics.

Fig. 4.

Prevalence of AMR across other Antimicrobials.

Fig. 6.

(a) Heat map of Pathogens against resistance of antibiotics in the Access category. (b) Heat map of Pathogens against resistance of antibiotics in the Watch category. (c) Heat map of Pathogens against resistance of antibiotics in the Reserve category.

Fig. 3.

Prevalence of AMR across Reserve antibiotics.

Fig. 2.

Prevalence of AMR across Watch Antibiotics.

In the Watch Category (Fig. 2), high levels of resistance were observed for the readily available Cefuroxime at 85.71%, Ceftazidime at 81.23%, and 72.51% for Meropenem. Some of the antibiotics in this class showed no resistance, especially Netilmicin and Ticarcillin/Clavulanate combination.

In the Reserve category (Fig. 3), the highest resistance was observed for Aztreonam (67.96%) and Fosfomycin (47.27%). The least resistance was observed for Polymyxin-B (9.71%). All the antibiotics in the Reserve category showed some level of resistance.

Figure 4 shows the resistance profile for other antimicrobials that were tested. The resistance patterns were similar with Fluconazole showing resistance of 52.88%, Voriconazole showed resistance of 51.06%), while Bacitracin showed resistance of 50%.

Figure 5 shows the collective resistance and susceptibility profiles of antibiotics across the AWaRe classification. Antibiotics in the Access category had a resistance prevalence rate of 44.29%; those in the Watch category had a resistance prevalence rate of 46.65%; while those in the Reserve category had a resistance prevalence rate of 34.95%. The average prevalence of AMR across all the antibiotics was 41.96% and susceptibility to the antibiotics was 58.04%.

Fig. 5.

Pattern of resistance and susceptibility across AWaRe antibiotics.

Multidrug resistance (MDR), extensively drug-resistance (XDR), and pandrug-resistance (PDR) within the aware classification

Figure 6a and b, and c shows heatmaps presenting organism-specific resistance profiles across all tested antibiotics (classified by AWaRe groups). It provides a clear visual summary of clinically relevant resistance patterns which could enable rapid assessment of organism-specific resistance patterns and facilitate easy and meaningful comparison to support evidence-based empirical therapy decisions.

Within the AWaRE classification, many microbes show particular concern for MDR and XDR. Acinetobacter spp. showed high resistance to one antibiotic (Tetracycline) in the Access category and multiple antibiotics in the Watch category. This trend was also noted for other microbes like Citrobacter spp., Corynebacterium spp., E. coli, Klebsiella spp., and Staphylococcus aureus.

Enterococcus spp., Klebsiella spp., Proteus spp., Pseudomonas spp., and Streptococcus spp. demonstrated particularly concerning resistance profiles across the AWaRe classification. These organisms exhibited high resistance levels to both Access and Watch antibiotics, with emerging resistance also observed among Reserve agents. No case of PDR was observed for any microbe within the dataset.

Access antibiotics showed mixed performance: while some first-line agents retained moderate susceptibility, others—most notably doxycycline—demonstrated very high resistance. Watch antibiotics displayed the most concerning trends, with several broad-spectrum agents exhibiting high resistance rates exceeding 70%, including Cefuroxime, Ceftazidime, Cefepime, Meropenem, and Norfloxacin. In contrast, Reserve antibiotics maintained comparatively lower resistance levels, indicating that they remain effective options when used judiciously.

Discussion

This study analyzed the prevalence of antibiotic resistance across AWaRe antibiotics grouping in two large hospitals in a resource-constrained setting. Results show susceptibility testing against a wide array of antibiotics in this classification; however, it is important to note that laboratories typically test only a panel of 8–10 antibiotics at a time as part of their routine protocol. This panel usually consists of commonly used or first-line antibiotics. If complete resistance is observed, the test is then repeated using second-, third-, or fourth-line options as appropriate. This limitation in routine testing reflects the challenges of cost, reagent availability, and workflow efficiency in resource-limited settings like Nigeria, which restricts laboratories from conducting comprehensive testing against an extensive antibiotic panel.

Findings showed high burden of resistance. Since there is a paucity of research on AMR across the different classes of Access, Watch and Reserve for antibiotics, the findings of this study offer new insights to support the need for institutionalized WHO-AWaRe and GLASS frameworks to ensure efficient stewardship of antimicrobials to curb resistance.

Across all samples analyzed, stool samples showed the least microbial growth. The low number of bacterial identifications from stool in our study is consistent with known limitations of stool culture: routine stool cultures often have very low diagnostic yield and conventional culture has intrinsically low sensitivity in stool because fastidious enteric bacteria are difficult to grow amid dense commensal flora^23,24.

The gram-negative bacterium Escherichia coli exhibited the highest resistance (23.78%), similar to a study by Bizimungu et al.²⁶, although our study had a sample size 32 times greater and targeting the AWaRe classification. With respect to the most burdened and vulnerable group, Staphylococcus aureus and Escherichia coli showed leading dominance in children. This dominant burden of Staphylococcus aureus in children is similar to a report by McNeil et al. in the United States of America²⁷. In addition to Staphylococcus aureus, our findings also show that Corynebacterium spp., Enterococcus spp., Escherichia coli, Klebsiella spp., Proteus spp., and Salmonella spp. were more prevalent in children. This has also been reported in the literature on antibiotic susceptibility patterns among children admitted to Nigerian hospitals²⁸. Evidence suggests that this could be due to some interacting reasons. Physiologically, children have immature immune systems and higher rates of mucosal and skin colonization, which raises susceptibility to both community-acquired and hospital-associated pathogens^29,30. Furthermore, enteric organisms like Salmonella, E. coli, Proteus, and Klebsiella species are common causes of diarrhoeal disease and bacteraemia in children where water, sanitation and hygiene (WASH) gaps and high burden of enteric infection increase exposure and transmission³¹. Also, high carriage of these extended-spectrum β-lactamase (ESBL)–producing Enterobacteriaceae have been reported in pediatric populations in Nigeria, further fueling this prevalence³². It is important to note that in our study, we were unable to estimate ESBL prevalence within our dataset because routine phenotypic ESBL testing was not performed in the participating laboratories. The antibiograms made available to us reflected standard susceptibility testing without confirmatory ESBL methods. This gap highlights the broader challenge of constrained microbiology capacity in many Nigerian health facilities and the need to expand routine ESBL detection to support more accurate surveillance and guide empirical treatment.

Among adult patients, Escherichia coli recorded the highest resistance, similar to the findings by Hossain et al. in Bangladesh, a similar resource-constrained setting like Nigeria³³. Staphylococcus aureus exhibited the highest multi-drug resistance; such multi-drug resistance has been reported in Ghana for Staphylococcus aureus and other pathogens like Escherichia coli, Klebsiella spp., Streptococcus spp., and Pseudomonas spp³⁴. This indicates similarities across health systems in resource-constrained settings within similar geographic locations. These comparisons suggest that the proliferation of AMR is without boundaries and is seen across various settings, necessitating enhanced infection control practices, reducing overuse and misuse of antibiotic through prioritized stewardship interventions, and cross-regional policies and global AMR strategies that are not implemented in siloes. There is need for serious stewardship action, as these bacteria are among the WHO’s global concern ESKAPE pathogens (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter spp.)³⁵.

Findings also showed over 70% multi-drug resistance to various antibiotics, including Ceftazidime, Cefuroxime, Cefepime, Doxycycline, Meropenem, and Norfloxacin. This compares with a recent study published in Nature Scientific Reports showing 75% resistance to Cefuroxime³⁶. Trimethoprim/Sulfamethoxazole also recorded high resistance of 69.64% similar to a study by Newman et al. in Ghana, with 73%³⁷ complicating infection treatment options in Nigeria and the West African region. Our study revealed over 60% resistance to three antibiotics—Ertapenem (60.98%), Aztreonam (67.96%), and Teicoplanin (75.0%)—which are not readily available in the Nigerian drug market. Similarly, a study in Bangladesh reported high resistance to Aztreonam, with 44.9% multi-drug resistance³⁸. These findings indicate a significant likelihood of cross-resistance or the importation of resistant pathogens that may render current medications in Nigeria ineffective. This poses a serious public health risk and underscores the urgent need for robust stewardship actions and the institutionalization of functional AWaRe and GLASS frameworks. These necessitates monitoring of emerging resistance patterns, which could support the development of comprehensive treatment strategies and fostering research and innovation against resistant strains.

Findings also revealed that certain antibiotics, including Ticarcillin-clavulanic acid, Tigecycline, Polymyxin-B, Netilmicin, and Amikacin, exhibited over 70% susceptibility rates to pathogens, highlighting their value in infection treatment and the need for careful stewardship. Among Access group antibiotics, Amikacin demonstrated a high susceptibility rate (75.37%), consistent with studies reporting low resistance to Amikacin³⁴.

While the Access group of antibiotics does not have the highest recorded resistance, its 44% resistance prevalence is concerning, particularly because WHO recommends that this group should constitute 60% of antibiotic consumption. This high resistance means many infections that could have been treated with Access antibiotics may now require treatment with antibiotics from the Watch or even Reserve class.

The Watch class of antibiotics, which infectious disease experts aim to preserve, has emerged as the epicenter of antimicrobial resistance (AMR) within the AWaRe classification. Findings indicate that antibiotics in this group exhibit the highest resistance, aligning with reports linking high resistance rates to elevated consumption levels³⁹. Cefuroxime, one of the most commonly used antibiotics in treatment regimens in Nigeria and beyond, showed the highest microbial multi-resistance within the Watch class, with a resistance prevalence of 85.71%. This suggests that many infections may now require treatment with Reserve class antibiotics.

Although findings indicate that the Reserve class has the lowest resistance, concerns remain. For example, Aztreonam, which is not included in Nigeria’s Essential Medicines List and is not readily available in the country⁴⁰, demonstrated a resistance prevalence of 67.98%. This high resistance poses significant risks to both local and global health, exacerbating infection severity and mortality rates and increasing the likelihood of an epidemic or even a pandemic caused by resistant pathogens. These findings underscore the urgent need for targeted stewardship efforts and enhanced monitoring frameworks to mitigate the threat of AMR.

Notably, a resistance rate of 52.88% was observed for fluconazole, one of the primary antifungal agents, and a similar resistance rate of 51.06% for Voriconazole. This finding poses a challenge with fungal infections such as candidiasis and candidemia, which are linked to higher morbidity and mortality rates, particularly among the elderly and immunosuppressed populations⁴¹. Although not a bacterium, we included Candida spp. in the analysis because testing for Candida spp. in this study was performed selectively when clinicians suspected fungal infections such as candidiasis. Additionally, it has clinical relevance, due to its role in polymicrobial infections that may influence disease severity and drug resistance^42,43. Therefore, results on Candida spp. likely does not reflect the true fungal epidemiology in these settings and should be interpreted with caution.

Generally, in interpreting these results, it is important to note that certain bacteria have intrinsic resistance profiles, which means that they naturally resist certain antibiotics due to inherent genetic traits like outer membranes, efflux pumps, or enzymes^44,45. For example, Escherichia coli has intrinsic resistance to Vancomycin; certain species of Citrobacter, Klebsiella, Enterobacter, Proteus, Pseudomonas and Serratia are resistant to Ampicillins, Amoxicillin-Clavulanate and Ampicillin-sulbactam combinations; some Enterococcus species exhibit intrinsic low susceptibility to cephalosporins, while some Pseudomonas species have intrinsic resistance to many β-lactams and quinolones^44,45. These profiles, distinct from acquired resistance, are vital and should be considered when interpreting these findings with respect to guiding appropriate antibiotic choices.

What these results imply is that AMR could be responsible for many deaths in Nigeria, further contributing to the over 1.27 million deaths directly caused by AMR globally⁴⁶. The findings of this research not only enhance existing knowledge on AMR prevalence but also serve as a crucial call to action to strengthen health system capacity to curb AMR and its debilitating effects. First, scaling up the Nigeria National Antimicrobial Stewardship Taskforce (NNAST)-led antimicrobial stewardship programmes in hospitals and clinics can promote rational antibiotic use such as prescribing only when necessary, at correct dosages, for appropriate durations, and de-escalating or stopping therapy when indicated⁴⁷. Second, enhancing pharmaceutical supply chains and the capacity of its handlers is crucial⁴⁸. This will help ensure a stable, quality-assured supply of essential antibiotics to avoid inappropriate switching or concomitant use driven by stockouts⁴⁹. Robust pharmaceutical supply chains and procurement systems could reduce the use of suboptimal or broad-spectrum agents⁴⁹. Additionally, expanding diagnostic capacity, including routine ESBL detection, would enable more accurate surveillance and targeted therapy based on actual pathogens rather than broad-spectrum use⁵⁰. Continuous education and training of prescribers, pharmacists and laboratory staff would reinforce stewardship principles, improve interpretation of diagnostic results, and reduce misuse of antibiotics⁵¹. Strengthening infection prevention and control, sanitation and hygiene in community settings could limit cross-transmission of resistant pathogens, reducing the overall burden of infections and the need for antibiotic therapy⁵². Finally, sustained national surveillance and monitoring of AMR patterns would allow timely detection of emerging resistance, evidence-based updating of treatment guidelines, and evaluation of intervention effectiveness⁵³.

Conclusion

This study assessed the prevalence of AMR within the AWaRe antibiotic classification framework by examining resistant pathogens among AWaRe antibiotics in two large tertiary hospitals in Nigeria. The findings revealed a high prevalence of AMR across all three antibiotic classes, emphasizing the urgent need for a robust response to address this critical public health challenge.

A larger population-based study is necessary to further explore AMR trends and patterns. Additionally, a comprehensive response is required, encompassing strong leadership and governance, improved service delivery related to antimicrobial use (including susceptibility testing, prescription practices, and consumption monitoring), institutionalization of evidence-based antimicrobial stewardship policies, and allocation of sufficient human and financial resources to ensure effective implementation and sustainability of these efforts.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Ejikemeuwa Benedict Eya: Conceptualization, Data curation, Validation, Formal analysis, Writing – original draft, Writing – review & editing. Blessing Enyanwu Ogbeche: Data curation, Formal analysis, Writing – review & editing. Otuto Amarauche Chukwu: Conceptualization, Validation, Formal analysis, Writing – review & editing, Supervision.

Data availability

The data that support the findings of this study will be made freely and readily available upon request to the corresponding author.

Declarations

Competing interests

The authors declare no competing interests.