Antibiogram of pathogenic bacterial isolates from various samples in Prince Sultan Hospital, Western Region, Saudi Arabia. Retrospective study from 2020 to 2024

Introduction

The incidence of antimicrobial resistance (AMR) incidence has a particularly severe impact on developing countries, such as Saudi Arabia. Insufficient infection control practices, combined with the misuse of antibiotics, can contribute to the spread of resistant pathogens in overcrowded hospitals (Salam et al., 2023). Healthcare-associated infections (HAIs) and limited resources for comprehensive antimicrobial antibiograms are factors contributing to the spread of AMR (Zaha et al., 2019).

The increasing resistance among Gram-negative organisms, such as Escherichia coli, Pseudomonas aeruginosa, Acinetobacter baumannii, Salmonella, Klebsiella pneumoniae, and Gram-positive organisms such as Staphylococcus aureus, presents a significant public health threat. Increasing antibiotic resistance makes it more difficult to treat serious infections caused by these pathogens, such as bloodstream infections, pneumonia, wound infections, and urinary tract infections (Ahmad et al., 2023; Elfadadny et al., 2024).

In order to combat AMR, antibiograms provide vital local data that can guide empirical therapy. Whenever access to highly effective, costly antibiotics is restricted, antibiograms provide clinicians with evidence-based recommendations for antibiotic use, ensuring that the most appropriate treatment is administered in a timely manner. As an alternative to inappropriate therapy that can exacerbate resistance development, antibiograms play a crucial role (Saleem et al., 2018).

There is an alarmingly high rate of antibiotic resistance in Saudi Arabia, according to several studies. According to a study by Abdallah et al. (2020), 30% of isolates were ESBL-positive, while 74% of the isolates were multidrug resistant. Due to unregulated antibiotic use and inadequate healthcare infrastructure, these resistance trends are often more severe in developing countries (Ehsan, 2025).

Regionally, AMR is well recognized; however, many tertiary care hospitals lack data on resistance patterns. This study aims to empower clinicians in their decision-making and to encourage the development of focused antimicrobial stewardship programs. Such efforts are essential for preserving the efficacy of the limited antibiotics that remain effective against MDR organisms in Saudi Arabia, ultimately improving patient outcomes, mitigating the further spread of resistance, and documenting changes in bacterial resistance patterns.

Material and methods

Study design and setting

A retrospective study (January 2020–September 2024) at Prince Sultan Hospital, Western Saudi Arabia, identified regional resistance patterns, which guided antibiotic policies in similar facilities.

Sample collection and processing

We included different samples, such as blood cultures (BCs), urine (URN) (voided, suprapubic), lower respiratory tract (LRT) samples (sputum, endotracheal tube, tracheal aspirate), and wound swabs (WS) (pus and discharge). We included a total of 5,626 clinical specimens (1,438 BCs, 3,022 URNs, 875 LRTs, and 291 WSs) that yielded significant growth during the study period.

Bacterial identification and antibiotic susceptibility testing

Our laboratory team identified all organisms using established microbiological techniques. Antibiotic susceptibility testing was performed in accordance with Clinical and Laboratory Standards Institute (CLSI) guidelines. Data were collected on the types of pathogenic organisms, specimen sources, and antibiotic susceptibility patterns.

Antibiotics tested

These antibiotics were chosen based on their clinical relevance and frequency of use in our setting: ampicillin (AM), penicillin (P), amikacin (AK), amoxicillin (AMOX), amoxicillin/clavulanic acid (AUG), amphotericin B (AMB), ampicillin/sulbactam (A/S), aztreonam (AZT), caspofungin (CAS), cefazolin (CFZ), cefepime (CPM), cephalothin (CF), cefotaxime (CFT), cefoxitin (CFX), ceftazidime (CAZ), ceftazidime/avibactam MIC (CAZ/AVI), ceftolozane/tazobactam (CEF/TAZ), ceftriaxone (CAX), cefuroxime sodium (CRM), chloramphenicol (C), oxacillin (OX), ciprofloxacin (CP), clindamycin (CD), erythromycin (E), fosfomycin (FOS), colistin (CT), gentamicin (GM), levofloxacin (LV), linezolid (LEN), imipenem (IMP), meropenem (MEM), nitrofurantoin (FD), fusidic acid (FA), piperacillin/tazobactam (PI/T), moxifloxacin (MX), rifampin (RIF), ticarcillin/clavulanic acid (TIM), teicoplanin (TP), tetracycline (TE), tigecycline (TGC), tobramycin (TO), trimethoprim/sulfamethoxazole (T/S), sulfamethoxazole/trimethoprim (SXT), polymyxin B (PB), vancomycin (V), minocycline (MN), and norfloxacin (NXN).

Statistical analysis

Descriptive statistics were used for the antibiogram analysis. We calculated the percentages of resistance to each combination of organisms and antibiotics for each year, then summed the percentages across all years for both men and women. Time-based trends in antimicrobial resistance were evaluated using binary logistic regression models applied separately to each antibiotic, with resistance status coded as a binary outcome (0 = sensitive, 1 = resistant) and year treated as a continuous predictor. Odds ratios per year with corresponding confidence intervals were estimated to assess the direction and magnitude of resistance changes over time. To provide an overall comparison across antibiotics, forest plots were constructed summarizing the odds ratios derived from individual models. In addition, temporal changes in the distribution of bacterial organisms and infection sites were assessed using multinomial logistic regression, given the multi-category nature of these outcomes. Graphical representations were generated to illustrate resistance trends and distributional changes over the study period.

Ethical considerations

The study was approved by the Institutional Review Board. As this was a retrospective study using anonymized laboratory data, the need for individual patient consent was waived. The authors declare no conflicts of interest related to this study. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Results

Analysis of antimicrobial resistance trends from 2020 to 2024 revealed marked heterogeneity in time-based patterns across the tested antibiotics, as summarized in the forest plot (Figure 1) and the corresponding summary of odds ratios per year (Table 1).

Table 1. Summary of temporal resistance trends by antibiotic (2020–2024).

On the one hand, antibiotics showed a clear increase in resistance over the years (OR >1 and CIs entirely above 1), as follows: TGC (tigecycline), FOS (fosfomycin), CAZ-AVI (ceftazidime-avibactam), CP (cephalothin), CFX (cefoxitin), and AZT (aztreonam). These antibiotics showed a significant rise in resistance over the five-year study period. On the other hand, some antibiotics with an increasing trend (OR >1, but the CIs cross 1) are as follows: NXN (norfloxacin), CFT (ceftriaxone), TE (tetracycline), OX (oxacillin), CRM (clarithromycin), CT (colistin), E (erythromycin), and RIF (rifampicin). There is a signal toward increasing resistance, with no statistical certainty for this trend. Antibiotics showing a decreasing trend (OR <1) are as follows: P (penicillin), LV (levofloxacin), AK (amikacin), A_S (ampicillin–sulbactam), MX (moxifloxacin), and MN (minocycline). The decreasing trend of these antibiotics showed a pattern but was not statistically confirmed. The remaining antibiotics illustrated a stable trend with no significant changes (OR ≈ 1 and narrow CIs crossing 1).

These contrasting patterns were further evident in the individual time-trend plots, which showed variability in both the magnitude and direction of year-to-year resistance changes among antibiotics (Supplementary Figure S1-S37). In some cases, resistance proportions increased steadily across consecutive years, whereas in others, fluctuations were observed without a consistent directional trend.

In parallel with resistance trends, the composition of bacterial organisms contributing to clinical infections changed over time (Figure 2; Table 2). The figure illustrated that certain species became more prominent during the later years of the study period, particularly E. coli. While others demonstrated a decline in relative frequency (K. pneumoniae, P. aeruginosa and S. aureus), reflecting time-based shifts in the bacterial population structure over a five-year period of five years. Non-duplicate bacterial isolates from community-acquired infections were provided by a clinical diagnostic laboratory. The dataset comprised 1,540 E. coli, 614 K. pneumoniae, 172 Acinetobacter sp., 151 P. aeruginosa, and 105 S. aureus.

Table 2. Distribution of bacterial organisms by study year (%).

Similarly, the distribution of infection sites varied across the study years, as illustrated in Figure 3; Table 3, showing changes in the relative contribution of different specimen sources. The changes were illustrated by an increase in urinary tract infections, a dramatic decrease in lower respiratory tract infections (LRTs), and a slight decrease in blood and wound infections over the five-year period. Collectively, these findings demonstrated that antimicrobial resistance trends occurred alongside concurrent time-based variation in both organism and infection site distributions, underscoring the dynamic nature of the clinical and microbiological landscape over the study period.

Table 3. Distribution of infection sites by study year (%).

Discussion

Many pathogenic bacteria were studied in multiple sites, but only those bacteria that showed significant antibiotic resistance were documented in this study. Many antibiotics showed an increasing pattern over the five-year period. We categorized these antibiotics as β-lactams, which are cell wall synthesis inhibitors, and non-β-lactams. The β-lactams illustrated in this study with an increasing pattern were aztreonam (a monobactam used for Gram-negative bacteria), cefoxitin (a second-generation cephalosporin), ceftazidime-avibactam, and piperacillin-tazobactam (β-lactams with β-lactamase inhibitors). These antibiotics are broad-spectrum options, and some are prescribed for the treatment of complicated UTI and sepsis as intravenous (IV) agents and are considered last-line β-lactams with β-lactamase inhibitors (Hidalgo et al., 2016). The IV broad spectrum aztreonam and piperacillin-tazobactam are often used for the treatment of UTI, specifically in patients with severe β-lactam allergy (Kaye et al., 2022).

Distribution of ESBL-producing isolates

The study investigated 2,033 isolates, identifying that 29% (n = 607) were ESBL producers. In this ESBL context, a reserve agent used is ceftazidime-avibactam (Tamma et al., 2024). The particular concern is antibiotics often used for MDR infections, such as tigecycline resistance (Park et al., 2020). These drugs may be losing reliability and should prompt reevaluation of empirical use (Kadri et al., 2018).

Antibiotics in the current study that demonstrated an increasing but non-significant trend were as follows: norfloxacin, tetracycline, rifampicin, β-lactams (ceftriaxone, oxacillin), clarithromycin, erythromycin, and colistin. Resistance to these antibiotics may need continued surveillance, but there is no recent classification indicating an increase in resistance. On the other hand, the resistance patterns of many carbapenems and aminoglycosides stayed the same over time. Therefore, the emergence of resistance in the antibiotics showing an increasing trend suggests therapeutic exhausting. This has been reported globally with a similar trend (Reynolds et al., 2022).

Resistance patterns differed across species and infection sites, underscoring the importance of developing and implementing up-to-date, site-specific treatment guidelines and pathogen-targeted therapeutic strategies. The current study indicates the potential spread of carbapenems-producing E. coli. E. coli showed fluctuations in total numbers over the five years; overall ESBL rates have been increasing over this time interval. This aligns with the global literature reports particularly on urinary tract infections over the last decade (Cantas et al., 2016). The prevalence in some regions has increased from 44% to 59% over the last 10 years (Mączyńska et al., 2023).

Escalating resistance to “last resort” antibiotics

It is especially concerning that tigecycline and ceftazidime-avibactam are becoming less effective, since these are often used to treat Enterobacterales that are resistant to carbapenems (CRE). These agents are frequently considered “last-line” therapies against carbapenem-resistant Enterobacterales (CRE) (European Centre for Disease Prevention and Control (ECDC), 2025). Results from the current study indicate a statistically significant increasing trend for both antibiotics, with tigecycline exhibiting an odds ratio (OR) of 1.442 (p = 0.0002) and ceftazidime-avibactam demonstrating an OR of 1.395 (p = 0.0373). Global studies indicate that this phenomenon may be influenced by the heightened use of these agents during the COVID-19 pandemic and the ensuing selective pressure (U. S. Centers for Disease Control and Prevention (CDC), 2022).

Impact of COVID-19 and selective pressure

The time frame of the current study (2020–2024) and its correlation with the COVID-19 outbreak suggest a relationship between pandemic-associated factors and the development of antibiotic resistance. This is believed to have contributed to the development of resistance to several antibiotic agents (Wang et al., 2026). During the early waves of the outbreak, a large proportion of patients were administered broad-spectrum antibiotics empirically, despite there being no confirmed cases of bacterial co-infection. As a result, strong selective pressure was exerted, resulting in the proliferation of resistant strains.

It has been recently established by medical researchers that resistance to tigecycline and ceftazidime-avibactam may develop within one month under the selective pressure induced by β-lactam and TGC treatment. Non-susceptibility to tigecycline in some clinical settings increased from virtually 0% in 2019 to 71% in 2022, whereas resistance to CAZ-AVI increased from 27% to 70%. This may be attributed to the reallocation of healthcare resources and the suspension of antimicrobial stewardships (ASPs) due to the focus on addressing the consequences of the COVID-19 outbreak (Han et al., 2021).

Stability and declining resistance trends

Although an increase occurred in some regions, many antibiotics had stable or even declining resistance trends. There were stable trends for most common antibiotics, such as the carbapenems: imipenem (IMP) and meropenem (MEM), with no significant increases each year (OR ≈1). Resistance trends showed a significant decline for amikacin (AK) (OR 0.798, p <0.001) and trimethoprim-sulfamethoxazole (T_S) (OR 0.935, p = 0.0239). According to regional surveillance studies conducted in Saudi Arabia and China, amikacin shows very high susceptibility, sometimes below 5%, while resistance to drugs such as ceftriaxone and fluoroquinolones is increasing (Khalid et al., 2025).

In conclusion, this study highlights a pivotal transformation in the epidemiological framework of antimicrobial resistance (AMR), marked by the rising incidence of extended-spectrum β-lactamase (ESBL)-producing E. coli and a notable increase in resistance to “last-resort” antibiotics. According to the study findings, traditional classes of drugs, such as aminoglycosides and carbapenems, have remained relatively stable. However, resistance to newer agents, such as ceftazidime-avibactam and tigecycline, has increased rapidly. Such trends, particularly the year-on-year increases in resistance to ceftazidime-avibactam (OR 1.395) and tigecycline (OR 1.442), indicate an emerging state of “therapeutic exhaustion,” similar to the global trend of declining treatment alternatives for carbapenem-resistant Enterobacterales (CRE).

It becomes evident that the emergence of antibiotic resistance is linked to the COVID-19 pandemic through the selective pressure caused by the overprescription of broad-spectrum antibiotics (U. S. Centers for Disease Control and Prevention (CDC), 2022). As increasing resistance rates are observed among last-line antimicrobial agents, it becomes necessary to reconsider current empiric treatment approaches. The use of these agents without prior confirmation of susceptibility may become untenable, underscoring the need to revitalize ASPs to ensure the continued effectiveness of the existing antibiotic arsenal. An interesting variation in the population structure of the studied bacteria was identified in the current study, specifically an increase in the proportion of E. coli isolated from UTIs, whereas other strains, such as K. pneumoniae, were less represented among the collected samples. This information should be considered during the development of strategies for future studies and treatment in the region. Increasing resistance to antibiotics prescribed to treat complex infections represents a potential threat to public health. It is essential that this shifting disease burden be considered when planning for future regional testing and treatment programs. The emergence of drug resistance among these infections is a public health concern. This study represents a significant first step toward local and regional monitoring of such diseases. It emphasizes that, without strict intervention and careful antibiotic use, the medical community may soon have to deal with infections for which no viable therapeutic options exist.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/frabi.2026.1872911/full#supplementary-material

Amikacin resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Ampicillin resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Amoxicillin–clavulanic acid resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Cefuroxime resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Ceftazidime resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Ceftriaxone resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Cefazolin resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Cephalothin resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Cefepime resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Extended-spectrum β-lactamase phenotype resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Gentamicin resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Meropenem resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Trimethoprim–sulfamethoxazole resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Tigecycline resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Ceftazidime–avibactam resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Cephalothin resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Cefoxitin resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Imipenem resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Piperacillin–tazobactam resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Ampicillin–sulbactam resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Aztreonam resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Colistin resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Fusidic acid resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Fosfomycin resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Clindamycin resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Erythromycin resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Levofloxacin resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Moxifloxacin resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Oxacillin resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Penicillin resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Rifampicin resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Tetracycline resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Clarithromycin resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Norfloxacin resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Tobramycin resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Minocycline resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Ticarcillin–clavulanic acid resistance time-trend analysis derived from an individual binary logistic regression model (2020–2024). Supplementary material at the source

Footnotes

Edited by: Ahmed Hegazi, Ahmed G. Hegazi, Egypt

Reviewed by: Samreen Soomro, Dow University of Health Sciences, Pakistan