Prevalence and Resistance Patterns of Uropathogens in Critically Ill Patients at a Tertiary Care Hospital in Tehran: Implications for Antimicrobial Stewardship in Developing Countries

Abstract

Background:

Antimicrobial resistance is a critical global threat in resource-limited settings with underdeveloped laboratory capacity and stewardship programs. Intensive care unit (ICU) patients are at high risk for complicated urinary tract infections (cUTIs) caused by multidrug-resistant (MDR) uropathogens. Local resistance data are essential to guide empirical therapy and design effective stewardship interventions.

Methods:

We conducted a retrospective, cross-sectional study (March 2020–December 2022) of 127 adult ICU patients with cUTIs at a tertiary hospital in Tehran, Iran. Urine isolates were identified by standard phenotypic methods, and antimicrobial susceptibility testing (AST) was performed via disk diffusion following Clinical and Laboratory Standards Institute guidelines. Resistance phenotypes—extended-spectrum beta-lactamase (ESBL) production, carbapenem-resistant Enterobacteriaceae, vancomycin-resistant enterococci (VRE), difficult-to-treat Pseudomonas, and pan-drug-resistant (PDR) Acinetobacter baumannii—were defined using current breakpoints.

Results:

Escherichia coli (52.2%) and Klebsiella pneumoniae (26.9%) predominated. Among Enterobacterales, 60.4% produced ESBL and 30.2% were carbapenem resistant. VRE comprised all enterococcal isolates; PDR A. baumannii occurred in one case. No significant associations were found between resistance profiles and sepsis, septic shock, or mortality. Multivariable analysis identified heart failure (odds ratio [OR] 2.45; 95% confidence interval [CI] 1.15–5.21; p = 0.017) and longer ICU stay (OR 1.03 per day; 95% CI 1.01–1.05; p = 0.012) as independent predictors of MDR infection.

Conclusions:

We report an alarming burden of MDR uropathogens in Tehran ICUs, underscoring the need for tailored empirical-therapy guidelines, enhanced antimicrobial stewardship programs, and multicenter surveillance to curb resistance and improve patient outcomes.

Keywords

multidrug-resistant organisms (MDROs)uropathogen antibiotic resistance patterns critically ill nosocomial infections

Introduction

Antimicrobial resistance (AMR) is a growing global public health challenge, particularly in developing countries with limited health care infrastructure.¹ Health care-associated infections (HAIs) are a key cause of morbidity and mortality worldwide, having a particularly significant impact in low-income regions where systems for infection prevention and care are often lacking.² Urinary tract infections (UTIs) stand out among HAIs, representing approximately 28% of infections acquired in intensive care unit (ICU) settings.³ Patients under critical care are more susceptible to UTIs due to factors such as prolonged hospital stays, use of invasive medical devices like catheters, and comorbidities such as diabetes and kidney disease.⁴

The development of multidrug-resistant microorganisms (MDROs) has significantly impacted UTI management, particularly in ICU.⁴ Gram-negative pathogens, such as Escherichia coli (E. coli) and Klebsiella pneumoniae (K. pneumoniae), show high resistance to antibiotics, including beta-lactams, fluoroquinolones, and carbapenems.⁵ In Iran, a 2020 study reported a pooled prevalence of 24.0% for carbapenem-resistant K. pneumoniae and 5.0% for carbapenem-resistant E. coli, with blaOXA-48 as a common resistance gene.⁵ A 2024 study from Palestine, a region with comparable health care challenges, found 62.8% of E. coli and 18.6% of K. pneumoniae in ICUs were extended-spectrum beta-lactamase (ESBL)-producing, with 7.4% of E. coli and 11.2% of K. pneumoniae showing carbapenem resistance.⁶

In resource-limited settings, AMR is worsened by over-prescription and misuse of antibiotics, limited access to laboratory diagnostics, and inadequate infection control measures [2].

Although this study is restricted to a single center, it provides crucial information on AMR patterns in a resource-constrained environment, which is especially relevant for other developing nations facing similar health care challenges. Therefore, this study aimed to evaluate the prevalence and resistance profiles of uropathogens in critically ill patients admitted to the ICU of Sina Hospital, a tertiary care hospital in Tehran, Iran, to offer practical advice regarding antimicrobial stewardship programs (ASP) in resource-limited settings.

Materials and Methods

Study design and population

This retrospective, cross-sectional study was conducted between March 2020 and December 2022, at Sina Hospital, a university-affiliated medical complex in Tehran, Iran, and a part of Tehran University of Medical Sciences.

Participants were adult ICU patients (≥18 years) with positive urine cultures for catheter-associated urinary tract infections (cUTIs) and a minimum ICU stay of 48 hours and had symptoms. Exclusion criteria included negative cultures, incomplete medical records, ICU stays <24 hours, immunocompromised status, non-cUTIs, and preexisting UTIs at catheter insertion.

Definition of cUTIs

Symptomatic cUTI was defined according to Centers for Disease Control and Prevention (CDC)/National Healthcare Safety Network (NHSN) criteria⁷ as (a) fever ≥38°C or new-onset altered hemodynamic status (e.g., hypotension or vasopressor requirement) occurring ≥48 hours after catheter insertion and (b) a positive urine culture. Asymptomatic bacteriuria (colonization) was not included.

Ethical consideration

Ethical approval was obtained from the hospital’s ethics committee (IR.TUMS.SINAHOSPITAL.REC.1401.076) prior to data collection. Written consent for access to patient information was secured from the hospital’s research department. All collected data were fully anonymized to ensure patient confidentiality throughout the study process.

Sample size calculation

The sample size was calculated using a formula for descriptive-analytical studies based on a prevalence of 20% and a 95% confidence interval (CI), which resulted in at least 126 patients.⁶

Data collection

Data collection was performed in two phases: first, urine analysis results were collected from the laboratory of the hospital using admission codes for positive cultures of urine. In the second phase, information about the antibiotics received by patients was collected according to both initial and follow-up cultures’ results.

Data were taken from the hospital information system (HIS) and included details about the patients (like age, sex, and body mass index [BMI]), their health conditions (such as other illnesses, how long they stayed in the ICU, and whether they survived), and lab results (including urinalysis, cultures, and tests for how bacteria respond to antibiotics).

All antimicrobial agents administered during the current hospitalization—including those given on the ward prior to ICU transfer—were extracted from the HIS. Data on outpatient (prehospital) antibiotic regimens were not electronically recorded, and due to the retrospective design, patient or family interviews to recover these data were not feasible.

Antimicrobial susceptibility testing (AST) and definitions

AST was performed in the hospital laboratory using disk diffusion on Mueller–Hinton agar (for most isolates) or automated platforms, following Clinical and Laboratory Standards Institute (CLSI) M100 and European Committee on Antimicrobial Susceptibility Testing (EUCAST) guidelines.^8,9 Fosfomycin testing was only done for E. coli and E. faecalis according to CLSI standards; for E. faecium and other Enterobacterales, we checked their susceptibility using antibiotics that are recommended for these bacteria, like vancomycin for enterococci and carbapenems for Gram-negatives. High-level gentamicin testing (using a 120-µg disk) was done for Enterococcus species, where a result of 10 mm or more meant the bacteria were susceptible, and 6 mm or less meant they were resistant.

Testing followed CLSI/EUCAST guidelines. Resistance was defined as:^8,9 -

ESBL: Resistance to ceftazidime or cefotaxime plus clavulanic acid.

CRE: Carbapenem-resistant Enterobacterales (CLSI criteria).

DTR: Resistance to ≥1 agent in all tested first-line antibiotic classes.

MDR: Non-susceptible to ≥3 antimicrobial classes.

PDR: Resistance to all tested antimicrobials.

Table 1 lists antibiotics tested, interpretive breakpoints, and their relevance to resistance pattern definitions.

Table 1.

Antibiotics Tested, CLSI Breakpoints, and Resistance Pattern Relevance

Antibiotic	Susceptible	Intermediate	Resistant	Relevance to resistance patterns
Ceftriaxone	≥23 mm/≤1 µg/mL	20–22 mm/2 µg/mL	≤19 mm/≥4 µg/mL	ESBL screening: resistance suggests possible ESBL
Ceftazidime	≥21 mm/≤4 µg/mL	18–20 mm/8 µg/mL	≤17 mm/≥16 µg/mL	ESBL definition: resistance + clavulanic acid test confirms ESBL
Meropenem	≥23 mm/≤1 µg/mL	20–22 mm/2 µg/mL	≤19 mm/≥4 µg/mL	CRE definition: resistance defines carbapenem resistance
Imipenem–Relebactam	≤2 µg/mL	4 µg/mL	≥8 µg/mL	CRE/DTR: alternative test for carbapenem-resistant isolates
Amikacin	≥17 mm/≤16 µg/mL	13–16 mm/32 µg/mL	≤12 mm/≥64 µg/mL	DTR component: resistance contributes to difficult-to-treat status
Ciprofloxacin	≥21 mm/≤1 µg/mL	16–20 mm/2 µg/mL	≤15 mm/≥4 µg/mL	DTR definition: resistance to all fluoroquinolones required for DTR
Piperacillin–Tazobactam	≥18 mm/≤8/4 µg/mL	14–17 mm/16 µg/mL	≤13 mm/≥32 µg/mL	DTR component: resistance to antipseudomonal β-lactams
Gentamicin (HLR)	≥10 mm	7–9 mm	≤6 mm	VRE synergy: high-level resistance negates aminoglycoside-cell wall agent synergy

Resistance pattern definitions: ESBL: resistance to ceftazidime OR ceftriaxone AND ≥5 mm increase in zone diameter with clavulanic acid; CRE: resistance to meropenem OR imipenem (per CLSI criteria); DTR: resistance to ≥1 agent in ALL categories: (a) antipseudomonal β-lactams (piperacillin–tazobactam/ceftazidime), (b) fluoroquinolones (ciprofloxacin), (c) aminoglycosides (amikacin); MDR: non-susceptibility to ≥3 antimicrobial classes; PDR: resistance to all tested antimicrobial agents.

Testing specifications: ESBL confirmation requires clavulanate synergy testing per CLSI M100-35th edition; Gentamicin HLR testing applies only to Enterococcus spp. (120 µg disk); Imipenem–relebactam breakpoints for Enterobacterales only.

CLSI, Clinical and Laboratory Standards Institute; CRE, carbapenem-resistant Enterobacterales; DTR, difficult-to-treat resistance; ESBL, extended-spectrum β-lactamase; HLR, high-level resistance (gentamicin 120 µg disk); MDR, multidrug-resistant; OR, odds ratio; PDR, pan-drug resistant; VRE, vancomycin-resistant Enterococcus.

All isolates were confirmed at the species level by traditional phenotypic assays as performed routinely in our hospital. involved checking the appearance of colonies on MacConkey and blood agar, using Gram staining, and conducting a series of biochemical tests, such as triple sugar iron, citrate utilization, indole, urease, motility, and oxidase reactions.

Statistical analysis

Statistical analyses were performed using SPSS software version 26 and STATA software version 16. Baseline characteristics including demographic data (age, sex, BMI), comorbidities, and clinical outcomes, were summarized using descriptive statistics. Continuous variables were expressed as mean ± standard deviation, and categorical variables were expressed as frequencies and percentages.

Logistic regression analysis was used to evaluate the relationship between antibiotic resistance and demographic/clinical factors. Univariate logistic regression was conducted to ascertain potential predictors of AMR, encompassing age, sex, BMI, comorbidities, and duration of ICU stay. Variables demonstrating a p-value <0.20 in the univariate analysis were incorporated into the multivariable logistic regression model to account for potential confounding variables. Odds ratios (OR) followed by 95% CI were computed to evaluate the strength of associations.

Subgroup analyses were also conducted to evaluate the impact of specific comorbidities and ICU stay on antibiotic resistance. Sensitivity analyses were performed to evaluate the robustness of findings by analysis of a series of alternate models, including age adjustment, sex adjustment, and comorbidity adjustment.

To evaluate the reliability of relationships, consistency between these models was confirmed by cross-checking results.

A p-value of less than 0.05 was considered statistically significant for all analyses. To maintain the strength of the analysis, a two-tailed test was used in all the tests, and listwise deletion was used to manage cases of missing data.

Results

Baseline characteristics

As shown in Figure 1, a total of 1,415 ICU patients were screened, among whom 313 had urine analysis results during their ICU stay. Of the 313-urine analysis, 186 were excluded: 159 with a negative culture, 20 with an incomplete medical record, and 7 with an ICU stays <24 hours. There was a final analysis sample of 127 patients that resulted.

FIG. 1.

Strengthening the Reporting of OBservational studies in Epidemiology (STROBE) flow diagram of Study.

The demographic analysis showed that 46.61% of the patients were female, and 50.39% were male. The average age of the patients was 62.29 ± 17.85 years, with no statistically significant difference between the age of patients with positive and negative cultures (61.15 ± 17.89 vs. 66.52 ± 17.02, respectively, p = 0.165). Notably, the average duration from ICU admission to the occurrence of UTI was 14.06 ± 13.53 days. Common morbidities among participants were hypertension (HTN) (46.46%), diabetes mellitus (DM) (33.28%), and urinary disorder (25.20%). More data are summarized in Table 2.

Table 2.

Baseline Data of Participants

Variable	Negative cultureN (%) = 27 (21.26)	Positive cultureN (%) = 100 (78.74)	TotalN (%) = 127 (100)	p-Value
Sex; n (%)
Male	14 (51.85)	50 (50.0)	64 (50.39)	0.864
Female	13 (48.15)	50 (50.0)	63 (49.61)	0.864
Age; mean ± SD	66.52 ± 17.02	61.15 ± 17.89	62.29 ± 17.85	0.165
BMI; mean ± SD	24.23 ± 4.69	25.00 ± 5.21	24.84 ± 5.10	0.484
Underling disease
HTN	12 (44.44)	47 (47.0)	59 (46.46)	0.183
DM	11 (40.74)	30 (30.0)	41 (32.28)	0.290
BPH	2 (7.41)	9 (9.0)	11 (8.66)	0.794
Dementia	4 (14.81)	7 (7.0)	11 (8.66)	0.200
Parkinson	1 (3.70)	6 (6.0)	7 (5.51)	0.643
CVA	3 (11.11)	15 (15.0)	18 (14.17)	0.607
CKD	1 (3.70)	5 (5.0)	6 (4.72)	0.778
Hypothyroidism	1 (3.70)	10 (10.0)	11 (8.66)	0.302
Seizure	0 (0)	9 (9.0)	9 (7.09)	0.106
HF	0 (0)	18 (18.0)	18 (14.17)	0.017

Bold value indicates a statistically significant difference (p < 0.05) between groups.

BMI, body mass index; BPH, benign prostate hypertrophy; CKD, chronic kidney disease; CVA, cerebrovascular accident; DM, diabetes mellitus; HF, heart failure; HTN, hypertension; SD, standard deviation.

We first examined key exposures and clinical variables known to predispose to MDR urinary pathogens (Table 3). Antimicrobial pressure was substantial, with 29.1% of patients having received ≥3 days of third-generation cephalosporins and 11.0% a similar duration of fluoroquinolones prior to UTI onset. Median durations of catheterization (7 days, interquartile range [IQR] 5–10) and ICU stay before culture (6 days, IQR 4–9) were also consistent with high-risk settings. One-third of patients had DM, and 18.1% had a documented prior MDR infection.

Table 3.

Baseline Antimicrobial Exposures and Clinical Risk Factors for MDR Pathogen Isolation in Catheter-Associated Urinary Tract Infections

Risk factor	n (%) or median (IQR)
≥3 days 3-GC use prior UTI	29.1
≥3 days FQ use prior UTI	11.0
Median catheter days (IQR)	7 (5–10) days
Median ICU days prior to culture (IQR)	6 (4–9) days
Diabetes mellitus (DM)	32.3
Prior MDR infection	18.1

3-GC, third-generation cephalosporins; DM, diabetes mellitus; FQ, fluoroquinolones; ICU, intensive care unit; IQR, interquartile range; MDR, multidrug-resistant; UTI, urinary tract infection.

Common pathogens

Table 4 demonstrates that the most common bacterial pathogens found were E. coli (52.24%), K. pneumoniae (26.86%), and Enterococcus spp. (11.94%). A total of 33 cultures were positive for fungi, of which candida albicans (C. albicans) accounted for 51.51% of the fungal pathogens.

Table 4.

Common Pathogens of Urine Cultures

Pathogen	BacterialN (%) = 67 (67)	FungalN (%) = 33 (33)	p-Value
Bacterial			0.28
E. coli	35 (52.24)	—
K. pneumoniae	18 (26.86)	—
E. faecium	6 (8.95)	—
E. Faecalis	2 (2.98)
P. aeruginosa	3 (4.48)	—
A. baumannii	1 (1.49)	—
Fungal			0.72
Candida albicans	—	17 (51.51)
Candida non-albicans	—	16 (48.49)

E. coli, Escherichia coli; K. pneumoniae, Klebsiella pneumoniae; P. aeruginosa, Pseudomonas aeruginosa; A. baumannii, Acinetobacter baumannii; E. faecium, Enterococcus faecium; E. faecalis, Enterococcus faecalis.

Antibiotic resistance and susceptibility patterns

Tables 5 and 6 analyze the percentages of AMR and susceptibility profiles of major pathogens. The incidence of ESBL-producing pathogens was 60.37% (28 cases of E. coli and 7 cases of K. pneumoniae out of 53 Enterobacterales spp.), while carbapenem-resistant Enterobacteriaceae (CRE) was found in 30.18% (7 cases of E. coli and 11 cases of K. pneumoniae out of 53 Enterobacterales spp.) of the cases. All six samples of Enterococcus faecium (E. faecium) were identified as vancomycin-resistant enterococci (VRE) pathogens exhibiting resistance to vancomycin, but all E. faecalis were susceptible to vancomycin. In addition, the DTR-Pseudomonas pathogens, including three separate cases of Pseudomonas aeruginosa (P. aeruginosa), were documented as exhibiting complete resistance to treatments. Furthermore, Acinetobacter baumannii (A. baumannii), which was detected in one case, was resistant to all antimicrobial options and was considered PDR.

Table 5.

Antimicrobial Resistance Pattern

Pattern	N (%)
ESBL	32 (60.37)
CRE	16 (30.18)
VRE	6 (100)
DTR-P	3 (100)
A. baumannii (PDR)	1 (100)

A. baumannii, Acinetobacter baumannii; CRE, carbapenem-resistant Enterobacteriaceae; DTR-P, difficult-to-treat Pseudomonas; ESBL, extended-spectrum beta-lactamase; PDR, pan-drug resistant; VRE, vancomycin-resistant Enterococcus.

Table 6.

Antimicrobial Susceptibility Profiles of Major Uropathogens

Antimicrobial agentsusceptibility; n (%)	E. coli(n = 35)	Klebsiella pneumoniae(n = 18)	Enterococcus faecalis(n = 2)	Enterococcus faecium(n = 6)	Pseudomonas aeruginosa(n = 3)	Acinetobacter baumannii(n = 1)
Nitrofurantoin	22 (62.85)	0 (0)	N/A	N/A	N/A	N/A
Fosfomycin	5 (42.85)	1 (5.55)	1 (50.00)	1 (30.30)	N/A	N/A
TMP–CMX	4 (11.42)	4 (22.22)	N/A	N/A	N/A	N/A
Ciprofloxacin	4 (11.42)	5 (27.77)	N/A	N/A	0 (0)	0 (0)
Ceftriaxone	7 (20)	3 (16.67)	N/A	N/A	N/A	N/A
Piperacillin–Tazobactam	8 (22.85)	1 (5.55)	N/A	N/A	0 (0)	0 (0)
Imipenem	27 (77.14)	13 (72.22)	N/A	N/A	0 (0)	0 (0)
Amikacin	24 (68.57)	12 (66.67)	N/A	N/A	0 (0)	0 (0)
Gentamicin	6 (17.14)	2 (11.11)	0 (0)	0 (0)	0 (0)	0 (0)
Vancomycin	N/A	N/A	2 (100)	0 (0)	N/A	N/A

TMP–SMX, Trimethoprim–Sulfamethoxazole.

Clinical outcomes

According to Table 7, no significant correlation regarding the resistance patterns and clinical outcomes—sepsis, septic shock, or mortality—potentially was found, which may be due to small sample size, timely start of empirical therapy itself, or confounding variables such as comorbid state and interventions in the ICU.

Table 7.

Outcomes Based on Antimicrobial Resistance Pattern

Parameter	Antimicrobial resistance pattern N (%)
Parameter	ESBL32 (60.37)	CRE16 (30.18)	VRE6 (100)	DTR-P3 (100)	PDR1 (100)	p-Value
Sepsis	15 (46.87)	8 (50.0)	4 (66.67)	1 (33.33)	1 (100)	1
Septic shock	18 (56.25)	9 (56.25)	3 (50.0)	1 (33.33)	1 (100)	1
28-days mortality	19 (59.37)	7 (43.75)	3 (50.0)	3 (100)	1 (100)	1
Overall mortality	26 (81.25)	12 (75)	5 (83.33)	3 (100)	1 (100)	1
LOS	28.44 ± 14.89	39.93 ± 32.74	25.17 ± 9.94	49.33 ± 14.06	8	1

CRE, carbapenem-resistant Enterobacteriaceae; DTR-P, difficult-to-treat Pseudomonas; ESBL, extended-spectrum beta-lactamase; LOS, length of stay; PDR, pan-drug resistant; VRE, vancomycin-resistant Enterococcus.

Clinical and demographic effects on antimicrobial resistance odds

The logistic regression analysis identified potential associations of demographic/clinical factors with resistance: patients with HF had 2.45-fold increased odds for AMR, with an OR of 2.45 (95% CI: 1.15–5.20, p = 0.017)—hence, the relation was statistically significant. Each additional day of stay in the ICU increased the odds of AMR by 3% (OR = 1.03, 95% CI: 1.01–1.05, p = 0.012). For instance, increasing the stay in the ICU by 10 days corresponded to roughly 34% higher odds of resistance.

While ESBL, CRE, VRE, DTR-P, and PDR A. baumannii showed increased odds for resistance, these associations were not statistically strong, probably because of the small sample size. In addition, no significant association was observed between age, sex, HTN, and DM and AMR. More data were presented in Table 8.

Table 8.

Logistic Regression Analysis of Factors Associated with Antimicrobial Resistance

Variable	OR	95% CI	p-Value
Underlying factors
Age	1.01	0.99–1.06	0.165
Sex (male)	1.06	0.45–2.46	0.864
DM	1.13	0.48–2.64	0.183
HTN	0.62	0.26–1.53	0.290
HF	2.45	1.15–5.21	0.017
Nosocomial factors
ICU stay (days)	1.03	1.01–1.05	0.012
ESBL	1.51	0.65–3.45	0.320
CRE	1.19	0.50–2.91	0.680
VRE	1.82	0.70–4.59	0.210
DTR-P	2.09	0.85–5.21	0.110
PDR (A. baumannii)	3.53	0.95–12.88	0.060

Bold values indicate a statistically significant difference (p < 0.05) between groups.

CI, confidence interval; CRE, carbapenem-resistant Enterobacteriaceae; DM, diabetes mellitus; DTR-P, difficult-to-treat; ESBL, extended-spectrum beta-lactamase; HF, heart failure; HTN, hypertension; ICU, intensive care unit; OR, odds ratio; PDR, pan-drug resistant; VRE, vancomycin-resistant Enterococcus.

Subgroup and sensitivity analyses

To account for potential confounding factors, subgroup analyses were performed based on demographic and clinical variables, including sex, age, BMI, comorbidities, and length of ICU stay. Sensitivity analyses were also conducted to assess the robustness of our findings. Subgroup analyses showed that patients with heart failure (HF) had much higher chances of developing AMR (OR = 2.45, 95% CI: 1.15–5.21, p = 0.017), and for each extra day in the ICU, the chances of resistance increased by 3% (OR = 1.03, 95% CI: 1.01–1.05, p = 0.012) (Table 9). Sensitivity analyses confirmed the robustness of these findings, as the associations remained consistent across different models. No significant associations were observed between other variables (e.g., age, gender, BMI, HTN, and DM) and AMR patterns (Table 8). These results suggest that HF and prolonged ICU stays are key predictors of AMR in critically ill patients with cUTIs.

Table 9.

Subgroup Analyses of Demographic and Clinical Variables in Relation to Antimicrobial Resistance

Variable	OR	95% CI	p-Value
Demographics
Sex (male)	1.06	0.45–2.46	0.864
Age (per year)	1.01	0.99–1.06	0.165
BMI (per unit increase)	1.02	0.95–1.10	0.620
Comorbidities
HTN	0.62	0.26–1.53	0.290
DM	1.13	0.48–2.64	0.183
BPH	1.05	0.20–5.50	0.950
Dementia	0.85	0.25–2.90	0.790
Parkinson	1.50	0.30–7.50	0.620
CVA	1.25	0.50–3.12	0.630
CKD	1.10	0.30–4.05	0.890
Hypothyroidism	1.20	0.40–3.60	0.750
Seizure	1.80	0.50–6.50	0.370
HF	2.45	1.15–5.21	0.017
ICU stay (per day)	1.03	1.01–1.05	0.012

Bold values indicate a statistically significant difference (p < 0.05) between groups.

BMI, body mass index; BPH, benign prostate hypertrophy; CI, confidence interval; CKD, chronic kidney disease; CVA, cerebrovascular accident; DM, diabetes mellitus; HF, heart failure; HTN, hypertension; ICU, intensive care unit; OR, odds ratio.

Discussion

High resistance rates among gram-negative pathogens

Gram-negative bacteria made up most of the bacteria found in our ICU urine samples (91%), with E. coli being the most common at 52% and K. pneumoniae at 27%. Alarmingly, 60.4% of Enterobacterales were ESBL-producers and 30.2% were carbapenem resistant. These figures echo and, in some cases, exceed regional reports. For instance, a 2024 meta-analysis showed an overall 60% prevalence of ESBL-producing Enterobacteriaceae in Egypt.¹⁰ In Gulf Persian hospitals, ESBL rates among Enterobacterales have been recognized at ∼21% to 29% overall (and ∼25% to 32% in ICU UTIs).¹¹ Our ICU’s ESBL rate greatly exceeds these values, highlighting high selective pressure from antibiotic usage. Our ICU shows significantly higher rates, which point toward strong selective pressure from antibiotic usage.

Carbapenem resistance is also especially troubling: almost 78% of K. pneumoniae in our study were resistant to imipenem, whereas Gulf region reports have described relatively low CRE levels.¹¹ Dominant resistance genes in the region appear to include blaCTX-M, blaSHV, blaTEM, and carbapenemases blaOXA-48/blaNDM-1, which suggests common molecular mechanisms.¹¹ High ESBL/CRE levels overall may signal shrinking treatment options. Although carbapenems maintained between 72% and 77% activity against the population examined in our study, their effectiveness continues to decline. Guidelines now recommend novel β-lactam/β-lactamase inhibitor combinations (e.g., ceftazidime–avibactam) for CRE infections, and recent data suggest adding an aminoglycoside may further improve outcomes.¹² In practice, our findings support using such advanced therapies empirically in high-risk ICU UTI cases and strongly emphasize the need for ICU antibiograms to be updated frequently to guide such decisions.

Emergence of resistant gram-positive pathogens

Gram-positive, resistant uropathogens were also identified. All E. faecium isolates were vancomycin- and ampicillin-resistant, whereas E. faecalis isolates were fully susceptible. This high incidence of VRE is consistent with broader patterns; western Asia (including Iran) has the highest regional VRE burden in Asia, with a pooled prevalence of ∼11.4%.¹³ The near commonness of VRE in our setting emphasizes the high priority of strict contact precautions, routine screening, and antimicrobial stewardship focused on enterococci. Empirical therapy for severe patients should reflect this: For example, vancomycin or linezolid may be required where VRE is strongly suspected. Overall, our findings on Gram-positive resistance show that ICU treatment plans cannot just rely on older medications (like ampicillin) and must consider the local levels of VRE.

Fungal infections: An overlooked challenge

Fungal pathogens were also commonplace (33% positive cultures), specifically candida, with C. albicans predominating. Critically ill patients, especially catheterized patients on broad-spectrum antibiotics, are at high risk for both candiduria and invasive candidiasis. Our ICU’s high burden of C. albicans suggests antifungal stewardship is warranted. While antibiotics and antifungal stewardship are not often mentioned together, in the ICU antifungal stewardship programs describe themselves as “an essential part to ensure responsible use of antifungals”¹⁴ by not starting unnecessary antifungal therapy, utilizing rapid diagnostics (e.g., (1,3)-β-D-glucan (BDG) or T2 Magnetic Resonance (T2MR) for antifungal stewardship), and by de-escalating therapy if cultures are negative. These data provide a good reminder for the clinician to make sure to include fungal infection in differential diagnoses when an ICU UTI is being considered, especially in patients with diabetes or prior length of antibiotics, and certainly it is worth the effort to inquire with the Infectious Diseases (ID) teams regarding input into stewardship on antifungals.

Implications for antimicrobial therapy and stewardship

In line with this, the susceptibility profiles speak to important stewardship priorities. To begin with, empirical choices in the ICU must account for likely MDR pathogens. Given the high rates of ESBL, carbapenems to empirically treat a severe ICU UTI should be restricted and de-escalated whenever possible. Older oral medications like fosfomycin have shown weak effectiveness (less than 50% against E. coli and about 5% against K. pneumoniae), indicating that there is a growing resistance to fosfomycin among ESBL-producing bacteria.¹ As a result, fosfomycin is generally not appropriate for empirical ICU c-UTI treatment in this context. Second, ongoing surveillance is important: hospitals need to routinely update ICU antibiograms and distribute the information as needed to supporting clinicians. Third, combination therapy may be warranted with CRE. Recent evidence published in 2024, with Veterans Affairs hospitals assessed whether adding an aminoglycoside to ceftazidime–avibactam led to lower mortality in CRE infection; notably, prescribing combinations trended toward less mortality.¹² In our unit we will ensure that a pathway exists for access to these combinations, reviewing therapeutic options that include colistin or more recent additions (meropenem–vaborbactam, cefiderocol) when managing pan-resistant patients in consultation with an infectious disease specialist with an aim to provide treatment that is consistent with South African practice guidelines. By integrating these local resistance rates into ICU antibiograms, ASP teams can tailor empirical-therapy pathways, set unit-specific formulary restrictions (e.g., reserve ceftazidime–avibactam for confirmed CRE), and prioritize audit-and-feedback on the highest-risk agents. Such data-driven stewardship ensures more judicious use of broad-spectrum agents and timely de-escalation.¹⁵

Lastly, systemic stewardship issues also need to be improved. We have demonstrated the opportunity for improving ASPs in the ICU. In terms of ASP generally, it is crucial that measures consistently include prescribing and review requirements, dose optimization, and de-escalation protocols, as well as clinician education. Importantly, an ASP in Saudi hospitals achieved significant implications on antimicrobial stewardship practices: their multidisciplinary initiatives (educational sessions, audits, IV-to-PO switch, and formulary restrictions initiated by educational sessions) led to a decrease in overall antimicrobial consumption and a reduction in cases with MDRO.¹⁶

Undoubtedly, establishing a similar ASP would reduce the selection pressure that contributes to our considerable rates of MDR organisms. Infection control is just as important: simple steps like glove use, hand hygiene, patient isolation, and timely catheter removal are known to prevent or reduce transmission of resistance.¹⁷ Even in resource-limited ICUs, low-cost interventions, for example, adhering to strict contact precautions, disinfection of the unit on a routine basis, and staff training, can have a huge impact. In conclusion, combining ICU ASPs with infection control practices and continuing feedback to prescribers is key to preventing resistant uropathogens in our context.

Clinical outcomes and risk factors

Interestingly, we did not demonstrate any significant difference in mortality, septic shock, or other outcomes between MDR and susceptible infections. There is literature that has shown worse outcomes with MDR infections in the ICU setting; however, the absence of an association in our study may reflect our protocol of providing early appropriate empirical therapy or residual confounding from underlying disease. For example, it has been reported that in patients with cancer, mortality due to MDR Enterobacter bloodstream infection was comparable with non-ESBL infections. Nonetheless, this reiterates the need for larger, prospective studies to better define the effect of resistance on the outcomes of ICU UTIs in our population.

Our analysis found HF and longer stays in ICU to be independent predictors of MDR UTI. These results make clinical sense. Patients with chronic HF are expected to have repeated unplanned hospital and ICU admissions, implantable devices, and be immunosuppressed, which increases their exposure to hospital flora. In addition, we know that prolonged ICU exposure is a known risk factor for acquisition of MDR gram-positive and gram-negative pathogens.¹⁸ These associations suggest that shorter ICU stay lengths (through early mobilization protocols and early weaning protocols) and careful use of devices in the ICU (such as limiting the use of catheters and IV lines) may mitigate some of the risk for acquiring MDR pathogens. While other factors (like age and DM) were not significant in our analysis, they were not a confounder in our cohort initially, likely because of our sample size and/or local epidemiology, but should be evaluated in future work.

Study limitations

Beyond the limitations noted above, this work has several important constraints. First, as a single-center, retrospective review, selection bias is inevitable—clinicians may preferentially culture the sickest patients. Second, we did not perform molecular typing or resistance-gene testing, so we cannot link isolates strain-to-strain or track transmission pathways. Third, although we captured all in-hospital and pre-ICU antibiotic exposures via the HIS, data on outpatient (pre-hospital) regimens were not recorded, and the retrospective design precluded patient/family-based recall—potentially introducing unmeasured confounding in resistance analyses. Fourth, we defined symptomatic cUTI as fever (≥38°C) or new-onset hemodynamic instability plus a positive catheterized urine culture—thereby excluding asymptomatic bacteriuria—but reliance on clinical documentation and culture results may still obscure early colonization versus true infection. Fifth, including only patients with positive cultures may overestimate the true prevalence of resistance. Lastly, follow-up ended at ICU discharge, so we did not capture late recurrences or longer-term outcomes. Despite these limitations, our data offer a comprehensive view of ICU urinary-pathogen resistance patterns in Tehran and serve as valuable input for antimicrobial stewardship initiatives.

Conclusions and stewardship priorities

In summary, our ICU faces a very serious problem with ESBL-producing and carbapenem-resistant Enterobacterales and vancomycin-resistant E. faecium. This scenario needs action. Empirical therapy should be revised (with preference for advanced beta-lactamase inhibitors, combination therapy, or beta-lactam-based therapy with VRE coverage as appropriate) and should still not overlook antifungal coverage in appropriate scenarios. More importantly, this situation requires full engagement in antimicrobial stewardship and infection control. A recent review found that starting the right antimicrobial treatment quickly, using infection control methods, and having good stewardship programs can reduce the rise of AMR and lead to better patient outcomes.¹⁷ The present study supports that message; ICU teams in developing countries, where situations may indeed be more dire, need to increase surveillance, stewardship education, and prevention now to combat the potential threat of MDR.

Authors’ Contributions

P.K.: Conceptualization, methodology, and writing—reviewing and editing; S.R.: draft preparation; F.A.M.: Writing—original draft preparation; S.H.: Writing—original draft preparation; P.P.: Writing—original draft preparation; R.M.: Data curation, software, and writing—original draft preparation; M.Q.: Data curation, and writing—original draft preparation; H.S.; M.M.: Supervision; F.N.: Conceptualization, methodology, writing—reviewing and editing, and supervision.

Footnotes

Acknowledgments

Language editing was performed using QuillBot® to improve readability; all scientific content was author-controlled.

Funding Information

No funding was received for this article.

Confirmation Statement

Each author confirms that their research is supported by an institution that is primarily involved in education or research.

Data Availability

De-identified data generated and analyzed during this study are available from the corresponding author upon reasonable request.

Disclosure Statement

All authors declare that they did not have any financial or non-financial conflict of interest during the conduct of this study.

References

World Health Organization. Antimicrobial resistance. [Internet]. 2023 [cited 2025 Jun 29]. Available from: https://www.who.int/news-room/fact-sheets/detail/antimicrobial-resistance

World Health Organization. Global antimicrobial resistance and use surveillance system (GLASS) report: 2022. [Internet]. 2022 [cited 2025 Jun 29]. Available from: https://pmc.ncbi.nlm.nih.gov/articles/PMC10843100/

Ling

, Ching

, Apisarnthanarak

, et al. APSIC guide for prevention of catheter associated urinary tract infections (CAUTIs). Antimicrob Resist Infect Control, 2023; 12(1):52; doi: 10.1186/s13756-023-01254-8

Arshad

, Toori

, Rahim

. Differential profiles of intensive care unit multidrug-resistant patients: Influence of prior antibiotic therapy on clinical features. Pak J Med Sci, 2025; 41(3):706–711; doi: 10.12669/pjms.41.3.10392

Haeili

, Barmar

, Ghodousi

, et al. Prevalence and mechanisms of carbapenem resistance in enterobacteriaceae in Iran: A systematic review and meta-analysis. Microb Drug Resist, 2020; 26(12):1491–1501; doi: 10.1089/mdr.2020.0130

Al Laham

, Al-Sadi

, Badran

, et al. Epidemiology and clinical characteristics of patients with healthcare-acquired multidrug-resistant Gram-negative bacilli: A retrospective study from a tertiary care hospital. Sci Rep, 2024; 14:53596; doi: 10.1038/s41598-024-53596-x

Centers for Disease Control and Prevention. Urinary tract infection (Catheter-associated urinary tract infection [CAUTI] and non-catheter-associated urinary tract infection [UTI]) events [Internet]. Atlanta, GA: CDC; January 2024 [cited 2025 Jul 13]. Available from: https://www.cdc.gov/nhsn/pdfs/pscmanual/7psccauticurrent.pdf

The European Committee on Antimicrobial Susceptibility Testing. Breakpoint tables for interpretation of MICs and zone diameters. Version 12.0, 2022. Available from: http://www.eucast.org

Clinical and Laboratory Standards Institute (CLSI). Performance standards for antimicrobial susceptibility testing. 31st ed. CLSI supplement M100. Wayne, PA: CLSI; 2021.

10.

Azzam

, Khaled

, Samer

, Nageeb

. Prevalence and molecular characterization of ESBL-producing Enterobacteriaceae in Egypt: A systematic review and meta-analysis of hospital and community-acquired infections. Antimicrob Resist Infect Control, 2024; 13(1):145; doi: 10.1186/s13756-024-01497-z

11.

Hadi

, Al-Hail

, Aboidris

, et al. Prevalence and genetic characterization of clinically relevant extended-spectrum β-lactamase-producing Enterobacterales in the Gulf Cooperation Council countries. Front Antibiot, 2023; 2:1177954; doi: 10.3389/frabi.2023.1177954

12.

Bulman

, Cao

, Curry

, et al. Ceftazidime/avibactam alone or in combination with an aminoglycoside for treatment of carbapenem-resistant Enterobacterales infections: A retrospective cohort study. Int J Antimicrob Agents, 2024; 64(5):107321; doi: 10.1016/j.ijantimicag.2024.107321

13.

Iqbal F, Alocious A, Joy S, Stanly E, Rajesh V, Unnikrishnan M, et al. Vancomycin-resistant enterococci: A rising challenge to global health. Clin Epidemiol Glob Health. 2024;28:101663; doi: 10.1016/j.cegh.2024.101663

14.

Vazquez

, Whitaker

, Zubovskaia

. Invasive candidiasis in the intensive care unit: Where are we now? J Fungi (Basel), 2025; 11(4):258; doi: 10.3390/jof11040258

15.

Al-Omari

, Al Mutair

, Alhumaid

, et al. The impact of antimicrobial stewardship program implementation at four tertiary private hospitals: Results of a five-years pre-post analysis. Antimicrob Resist Infect Control, 2020; 9(1); doi: 10.1186/s13756-020-00751-4

16.

Alshehri

, Aldali

, Abdelhamid

, et al. Implementation of antimicrobial stewardship programs in Saudi Arabia: A systematic review. Microorganisms, 2025; 13(2):440; doi: 10.3390/microorganisms13020440

17.

Salarvan

, Abdollahi

, Afarinesh Khaki

, et al. Antibiotic resistance pattern in intensive care units in a large referral hospital in Iran. Iran J Pathol, 2023; 18(4):433–438; doi: 10.30699/IJP.2023.1990807.3073

18.

Itani

, Khojah

HMJ

, Karout

, et al. Acinetobacter baumannii: Assessing susceptibility patterns, management practices, and mortality predictors in a tertiary teaching hospital in Lebanon. Antimicrob Resist Infect Control, 2023; 12(1):136; doi: 10.1186/s13756-023-01343-8