Antimicrobial resistance patterns and molecular characterization of Acinetobacter baumannii isolates in non-COVID-19 patients admitted to intensive care units during the pandemic: a retrospective study at a tertiary hospital in Tehran, Iran

Abstract

Background

Acinetobacter baumannii is one of the most common causes of healthcare-acquired infections, particularly among the critically ill patients in intensive care units (ICUs). Its multidrug‐resistance (MDR) nature is a major factor contributing to treatment difficulties, and ultimately increased patient mortality. This study aimed to investigate the prevalence, antimicrobial resistance patterns, and molecular characteristics of A. baumannii strains isolated from ICU patients (non-COVID-19) in Tehran, Iran.

Results

The infection rate of A. baumannii among ICU patients was 2.25% (75/3340). More than 90% of isolates showed resistance to key antibiotics including imipenem, meropenem, cefotaxime, and ciprofloxacin. Colistin was the most effective drug, with a susceptibility rate of 61.3%. The majority of the isolates (92%) were categorized as extensively drug-resistant (XDR) and harbored multiple antibiotic resistance genes (ARGs). Among carbapenem resistance genes, bla_OXA-23-like was the most prevalent, detected in 72% (54/75) of isolates, followed by bla_OXA-24-like in 49.3% (37/75). Regarding aminoglycoside resistance, the genes aac(6')-Ib and ant(2')-Ia were identified in 66.6% (50/75) and 32% (24/75) of the isolates, respectively. Genotyping by multiplex PCR revealed that most of isolates (85.3%) belonged to GC2. Furthermore, REP-PCR genotyping identified eight different genotypes, with one dominant genotype accounting for 81% (51/63) of isolates.

Conclusion

The present study demonstrates the predominance of XDR A. baumannii strains carrying diverse ARGs in our ICU setting. These findings raise concern, as such strains complicate treatment, limit therapeutic options, and may contribute to poor outcomes in critically ill patients. Mortality was high and appeared more associated with extensive resistance than with comorbidities. The clonal dissemination of these strains further emphasizes the need for robust infection control measures. Implementation of strict infection control measures and robust antibiotic stewardship programs remains essential to limit the spread of these highly resistant pathogens.

Introduction

Acinetobacter baumannii, an opportunistic pathogen and member of the ESKAPE group (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter spp.), has rapidly emerged as a major global concern in healthcare settings [1]. It is a leading cause of severe healthcare-acquired infections, including pneumonia, bacteremia, and wound and urinary tract infections, particularly in intensive care units (ICUs) [2, 3]. A. baumannii is especially prevalent in lower- and middle-income countries, where ICU infections account for up to 20.9% of nosocomial cases and as much as 42% of ventilator-associated pneumonia cases [4]. In recent years, clinical isolates have shown rising resistance to a wide spectrum of antibiotics, including cephalosporins, aminoglycosides, fluoroquinolones, and carbapenems. Carbapenem-resistant A. baumannii (CRAB poses a particularly severe challenge due to its extensive resistance profile and has been designated a critical priority pathogen by the World Health Organization [5]. The pathogen's remarkable ability to persist and spread in hospital environments, including on mucosal surfaces and healthcare equipment, facilitates its transmission, especially in ICU settings and among healthcare personnel [6]. Infections caused by multidrug-resistant (MDR) A. baumannii, particularly CRAB associated with prolonged hospital stays, elevated mortality and morbidity rates especially in ICU patients and a significant economic burden on healthcare systems [7]. Consequently, considerable attention has been directed toward understanding its outbreak potential and its escalating resistance to antibiotics [8]. Addressing these challenges requires rapid diagnostic tools, novel effective antimicrobials, and stringent infection control practices. Moreover, molecular typing of clinical A. baumannii isolates is essential for identifying the sources and transmission pathways of epidemic strains. Although the molecular epidemiology of A. baumannii in Iran has been widely studied, limited data exist on isolates recovered specifically from non-COVID ICU patients during the pandemic period. To address this gap, we aimed to characterize the phenotypic and genotypic features of A. baumannii isolates in a tertiary care hospital in Tehran.

Results

Incidence of A. baumannii infections in patients admitted to ICU

During the study period (April—November 2021), 3,340 patients were admitted to ICUs, of whom 75 (2.25%) were infected with A. baumannii. All isolates were confirmed as A. baumannii through amplification of the gltA gene. The respiratory tract was the most common site of infection, 66.6% (50/75), followed by 18.6% (14/74) from blood, 8.2% (6/75) from urine and Central Venous Catheter (CVC), and 6.6% (5/75) from wound samples. The majority of A. baumannii isolates were obtained from male patients (45/75). Patient ages ranged from 1 to 89 years, with a mean age of 58.3 years. Comorbidities were present in 92% of the patient population, with hypertension (30.4%), autoimmune diseases (26%), and cardiovascular diseases (23.1%) being the most prevalent conditions. Among the 75 patients with A. baumannii infections, 63 (84%) died during hospitalization, with a median ICU stay of 15.5 days (Table 1). To further assess factors associated with patient outcomes, demographic characteristics, comorbidities, antimicrobial resistance profiles, resistance genes, and global clone distribution were compared between expired and discharged patients (Table 2). The analysis showed that comorbidities were not significantly associated with outcome (p > 0.05), consistent with the univariate analysis. However, resistance to several key antibiotics, including imipenem, meropenem, ciprofloxacin, cefotaxime, and piperacillin-tazobactam, was significantly associated with mortality (p < 0.05). Among resistance genes, only the aminoglycoside-modifying enzyme gene aac(6′)-Ib demonstrated a significant association with fatal outcome. All patients received meropenem as part of their treatment regimen, and 86.6% also received colistin in combination therapy.

Table 1.

Demographic characteristics and laboratory findings of patients (n = 75) with A. baumannii infections admitted in ICU

Parameters		(Mean ± SD)
Age		58.43 ± 18.97
Duration of Hospital Stay		15.5 ± 17.4
Laboratory findings on admission	CRP (mg/L)	110 ± 42.6
	ESR	89 ± 36.3
	WBC	11800 ± 19591
	Platelet count	164 ± 102
	Parameters	n (%)
Gender	Male	45 (60)
	Female	30 (40)
Outcome	Expired	63 (84)
	Discharged	12 (16)
Sample source	Respiratory tract samples	50 (66.6)
	Blood	14 (18.6)
	Wound	5 (6.6)
	Other	6 (8.2)
Comorbidities	Hypertension	21 (30.4)
	Autoimmunity	18 (26)
	Cardiovascular disease	16 (23.1)
	Diabetes	15 (21.7)
	Cancer	15 (21.7)
	addiction	6 (8.7)
	Chronic kidney disease	6 (8.7)
	Chronic respiratory disease	5 (7.2)
	Trauma	4 (5.8)

Table 2.

Comparison of demographic, clinical characteristics, antibiotic resistance, resistance genes, and global clones between expired and discharged ICU patients infected with A. baumannii

Variable		Expired	Discharged		p value
		n (%)
Gender	Male	37 (49.3)	4 (5.3)	0.818
	Female	26 (34.6)	8 (10.6)
Comorbidities	Hypertension	18 (24)	3 (4)		0.801
	Autoimmunity	17 (22.7)	1 (1.3)		0.166
	Cardiovascular disease	15 (20)	1 (1.3)		0.230
	Diabetes	13 (17.3)	2 (2.6)		0.753
	Cancer	15 (20)	0		0.059
	addiction	7 (9.3)	0		0.225
	Chronic kidney disease	6 (8)	0		0.265
	Chronic respiratory disease	5 (6.67)	0		0.312
	Trauma	3 (4)	1 (1.33)		0.614
Antibiotic resistance	Imipenem	63 (84)	11 (14.6)		0.021
	Meropenem	63 (84)	11 (14.6)		0.021
	Ciprofloxacin	62 (82.6)	10 (13.3)		0.015
	Gentamicin	60 (80)	10 (13.3)		0.130
	Amikacin	61 (81.3)	10 (13.3)		0.057
	Ceftazidime	58 (77.3)	9 (12)		0.079
	Cefotaxime	62 (82.6)	10 (13.3)		0.015
	Cefepime	61 (81.3)	11 (14.6)		0.403
	Ampicillin-sulbactam	54 (72)	11 (14.6)		0.578
	Piperacillin-tazobactam	62 (82.6)	10 (13.3)		0.015
	Trimethoprim-sulfamethoxazole	54 (72)	10 (13.3)		0.831
	Colistin	25 (33.3)	4 (5.3)		0.679
Antibiotic resistance genes	blaOXA-23-like	47 (62.6)	7 (9.3)		0.250
	blaOXA-24-like	31 (41.3)	6 (8)		0.960
	blaOXA-58-like	3 (4)	2 (2.6)		0.130
	blaNDM	16 (21.3)	2 (2.6)		0.516
	blaVIM-1	9 (12)	2 (2.6)		0.831
	blaTEM	29 (38.6)	4 (5.3)		0.417
	blaKPC	4 (5.3)	0		0.370
	aac(6′)-Ib	45 (60)	5 (6.6)		0.045
	ant(2″)-Ia	23 (30.6)	1 (1.3)		0.055
	aph(3′)-Ia	20 (26.6)	3 (4)		0.642
	aac(3)-Ia	1 (1.3)	0		0.660
Global clones	1	53 (70.6)	11 (14.6)		0.769
	2	9 (12)	1 (1.3)
	3	1 (1.3)	0

Antimicrobial susceptibility tests

Resistance rates were extremely high, with imipenem and meropenem showing the highest (98.7%). Piperacillin-tazobactam, cefotaxime, cefepime, and ciprofloxacin each exhibited 96% resistance, followed by amikacin (94.7%), gentamicin (93.3%), and ceftazidime (89.3%). Ampicillin-sulbactam and trimethoprim-sulfamethoxazole also showed high resistance (86.7% and 85.3%, respectively (Fig. 1). In contrast, colistin remained the most effective drug, with 61.3% of isolates susceptible. Overall, the majority of isolates (92%) were classified as extensively drug-resistant (XDR), while five (7%) exhibited multidrug-resistant (MDR) phenotypes. Notably, one isolate (AB117) was resistant exclusively to colistin but susceptible to all other tested antibiotics; according to standard definitions, this isolate does not meet MDR or XDR criteria and was therefore categorized as non-MDR/XDR.

Fig. 1.

The demographic, molecular characteristics and antimicrobial resistance profile of 75 Acinetobacter baumannii isolates obtained from patients admitted in ICU

Detection of antimicrobial resistance genes (ARGs)

The overall prevalence ARGs ranged from 0% (bla_IMP) to 72% (bla_OXA-23-like) (Fig. 1). More than half of the isolates carried bla_OXA-23-like, bla_TEM, and aac(6′)-Ib genes. bla_OXA-24-like, bla_NDM, and bla_VIM-1 were also frequently detected, whereas bla_OXA-58-like, bla_KPC, and aac(3)-Ia were less common. Additionally, the aminoglycoside-modifying enzyme genes ant(2″)-Ia and aph(3′)-Ia were present in nearly one-third of the isolates.

GC lineages

Multiplex PCR for global clone (GC) identification showed that 85.3% (64/75) of A. baumannii isolates belonged to GC2, 13.3% (10/75) to GC1, and 1.3% (1/75) to GC3 (Fig. 1).

REP-PCR genotyping

REP-PCR analysis generated between 4 and 9 bands per isolate, with fragment sizes ranging from 200 base pairs to 1.3 kilobases. Using a similarity cutoff of ≥ 80%, eight distinct genotypes (designated A–H), were identified. Twelve isolates did not yield detectable bands and were therefore considered non-typeable. Among the 63 typeable isolates, genotype G predominated, accounting for 81% (51/63). Notably, four genotypes (B, C, F, and H) were represented by single isolates only. Further analysis of genotype G revealed significant heterogeneity in both ARGs and phenotypic resistance profiles. In addition, 90% (46/51) of genotype G isolates exhibited XDR phenotype. The REP-PCR dendrogram depicting the genetic relationships among the 63 typeable A. baumannii isolates is shown in Fig. 2.

Fig. 2.

Dendrogram showing the genetic relatedness of 63 typeable strains of Acinetobacter baumannii determined by REP-PCR analysis using the Dice similarity coefficient. The vertical line displays the 80% similarity cut-off value. Based on a similarity index ≥ 80%, 8 genotypes were found. Each genotype were labelled A to H. Numbers at the terminal branches are strain names

Discussion

In this retrospective observational cohort study, the prevalence of A. baumannii infections in patients admitted to ICU was 2.25%, which was lower than other reports from Saudi Arabia (6.2%) [9], Nigeria (7%) [10], Morocco (8.4%) [11], Kenya (22.7%) [12] and Poland (31%) [13]. This variation in reports from different regions might be associated with several factors including: the dissimilarities in healthcare systems and infection control practices [14], study design and methodology [15], patient demographics [16], diagnostic methods and surveillance [17] and regional epidemiology [18]. Based on the demographic data of patients in present study, the number of male patients infected with A. baumannii in ICU was higher than the number of female patients. Our finding is in accordance with other studies [19–22]. This may be attributed to immune responses because testosterone suppresses immune defenses in males, while estradiol enhances pro-inflammatory responses in females [9]. In present study, the mortality rate among patients infected with A. baumannii was 84%, which is higher than that reported rates from China (34.4%) [23], Ecuador (35%) [24], Turkey (58.5%) [25] and Egypt (50%) [26]. In the statistical analysis of outcome-associated factors (Table 2), comorbidities, although highly prevalent among the patients, did not demonstrate a significant association with mortality. Instead, resistance to several critical antibiotics including carbapenems, ciprofloxacin, cefotaxime, and piperacillin-tazobactam was associated with fatal outcomes. This finding suggests that limited therapeutic options and treatment failure due to extensive resistance may contribute more substantially to mortality than underlying health conditions. Notably, the aminoglycoside-modifying enzyme gene aac(6′)-Ib also showed a significant association with mortality, highlighting the clinical relevance of genotypic resistance determinants in predicting poor outcomes. In this study, the median length of stay in the ICU was lower (15.5 days) compared with other study from Turkey (21 days) [20] but higher than a report from Lebanon (12.2 days) [27]. The difference in the median length of stay might be related to several factors, including differences in healthcare systems, the prevalence and resistance patterns of A. baumannii, treatment protocols, and patient demographics [28, 29]. In the present study, the majority of A. baumannii isolates were highly resistant to different antibiotic class including carbapenems, cephalosporins aminoglycosides, quinolones and 92% of the isolates were XDR. The high rate of antibiotic resistance observed in our hospital may be attributed to the overuse or misuse of antibiotics, which exerts selection pressure leading to the development of resistance [30]. Colistin is often regarded as one of the most effective antimicrobials against XDR A. baumannii infections [31]. Our study revealed that 38.7% A. baumannii isolates were colistin-resistant. Similar findings have been reported by Gerson et al. in Germany (48%) [32] and Papathanakos et al.in Greece (41%) [33]; however, a significantly higher resistance rate was reported by Al-Kadmy et al. in Iraq (76%) [34] and by Rahimzadeh et al. in Iran (84.2%) [35]. Lower rate of resistance (4.3%) was reported in Serbia [36] and Iran (0%) [37]. The high prevalence of resistance to colistin in the current study may be attributed to inappropriate use of bactericidal antibiotics in ICU and inadequate infection control practices in our hospital [38]. Despite high resistance rate to colistin, no isolates in the current study contained the mcr-1 gene, suggesting that other resistance mechanisms (mutations in the PmrAB system and loss of lipopolysaccharide (LPS) production) may be responsible for colistin resistance, which were not investigated in current study [39, 40]. In this study, a range of ARGs were detected among A. baumannii isolates. The most commonly detected resistance genes among the isolates were the class D carbapenemase genes bla_OXA-23-like (72%) and bla_OXA-24-like (49.3%), the AME gene aac(6’)-Ib (66.6%), the class A beta-lactamase gene bla_TEM (44%) and the MBL gene bla_NDM (24%). The distribution of ARGs varies significantly across geographical regions. For instance, a study conducted in Egypt by Hassan et al. reported the presence of OXA_−23-like (77.7%), bla_NDM (11.7%), bla_KPC (10.7%), and bla_OXA-58-like (1.9%) among ICU-derived A. baumannii isolates [41]. Similarly, a study from Pakistan found high prevalence rates for bla_OXA-23-like (73%), bla_NDM (92.2%); however, bla_OXA-24-like and bla_OXA-58-like genes were not detected in that study [42]. Studies from Iran have reported varying prevalence rates of carbapenemase genes in A. baumannii. Vahhabi et al. identified bla_OXA-23-like (82.1%), bla_OXA-24-like (36.6%), bla_NDM (6.2%), and bla_IMP (4.4%) [43]. Farajzadeh et al. found bla_OXA-23-like (65.7%), bla_VIM (31.4%), and bla_IMP (25.7%) [44]. Hashemizadeh et al. reported bla_OXA-24-like (55.3%), bla_OXA-23-like (41.7%), bla_VIM (41.7%), and bla_KPC (0.6%) [45]. Azimi et al. detected bla_OXA-23-like (76.5%), bla_OXA-24-like (65.8%), bla_NDM (9.4%), bla_VIM (2%), and bla_IMP (0.3%), while bla_OXA-58-like was not detected [46]. In Kuwait, the predominant carbapenemase genes identified were bla_OXA-23-like (85.1%), bla_IMP (13%), and bla_OXA-58-like (0.7%), while bla_NDM, bla_KPC, and bla_VIM were not found in any of the isolates studied [47]. In a study from Jordan, the most prevalent ARGs were bla_OXA-23-like (96.7%), bla_VIM (56.8%) and bla_NDM (7.4%) [48]. Other study from Saudi Arabia, bla_OXA-23-like, bla_IMP, bla_OXA-24-like and bla_TEM were the most prevalent genes, in which detected in 100%, 76.1%, 64.1%, 25.3% of isolates, respectively [49]. As noted earlier, the prevalence of ARGs exhibited substantial variation across different countries. These discrepancies may be attributed to horizontal transmission of resistance determinants, diverse patterns of antimicrobial prescription, the dissemination of particular clones carrying distinct types of ARGs and varying numbers of isolates tested [50]. Overall, β-lactamase genes were identified in 98.6% of the isolates. In the single isolate lacking these genes, β-lactam resistance may be attributed to alternative mechanisms, such as the activation of efflux pumps or reduced outer membrane permeability, which were not investigated in this study. In line with previous reports from Iran and the region, our study confirms GC2 as the prevailing clone responsible for XDR A. baumannii infections. What our data add is a snapshot of the clonal structure and resistance gene distribution specifically among non-COVID ICU patients during the pandemic, thereby complementing earlier national studies [51–53]. To further investigate the clonal relatedness of A. baumannii isolates, the REP-PCR typing method was employed. In this study, 75 isolates were analyzed, resulting in the identification of 8 distinct patterns among the 63 typeable isolates. A significant majority (80%) of these isolates grouped into a single clone (clone G) (Fig. 1), suggesting the presence of a dominant strain or a potential outbreak within the sample population. When compared to previous studies, these results reveal some differences in the genetic diversity among clinical isolates of A. baumannii. The study by Nogbou et al. from South Africa identified four clones among A. baumannii isolates, with clone A being the most prevalent, accounting for 69% of the cases [54]. In a study by Al Jabur et al. from Iran, REP-PCR results revealed 17 distinct clones, with clone A being the most prevalent, comprising 14% of the isolates [55]. In contrast, the current study shows a lower level of genetic diversity, with a significant proportion of isolates belonging to a single clone. Additionally, 90% (46/51) of genotype G isolates were XDR phenotype, underscoring the potential clinical importance of this clone. Interestingly, even though clonally related, genotype G isolates were highly heterogeneous in terms of ARGs as well as phenotypic resistance patterns. Intra-clonal heterogeneity could potentially represent horizontal transfer of genes, acquisition by plasmids, or selective pressure within the hospital, leading to genotypic and phenotypic diversification [56, 57]. It is worth noting that twelve isolates did not yield clear REP-PCR profiles, which should be considered when interpreting the observed patterns of clonal dominance. Despite this, REP-PCR successfully identified dominant profiles among the 63 typeable isolates. Consequently, conclusions regarding clonal dominance are based on these typeable isolates, and further confirmation using complementary methods, such as Multilocus Sequence Typing (MLST) and Whole Genome Sequencing (WGS), would strengthen these findings. Despite of providing useful information regarding the prevalence, phenotypic and genotypic characteristics of A. baumannii isolates, present study has several limitations. Although the study reflects the actual incidence of isolates during the investigation period, its single-center design and limited sample size (n = 75) inevitably reduce statistical power and restrict the generalizability of the findings. Consequently, the results should be interpreted with caution, as they may not fully represent the epidemiological situation in other hospitals or regions of Iran. In addition, the limited sample size and single-center design restricted the analyses to univariate tests; more robust multivariate or regression approaches were not feasible, limiting the ability to control for potential confounders. Second, the presence of other carbapenem resistance conferring mechanisms including efflux pumps and low outer membrane permeability and also colistin resistance mechanisms as mentioned earlier, were not investigated and should be considered in further researches. Third, since REP-PCR provides relatively low resolution, future studies should incorporate higher-resolution typing methods such as MLST or WGS for more precise genetic characterization of these isolates. In addition, band patterns were analyzed manually by two independent researchers to minimize bias; however, this approach may still introduce a degree of subjectivity that should be acknowledged.

Material and methods

Setting and sampling

This single-center, retrospective observational cohort study was conducted at a tertiary referral hospital in Tehran, Iran, from April to November 2021. Patients were included based on a negative RT-PCR test for SARS-CoV-2 from a nasopharyngeal swab and the absence of associated clinical symptoms. Hospital-acquired infections were defined as those diagnosed ≥ 48 h after ICU admission. Infections were confirmed in patients with positive cultures from clinically relevant sites (blood, sputum, bronchoalveolar lavage (BAL), urine, wounds) accompanied by clinical signs of infection (fever, respiratory distress, sepsis). Patients with positive cultures from non-infectious sites (throat, rectal swabs) without clinical symptoms were classified as colonized. Patient data were retrieved from the hospital's computerized databases and included clinical, demographic, and laboratory information, as well as underlying conditions, microbiological results, and ICU length of stay.

Isolate identification

A single isolate from each patient was included in the study. Isolate identification was performed using conventional microbiological and biochemical methods, including Gram staining, oxidase and catalase tests, oxidative/fermentative (O/F) reactions, motility testing, and assessment of growth at 42 °C [58]. Species-level identification was confirmed by polymerase chain reaction (PCR) amplification of the gltA gene, which encodes the species-specific citrate synthase, as described previously [59]. The A. baumannii isolates were stored at − 70 °C in trypticase soy broth supplemented with 20% glycerol until further analysis. Prior to testing, all isolates were subcultured to ensure viability and purity.

Antibiotic susceptibility testing

The Antimicrobial susceptibility testing of A. baumannii isolates was performed using the Kirby-Bauer disk diffusion method, following the 2022 Clinical and Laboratory Standards Institute (CLSI) guidelines [60]. Eleven antibiotic disks (MAST, United Kingdom) were tested, including amikacin, gentamicin, cefepime, ceftazidime, cefotaxime, ciprofloxacin, piperacillin-tazobactam, trimethoprim-sulfamethoxazole, ampicillin-sulbactam, meropenem, and imipenem. The minimum inhibitory concentration (MIC) of colistin was determined using the broth microdilution method in accordance with CLSI guidelines. Quality control was ensured using two standard reference strains, Escherichia coli ATCC 25922 and Pseudomonas aeruginosa ATCC 27853. Isolates resistant to three or more classes of antibiotics, excluding carbapenems, were classified as multidrug-resistant (MDR). MDR isolates exhibiting resistance to meropenem were further categorized as extensively drug-resistant (XDR) [61].

DNA extraction

Genomic DNA was extracted from a 24-h A. baumannii cultures grown on blood agar using the boiling method, as described previously [62–64]. Following extraction, the supernatant was stored at − 20 °C and subsequently utilized as the DNA template for PCR assays.

Detection of ARGs

Multiplex PCR was performed using previously reported primers to detect genes encoding class D carbapenemases (bla_{OXA−23−like}, bla_{OXA−24−like}, and bla_{OXA−58−like}) [65]. In addition, single PCR assays were conducted to detect MBL genes (bla_NDM, bla_IMP, bla_VIM), class A carbapenemase genes (bla_KPC, bla_TEM), aminoglycoside resistance genes (aac(6′)-Ib, aac(3)-Ia, ant(2″)-Ia, and aph(3')-Ia), and the colistin resistance gene (mcr-1), following previously studies [66–70].

Identification of global clone (GC) lineages

Global clones (GCs)1, 2, and 3 of A. baumannii were identified using multiplex PCR targeting three alleles of ompA, csuE, and bla_OXA-51-like, as previously described [71]_.

Repetitive extragenic palindromic element PCR (Rep-PCR) genotyping

To assess the genetic diversity of A. baumannii isolates, REP-PCR was performed using the primers REP1R-I (IIIICGICGICATCIGGC) and REP2-I (ICGICTTATCIGGCCTAC) [72]. PCR amplification was conducted in a thermal cycler under the following conditions: initial denaturation at 95 °C for 10 min; 30 cycles of denaturation at 95 °C for 1 min, annealing at 45 °C for 1 min, and extension at 72 °C for 1 min; followed by a final extension at 72 °C for 16 min. Due to the lack of appropriate software, REP-PCR band patterns were analyzed manually based on the presence or absence of bands. A dendrogram was subsequently generated using the online tool available at http://insilico.ehu.eus/dice_upgma/, which employs the Dice similarity coefficient and UPGMA clustering method. Isolates exhibiting ≥ 80% similarity were considered clonally related. REP-PCR banding patterns were independently evaluated by two experienced researchers using standardized criteria to ensure reproducibility and minimize analytical bias.

Statistical analysis

Data on microbial results, clinical parameters, and demographic characteristics were analyzed using descriptive statistics in IBM SPSS Version 26 (Armonk, NY, USA). Statistical analyses, including Chi-square and Fisher’s exact tests, were conducted to assess associations, with a p-value ≤ 0.05 considered indicative of statistical significance.

Conclusion

The present study demonstrates the predominance of XDR A. baumannii strains carrying diverse ARGs in our ICU setting. These findings raise concern, as such strains complicate treatment, limit therapeutic options, and may contribute to poor outcomes in critically ill patients. Mortality was high and appeared more associated with extensive resistance than with comorbidities. The clonal dissemination of these strains further emphasizes the need for robust infection control measures. Implementation of strict infection control measures and robust antibiotic stewardship programs remains essential to limit the spread of these highly resistant pathogens.

Abbreviations

A. baumannii

Acinetobacter baumannii

CRAB

Carbapenem-resistant A. baumannii

MDR

Multi-drug resistant

XDR

Extensively drug resistant

ICUs

Intensive care units

WHO

World health organization

MIC

Minimum inhibitory concentration

CLSI

Clinical laboratory standards institute

MHB

Mueller Hinton broth

AGRs

Antibiotic resistance genes

REP-PCR

Repetitive extragenic palindromic element PCR

GCs

Global clones

Authors’ contributions

R.B., F.J., and M.E. conceived and designed the experiment; M.GH. performed the experiments; M.GH., F.J., M.E. and R.B. analyzed and interpreted the data; M.GH and R.B. wrote and edited the manuscript. All authors read and approved the final manuscript.

Funding

This research was financially supported by Tehran University of Medical Sciences & health Services (grant code: 1400–2-101–54321).

Data availability

All data generated or analyzed during this study are included here and are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate declaration

The study was approved by the Ethics Committee of Tehran University of Medical Sciences, Tehran, Iran (approval code: IR.TUMS.MEDICINE.REC.1400.912). Sampling and data collection were conducted in accordance with the committee’s regulations and ethical standards. The Ethics Committee of Tehran University of Medical Sciences reviewed the study and waived the requirement for informed consent, as it was a retrospective analysis involving only anonymized bacterial isolates and did not include any direct human participation.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.