Abstract
Introduction:
Hospital-acquired respiratory infections caused by antimicrobial-resistant bacteria have become a major public health problem worldwide. This study aimed to evaluate the distribution and antimicrobial resistance profiles of bacterial strains isolated from adult patients treated with a diagnosis of lower respiratory tract infection in our hospital.
Materials and Methods:
Respiratory tract samples of adult patients hospitalized in our hospital were included. The identification of the isolated strains and their antibiotic susceptibilities were studied by the PHOENIX 100 (Becton Dickinson, USA) automated system, and results were evaluated according to the European Committee on Antimicrobial Susceptibility Testing criteria.
Discussion and Conclusions:
In this study, we observed that Gram-negative bacteria were more common than Gram-positive bacteria in lower respiratory tract samples. The multidrug resistance rate in Klebsiella pneumoniae, Pseudomonas aeruginosa, and Acinetobacter baumannii isolates is detected as 59%, 12.9%, and 96.4%, respectively. Carbapenem resistance rates for K. pneumoniae (n = 261), A. baumannii (n = 250), and P. aeruginosa (n = 132) were 88.9%, 97.2%, and 43.2%, respectively. The multidrug resistance and carbapenem resistance rates detected in lower respiratory tract samples in our study are alarming. It is noteworthy that the resistance rates we found are higher than the rates reported in other studies conducted in our country.
Keywords
Introduction
Respiratory tract infections are among the leading causes of morbidity and mortality worldwide. Although most of these infections are community-acquired, there has been a significant increase in hospital-acquired infections in recent years, and hospital-acquired infections have become a major public health problem worldwide.1,2 Intensive care units (ICUs) are units where resistant microorganism infections are frequently seen due to the use of invasive procedures, prolonged hospital stays, surfaces that are touched by hands, and so forth.1,3 The risk of infection is higher for patients in ICUs than for patients in the general hospital population, and the risk increases with the duration of intensive care. 2 The latest surveillance report from the U.S. Centers for Disease Control and Prevention showed that hospital-acquired pneumonia, usually observed as a result of endotracheal intubation and mechanical ventilation, is the most typical infection in ICUs.2,4 The widespread and nonspecific use of broad-spectrum antibiotics in ICUs is an important factor that causes the development of multidrug-resistant (MDR) bacteria and complicates treatment options. Methicillin-resistant Staphylococcus aureus (MRSA), vancomycin-resistant Enterococcus spp., extended-spectrum beta-lactamase (ESBL), and carbapenem-resistant Escherichia coli, Klebsiella pneumoniae, Acinetobacter baumannii, and Pseudomonas aeruginosa are the resistant microorganisms frequently encountered in intensive care infections.5,6
This retrospective study aims to evaluate the distribution and antimicrobial resistance (AMR) profile of bacterial strains isolated from tracheal aspirate and sputum samples of patients with lower respiratory tract infections treated in the adult ICUs and adult wards of our tertiary center.
Methods
In our study, 1,414 tracheal aspirate samples and 1,239 sputum samples sent to the Medical Microbiology Laboratory from patients hospitalized in the adult ward and ICUs of the University of Health Sciences, Izmir Bozyaka Education and Research Hospital, during January 2021 and January 2023 were evaluated. Since sputum samples can be contaminated with oropharyngeal flora, a medical microbiology expert examined the Gram stain of the respiratory tract samples microscopically and unqualified sputum samples were excluded from the evaluation by the Bartlett score. Tracheal aspirate and sputum samples were cultivated on blood agar, eosin methylene blue agar, and chocolate agar. All plates were incubated at 37°C for 24–72 hours. The identification of the isolated strains and their antibiotic susceptibilities was determined by a PHOENIX 100 (Becton Dickinson, USA) automated system. Antibiotic susceptibility results were evaluated according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) criteria (www.eucast.org).
Ethical approval statement
This study was approved by the Health Sciences University, Izmir Bozyaka Training and Research Hospital Ethics Committee (ethics reference: 06.09.2023; 2023/140). The study was carried out in compliance with relevant laws and guidelines and with the ethical standards of the Declaration of Helsinki.
Results
Lower respiratory tract samples from 545 male and 347 female patients aged between 18 and 96 years were analyzed. Among the male patients, 288 were hospitalized in ICUs, 105 in COVID-19 ICUs, 8 in COVID-19 wards, 36 in surgical wards, and 108 in internal medicine wards. Of the female patients, 207 were in ICUs, 65 in COVID-19 ICUs, 6 in COVID-19 wards, 26 in surgical wards, and 43 in internal medicine wards.
No microbial growth was detected in 153 (5.8%) of the lower respiratory tract samples submitted to the microbiology laboratory. Bacterial colonization was observed in 198 samples (7.5%), and upper respiratory tract commensal flora grew in 798 samples (30.1%). Six (0.2%) sputum samples were excluded due to poor specimen quality based on the Bartlett score. A total of 893 culture-positive (33.7%) samples were included in the evaluation. Only the first clinical isolate from each patient was included in the analysis. Resistance rates were calculated for bacterial species with ≥30 clinical isolates.
Yeast growth was detected in 93 (6.6%) tracheal aspirate samples, all from ICU patients. The most commonly isolated yeast species was Candida albicans (n = 72, 77.4%).
Across all lower respiratory tract samples, K. pneumoniae (n = 261) was the most frequently isolated bacterium, followed by A. baumannii (n = 250) and P. aeruginosa (n = 132). Streptococcus pneumoniae was identified in only eight patients, four of whom were in the ICU, three in internal medicine wards, and one in a surgical ward. None of the S. pneumoniae isolates showed resistance to penicillin. No vancomycin resistance was detected in Enterococcus spp. or S. aureus isolates. Among Stenotrophomonas maltophilia isolates (n = 52), trimethoprim/sulfamethoxazole resistance was found in only two isolates (3.8%). The methicillin resistance (MRSA) rate among S. aureus isolates was 18.4% (n = 49).
The ESBL resistance rates were 63.6% for E. coli (n = 44) and 89.7% for K. pneumoniae (n = 261). Among the E. coli isolates, 16 (36.4%) were susceptible to all tested antibiotics, 6 (13.6%) exhibited both ESBL and carbapenem resistance, and 22 (50%) showed ESBL resistance alone. For K. pneumoniae, 23 isolates (8.8%) had ESBL resistance only, 1 (0.4%) had carbapenem resistance only, and 211 isolates (80.8%) exhibited both ESBL and carbapenem resistance, while 26 isolates (10%) were fully susceptible.
Carbapenem resistance rates were as follows: E. coli 15.9%, K. pneumoniae 88.9%, A. baumannii 97.2%, and P. aeruginosa 43.2%. For P. aeruginosa, resistance rates to cefepime, ceftazidime, and fluoroquinolones were 45.5%, 34.8%, and 49%, respectively. The overall ESBL rate among all Enterobacterales was 67.7%. The rates of MDR were 59.0% for K. pneumoniae, 12.9% for P. aeruginosa, and 96.4% for A. baumannii.
Among COVID-19 patients hospitalized between 2021 and 2022, the most frequently isolated secondary lower respiratory tract pathogens in COVID-19 ICUs were A. baumannii, followed by K. pneumoniae and P. aeruginosa. All A. baumannii isolates were MDR. Among COVID-19 ward samples, three isolates exhibited MDR, and one isolate showed no resistance. Of the 50 K. pneumoniae isolates from COVID-19 ICU patients, 40 (80%) were MDR. Three isolates (6%) were fully susceptible, eight were only susceptible to aminoglycosides, and three were susceptible to both aminoglycosides and fluoroquinolones. Only one K. pneumoniae isolate was recovered from the COVID-19 ward, which was resistant to all tested antibiotics except aminoglycosides. Among 21 P. aeruginosa isolates from COVID-19 ICU samples, 6 were MDR, 2 exhibited carbapenem resistance only, and 2 were resistant to both carbapenems and cefepime/ceftazidime. In contrast, the four isolates from the COVID-19 ward did not show MDR; however, two showed fluoroquinolone resistance, and two exhibited both fluoroquinolone and cefepime/ceftazidime resistance.
The distribution of MDR isolates across ICUs and hospital wards was evaluated. For A. baumannii, MDR isolates were most frequently found in general ICUs (n = 66, 61.1%), followed by COVID-19 ICUs (n = 22, 20.4%), COVID-19 wards (n = 2, 1.9%), internal medicine wards (n = 8, 7.4%), and surgical wards (n = 10, 9.2%). For K. pneumoniae, MDR isolates were distributed as follows: general ICUs (n = 100, 64.5%), COVID-19 ICUs (n = 29, 18.7%), internal medicine wards (n = 21, 13.5%), and surgical wards (n = 5, 3.2%). No MDR K. pneumoniae isolates were detected in COVID-19 wards. Among P. aeruginosa isolates, MDR isolates were most commonly found in general ICUs (n = 12, 66.7%), COVID-19 ICUs (n = 4, 22.2%), and internal medicine wards (n = 2, 11.1%). No MDR P. aeruginosa isolates were detected in surgical or COVID-19 wards.
Table 1 presents the overall distribution (%) of bacteria isolated from lower respiratory tract samples across various hospital departments. Tables 2, 3, and 4 detail the AMR percentages for the most frequently isolated bacteria. Table 5 shows the distribution (%) of MDR Gram-negative bacteria across hospital wards and ICUs. Figures 1–3 illustrate the annual resistance change for P. aeruginosa, A. baumannii, and K. pneumoniae, respectively, between 2021 and 2022. The definition of MDR included resistance to third-generation cephalosporins, cefepime, carbapenems, fluoroquinolones, and aminoglycosides.
Distribution of Bacteria Grown in Lower Respiratory Tract Samples Between Hospital Wards and Intensive Care Units
ICU, intensive care units; n, number of isolates.
Antibiotic Resistance Pattern of Staphylococcus aureus (Methicillin-Resistant Staphylococcus aureus and Methicillin-Sensitive Staphylococcus aureus) Isolates
MRSA, methicillin-resistant Staphylococcus aureus; MSSA, methicillin-sensitive Staphylococcus aureus; NRI, number of resistant isolates.
The Antimicrobial Resistance Pattern of Staphylococcus aureus Isolates (n = 49)
The antimicrobial resistance pattern was determined by examining the resistance of all Staphylococcus aureus isolates.
Distribution of Antimicrobial Resistance Percentages of the Most Frequently Isolated Gram-Negative Bacteria from Lower Respiratory Tract Samples
n = 126.
n = 130.
AK, Amikacin; AMC, Amoxicillin/Clavulanic acid; AMP, Ampicillin; CAZ, Ceftazidime; CEF, Cefepime; CIP, Ciprofloxacin; CRO, Ceftriaxone; CXM, Cefuroxime; CZ, Cefazolin; ETP, Ertapenem; GN, Gentamicin; IPM, Imipenem; LEV, Levofloxacin; MEM, Meropenem; SAM, Ampicillin/Sulbactam; SXT, Trimethoprim/Sulfamethoxazole; TIG, Tigecycline; TPZ, Piperacillin/Tazobactam.
Distribution of Multidrug-Resistant Gram-Negative Bacilli Grown in Lower Respiratory Tract Samples Between Hospital Wards and the Intensive Care Unit
The multidrug resistance group includes carbapenems, fluoroquinolones, and aminoglycosides. The multidrug resistance patterns were determined by examining all Gram-negative bacilli isolated from lower respiratory tract samples. The nonfermenting Gram-negative bacilli group includes Acinetobacter baumanii and Pseudomonas aeruginosa. N, number of multidrug-resistant isolates.

Resistance % changes in Pseudomonas aeruginosa isolates in the hospital between 2021 and 2022. The carbapenem group includes imipenem and meropenem. The fluoroquinolone group includes ciprofloxacin and levofloxacin. The aminoglycoside group includes only amikacin. The multidrug resistance group includes ceftazidime, cefepime, carbapenems, fluoroquinolones, and aminoglycosides.

Resistance % changes in Acinetobacter baumannii isolates in the hospital between 2021 and 2022. The carbapenem group includes imipenem and meropenem. The fluoroquinolone group includes ciprofloxacin and levofloxacin. The aminoglycoside group includes gentamicin and amikacin. The multidrug resistance group includes carbapenems, fluoroquinolones, and aminoglycosides.

Resistance % changes in Klebsiella pneumoniae isolates in the hospital between 2021 and 2022. The third-generation cephalosporins group includes ceftazidime and ceftriaxone. The carbapenem group includes imipenem, meropenem, and ertapenem. The fluoroquinolone group includes ciprofloxacin and levofloxacin. The aminoglycoside group includes gentamicin and amikacin. The multidrug resistance group includes third-generation cephalosporins, carbapenems, fluoroquinolones, and aminoglycosides.
Discussion
In this study, the prevalence and AMR profiles of bacteria isolated from clinical samples of patients with lower respiratory tract infection in our hospital between January 2021 and January 2023 were determined among hospital wards and ICUs. We observed that Gram-negative bacteria were more frequently isolated than Gram-positive bacteria in respiratory tract specimens from patients treated in both the wards and ICUs. The most commonly isolated bacteria were K. pneumoniae, E. coli, A. baumannii, P. aeruginosa, and S. aureus, while yeast growth was less frequently observed (6.6%), with C. albicans being the predominant species.
Bacteria present in hospital environments are generally MDR and difficult to treat with available antibiotics. In particular, the widespread and nonspecific use of broad-spectrum antibiotics in ICUs has contributed significantly to the rapid dissemination of MDR strains and increased rates of AMR. 4
This rapid rise was further accelerated by the impact of the COVID-19 pandemic. In line with existing literature, our study also found higher rates of pathogenic bacteria and AMR in ICUs compared to general hospital wards, reinforcing previous findings on the prevalence of respiratory tract pathogens and resistance in critical care settings.1–3,7–11
Furthermore, by identifying the distribution and resistance profiles of these bacterial pathogens, especially those isolated from ICU patients, we aimed to guide clinicians in choosing effective empirical therapy. Accurate antibiotic selection is essential not only for successful treatment outcomes but also for reducing unnecessary antibiotic consumption and preventing further resistance development.
Our study reflects the period during which the global COVID-19 pandemic and its consequences became apparent. Gram-negative bacterial infections accounted for the majority of secondary infections in COVID-19 cases. The development of bacterial superinfections complicates diagnosis, treatment, and prognosis, contributing to worsened outcomes, prolonged ICU stays, and increased use of high-dose broad-spectrum antibiotics. This challenging treatment process has resulted in the development of high rates of resistance to different groups of antimicrobial drugs.12–15
It was estimated that in 2019 alone, antimicrobial-resistant bacterial infections were responsible for approximately 5 million deaths globally, with lower respiratory tract infections accounting for the majority. 16 Among Gram-negative pathogens, drug-resistant infections caused by E. coli, K. pneumoniae, A. baumannii, and P. aeruginosa were associated with over 250,000 deaths. The incidence and mortality rates of these pathogens have shown significant regional variation. 17
In our study, the most frequently isolated bacteria from secondary lower respiratory tract infections in hospitalized COVID-19 patients were A. baumannii, K. pneumoniae, and P. aeruginosa. All A. baumannii isolates exhibited multidrug resistance, followed by K. pneumoniae (80%) and P. aeruginosa (24%) isolates. MDR strains were predominantly isolated from ICUs, followed by COVID-19 ICUs. In a study conducted in our country by Berna Erdal et al., it was reported that methicillin-resistant coagulase-negative staphylococci (n = 21, 16.3%) were the most common cause of secondary infection in patients, followed by A. baumannii (n = 19, 14.7%) and C. albicans (n = 14, 10.9%). Carbapenem resistance rates of A. baumannii isolates were determined to be 94.7%. 18 In our study, A. baumannii was in the first place, and all isolates were found to be multidrug resistant.
Another notable finding of our study was the increasing resistance of K. pneumoniae, A. baumannii, and particularly P. aeruginosa to various antimicrobial classes, which may largely be attributed to the pandemic period. While similar findings have been reported in studies from other countries, our resistance rates were higher than those reported in some international and national studies.10,11
Specifically, carbapenem resistance in K. pneumoniae increased from 73.9% in 2021 to 82.9% in 2022, and third-generation cephalosporin resistance rose from 83.3% to 89.4%. The rate of MDR increased from 54.4% to 66.7% during the same period. In a study from Romania, the MDR rate in Klebsiella spp. isolates was reported as 70.0%, with over 70% resistance to third-generation cephalosporins and over 50% resistance to meropenem and ertapenem. 19 Similarly, a study in Bangladesh reported that 71.1% of K. pneumoniae isolates were MDR. 20 However, in a Turkish study by Caskurlu et al., carbapenem resistance in K. pneumoniae isolates remained relatively stable over 5 years. 11 In contrast, our study demonstrated a 9% increase in carbapenem resistance within just 2 years, potentially reflecting the impact of the COVID-19 pandemic.
In our study, the rates of aminoglycoside (96.3%), fluoroquinolone (96.3%), carbapenem (96.3%) group drugs, and multidrug resistance (96.3%), observed in A. baumannii isolates, increased slightly in 2022, and rates close to the previous year were observed. In the study conducted by Gazi et al. in our country, the MDR rates of A. baumannii were examined, and it was stated that while the MDR rate was 22.7% in 2009, this rate increased to 70.1% in 2010 (p < 0.001). 21 This sharp increase between the 2 years is a sign of the threat of MDR Acinetobacter spp., which began 15 years ago and continues to increase today.
In a more recent, multicenter surveillance study conducted by Güçlü et al. in our country, covering the years 2016–2019 with the participation of 17 hospitals, the A. baumannii multidrug resistance rate was found to be 25.1%. Imipenem and meropenem resistance was detected as 92.8% and 93.1%, respectively, in A. baumannii isolates. 10 In our study, the A. baumannii multidrug resistance rate was observed at a very high rate of 96.6% in 2022. While the multidrug resistance rate reported by Güçlü et al. 10 is lower than the value in our study, carbapenem resistance rates are close to the rate in our study.
Resistance rates reported from other countries also highlight regional variability. MDR rates for Acinetobacter spp. were reported as 77.5% in Nepal, 1 85.88% in Romania, 19 and 60% in Bangladesh. 9 Sarwar M et al. reported in a cross-sectional study covering July 2018–2020 that carbapenem resistance of A. baumannii strains isolated from lower respiratory tract samples collected from ICUs of many centers across Pakistan increased from 92% to 97.4%. 22 All these findings support the notion that AMR rates vary not only between countries but also within the different cities in the same country.23,24
The most concerning resistance shift in our study was observed in P. aeruginosa. Resistance of isolates to third-generation cephalosporins and cefepime increased almost threefold (from 16.3% to 44.2%) in 1 year. While resistance rates to aminoglycoside drugs were similar between the 2 years, resistance rates to fluoroquinolones more than doubled (33.8–73.1%). Resistance to carbapenems more than tripled (13.8–44.2%). Caskurlu et al., in their study on ICU patients between January 2014 and December 2018, observed that meropenem resistance in P. aeruginosa increased from 36.5% to 69.2% in 5 years. 11 When Gazi et al. evaluated the AMR profile by year in their study, it was reported that the rate of multidrug resistance in P. aeruginosa strains increased from 6.1% in 2009 to 23.4% in 2010. 21 The AMR rates reported in P. aeruginosa strains in these two studies conducted in our country were higher than the rates we found in our study, but the sharp acceleration in the resistance increase rate determined in just 2 years in our study is quite alarming. This increase was seen as the effect of the COVID-19 pandemic. Some studies from other countries have reported that P. aeruginosa multidrug resistance rates were higher than in our study. In a study conducted in Romania, the multidrug resistance rate in endotracheal aspirate samples taken from intensive care patients was found to be 66.25% for P. aeruginosa. It was reported that more than 60% of P. aeruginosa strains were resistant to carbapenems. 19 In the study by Dharm Raj Bhatta et al., 63.1% of P. aeruginosa in lower respiratory tract samples from patients treated in intensive care were reported to be multidrug resistant. 1 In the Philippines, Alain C. Juayang et al. study, resistance to cefepime and ceftazidime was found to be 65.8%, resistance to ciprofloxacin and levofloxacin was found to be 40.6%, and resistance to carbapenem was found to be 25% in P. aeruginosa isolates. 25 In the study by Sabrina Haque et al., the multidrug resistance rate of Pseudomonas spp. isolates isolated from tracheal aspirate cultures was determined as 60%.
In our study, the ESBL resistance rate was found to be 63.6% in E. coli isolates and 89.7% in K. pneumoniae isolates. In the study conducted in our country by Gürbüz et al., the prevalence of ESBL was reported as 40.6% and 25.0% in K. pneumoniae and E. coli isolates in 2021, respectively. 26 The rates we found in our study were considerably higher than those in the study by Gürbüz et al. In the study by Enno Jacobs et al., lower respiratory tract samples collected from 29 centers in the eastern, western, northern, and southern regions of Germany were examined, 1,859 pathogens were isolated, and their AMR profiles were investigated. While the ESBL prevalence was found to be 38.8% in Eastern Germany, this rate was determined to vary between 4.7% and 7.1% in Southern, Northern, and Western Germany. 27
In the Philippines, Alain C. Juayang et al. 25 reported the ESBL rate as 78.3% and the carbapenemase rate as 10.26% in Enterobacteriaceae. In our study, carbapenem resistance was determined as 15.9% (n = 44) in E. coli isolates and 88.9% in K. pneumoniae isolates.
S. aureus is an important cause of hospital-acquired lower respiratory tract infections. S. aureus, which continues to be a problem today with its methicillin resistance, was the most common bacterium among Gram-positive bacteria in our study. In our study, no vancomycin and linezolid resistance was found in all S. aureus isolates, but the MRSA rate was determined as 18.4%. This is considerably lower than the 78.2% and 61% MRSA rates reported by Bhatta et al. 1 and Upadhyay et al., 7 respectively.
In our country, Caskurlu et al. reported methicillin resistance rates in S. aureus as 67.7% in 2015 and 28.9% in 2018 in their study on ICU patients between January 2014 and December 2018. 11 In our study, increases in resistance rates were observed in the 2 years, but the MRSA rate we determined was lower than the rates reported in these studies. In another study conducted in our country, including 2021 data, Gürbüz et al. determined methicillin resistance (MRSA) in 35.3% of S. aureus isolates. Penicillin resistance was found in 82.3% of S. aureus isolates, while no resistance was observed to vancomycin, teicoplanin, and linezolid. 26 When compared with our study, it is seen that our MRSA rate is lower than that of Gürbüz et al., and our penicillin resistance rate is higher. Similar to our findings, no vancomycin or linezolid resistance was observed. In the study conducted by Sofia Maraki et al. in Greece between 2017 and 2022, resistance rates to oxacillin were 37.2%, and 62% of the methicillin-resistant strains were MDR. Low resistance rates were detected to trimethoprim/sulfamethoxazole (3.3%) and gentamicin (2.8%). It was reported that all isolates were susceptible to linezolid, daptomycin, tigecycline, teicoplanin, and vancomycin. 28 In our study, low resistance rates were detected to trimethoprim/sulfamethoxazole (4.1%) and gentamicin (8.2%), which is consistent with this study.
In our study, AMR of all isolates was determined using the PHOENIX 100 (Becton Dickinson) automated system, and the results were evaluated according to the EUCAST criteria (www.eucast.org). Since the colistin microdilution test and ceftazidime–avibactam antibiogram test had not yet been performed in our laboratory during the period covered by our study, we could not provide antimicrobial susceptibility results in our study. This was a limitation of our study.
In conclusion, the increase in carbapenem resistance and multidrug resistance determined in our study over a period of 2 years is alarming. The determination of high rates of MDR bacteria in ICUs in our study suggests that the treatment difficulties brought about by secondary bacterial infections developing during the COVID-19 pandemic period, the effectiveness of the applied empirical treatment, the presence of high doses of broad-spectrum antibiotics, and so forth, may have made the increase in resistance inevitable. The resistance rates we found to be higher than those in other studies conducted in our country demonstrated the need to monitor increases in resistance with such cross-sectional studies. We think that it is necessary to see the presence of MDR bacteria isolated from lower respiratory tract samples and to monitor increases in resistance, especially after the COVID-19 pandemic nightmare we are experiencing worldwide. With our study, we have once again emphasized the inevitable reality of measures to be taken against the AMR epidemic, starting at the hospital level.
Authors’ Contributions
G.A.: Conceptualization, formal analysis, methodology, investigation, project administration, resources, writing—original draft, and writing—review and editing. G.E.: Conceptualization, methodology, and writing—review and editing. M.Y.: Investigation and formal analysis. All authors have read and agreed to the published version of the article.
Footnotes
Funding Information
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Availability of Data and Materials
The data in the current study can be requested at any time.
Disclosure Statement
The authors declare that they have no conflicts of interest.
