An-Najah National University 
 

Faculty of Graduate Studies 

 

 
 

 

 

EPIDEMIOLOGY OF ACINETOBACTER 

BAUMANNII INFECTIONS AMONG PATIENTS 

ADMITTED TO THREE INTENSIVE CARE 

UNITS IN PALESTINE: A RETROSPECTIVE 

CHART REVIEW STUDY 
 
 

 

 
By 

Shaden Soleman Jomaa Arjan 
 

 

 

 

 

 

 

Supervisors 

Dr. Rasha Khayyat 

Dr. Hadi Rabee 

 
This Thesis is Submitted in Partial Fulfillment of the Requirements for the Degree 

of Master of Infectious Diseases Prevention and Control, Faculty of Graduate 

Studies, An-Najah National University, Nablus, Palestine. 

 
2024



ii 

 

 

 

 

 

 

 

 

 
 

 
EPIDEMIOLOGY OF ACINETOBACTER 

BAUMANNII INFECTIONS AMONG PATIENTS 

ADMITTED TO THREE INTENSIVE CARE 

UNITS IN PALESTINE: A RETROSPECTIVE 

CHART REVIEW STUDY 
 
 

 
By 

Shaden Soleman Jomaa Arjan 
 

 

 
 

This Thesis was Defended Successfully on 31/10/2024 and approved by 

Dr. Rasha Khayyat 

Supervisor 

Dr. Hadi Rabee 

Co-Supervisor 

Dr. Kamal Dumaidi 

External Examiner 

Dr. Adham Abu Taha 

Internal Examiner 



iii 

Dedication 

I gratefully dedicate this thesis to my family, whose unwavering support and 

encouragement have been my greatest strength throughout this journey.  

To my supervisors and advisors, thank you for your guidance, wisdom, and belief in my 

work. Your insights and encouragement have been invaluable. 

To my friends and colleagues, whose camaraderie and shared passion have made this 

research both enjoyable and rewarding, your support has been greatly appreciated. 

And to all the patients and healthcare professionals who have faced the challenges 

posed by Acinetobacter baumannii, this work is a tribute to your resilience and 

dedication. 

Shaden Arjan 

 

 

 



iv 

Acknowledgments 

I wish to begin by expressing my deepest gratitude to ALLAH, whose guidance, 

wisdom, and strength have been the foundation of this journey. I am profoundly 

thankful for the blessings and grace that have illuminated my path throughout this 

research. 

I extend my heartfelt thanks to my supervisors, Dr. Rasha Khayyat and Dr. Hadi Rabee, 

for their invaluable support, mentorship, and insightful feedback. Your guidance has 

been instrumental in shaping this thesis. 

Special thanks to An-Najah National University’s Faculty of Medicine and Health 

Sciences and all of its doctors for providing the resources and environment necessary 

for this study. 

Lastly, I am deeply appreciative of my family and friends for their unwavering support, 

patience, and encouragement. Your belief in me has been a constant source of strength. 

Thank you all for your contributions and support. 

Shaden Arjan 



v 

Declaration 

I, the undersigned, declare that I submitted the thesis entitled: 

EPIDEMIOLOGY OF ACINETOBACTER 

BAUMANNII INFECTIONS AMONG PATIENTS 

ADMITTED TO THREE INTENSIVE CARE 

UNITS IN PALESTINE: A RETROSPECTIVE 

CHART REVIEW STUDY 
 

I declare that the work provided in this thesis, unless otherwise referenced, is the 

researcher’s own work, and has not been submitted elsewhere for any other degree or 

qualification. 

 

 

 

 



vi 

List of Contents

 

Dedication ........................................................................................................................ iii 

Acknowledgments ........................................................................................................... iv 

Declaration ........................................................................................................................ v 

List of Contents ................................................................................................................ vi 

List of Tables ................................................................................................................. viii 

List of Figures .................................................................................................................. ix 

List of Appendices ............................................................................................................ x 

Abstract ............................................................................................................................ xi 

Chapter One: Introduction and Theoretical Background .................................................. 1 

1.1 Theoretical basis ......................................................................................................... 1 

1.2 Problem statement ....................................................................................................... 1 

1.3 Study questions/general aims ...................................................................................... 2 

1.4 Objectives ................................................................................................................... 3 

1.5 Importance of the study .............................................................................................. 3 

1.6 Literature review ......................................................................................................... 4 

1.6.1 Prevalence of A. baumannii ..................................................................................... 7 

1.6.2 Antimicrobial resistance and treatment options ....................................................... 9 

1.6.3 Prevalence of A. baumannii infection in ICU: ....................................................... 13 

Chapter Two: Methods ................................................................................................... 15 

2.1 Study design and settings .......................................................................................... 15 

2.2 Study period .............................................................................................................. 15 

2.3 Population ................................................................................................................. 15 

2.4 Selection of the patients and antimicrobial resistance .............................................. 16 

2.5 Sample size and sampling method ............................................................................ 16 

2.5.1 Inclusion criteria .................................................................................................... 16 

2.5.2 Exclusion criteria ................................................................................................... 16 

2.6 Data Collection and study tools ................................................................................ 17 

2.7 Microbiological methods .......................................................................................... 17 

2.8 Analysis plan ............................................................................................................. 18 



vii 

Chapter Three: Results .................................................................................................... 19 

3.1 Prevalence of A. baumannii infections in the different study hospitals .................... 19 

3.2 Demographics and clinical variables of the patients who acquired A. baumannii 

infections .................................................................................................................. 20 

3.3 Use of antibiotics, previous infections with MDR strains, hospitalization, and 

intensive care unit admission ................................................................................... 24 

3.4 Onset of infection, source, susceptibility/resistance pattern, treatment, and outcomes 

of the patients ........................................................................................................... 27 

3.5 Risk of death among the patients .............................................................................. 35 

Chapter Four: Discussion ................................................................................................ 41 

4.1 Strengths of the study ............................................................................................... 45 

4.2 Limitations of the study ............................................................................................ 46 

4.3 Recommendations ..................................................................................................... 48 

4.4 Conclusion ................................................................................................................ 48 

List of Abbreviations ...................................................................................................... 49 

References ....................................................................................................................... 50 

Appendices ...................................................................................................................... 57 

 ب ............................................................................................................................... الملخص

 

 

 

 

 



viii 

List of Tables 

 

Table 1: Demographics, comorbidities, and use of medications .................................... 23 

Table 2: Use of antibiotics, previous infections with MDR strains, and intensive care 

unit admission ................................................................................................... 26 

Table 3: Hospitalization and intensive care unit admission reason ................................ 27 

Table 4: Onset of infection, source, and type pf colonization ........................................ 33 

Table 5: Susceptibility/resistance pattern, treatment, and outcomes of the patients ...... 34 

Table 6: Univariate analysis of associations between survival/mortality outcomes with 

the other variables of the patients ..................................................................... 36 

Table 7: Multivariate logistic regression analysis of associations between survival/ 

mortality outcomes with the other variables of the patients ............................. 38 

Table 8: Univariate analysis of associations between survival/mortality outcomes with 

age, WBCs, platelets, duration of hospitalization, and length of treatment ..... 39 

Table 9: Multiple linear regression analysis of associations between survival/mortality 

outcomes with age, WBCs, platelets, duration of hospitalization, and length of 

treatment ........................................................................................................... 40 

 

 



ix 

List of Figures 

Figure 1: Prevalence of A. baumannii infections in the different three major 

governmental hospitals in the north, center, and south of the West Bank of 

Palestine ............................................................................................................ 19 

Figure 2: Hospital distribution of the patients with A. baumannii infections ................. 20 

Figure 3: Age distribution of the patients with A. baumannii infections ....................... 21 

Figure 4: Comorbidities of the patients with A. baumannii infections ........................... 22 

Figure 5: Prevalence of respiratory disease among the patients who admitted to the three 

hospitals ............................................................................................................ 24 

Figure 6: Percentages of patients with previous infections with multidrug-resistant 

strains bacteria in Intensive care units .............................................................. 25 

Figure 7: Proportion of A. baumannii Infections: Admission vs. ICU-acquired ............ 28 

Figure 8: Antibiotic Resistance Type Across Hospitals ................................................. 29 

Figure 9: Frequency of Antibiotic susceptibility for A. baumannii Patients at SICU, 

NICU, CICU Hospitals ..................................................................................... 30 

Figure 10: Patient Outcomes for A. baumannii Infections Across Hospitals ................. 32 

 

 

 



x 

List of Appendices 

 

Appendix A: Data collection form .................................................................................. 57 

Appendix B: Study approval 1 ........................................................................................ 60 

Appendix C: Study approval 2 ........................................................................................ 60 

 

 

 

 

 

 



xi 

EPIDEMIOLOGY OF ACINETOBACTER BAUMANNII 

INFECTIONS AMONG PATIENTS ADMITTED TO THREE 

INTENSIVE CARE UNITS IN PALESTINE: A RETROSPECTIVE 

CHART REVIEW STUDY 

By 

Shaden Soleman Jomaa Arjan 

Supervisors 

Dr. Rasha Khayyat 

Dr. Hadi Rabee 

Abstract 

Background: Multidrug-resistant (MDR) strains of Acinetobacter baumannii pose 

significant treatment challenges. This study aimed to evaluate the sociodemographic 

and clinical characteristics of patients with A. baumannii infections in medical intensive 

care units (ICUs) across three hospitals in the West Bank, Palestine. It also sought to 

determine the prevalence of multidrug-resistant and extensively drug-resistant strains 

and assess patient survival and mortality rates. 

Methods: A retrospective cohort design was conducted from 2019 to 2020, spanning a 

period of two years, and including patients from adult medical ICUs from three 

hospitals of the north (NICU), the center (CICU), and the south (SICU) of in the West 

Bank, Palestine. Data on infections acquired during ICU stays or present at admission 

were analyzed using patient medical records, where sociodemographic, clinical, and 

treatment-related information, including comorbidities, infection history, diagnostic 

results, therapies, device usage, and outcomes such as ICU length of stay and mortality, 

were collected, and prevalence was calculated.  

Results: The study encompassed 231 patients, with 136 (58.9%) from CICU Hospital, 

56 (24.2%) from SICU Hospital, and 36 (16.9%) from NICU Hospital. The median 

patient age was 63 years. Prevalence rates of A. baumannii infections were (11.5%) at 

CICU Hospital, (6.5%) at NICU Hospital, and (5.3%) at SICU Hospital, with an overall 

rate of (8.2%). Of the patients, 46 (19.9%) had infections upon admission, while 185 

(80.1%) developed infections during their ICU stay. The isolates included 154 (66.7%) 

extensively drug-resistant (XDR), 61 (26.4%) multidrug-resistant (MDR), and 16 

(6.9%) non-resistant strains. Factors associated with increased mortality included recent 

ICU admission (2.65 times more likely), heart failure (4.95 times more likely), and 



xii 

central line catheter use (3.46 times more likely). Higher white blood cell counts, lower 

platelet counts, longer ICU and mechanical ventilation durations, and shorter hospital 

stays, shorter therapy were linked to survival. 

Conclusions: This study provides a comprehensive understanding of A. baumannii 

infections in the medical ICUs, highlighting the critical need for multidisciplinary 

infection control, especially for critically ill patients with comorbidities. The rise of 

drug-resistant bacteria underscores the importance of antimicrobial stewardship and 

infection control. Findings suggest a need for ongoing research and collaboration to 

address this growing hospital-acquired issue effectively. 

Keywords: Acinetobacter species, A. baumannii, Intensive care unit, infection, Hospital 



1 

Chapter One 

Introduction and Theoretical Background 

1.1 Theoretical basis 

A. baumannii is a gram-negative bacterium that pertains to the family Moraxellaceae 

(1). This bacterium is known to withstand harsh environmental conditions like survival 

for extended period over dry surfaces and human skin (2). These abilities potentiate its 

persistence and wide spread in hospitals and other healthcare facilities (3, 4). Over the 

years, misuse of antibiotics and poor management measures have contributed to the 

appearance of strains including A. baumannii strains that exhibit significant resistant to 

different antibiotics (5, 6). Therefore, the Centers for Disease Control and Prevention 

(CDC) have recognized A. baumannii as an important source of epidemics and 

outbreaks because of its mounting resistance to many of the available antibiotics (7). 

Additionally, A. baumannii was shown to cause life-threatening infections, including 

urinary tract infections, meningitis, bloodstream infections, pneumonia, and skin/wound 

infections (8).  

1.2 Problem statement 

It is well-established that multidrug resistant strains of A. baumannii are difficult to 

treat. MDR refers to bacteria that are resistant to at least one antibiotic in three or more 

different classes of antibiotics. Therefore, infections with multidrug resistant strains of 

A. baumannii present a major challenge to providers of healthcare for hospitalized 

patients, notably, those admitted to the intensive care units. Because patients in the 

intensive care units are fragile, they could be at higher risk for increased morbidity and 

mortality in case they acquire A. baumannii (9-11).  

Lately, description of epidemiology of hospital infections have received considerable 

attention. In a previous epidemiological analysis of sepsis syndrome among patients in 

an intensive care unit in Palestine, Acinetobacter species caused the majority of the 

hospital-acquired infections (12). In another respective analysis, medical records of 

intensive care unit patients with A. baumannii infections at An-Najah National 

University Hospital who were admitted between the year 2015 and 2017 were reviewed 

for sensitivity tests (13). The analysis was a single center study, included a small 



2 

number of cases (99 cases), was conducted over a span of 3 years, and did not include 

analysis of clinical manifestations and survival/mortality rates. Additionally, the 

previous study did not classify the patients using sequential organ failure assessment 

score, a tool used to evaluate the extent of organ dysfunction or failure in critically ill 

patients. Moreover, the study did not include some variables like white blood cells 

count, platelets count. Similarly, the study did not collect data on the use of 

corticosteroids, use of antibiotics, immunosuppressant drugs, continuous renal 

replacement therapy and/or mechanical ventilation. Information of placement of devices 

like inserted central venous catheter, inserted urinary catheter, and inserted nasogastric 

tube were not collected. The duration of mechanical ventilation, length of stay in the 

intensive care unit, length of stay in the hospital, duration of antibiotic therapy, 

mortality at discharge were also not included in the analysis.  

Little research was conducted on the epidemiology of A. baumannii infections, 

specifically among patients admitted to the intensive care units in Palestine. 

Additionally, little is known on the survival/mortality among patients with A. baumannii 

infections in the Palestinian intensive care units. Furthermore, there is a notable lack of 

research on extensively drug-resistant (XDR) A. baumannii, which refers to strains 

resistant to at least one agent in all but two or fewer antibiotic categories, complicating 

treatment options significantly. Therefore, an epidemiological analysis of these 

infections using recent data would be interesting and informative to the infectious 

diseases practice in Palestine.  

1.3 Study questions/general aims 

The research questions of this thesis were posed in the context of the Palestinian 

healthcare system in the West Bank. This thesis attempted to answer the following 

research questions: 

 What were the sociodemographic and clinical variables of the patients admitted with 

A. baumannii infections or acquired during ICU stay in the medical intensive care 

units in three different hospitals in the West Bank of Palestine?  

 What was the prevalence of multidrug resistant and extensive drug-resistant A. 

baumannii strains in the medical intensive care units in three different hospitals in 

the West Bank of Palestine?  



3 

 What were the clinical manifestations of A. baumannii infections among the patients 

admitted to in the medical intensive care units in three different hospitals in the West 

Bank of Palestine in relation to their sociodemographic and clinical variables?  

 What were the survival/mortality rates among patients with A. baumannii infections 

in the medical intensive care units in three different hospitals in the West Bank of 

Palestine and what are the predictors of higher survival/mortality?  

1.4 Objectives 

The study was conducted with the following objectives: 

 Assess the sociodemographic and clinical variables of the patients who acquired A. 

baumannii infections in the medical intensive care units in three different hospitals in 

the West Bank of Palestine. 

 Assess the prevalence of multidrug resistant and extensive drug-resistant A. 

baumannii strains in the medical intensive care units in three different hospitals in 

the West Bank of Palestine. 

 Describe the clinical manifestations of A. baumannii infections among the patients 

admitted to in the medical intensive care units in three different hospitals in the West 

Bank of Palestine in relation to their sociodemographic and clinical variables. 

 Determine the survival/mortality rates among patients with A. baumannii infections 

in the medical intensive care units in three different hospitals in the West Bank of 

Palestine and what are the predictors of higher survival/mortality. 

1.5 Importance of the study 

In 2019, an outbreak of a novel coronavirus that caused severe acute respiratory 

syndrome coronavirus 2 (SARS-CoV-2) was reported in Wuhan, China. Later, the virus 

spread globally and led the World Health Organization (WHO) to declare this global 

widespread of the virus as COVID-19 pandemic (14). Patients who experienced severe 

symptoms and manifestations of COVID-19 required admissions to the intensive care 

units. During the pandemic, the numbers of admissions to the intensive care unit 

skyrocketed in all healthcare systems around the globe. Patients admitted to the 

intensive care units often received empiric broad-spectrum antibiotics and other drugs 

that reduce cytokines (15). The use of empiric broad-spectrum antibiotics and cytokines 

reducing agents increased the risk of superinfections that are caused by multi-drug-



4 

resistant bacteria (15). Infections caused by A. baumannii represent a major health 

concern in patients admitted to the intensive care units. Infections causes by multi-drug-

resistant strains were shown to be associated with high mortality rates among patients 

admitted to the intensive care units. Description of the epidemiology of A. baumannii 

infections among patients admitted to the intensive care units in Palestine could be 

informative to the clinicians, intensivists, infectious diseases specialists, 

epidemiologists, and other healthcare professionals and decision makers who might be 

involved in developing national and local protective measures and antimicrobial use 

policies.    

1.6 Literature review  

Acinetobacter species are ubiquitous environmental bacteria that are non-fermentative, 

aerobic, gram-negative coccobacilli (16). The majority of infections caused by 

Acinetobacter species are opportunistic and mostly affect patients with severe 

underlying diseases who need prolonged treatment with mechanical ventilation and 

antimicrobials while staying in the intensive care unit (17). In managing patients with A. 

baumannii in intensive care units, the key controllable factor is antimicrobial treatment. 

It's essential to avoid unnecessary antibiotics to reduce resistance. Transmission of A. 

baumannii is commonly associated with ventilatory equipment and can be facilitated by 

contaminated hands of nursing and respiratory staff (16). 

The epidemiology of nosocomial respiratory infections with A. baumannii has become 

complex due to the presence of both epidemic and sporadic cases from different strains. 

This complexity underscores the importance of using molecular typing techniques to 

identify and control microepidemics early. Nosocomial infections caused by 

Acinetobacter species are associated with high fatality rates, ranging from (30%) to 

(75%), with the highest mortality observed in ventilator-dependent patients. (18). Like 

other gram-negative bacilli that take advantage of opportunities, the growing resistance 

to antibiotics has made it difficult to effectively treat nosocomial infections caused by 

Acinetobacter species.  

Usually species of the Acinetobacter genus are often found in the soil, water, and 

animals. A. baumannii is mostly discovered in the hospital setting, namely in critical 

care units (19). In veterinary practice, very ill animals in clinics and intensive care units 



5 

are increasingly observed to become infected and colonized by A. baumannii. Although 

there are occasional reports of A. baumannii in nonhuman sources such as animals, lice, 

vegetables, aquaculture, and soil, it remains unclear whether these findings are due to 

environmental contamination from hospital reservoirs or if they point to alternative 

natural reservoirs for the bacterium (20). This requires further investigation. 

The significance of A. baumannii in clinical settings has significantly grown since the 

late 1980s due to the appearance and widespread transmission of three dominant clones 

(referred to as international clonal lineage) that have the ability to cause hospital 

outbreaks globally (21). Among these clones, international clonal lineage 1 and 

international clonal lineage 2 are multidrug-resistant. Currently, the multilocus sequence 

typing of A. baumannii reveals around 400 MLST sequence types. A recent 

investigation has shown the presence of at least six significant international clonal 

lineages that are spread throughout different continents globally (21). The six 

international clonal lineages consist of the three primary outbreak clones that were first 

detected, and they indicate the concerning appearance of additional epidemic clones. 

The developing population structure is believed to be caused by a low initial 

phylogenetic diversity of A. baumannii, followed by a fast dissemination after a severe 

evolutionary bottleneck (21, 22). 

A. baumannii is a widely distributed bacterium that thrives in both hospital and 

community settings, making it a significant nosocomial pathogen. Known for its 

resistance to last-resort antibiotics such as colistin, tigecycline, and carbapenems, it is a 

major concern for public health, as highlighted by the WHO and CDC. Its ability to 

form biofilms, survive harsh conditions, and resist host defenses underscores its 

persistence and virulence. The plasticity of the bacterium's genome leads to an increased 

genetic diversity within its strains, which is a major problem in the study and 

management of this organism. This combination of special features makes A. baumannii 

a dangerous global disease occurrence that needs further, close surveillance (23). 

Traditional paradigm regards A. baumannii as a Gram-negative bacterium that possesses 

all of the exhaustion factors as such but very few of the classic virulence determinants. 

It has been elucidated quite a number of these bacterial structures and more so its 

mechanisms of surface coating. This paper contains a survey of immune responses 



6 

elicited to A. baumannii, outlines the current status of the problem and gap which exists, 

and evaluates the current state of different animal models utilized in the exploration of 

A. baumannii (24). 

Acinetobacter species especially A. baumannii are highly prevalent in the hospitals as 

well as normal skin, throat and rectal flora or even in food and lice. Since they are often 

found in intensive care units, they can also transmit from one patient to another through 

surfaces, equipment and even through wounds. Even though A. baumannii is 

extensively harboring in many patients, invasive pathologies like ventilator-associated 

pneumonia and bacteremia do occur. Multidrug resistance particularly towards the 

carbapenems and cephalosporins derives its root from the mechanisms of β-lactamase 

production, membrane protein alterations, and efflux proteins (25). 

A. baumannii is successful multicultural boreal or very poorly acid ecology shape out, 

undergoes evolution to the harsh test of loyalty through fierce battles of change in 

surrounding environment. 

A. baumannii causes severe hospital-acquired infections like pneumonia and 

bloodstream infections, especially in patients with preexisting conditions or significant 

surgery. Infections are linked to prolonged hospital stays and are more common in older 

males. (10). 

A. baumannii can cause community-acquired infections like pneumonia, which 

constitutes (85%) of such cases, as well as bacteremia, skin infections, and secondary 

meningitis. These infections are more common in older males with underlying 

conditions such as diabetes and COPD. Community-acquired pneumonia is often 

severe, with rapid progression and fatality rates up to (60%), predominantly reported in 

tropical and subtropical regions during summer (1, 26).  

Recent developments link A. baumannii infections to injuries in conflict zones and 

natural disasters, such as the Marmara earthquake. Data suggest that morphine, often 

used for pain management in these situations, might worsen A. baumannii infections 

due to its immunosuppressive effects. Additionally, the high demand on emergency 

units during crises can compromise infection control, leading to rapid A. baumannii 

spread and outbreaks, although this is not always a direct cause (27). 



7 

The severity of A. baumannii infections varies widely, with fatality rates averaging 

between (8%) and (35%) depending on the strain and infection type (16, 28, 29). 

Studies using mouse models have shown significant differences in pathogenicity among 

strains. For example, a multidrug-resistant strain from a blood culture caused an (80%) 

mortality rate in mice, while a strain causing meningitis resulted in only a (13%) 

mortality rate. Additionally, strains from international clone 1 and sporadic isolates 

showed higher virulence compared to those from international clone 3. Variations in 

mortality rates are linked to different virulence factors; for instance, a mucoid strain had 

a (48%) mortality rate in a pneumonia model, while a nonmucoid strain had a (19%) 

rate, despite having the same PFGE profile. 

1.6.1 Prevalence of A. baumannii 

In a retrospective study, A. baumannii infections were analyzed over a period of 3-years 

(2015-2017) (13). The medical records of intensive care unit patients in a single tertiary 

hospital in Palestine were reviewed for culture and sensitivity test results. The study 

included 99 patients with a mean age of 59.9 ± 18.8 years. Of the patients, (58.6%) were 

male and Colistin was the most effective antibiotic. The study did not identify any 

significant relationships between sensitivity profiles and age, gender, and number of 

comorbidities. 

Ayobami et al. (2019) have also undertaken a detailed systematic review and a meta-

analysis of the incidence and prevalence of A. baumannii nosocomial infections across 

Africa, Eastern Mediterranean and Europe. In particular, they underlined the threat these 

infections cause to health, especially those achieving resistance to carbapenem 

antibiotics. Moreover, analyzing 24 studies (total, 3,340 records), they demonstrated 

that A. baumannii ICU-associated infections were at 56.5 cases per 1000 young patients 

and 4.4 cases per 1000 patient-days. It was also reported that rates of healthcare-

associated bloodstream infections were between 0.85 and 5.6 cases per 1000 patients. 

Data on the pre-occupying issue of A. baumannii emerged from infection by 

carbapenems was presented as 41.7/1000 patients and 2.1/1000 patient-days. While A. 

baumannii were responsible for (20.9%) of all infections ‘picked up’ hospital and 

(13.6%) of carbapenem resistant strains were identified as the causative agent. These 

results highlight the continuing relevance of A. baumannii infections in clinical practice 

today, especially in hospital settings (18). 



8 

Alotaibi et al. (2021) evaluated A. baumannii infection in ICU patients accorded at King 

Fahad University Hospital in Saudi Arabia from the years of 2013 to 2017. Of the 198 

patients assessed, the infection had a (3.37%) prevalence with a (40.8%) mortality rate. 

Most of the patients were male in the mean age of 49 years and younger in the average 

age of survivors. Therapy was different: 65 patients were on colistin treatment, 22 were 

on combined treatment with carbapenems and colistin and 18 on carbapenems only. A 

greater percentage of patients treated with carbapenems survived. The study noted the 

still great problem how to treat patients with multi-drug-resistant A. baumannii as it was 

health care associated complication with high mortality and long ICU care (30). 

Uwingabiye et al. (2017) conducted a study in Morocco regarding A. baumannii 

infections of the Intensive Care Unit patients at a teaching hospital between January 

2015 to July 2016. Another retrospective cohort study by Seng et al. conducted within 

964 invasive ICU patients reported that 81 (8.4%) of the subjects were found to have A. 

baumannii infections. Several factors responsible for prolonged ICU stay were 

identified and especially included central lines, mechanical ventilation and prior use of 

antibiotics, operative room admissions were however beneficial. High mortality rate due 

to A. baumannii infections was (74.1%). Moreover, multivariate analysis has shown 

higher risk of death associated with septic shock (OR=19.2) and elderly patients aged 

≥65 years (OR=4.9). The study concludes that reducing length of stay at the ICU, better 

use of medical apparatus and correct prescriptions of antimicrobials may reduce 

infection rates and improve the overall outcome (31). 

In another study, El Kettani et al (2017) conducted a study on the detection of A. 

baumannii from blood cultures and its patterns of antibiotics resistance in the intensive 

care unit of Ibn Rochd University Hospital in Morocco (Casa Blanca). A distinct but 

complementary design was longitudinal retrospective in nature and it involved the 

abstraction and analysis of blood cultures from all critical care units at Ibn Rochd 

University Hospital from 2010 to 2014. Out of the 4232 samples sent to the laboratory, 

2402 was found positive for the test. Most of the A. baumannii strains that were 

identified in this study were resistant to the majority of antibiotics tested which include 

but are not limited to imipenem (75.7%), ceftazidime (85.4%), cefotaxime (98.6%), 

gentamicin (78.1%), amikacin (63.5%), and finally ciprofloxacin (88.2%) (32). 



9 

A study in Tehran, Iran, from July 2005 to November 2006 (33), examined the 

prevalence of resistant Acinetobacter baumannii in a large hospital. Out of 88 strains 

recovered, most were from respiratory specimens. The bacterium showed high 

resistance to several antibiotics, including ceftriaxone, piperacillin, ceftazidime, 

amikacin, and ciprofloxacin. Imipenem was the most effective, with a low resistance 

rate of (4.5%), while tobramycin had a (44.3%) resistance rate. The study found a 

correlation between patient demographics, such as age and hospital stay, and the level 

of antibiotic resistance. 

A retrospective study in Turkey from 2003 to 2007 (34), analyzed imipenem-resistant 

A. baumannii (IRAB) infections. Out of 720 patients with 925 infections, most cases 

were in critical care units, with an occurrence rate of 6.2 per 1000 admissions. 

Pneumonias and bloodstream infections were most common. The prevalence of 

imipenem resistance increased significantly from (43.3%) to (72.9%) over the study 

period. Factors linked to higher mortality included imipenem resistance, extended ICU 

stays, female gender, older age, and pneumonia. Infections were notably associated with 

hemodialysis, malignancy, and mechanical ventilation. Strains from pneumonia patients 

had higher imipenem resistance. Cefoperazone-sulbactam and netilmicin were the most 

effective treatments. The study highlighted a rising trend in A. baumannii infections and 

imipenem resistance, emphasizing the need for vigilance in patients with specific risk 

factors. 

1.6.2 Antimicrobial resistance and treatment options 

A. baumannii poses significant treatment challenges due to its ability to endure in 

various environments and cause severe healthcare-associated infections. The rise in 

antimicrobial resistance, particularly to carbapenems, limits treatment options. 

Addressing this issue requires developing new therapies, evaluating current treatments, 

and enhancing infection prevention strategies in healthcare settings (35).  

The prevalence of certain A. baumannii lineages has been associated with the 

multidrug-resistant phenotype of the infecting strains (11, 34, 36, 37). Currently, it’s 

unclear if multidrug resistance is a factor in the epidemic spread of A. baumannii or a 

trait of specific strains. Since the 1970s, antimicrobial resistance has increased, with up 



10 

to (70%) of isolates being multidrug-resistant by 2007, including resistance to 

carbapenems, affecting both military and civilian settings. 

Colistin seems to be the most consistently successful antibiotic in laboratory tests 

against multidrug-resistant A. baumannii (38). However, the use of colistin is linked to 

various adverse effects and is not appropriate for treating all kinds of infections. There 

are already global reports of colistin resistance, leading to the development of strains 

that are resistant to all known antibiotics in some regions. Concerningly, there are 

indications that resistance is becoming more widespread throughout the population. 

There have been reports of Acinetobacter species that are resistant to carbapenem in 

cattle and other animals, as well as in the Seine River in Paris, France. 

In recent decades, A. baumannii has developed resistance to a wide range of 

antimicrobials (4). The ability of this bacteria to acquire resistance genes, sometimes by 

horizontal gene transfer, largely determines it’s potential. Recent research indicates that 

the development of the multidrug-resistant phenotype is a crucial aspect in the success 

of A. baumannii as a pathogen in healthcare settings. 

A. baumannii is often seen in ICU patients and those with weakened immune systems. It 

causes severe infections like ventilator-associated pneumonia and bloodstream 

infections, leading to higher mortality and longer hospital stays. Key risk factors include 

previous antibiotic use and prolonged ICU stays. Research is ongoing into new 

treatments to address its resistance. (2). 

A study by Abbo et al. (2005) investigated multidrug-resistant Acinetobacter baumannii 

using epidemiologic methods and PFGE. Analyzing 118 patients across 27 wards, the 

study identified clusters of infections and ten PFGE clones, with two being most 

common. Key risk factors included male gender, cardiovascular disease, mechanical 

ventilation, and metronidazole use, while penicillins provided some protection. The 

study underscores the challenges in managing A. baumannii due to its complex 

epidemiology and resistance patterns. (9). 

The "antibiotic era" may be approaching its end in the near future. The early and 

apparently unstoppable triumph of antibiotics, a product of human ingenuity, has been 

met by an increase in resistance mechanisms in bacteria. This situation has been 



11 

characterized as a "unwinnable conflict". The monitoring efforts have yielded data that 

demonstrate the rise of several species of bacteria that exhibit resistance to all 

antibiotics. The genus Acinetobacter exemplifies this pattern and warrants further 

scrutiny. Acinetobacter species exhibit resistance mechanisms against all currently 

available classes of antibiotics, and possess a remarkable ability to acquire novel 

resistance determinants. The increasing success in treating multidrug-resistant A. 

baumannii in the clinical setting is a daunting truth (39).  

A. baumannii infections mainly affect patients with weakened immune systems, severe 

illnesses, and those with recent invasive procedures or broad-spectrum antibiotic use 

(34). Therefore, A. baumannii infections are prevalent in critical care units, causing 

ventilator-associated pneumonia, urinary tract infections, and bacteremia. They can also 

lead to complex skin, soft tissue, gastrointestinal, and central nervous system infections. 

Recently, A. baumannii has emerged as a significant pathogen in conflict-related 

wounds, with ongoing research focusing on factors influencing its colonization, 

pathogenicity, and invasion. 

Distinguishing between infection and colonization by Acinetobacter baumannii can be 

challenging, and there is ongoing debate about the clinical significance of its presence. 

Some physicians view the detection of A. baumannii in hospitalized patients as an 

indicator of severe illness, with a fatality rate around (30%). Recent studies, including 

cases from northern Australia and southern Asia, have reported high fatality rates for 

community-acquired pneumonia caused by A. baumannii. Many patients in these 

studies also experienced severe complications such as bacteremia, acute respiratory 

distress syndrome, and disseminated intravascular coagulation. These findings suggest 

that some strains of A. baumannii may have developed enhanced virulence traits, 

distinguishing them from other Acinetobacter species (4, 11, 18, 28, 29). 

Acinetobacter species, particularly A. baumannii, have developed resistance to multiple 

classes of antibiotics. These bacteria are notable for their high capacity for genetic 

exchange and are classified as "naturally transformable," with A. baylyi strain ADP1 

showing exceptionally high natural competence and homologous recombination 

abilities. Strains lacking the mutS gene, which helps repair DNA mismatches, exhibit 

increased mutation rates. The presence of competence genes such as comFECB and 



12 

comQLONM aids in the efficient uptake of external DNA. It remains unclear whether 

other Acinetobacter species, including A. baumannii, naturally acquire foreign DNA or 

if environmental factors could be exploited to enhance their pathogenicity and 

resistance (6, 10, 11, 15, 19, 29, 34). 

Recent research analyzing the genome sequences of both susceptible SDF and resistant 

AYE A. baumannii isolates has uncovered a high prevalence of resistance genes in the 

bacterium. The study identified the AbaR1 resistance island in the multidrug-resistant 

AYE isolate, which contains 45 resistance genes, including major beta-lactamases 

(VEB-1, AmpC, OXA-10), aminoglycoside-modifying enzymes, and tetracycline efflux 

pumps. This resistance island features mobile genetic elements like transposons and 

genes previously associated with other bacterial species such as Pseudomonas, 

Salmonella, and E. coli. In contrast, the susceptible SDF strain had a 20-kb genomic 

island at the same location, but it lacked resistance genes. The distribution and presence 

of this extensive resistance region across different multidrug-resistant A. baumannii 

strains remain unclear (4, 27, 37). 

A different modern study presents the examination of the A. baumannii ATCC 17978 

genomic sequence (40). Researchers combined sequencing, insertional mutagenesis, and 

C. elegans virulence tests to create a new pathogenicity assay using Dictyostelium 

discoidium. They found that a 1951 genome sample had 3.98 million base pairs with 

3,830 open reading frames, (17%) of which were from potential foreign DNA. Many 

genes were part of pathogenicity islands, including one similar to type IV secretion 

systems in Legionella and Coxiella burnetii. Six of these foreign islands' pathogenic 

features were confirmed in C. elegans using insertional mutagenesis. 

A sequence with a similar insertion pattern was identified. The resistance island had a 

length of 13,277 base pairs and consisted of nine genes. It was situated between the 5′ 

and 3′ ends of the presumed ATPase gene. Remarkably, just a single gene associated 

with drug resistance, sulI, was discovered inside this specific region of genetic insertion. 

A. baumannii ATCC 17978 isolate has 74 putative drug resistance genes, which consist 

of 32 efflux pump genes, 11 permease genes, and 26 genes that encode resistance to 

heavy metals (41, 42). These results suggest that genes have inserted into this organism 

in locations other than the specified "hot spot". Due to the unavailability of the whole 



13 

sequences of the two French isolates, it is currently not possible to compare the 

resistance and pathogenicity islands. 

Resistance to beta-lactams in A. baumannii involves multiple mechanisms: beta-

lactamases hydrolyze the antibiotics (1). Changes in penicillin-binding proteins (PBPs) 

hinder their action, structural changes in porin proteins limit antibiotic entry, and efflux 

pumps expel antibiotics from the cell. Notably, class A extended-spectrum beta-

lactamases (ESBLs), such as PER-1, are newly identified in A. baumannii, conferring 

resistance to penicillins and cephalosporins but not to carbapenems. PER-1 is prevalent 

in Turkey, Korea, France, Belgium, and Bolivia, and was recently reported in the U.S. 

Other ESBLs found in A. baumannii include VEB-1 (in France and Belgium), SHV-12 

(in China and the Netherlands), TEM-92 (in Italy), and CTX-M-2 (in Japan and 

Bolivia), though the spread of CTX-M-2 is less extensive compared to 

Enterobacteriaceae (43). Clinical identification of ESBLs is challenging due to the 

presence of chromosomal cephalosporinases and the variability in clinical standards (39, 

44). 

1.6.3 Prevalence of A. baumannii infection in ICU:

A recent study (2019-2021) in an ICU examined multi-drug-resistant A. baumannii 

strains among patients with COVID-19 and other conditions. It found that A. baumannii 

colonization, high serum lactate levels, and steroid use were linked to increased risk of 

infection and 30-day mortality in COVID-19 patients (45). 

In another study, the demographic characteristics, clinical variables, and outcomes of 

the patients who had A. baumannii infections in the intensive care units in Indian were 

described (46). It was a retrospective study of A. baumannii infections in ICU patients 

from January to December 2009 identified 108 infections in 94 patients, with (76.9%) 

affecting the respiratory tract. The study found that medical patients were more prone to 

these infections than surgical patients were, with (63.8%) acquired in the ICU. Infected 

patients had longer intubation durations, and (70%) of the infections were from 

multidrug-resistant strains. The overall mortality rate exceeded (70%), linked to 

improper antimicrobial use and prolonged intubation 



14 

In Poland a 6-years single-center retrospective study was conducted to investigate A. 

baumannii infections among patients admitted to the intensive care unit (29). The study 

also aimed to investigate how the susceptibility of A. baumannii to antimicrobials 

changed over the 6 years period. During the study, there were 183 A. baumannii 

infections. The incidence of A. baumannii infections was amounted to 6.4/1000 patient-

days. The study showed increase in the percentages of A. baumannii infections over the 

years from 2011 to 2016. The respiratory tract was the primary site of A. baumannii 

infections and the most common infections (73.8%) involved ventilator-associated 

pneumonia. The majority of A. baumannii were considered multidrug-resistant. All A. 

baumannii strains were susceptible to colistin. On the other hand, susceptibility 

percentages to imipenem, meropenem, amikacin, and ciprofloxacin were (12.3%), 

(11.5%), (10.7%), and (2.4%). 



15 

Chapter Two 

Methods 

2.1 Study design and settings  

This study was conducted in a retrospective cohort design. Patients admitted to the adult 

medical ICUs at three major governmental hospitals in the north, center, and south of 

the West Bank of Palestine with A. baumannii infections and who acquired infection 

during ICU stay, (defined as an infection that occurs 48 hours or more after a patient has 

been admitted to the Intensive Care Unit) were included. The data were collected from 

three governmental hospitals, south hospital (SICU), center hospital (CICU), and north 

hospital (NICU).  

2.2 Study period 

In this study, the files of the patients admitted to the medical intensive care units in the 

three hospitals in the period between January 2019 and December 2020 were reviewed. 

This time frame allowed to compare A. baumannii infections before and during the 

COVID-19 pandemic. Also, to assess the impact of the pandemic on ICU infection 

rates, as the COVID-19 crisis resulted in significant changes to healthcare systems, 

infection control practices, and the prevalence of hospital-acquired infections, including 

A. baumannii. In Palestine, the first case of COVID-19 was diagnosed on March 5, 

2020. Therefore, the period from the beginning of 2019 up to March 2020 was 

considered “period before COVID-19” and the period from then until the end of 2020 

was considered “period during COVID-19”. 

2.3 Population 

The study population was adult patients (> 18 years) admitted to the medical intensive 

care units in the three hospitals in the period from the beginning of January 2019 and 

end December 2020 with A. baumannii infections whether admitted with the infection 

or acquired during the intensive care units stay. 



16 

2.4 Selection of the patients and antimicrobial resistance 

Adult patients with A. baumannii infections admitted with the infection or acquired 

during the intensive care units stay were included in this study. Clinical presentation of 

A. baumannii infections were considered only if confirmed by a positive culture. 

Multidrug resistant (MDR) A. baumannii strains were considered when the isolates 

were resistant to three classes or more of antimicrobials: 1) penicillins/cephalosporins, 

2) aminoglycosides, and 3) fluoroquinolones (47, 48). Extensively resistant (XDR) A. 

baumannii strains were considered when the isolates were resistant to at least one agent 

in all but two or fewer antimicrobial categories (i.e., bacterial isolates remain 

susceptible to only one or two categories) (49). Resistance to carbapenems was also 

investigated in this study.  

2.5 Sample size and sampling method 

In this study, a purposive total population sampling method was used. This means that 

all patients who were admitted to the medical intensive care units in the three hospitals 

in the study period were selected. The inclusion and exclusion criteria were applied on 

the total population of the patients.  

2.5.1 Inclusion criteria 

The inclusion criteria were: 

All adult patients (> 18 years) who were admitted to the medical intensive care units in 

the three hospitals and had A. baumannii infection confirmed by a positive culture either 

admitted with the infection or acquired during ICU stay regardless of the reason for 

admission to the intensive care units. 

2.5.2 Exclusion criteria 

Patients with incomplete files were excluded from this study.  



17 

2.6 Data Collection and study tools 

The data collection form was developed to collect the following information: 

1. Sociodemographic data of the patients like age and gender. 

2. Clinical data of the patient, 

a. History of hospitalization, hospital to which the patient was admitted, history of 

admission to the intensive care unit, reason for admission to the intensive care unit. 

b. Presence of comorbidities like cardiovascular disease, heart failure, hypertension, 

hyperlipidemia, diabetes mellitus, chronic kidney disease, chronic liver disease, 

neurological disease, respiratory disease, COVID-19, cancer. 

c. History of infections like previous infection with A. baumannii, multidrug-resistant, 

bloodstream infections, sepsis, other infections and also source of infection and site 

of infection and A. baumannii colonization. 

d. Scoring and test results like sequential organ failure assessment score, white blood 

cells count, platelets count, results of culture and susceptibility tests. 

e. Drug use and therapies like chronic use of corticosteroids, use of antibiotics, use of 

immunosuppressant drugs, continuous renal replacement therapy, use of mechanical 

ventilation 

f. Placement of devices like inserted central venous catheter, inserted urinary catheter, 

inserted nasogastric tube.  

g. Duration of mechanical ventilation, length of stay in the intensive care unit, length of 

stay in the hospital, duration of steroidal therapy, duration of antibiotic therapy, 

mortality at discharge from ICU. 

2.7 Microbiological methods 

In the South Hospital, microbiological samples for culture testing were processed 

manually. The method involved the direct inoculation of clinical samples, such as 

blood, urine, sputum, and wound swabs, onto various types of agar plates (e.g., 

MacConkey agar, Blood agar, and Chocolate agar). The plates were then incubated at 

specific temperatures based on the suspected organism, and colonies were examined for 

growth. Bacterial identification was performed through colony morphology, Gram 

staining, and biochemical testing, which included catalase, oxidase, and urease tests, 

among others. This traditional method relies heavily on the experience of 

microbiologists for accurate interpretation of results. 



18 

In contrast, the North and Center Hospitals uses the Vitek system for microbiological 

testing. The Vitek 2 Compact System is an automated microbiological identification and 

susceptibility testing system that provides faster and more accurate results. After 

inoculating the clinical samples into specialized Vitek cards, which contain a series of 

wells with substrates for biochemical reactions, the system performs the tests 

automatically. The Vitek machine measures changes in the color of the wells, 

corresponding to bacterial metabolism, and uses this information to identify the 

microorganism and its antibiotic susceptibility profile. This method significantly 

reduces the turnaround time for results and increases diagnostic accuracy, compared to 

manual testing.

2.8 Analysis plan 

The data were tabulated using MS Excel and were then entered into IBM SPSS v.22.0. 

Descriptive statistics and cross-tabulations were used. The data were assessed for 

normality of distribution using Kolmogorov–Smirnov test. Because the data were not 

normally distributed, nonparametric tests were used. Categorical data were presented as 

frequencies (n) and percentages (%). On the other hand, continuous data were expressed 

using the median with the interquartile range (IQR = Q1, Q3). Distribution of the 

categorical variables were compared using Chi-square tests or Fisher’s exact tests, as 

appropriate. The variables that were significantly associated in the Chi-square tests or 

Fisher’s exact tests were included in a multivariate analysis to calculate the odds ratios 

with their (95%) confidence intervals (95% CI).  

On the other hand, continuous data were compared using Mann-Whitney U tests or 

Kruskal-Wallis tests, as appropriate. The variables that were significantly associated in 

the Mann-Whitney U tests or Kruskal-Wallis tests were included in a multiple linear 

regression model. A p value of < 0.05 was considered as statistically significant.  



19 

Chapter Three 

Results 

3.1 Prevalence of A. baumannii infections in the different study hospitals  

The total number of admitted patients to medical ICU in the three hospitals was 2832. 

Among these patients, 231 patients met the inclusion criteria and their data were 

analyzed. The overall prevalence rate across all hospitals was (8.2%) (231/2832) 

(Figure 1). The prevalence rates of A. baumannii infections at the CICU was (11.5%) 

(136/1187), at the NICU was (6.5%) (39/598), and at the SICU was (5.3%) (56/1047). 

Figure 1 

Prevalence of A. baumannii infections in the different three major governmental hospitals in the 

north, center, and south of the West Bank of Palestine 

 



20 

3.2 Demographics and clinical variables of the patients who acquired A. baumannii 

infections  

The total number of patients from the different ICUs was 231 patients. Patients from 

CICU presented (58.9%) (136) of the patients, while (24.2%) (56) were from SICU and 

(16.9%) (39) were from NICU (Figure 2). 

Figure 2 

Hospital distribution of the patients with A. baumannii infections  

 

The median age of the patients was 63.0 [IQR = 52.0, 75.0] years. There was no 

statistical difference (p = 0.769) between the age of the patients who were included 

from CICU, SICU, and NICU Hospital (62.5 [IQR = 54, 73], 62 [IQR = 37.5, 75.5], and 

66 [IQR = 48, 75], respectively). Of the patients, 148 (64.1%) were males, 83 (35.9%) 

were females, and 87 (37.7%) were smokers (Table 1) (Figure3). 



21 

Figure 3 

Age distribution of the patients with A. baumannii infections  

 

The majority of the patients suffered from comorbidities, including hypertension (n = 

166, 71.9%), hyperlipidemia (n = 150, 64.9%), diabetes mellitus (n = 136, 58.9%), heart 

failure (n = 105, 45.5%), neurological diseases (n = 80, 34.6%), and respiratory diseases 

(n = 91, 39.4%). Corticosteroids were administered to 91 (39.4%) patients. The details 

of the comorbidities that the patients had are shown in Table 1 and Figure 4. 



22 

Figure 4 

Comorbidities of the patients with A. baumannii infections  

 



23 

Table 1 

Demographics, comorbidities, and use of medications  

Hospital 
CICU (n 

= 136) 

SICU (n 

= 56) 

NICU (n 

= 39) 

Total (n = 

231) 
p* 

Variable 
     

Gender 
     

Male 91 (66.9) 32 (57.1) 25 (64.1) 148 (64.1) 
0.439 

Female 45 (33.1) 24 (42.9) 14 (35.9) 83 (35.9) 

Smoking status 
     

No 79 (58.1) 37 (66.1) 28 (71.8) 144 (62.3) 
0.239 

Yes 57 (41.9) 19 (33.9) 11 (28.2) 87 (37.7) 

Comorbidities 
     

Coronary artery disease 41 (30.1) 17 (30.4) 14 (35.9) 72 (31.2) 0.783 

Heart failure 62 (45.6) 24 (42.9) 19 (48.7) 105 (45.5) 0.852 

Hypertension 97 (71.3) 37 (66.1) 32 (82.1) 166 (71.9) 0.229 

Hyperlipidemia 93 (68.4) 33 (58.9) 24 (61.5) 150 (64.9) 0.408 

Diabetes mellitus 79 (58.1) 35 (62.5) 22 (56.4) 136 (58.9) 0.804 

Chronic kidney disease 26 (19.1) 11 (19.6) 8 (20.5) 45 (19.5) 0.981 

Chronic liver disease 20 (14.7) 4 (7.1) 6 (15.4) 30 (13) 0.325 

Neurological disease 48 (35.3) 18 (32.1) 14 (35.9) 80 (34.6) 0.902 

Respiratory disease 44 (32.4) 9 (16.1) 20 (51.3) 73 (31.6) 0.001 

COVID-19 6 (4.4) 5 (8.9) 6 (15.4) 17 (7.4) 0.136 

Cancer 12 (8.8) 3 (5.4) 3 (7.7) 18 (7.8) 0.717 

Chronic use of corticosteroids 60 (44.1) 17 (30.4) 14 (35.9) 91 (39.4) 0.184 

Use of immunosuppressant drugs 21 (15.4) 5 (8.9) 4 (10.3) 30 (13) 0.396 

Continuous renal replacement 

therapy 
21 (15.4) 12 (21.4) 4 (10.3) 37 (16) 0.330 

Note. COVID-19: Coronavirus disease, CICU: center intensive care unit, NICU: north intensive care unit, 

SICU: south intensive care unit, *Chi-square/Fisher’s exact test  

The patients included from the different hospitals were not different in terms of 

demographics, comorbidities, and use of medications, except for the prevalence of 

respiratory disease among the patients. The patients who were included from NICU 

Hospital had significantly more respiratory diseases compared to those included from 

the other two hospitals (Table 1 and Figure 5). 



24 

Figure 5 

Prevalence of respiratory disease among the patients who admitted to the three hospitals 

 

3.3 Use of antibiotics, previous infections with MDR strains, hospitalization, and 

intensive care unit admission 

Of the patients, 156 (67.5%) had used antibiotics in the last 90 days whether 

intravenous or oral antibiotics and 133 (57.6%) had previous infections with multidrug-

resistant strains. There were no statistical differences in the use of antibiotics and having 

previous infections with multidrug-resistant strains between the patients who were 

included from the different hospitals (Figure 6). 



25 

Figure 6 

Percentages of patients with previous infections with multidrug-resistant strains bacteria in 

Intensive care units 

 

Of the patients, 64 (27.7%) were given previously vancomycin and 40 (17.3%) used 

piperacillin/tazobactam. The patients from SICU Hospital had received previously 

(before infection with A. baumannii) significantly more piperacillin/tazobactam 

compared to those from the other two hospitals. On the other hand, the patients who 

were included from NICU Hospital had used previously colistin, linezolid, and 

rifampicin (rifampin) compared to those who were from the other two hospitals. Of the 

patients, 149 (64.5%) were hospitalized in the last 90 days and 56 (24.2%) were 

admitted to the intensive care unit in the last 90 days, primarily due to infection-related 

issues. The patients who were included from NICU Hospital had significantly more 

previous admission to the intensive care unit compared to those who were included 

from the other two hospitals. There was no statistically significant difference between 

the intensive care unit admission rates before and after COVID-19. These details are 

shown in Table 2 and Table 3.  



26 

Table 2 

Use of antibiotics, previous infections with MDR strains, and intensive care unit admission  

Hospital 
CICU  

(n = 136) 

SICU  

(n = 56) 

NICU  

(n = 39) 

Total  

(n = 231) 
p* 

Variable 
     

Use of antibiotics in the last 90 

days 
89 (65.4) 36 (64.3) 31 (79.5) 156 (67.5) 0.214 

Previous infection with multidrug-

resistant strains 
82 (60.3) 35 (62.5) 16 (41) 133 (57.6) 0.069 

Type of previous multidrug-

resistant strain      

Escherichia coli ESBL 29 (21.3) 15 (26.8) 6 (15.4) 50 (21.6) 0.410 

MRSA 42 (30.9) 20 (35.7) 7 (17.9) 69 (29.9) 0.163 

A. baumannii 9 (6.6) 4 (7.1) 2 (5.1) 15 (6.5) 0.922 

Klebsiella pneumoniae 23 (16.9) 6 (10.7) 7 (17.9) 36 (15.6) 0.507 

Pseudomonas aeruginosa 20 (14.7) 10 (17.9) 2 (5.1) 32 (13.9) 0.190 

Acinetobacter species 1 (0.7) 1 (1.8) 0 (0) 2 (0.9) 0.631 

Morganella morganii 1 (0.7) 1 (1.8) 0 (0) 2 (0.9) 0.631 

Citrobacter species 1 (0.7) 0 (0) 0 (0) 1 (0.4) 0.704 

Proteus mirabilis 2 (1.5) 0 (0) 0 (0) 2 (0.9) 0.494 

CRE 7 (5.1) 4 (7.1) 0 (0) 11 (4.8) 0.260 

VRE 6 (4.4) 2 (3.6) 1 (2.6) 9 (3.9) 0.862 

Antibiotics used in previous three months 

Penicillin      

Ampicillin 1 (0.7) 0 (0) 0 (0) 1 (0.4) 0.704 

Beta-lactams inhibitors      

Piperacillin/tazobactam 24 (17.6) 14 (25) 2 (5.1) 40 (17.3) 0.041 

Cephalosporin      

Ceftazidime 14 (10.3) 6 (10.7) 0 (0) 20 (8.7) 0.108 

Cefuroxime 1 (0.7) 2 (3.6) 0 (0) 3 (1.3) 0.212 

Carbapenems      

Meropenem 13 (9.6) 4 (7.1) 6 (15.4) 23 (10) 0.407 

Fluoroquinolones      

Levofloxacin 1 (0.7) 1 (1.8) 0 (0) 2 (0.9) 0.631 

Ciprofloxacin 4 (2.9) 1 (1.8) 0 (0) 5 (2.2) 0.525 

Aminoglycosides      

Amikacin 15 (11) 4 (7.1) 2 (5.1) 21 (9.1) 0.446 

Gentamicin 0 (0) 0 (0) 1 (2.6) 1 (0.4) 0.084 

Tetracyclines      

Doxycycline 0 (0) 1 (1.8) 0 (0) 1 (0.4) 0.208 

Tetracycline 1 (0.7) 0 (0) 2 (5.1) 3 (1.3) 0.063 

Tigecycline 1 (0.7) 0 (0) 0 (0) 1 (0.4) 0.704 

Others      

Vancomycin 42 (30.9) 17 (30.4) 5 (12.8) 64 (27.7) 0.074 

Trimethoprim/sulfamethoxazole 4 (2.9) 3 (5.4) 0 (0) 7 (3) 0.324 

Colistin 2 (1.5) 0 (0) 5 (12.8) 7 (3) < 0.001 

Linezolid 3 (2.2) 0 (0) 6 (15.4) 9 (3.9) < 0.001 

Nitrofurantoin 1 (0.7) 0 (0) 0 (0) 1 (0.4) 0.704 

Rifampicin (rifampin) 0 (0) 0 (0) 2 (5.1) 2 (0.9) 0.007 

Clindamycin 0 (0) 0 (0) 1 (2.6) 1 (0.4) 0.084 

Note. CICU: center intensive care unit, NICU: north intensive care unit, SICU: south intensive care unit, CRE: Carbapenem-
resistant Enterobacteriaceae, ESBL: Extended-spectrum beta-lactamases, MRSA: Methicillin-resistant Staphylococcus aureus, 
*Chi-square/Fisher’s exact test  



27 

Table 3 

Hospitalization and intensive care unit admission reason  

Hospital 
CICU  

(n = 136) 

SICU  

(n = 56) 

NICU  

(n = 39) 

Total  

(n = 231) 
p* 

Variable 
     

Previous hospitalization in the 

last 90 days      

Hospital admission 89 (65.4) 32 (57.1) 28 (71.8) 149 (64.5) 0.319 

ICU admission 33 (24.3) 12 (21.4) 11 (28.2) 56 (24.2) 0.750 

Reason for ICU admission 
     

Infectious cause 88 (64.7) 27 (48.2) 36 (92.3) 151 (65.4) 
< 

0.001 

Post operation 11 (8.1) 10 (17.9) 0 (0) 21 (9.1) 
 

Burn 0 (0) 6 (10.7) 0 (0) 6 (2.6) 
 

Other medical reason 37 (27.2) 13 (23.2) 3 (7.7) 53 (22.9) 
 

ICU admission year 
     

2019- march 2020 (before 

COVID-19) 
92 (67.6) 33 (58.9) 26 (66.7) 151 (65.4) 0.505 

March 2020 (COVID-19) 44 (32.4) 23 (41.1) 13 (33.3) 80 (34.6) 
 

Note. CICU: center intensive care unit, NICU: north intensive care unit, SICU: south intensive care unit, 
*Chi-square/Fisher’s exact test. COVID-19: Coronavirus disease 

3.4 Onset of infection, source, susceptibility/resistance pattern, treatment, and 

outcomes of the patients 

In this study, 46 (19.9%) patients had A. baumannii infection upon admission and 185 

(80.1%) patients had ICU-acquired. baumannii infection (Table 4, Table 5, and Figure 

7). 



28 

Figure 7 

Proportion of A. baumannii Infections: Admission vs. ICU-acquired 

 

Percentage of sputum, wound, urine, and blood cultures were the most common sources 

of A. baumannii isolates. Of the isolates, 154 (66.7%) were XDR, 61 (26.4%) were 

MDR, and 16 (6.9%) were non-resistant to commonly used antibiotics. The strains 

isolated from the patients who were included from NICU Hospital were significantly 

more XDR compared to those who were included from the other two hospitals (Figure 

8). 



29 

Figure 8 

Antibiotic Resistance Type Across Hospitals 

  

The vast majority of the patient’s culture results (99.6%) susceptible to colistin and 62 

(26.8%) susceptible to trimethoprim/sulfamethoxazole. The patients who were included 

from NICU Hospital susceptible significantly less trimethoprim/sulfamethoxazole 

compared to those who were included from the other two hospitals (Figure 9). 



30 

Figure 9 

Frequency of Antibiotic susceptibility for A. baumannii Patients at SICU, NICU, CICU 

Hospitals 

 

The patients received antibiotics for a median duration of 9 [IQR = 4, 14] days. There 

was no significant difference (p = 0.206) between the duration of antibiotic therapy 

received by the patients who were included from CICU, SICU, and NICU Hospital (10 

[IQR = 4, 14], 7.5 [IQR = 2, 13.5], and 8 [IQR = 1, 17], respectively). 

Of the patients, 161 (69.7%) received mechanical ventilation. Of the patients, 198 

(85.7%) had a central venous catheter inserted. The vast majority of the patients 

(98.3%) had a urinary catheter inserted. Of the patients, 164 (71.0%) had a nasogastric 

tube inserted. Significantly, more patients who were included from CICU received 

central venous catheters, urinary catheters, and nasogastric tubes compared to those who 

were included from the other two hospitals. 

The patients received mechanical ventilation for a median duration of 8 [IQR = 0, 17] 

days, stayed in the intensive care unit for a median of 15 [8, 23] days, and stayed in the 

hospital for a median of 18 [IQR = 12, 26] days. The patients who were included from 

CICU received mechanical ventilation for a significantly (p = 0.006) longer period 



31 

compared to those who were included from SICU and NICU (10 [IQR = 1.5, 18] vs 4.5 

[IQR = 0, 10] and 4 [IQR = 0, 22], respectively). On the other hand, the patients who 

were included from SICU Hospital stayed in the intensive care unit for shorter durations 

(p < 0.001) compared to those who were included from CICU and NICU hospital (9 

[IQR = 6, 16] vs 16 [IQR = 10, 24] and 16 [IQR = 7, 27.5], respectively). Similarly, the 

patients who were included from SICU Hospital also stayed in the hospital for shorter 

durations (p = 0.023) compared to those who were included from CICU and NICU 

hospital (16 [IQR = 9.5, 21] vs 19 [IQR = 13, 26.5] and 20 [IQR = 11.5, 28], 

respectively).  

During their stay, the median sequential organ failure assessment score was 12 [IQR = 

10, 12]. The patients who were included from SICU Hospital had a significantly lower 

(p = 0.006) sequential organ failure assessment score compared to those who were 

included from CICU and NICU hospital (11 [IQR = 8, 12] vs 12 [IQR = 11, 12] and 12 

[IQR = 11, 13], respectively). The median white blood cells count was 16.5 [IQR = 

11.2, 21.7]. The patients who were included from NICU Hospital had a lower white 

blood cells count (p = 0.001) compared to those who were included from CICU and 

SICU hospital (11.2 [IQR = 8.4, 16.4] vs 16.8 [IQR = 12.9, 21.7] and 17.7 [IQR = 10.6, 

25.5], respectively). The median platelets count was 180 [IQR = 120, 266]. There were 

no statistically significant differences (p = 0.100) in the platelets counts between the 

patients who were included from the different hospitals (185.5 [IQR = 131.5, 267] vs 

213 [IQR = 112.5, 265.5], and 158.5 [IQR = 79, 220], respectively). Of the patients, 163 

(70.6%) survived and were discharged alive and 68 (29.4%) died in the intensive care 

unit (Figure 10). 



32 

Figure 10 

Patient Outcomes for A. baumannii Infections Across Hospitals 

  



33 

Table 4 

Onset of infection, source, and type pf colonization 

Hospital 
CICU  

(n = 136) 

SICU  

(n = 56) 

NICU  

(n = 39) 

Total  

(n = 231) 
p* 

Variable 
     

Onset of infection 
     

Present on admission 22 (16.2) 15 (26.8) 9 (23.1) 46 (19.9) 
0.213 

ICU-acquired 114 (83.8) 41 (73.2) 30 (76.9) 185 (80.1) 

Sources of A. baumannii 

isolated from patients      

Nasal swab colonization 22 (16.2) 9 (16.1) 1 (2.6) 32 (13.9) 0.026 

Rectal swab colonization 16 (11.8) 7 (12.5) 1 (2.6) 24 (10.4) 
 

Vaginal swab 1 (0.7) 0 (0) 0 (0) 1 (0.4) 0.702 

Throat swab 0 (0) 3 (5.4) 0 (0) 3 (1.3) 0.009 

Sputum culture 88 (64.7) 19 (33.9) 18 (46.2) 125 (54.1) < 0.001 

Urine culture 22 (16.2) 16 (28.6) 7 (17.9) 45 (19.5) 0.145 

Blood culture 23 (16.9) 5 (8.9) 15 (38.5) 43 (18.6) 0.005 

Wound culture 32 (23.5) 19 (33.9) 7 (17.9) 58 (25.1) 0.173 

PEG tube culture 0 (0) 0 (0) 1 (2.6) 1 (0.4) 0.085 

Central line culture 0 (0) 0 (0) 1 (2.6) 1 (0.4) 0.085 

Missing data 88 (64.7) 33 (58.9) 37 (94.9) 158 (68.4) 
 

Type of colonization (Swabs 

done upon admission to ICU)      

Nasal 22 (16.2) 9 (16.1) 1 (2.6) 32 (13.9) 
0.026 

Rectal 16 (11.8) 7 (12.5) 1 (2.6) 24 (10.4) 

Note. CICU: center intensive care unit, NICU: north intensive care unit, SICU: south intensive care unit, MDR: 

Multi-drug resistant, XDR: extensively drug-resistant, *Chi-square/Fisher’s exact test  



34 

Table 5 

Susceptibility/resistance pattern, treatment, and outcomes of the patients 

Hospital 
CICU  

(n = 136) 

SICU  

(n = 56) 

NICU  

(n = 39) 

Total 

(n = 231) 
p* 

Variable 
     

Susceptibility/Resistance to 

antibiotics      

Non-resistant 10 (7.4) 5 (8.9) 1 (2.6) 16 (6.9) 

0.005 MDR 44 (32.4) 15 (26.8) 2 (5.1) 61 (26.4) 

XDR 82 (60.3) 36 (64.3) 36 (92.3) 154 (66.7) 

Antibiotics susceptibility for 

patients with A. baumannii      

Beta-lactams      

Beta-lactamase Inhibitors      

Piperacillin/tazobactam 5 (3.7) 3 (5.4) 0 (0) 8 (3.5) 0.365 

Cephalosporin      

Cefotaxime 5 (3.7) 3 (5.4) 0 (0) 8 (3.5) 0.365 

Ceftazidime 5 (3.7) 3 (5.4) 0 (0) 8 (3.5) 0.365 

Carbapenems      

Imipenem 5 (3.7) 3 (5.4) 0 (0) 8 (3.5) 0.365 

Meropenem 5 (3.7) 4 (7.1) 0 (0) 9 (3.9) 0.204 

Aminoglycosides      

Gentamicin 11 (8.1) 6 (10.7) 1 (2.6) 18 (7.8) 0.339 

Amikacin 6 (4.4) 4 (7.1) 1 (2.6) 11 (4.8) 0.562 

Fluoroquinolones      

Ciprofloxacin 5 (3.7) 4 (7.1) 0 (0) 9 (3.9) 0.204 

Levofloxacin 5 (3.7) 3 (5.4) 0 (0) 8 (3.5) 0.365 

Ofloxacin 0 (0) 0 (0) 1 (2.6) 1 (0.4) 0.084 

Others      

Colistin 135 (99.3) 56 (100) 39 (100) 230 (99.6) 0.704 

Trimethoprim/ 

sulfamethoxazole 
46 (33.8) 15 (26.8) 1 (2.6) 62 (26.8) 0.001 

Tigecycline 11 (8.1) 4 (7.1) 0 (0) 15 (6.5) 0.190 

Inserted devices      

Use of mechanical ventilation 103 (75.7) 32 (57.1) 26 (66.7) 161 (69.7) 0.035 

Inserted central venous 

catheter 
122 (89.7) 44 (78.6) 32 (82.1) 198 (85.7) 0.104 

Inserted urinary catheter 136 (100) 54 (96.4) 37 (94.9) 227 (98.3) 0.046 

Inserted nasogastric tube 107 (78.7) 32 (57.1) 25 (64.1) 164 (71) 0.007 

Outcome 
     

Discharged 106 (77.9) 30 (53.6) 27 (69.2) 163 (70.6) 
0.003 

Died 30 (22.1) 26 (46.4) 12 (30.8) 68 (29.4)
Note. CICU: center intensive care unit, NICU: north intensive care unit, SICU: south intensive care unit, MDR: 

Multi-drug resistant, XDR: extensively drug-resistant, *Chi-square/Fisher’s exact test  



35 

3.5 Risk of death among the patients 

Significant associations between smoking status, intensive care unit admission within 

the last 90 days, coronary artery disease, heart failure, hypertension, hyperlipidemia, 

diabetes mellitus, chronic kidney disease, liver disease, and respiratory disease were 

more likely to die compared to the other patients (Table 6). Similarly, the isolation of A. 

baumannii from sputum and wound culture, continuous renal replacement therapy, 

hospitalization for infection, mechanical ventilation, a central catheter, and a nasogastric 

tube significantly correlated with increased mortality. Table 6 displays these details. 



36 

Table 6 

Univariate analysis of associations between survival/mortality outcomes with the other 

variables of the patients 

  
Outcome 

 

Variable Category 
Survived 

n (%) 

Died 

n (%) 
p* 

Smoking 
No 109 (47.2) 35 (15.2) 

0.028 
Yes 54 (23.4) 33 (14.3) 

Admission to ICU in the last 

90 days 

No 117 (50.6) 58 (25.1) 
0.029 

Yes 46 (19.9) 10 (4.3) 

Coronary artery disease 
No 100 (43.3) 59 (25.5) 

< 0.001 
Yes 63 (27.3) 9 (3.9) 

Heart failure 
No 71 (30.7) 55 (23.8) 

< 0.001 
Yes 92 (39.8) 13 (5.6) 

Hypertension 
No 38 (16.5) 27 (11.7) 

0.012 
Yes 125 (54.1) 41 (17.7) 

Hyperlipidemia 
No 47 (20.3) 34 (14.7) 

0.002 
Yes 116 (50.2) 34 (14.7) 

Diabetes mellitus 
No 59 (25.5) 36 (15.6) 

0.018 
Yes 104 (45) 32 (13.9) 

Chronic kidney disease 
No 124 (53.7) 62 (26.8) 

0.008 
Yes 39 (16.9) 6 (2.6) 

Liver disease 
No 137 (59.3) 64 (27.7) 

0.038 
Yes 26 (11.3) 4 (1.7) 

Respiratory disease 
No 105 (45.5) 53 (22.9) 

0.044 
Yes 58 (25.1) 15 (6.5) 

Sputum culture 
No 64 (27.7) 42 (18.2) 

0.002 
Yes 99 (42.9) 26 (11.3) 

Wound culture 
No 128 (55.4) 44 (19) 

0.023 
Yes 34 (14.7) 24 (10.4) 

Continuous renal replacement 

therapy 

No 130 (56.3) 64 (27.7) 
0.007 

Yes 33 (14.3) 4 (1.7) 

Reason for hospital admission 

Infection (sepsis) 118 (51.1) 33 (14.3) 

< 0.001 
Post operation 7 (3) 14 (6.1) 

Burn 1 (0.4) 5 (2.2) 

Other medical cause 37 (16) 16 (6.9) 

Mechanical ventilation 
No 31 (13.4) 39 (16.9) 

< 0.001 
Yes 132 (57.1) 29 (12.6) 

Central catheter 
No 12 (5.2) 21 (9.1) 

< 0.001 
Yes 151 (65.4) 47 (20.3) 

Nasogastric tube 
No 31 (13.4) 36 (15.6) 

< 0.001 
Yes 132 (57.1) 32 (13.9) 

Note. ICU: intensive care unit, *Chi-square/Fisher’s exact test  



37 

The variables that were significantly associated in the univariate analysis were included 

in a multivariate logistic regression model (Table 7). The model showed that being 

admitted to the intensive care unit in the last 90 days was 2.65-times (95% CI = 1.00 to 

7.03) more likely to die compared to the patients who were not admitted to the intensive 

care unit in the last 90 days. In addition, the patients who had heart failure were 4.95-

times (95% CI = 1.64 to 14.97) more likely to die compared to the patients who did not 

have heart failure. Moreover, the patients who received central line catheter were 3.46-

times (95% CI = 1.03 to 11.61) more likely to die compared to the patients who did not 

receive central line catheter. 



38 

Table 7  

Multivariate logistic regression analysis of associations between survival/mortality outcomes 

with the other variables of the patients 

       

95% C.I. for 

OR 

Variable Category β S.E. Wald p OR Lower Upper 

Smoking 
No Reference category 

Yes -0.45 0.41 1.23 0.268 0.64 0.29 1.42 

Admission to ICU in 

the last 90 days 

No Reference category 

Yes 0.98 0.50 3.85 0.050 2.65 1.00 7.03 

Coronary artery 

disease 

No Reference category 

Yes 0.34 0.62 0.31 0.576 1.41 0.42 4.73 

Heart failure 
No Reference category 

Yes 1.60 0.56 8.04 0.005 4.95 1.64 14.97 

Hypertension 
No Reference category 

Yes -0.25 0.75 0.11 0.740 0.78 0.18 3.37 

Hyperlipidemia 
No Reference category 

Yes 0.44 0.71 0.39 0.530 1.56 0.39 6.23 

Diabetes mellitus 
No Reference category 

Yes 0.19 0.48 0.16 0.692 1.21 0.47 3.10 

Chronic kidney 

disease 

No Reference category 

Yes -0.36 1.12 0.10 0.747 0.70 0.08 6.24 

Liver disease 
No Reference category 

Yes 1.23 0.74 2.77 0.096 3.43 0.80 14.63 

Respiratory disease 
No Reference category 

Yes -0.21 0.49 0.18 0.669 0.81 0.31 2.11 

Sputum 
No Reference category 

Yes 0.13 0.48 0.08 0.782 1.14 0.45 2.92 

Wound 
No Reference category 

Yes 0.12 0.56 0.05 0.825 1.13 0.38 3.39 

Continuous renal 

replacement therapy 

No Reference category 

Yes 1.10 1.31 0.71 0.398 3.02 0.23 39.04 

Reason for hospital 

admission 

Infection Reference category 

Post 

operation 
-0.34 0.51 0.44 0.507 0.71 0.27 1.93 

Burn 1.24 0.82 2.26 0.133 3.45 0.69 17.36 

Other 

medical 

cause 

1.80 1.30 1.90 0.168 6.03 0.47 77.51 

Mechanical ventilation 
No Reference category 

Yes 1.43 1.01 2.01 0.156 4.19 0.58 30.34 

Central catheter 
No Reference category 

Yes 1.24 0.62 4.03 0.045 3.46 1.03 11.61 

Nasogastric tube 
No Reference category 

Yes 0.38 0.96 0.16 0.693 1.46 0.22 9.60 

Note. ICU: intensive care unit 



39 

On the other hand, the patients who died were significantly younger, had lower SOFA 

score, lower WBCs, higher platelets, and received treatment for a longer period. These 

details are shown in Table 8. 

Table 8 

Univariate analysis of associations between survival/mortality outcomes with age, WBCs, 

platelets, duration of hospitalization, and length of treatment  

Variable Survival/Mortality Median [Q1, Q3] p* 

Age 
Survived 66 [55, 75] 

< 0.001 
Died 55 [35, 69] 

SOFA 
Survived 12 [11, 13] 

< 0.001 
Died 8 [7, 10] 

WBCs 
Survived 16.8 [12.2, 22.7] 

0.005 
Died 12.9 [9.6, 20] 

Platelets 
Survived 166 [98, 231] 

< 0.001 
Died 261 [183, 299] 

Duration of mechanical 

ventilation (days) 

Survived 11 [4, 20] 
< 0.001 

Died 0 [0, 7] 

Length ICU stay (days) 
Survived 16 [10, 24] 

0.002 
Died 10 [6, 19] 

Length of hospital stay (days) 
Survived 18 [12, 26] 

0.779 
Died 19 [12, 26] 

Duration of therapy (days) 
Survived 8 [3, 14] 

0.047 
Died 12 [7, 14] 

Note. SOFA: sequential organ failure assessment score, WBCs: white blood cells, ICU: intensive care 

unit, Q1: first quartile, Q3: third quartile, *Mann-Whitney U test  

Similarly, the variables that were significantly associated in the univariate analysis were 

included in a multiple linear regression model (Table 9). The model showed that 

survival was significantly associated with higher SOFA, higher WBCs, lower platelets, 

longer duration of mechanical ventilation, duration of intensive care unit stay, shorter 

duration of therapy.  



40 

Table 9 

Multiple linear regression analysis of associations between survival/mortality outcomes with 

age, WBCs, platelets, duration of hospitalization, and length of treatment  

Variable 
Unstandardized 

Coefficients 
S.E. 

Standardized 

Coefficients 
t p 

Age 0.00 0.00 -0.07 -1.67 0.097 

SOFA -0.14 0.01 -0.66 -14.70 < 0.001 

WBCs -0.01 0.00 -0.09 -2.24 0.026 

Platelets 0.00 0.00 0.10 2.37 0.019 

Duration of mechanical 

ventilation 
0.00 0.00 -0.22 -3.16 0.002 

Length ICU stay -0.01 0.00 -0.43 -2.28 0.024 

Duration of therapy 0.01 0.00 0.22 3.44 0.001 

Note: SOFA: sequential organ failure assessment score, WBCs: white blood cells, ICU: intensive care 

unit, SE: standard error, t: t-statistics, outcome: mortality 

 



41 

Chapter Four 

Discussion

Acinetobacter baumannii is one of the most important threats in ICUs, especially for 

immunocompromised and critically ill patients. Its ability to survive in healthcare 

environments and increasing multidrug resistance, including pandrug-resistant strains, 

has been associated with increased transmission and high mortality rates. With the 

limited availability of treatment options, combination therapies are generally 

recommended, which further increases the need for better infection control and 

therapeutic strategies (50). 

The results show an overall incidence of A. baumannii infections of (8.2%) in medical 

ICU patients in three major governmental hospitals located in the West Bank of 

Palestine. This prevalence is significantly higher than the (3.37%) rate reported in a 

study from Saudi Arabia, indicating a public health concern in the region (51, 52). 

Incidence rates of infection varied between hospitals: (11.5%) at CICU, (6.5%) at NICU 

Hospital, and (5.3%) at SICU Hospital. This may be attributed to differences in 

infection control practices, the populations served, or the resources available to the 

hospitals. One study done in Taiwan found that ICU admission rates may vary between 

hospitals due to variations in infection control practices (53).  

The median age of the patients with the infection was 63 years; (64.1%) of them were 

men. This is in agreement with reports from other countries' current research that reveal 

older individuals and males get affected more due to A. baumannii (51). 

Hypertension was the most frequent comorbidity among the patients (71.9%), followed 

closely by hyperlipidemia (64.9%) and diabetes mellitus (58.9%). A very well-known 

risk factor for A. baumannii infections, and consequently for acquiring this disease, is 

the presence of multiple comorbidities, since these diseases can compromise the 

immune system and lead to susceptibility (54). 

The findings concerning the use of antibiotics, previous infections caused by multidrug-

resistant (MDR) strains, and episodes of hospitalization reveal important determinants 

related to the development of A. baumannii infections in critically ill patients. A 



42 

significant number (67.5%) of the subjects in this study had used antibiotics in the 

previous 90 days, and (57.6%) had previous infections caused by multidrug-resistant 

strains. The findings are in the line with the known association between recent antibiotic 

exposure and the risk of developing multidrug-resistant infections. Previous antibiotic 

therapy is known to disrupt the normal flora, hence predisposing to the colonization of 

resistant species, including A. baumannii (55, 56). 

Another important finding that can be drawn from the data is the increased use of 

certain antibiotics in patients across different healthcare settings. In contrast, patients 

from SICU Hospital had much exposure to piperacillin/tazobactam, while patients from 

NICU had more frequent exposure to colistin, linezolid, and rifampicin. This present 

finding signifies that regional differences in antibiotic usage may be the major cause 

resulting in resistance to A. baumannii. From literature, it is enlightened that A. 

baumannii develops resistance to commonly used antibiotics, specifically 

piperacillin/tazobactam. Also, treatment with colistin and rifampicin has been reported 

to have efficacy against the multidrug-resistant strains; however, resistance to these 

drugs is increasing at the same time (57, 58). 

The high percentages of previous hospitalization (64.5%) and intensive care unit 

admissions (24.2%) underlined the fact that critically ill patients were easily afflicted 

with infections due to A. baumannii. Nosocomial infections, particularly those caused 

by multidrug-resistant organisms, are more likely to develop in patients hospitalized and 

in the ICU, especially those with prolonged stays (59). The results are in line with 

previous studies showing that patients in the ICU are more prone to acquiring A. 

baumannii infections because of invasive procedures, prolonged antibiotic use, and 

frequent contacts with healthcare staff (60). 

Although there were no significant differences in antibiotic use rates or preceding MDR 

infections between hospitals, differences in specific antibiotic use and ICU admissions 

suggest that hospital-specific factors may influence the prevalence and spread of A. 

baumannii and suggests discrepancies in hospital antibiotic policies through the 

implementation of Antibiotic stewardship program. These findings strongly support the 

need for hospital-level infection control measures and tailored antibiotic stewardship 

programs to limit the spread of resistance organisms (61). 



43 

This will bring out critical results in relation to the challenges encountered in the 

treatment of such infections in ICUs. The majority of the infections, (80.1%), are 

acquired in the ICU setting; hence, it poses a very high risk of nosocomial infections in 

critically ill patients, especially for those who require longer ICU stays, more invasive 

procedures, and mechanical ventilation. This is consistent with other studies showing 

that ICU-acquired infections are more likely to be caused by MDR organisms, including 

A. baumannii (62). 

The most frequent sites of infection were sputum, wound, urine, and blood cultures. 

This was similar to the previous study of A. baumannii is often associated with 

respiratory infections, bloodstream infections, and wounds, especially in mechanically 

ventilated patients and those receiving invasive medical devices (63). The high rate of 

extended drug-resistant (XDR) strains, (66.7%) of the isolated samples, is of great 

concern. In relation, the rise in XDR A. baumannii has been linked to inadequate 

treatment options, thus placing a rising threat to health globally (64). 

The study has also demonstrated important diversities in resistance patterns across the 

hospitals. Indeed, patients in NICU Hospital showed a significantly higher proportion of 

XDR strains compared with the rest of the hospitals. This is indicative of the need for 

more localized epidemiological surveillance and specific infection control strategies in 

preventing the spread of resistant strains (61). This presents the variation in 

susceptibility to trimethoprim/sulfamethoxazole with the most striking finding from 

NICU, showing reduced susceptibility, and thus perhaps geographical differences in 

resistance findings possibly related to antibiotic stewardship strategies and the selection 

pressure from previous antibiotic treatments. 

Most patients were treated with an antibiotic average duration of 9 days; however, no 

significant difference was observed among the different institutes in the length of 

treatment. Although the causes of mechanical ventilation, central venous catheters, and 

urinary catheters were similar, their incidence was higher in patients treated at CICU, 

hence explaining the higher duration of mechanical ventilation in this group. The length 

of both ICU and hospital stays was longer at CICU compared with those at SICU and 

NICU Hospitals, reflecting the complex clinical course that developed in patients with 



44 

A. baumannii In general, A. baumannii infections and the intensity of pre-existing 

comorbid conditions usually require longer periods of critical care intervention (65). 

These findings underline the importance of good infection control practices, appropriate 

use of antibiotics, and close monitoring of critically ill patients in order to prevent A. 

baumannii infections. The geographic differences in resistance mechanisms and 

therapeutic strategies further underline the need for institution-specific approaches to 

fight the spread of resistant A. baumannii strains. 

The clinical and demographic variables that were found to have a significant influence 

on mortality risk among patients with Acinetobacter baumannii infections included 

smoking, recent ICU hospitalization, and comorbidities like coronary artery disease, 

heart failure, diabetes mellitus, and chronic kidney disease. The risks of invasive 

interventions, such as central venous catheterization, nasogastric tube insertion, and 

mechanical ventilation, showed an increased mortality rate, which corroborated 

previous studies from Europe and the Middle East that identified these as prominent risk 

factors (66, 67). Multivariate analysis showed that ICU admission in the last 90 days 

was an independent predictor of mortality (OR: 2.65), along with heart failure (OR: 

4.95) and use of central line catheter (OR: 3.46), underlining the strong influence of 

ICU-related exposures and pre-existing cardiovascular conditions on outcomes (68, 69).  

Younger patients with low SOFA scores and WBC counts but high platelet levels had 

high mortality, which was contrary to the prediction and may reflect complexity in 

disease progression and challenges in treatment. The relationship between longer ICU 

stays and mechanical ventilation underscores the importance of increased critical care, 

whereas the relationship between longer antibiotic use and increased mortality points to 

the need for preventing complications such as drug toxicity and resistance (70, 71). 

These findings flag the need for focused infection control measures, especially in 

institutions with a high prevalence. Implementation of strict infection control measures, 

including hand hygiene practices, environmental sanitation, and antimicrobial 

stewardship, will be needed to decrease A. baumannii infections (61). The results also 

have brought into the forefront the real need use invasive devices judiciously, and tailor 

therapeutic strategies to reduce mortality risks in patients with MDR A. baumannii 

infections at both regional and global levels. 



45 

4.1 Strengths of the study 

This study had several important strengths. The retrospective cohort design is an 

appropriate method for analyzing the results of a study conducted over a certain time 

period, as it enables researchers to use previously collected data to investigate the 

development and effects of A. baumannii infections over time. Additionally, the 

inclusion of three large public hospitals from different regions of the West Bank—

north, center, and south—adds diversity to the patient group and enhances the 

generalizability of the results. 

The study spans two years, from January 2019 to December 2020, providing a thorough 

understanding of A. baumannii infections before and after the COVID-19 pandemic. 

This timeframe offers valuable insights into how the pandemic influenced infection 

rates and outcomes. The inclusion and exclusion criteria are clearly defined, ensuring 

that the patient group is well-defined and relevant to the research. 

The data collection form is comprehensive, incorporating a wide range of 

sociodemographic, clinical, and treatment-related factors. This thorough approach 

enables an in-depth examination of the elements that influence A. baumannii infections 

and patient outcomes. The use of reliable statistical methods, such as Chi-square, 

Fisher's exact test, Student t-test, analysis of variance, Pearson's correlations, logistic 

regression models, and linear regression models, ensures that the findings are robust and 

trustworthy. 

The study specifically focuses on the challenging area of multidrug-resistant (MDR) 

and extensively drug-resistant (XDR) strains of A. baumannii. This is important for 

understanding resistance patterns and the effectiveness of treatments. Analyzing data 

from multiple hospitals also allows the identification of variations in infection rates, 

treatment approaches, and outcomes, which can guide hospital-specific actions and 

policies. 

By considering comorbidities and medication usage, such as corticosteroids and 

antibiotics, the study enhances its understanding of the underlying factors that affect 

patient outcomes and the complexity of treating A. baumannii infections. Furthermore, 

the research distinguishes between infections that were present at the time of admission 



46 

and those that developed during the hospital stay, which is crucial for understanding the 

epidemiology and transmission dynamics of A. baumannii. 

Finally, the study provides precise outcome indicators, such as the length of ICU and 

hospital stays, the duration of mechanical ventilation, and death rates. These indicators 

offer a comprehensive view of the impact of A. baumannii infections on both patient 

health and healthcare system resources 

4.2 Limitations of the study 

This study presents several limitations that should be considered. One of the potential 

issues is related to the inclusion criteria, which might unintentionally exclude 

significant cases, such as patients with incomplete medical records, or include cases that 

do not entirely reflect the target population, like patients who were admitted to the ICU 

but died on the same day. This could lead to selection bias, which in turn might affect 

the generalizability of the results. 

Another limitation is the study’s limited geographical scope. Although the research 

incorporates data from three hospitals in different parts of the West Bank, it may not 

fully represent other geographical regions or healthcare systems. This could restrict the 

ability to apply the findings to broader settings, particularly those with different 

healthcare infrastructures or patient populations. 

The quality and completeness of the data is another concern, as the study relies heavily 

on hospital records. This dependence on existing records may result in missing or 

incomplete data, which could compromise the reliability of the findings and lead to 

potentially skewed conclusions. Furthermore, despite employing statistical methods to 

control for various factors, there may still be unmeasured confounders influencing the 

results, such as differences in healthcare practices, environmental factors, or patient 

behaviors. 

Changes over time also present a challenge. The study was conducted between January 

2019 and December 2020, a period during which medical practices, antibiotic resistance 

patterns, and healthcare regulations likely evolved. These changes could affect the 

comparability of data collected at different points in time, making it difficult to draw 

consistent conclusions across the entire study period. 



47 

The generalizability of the study’s findings may also be limited. Given that the data 

comes from hospitals in the West Bank, it might not be directly applicable to other 

regions or countries with different healthcare systems, patient demographics, or local 

strains of A. baumannii. Additionally, there could be biases in how certain factors, such 

as comorbidities or medication usage, are reported in hospital records, leading to 

inaccurate data or omissions. 

The study also faces challenges in measuring outcomes. For example, the presence of an 

A. baumannii infection is not the only factor influencing outcomes such as the duration 

of mechanical ventilation or ICU stay. Other variables could affect these outcomes, 

making the interpretation of the data more complex. Moreover, pinpointing the exact 

source of infection can be difficult, particularly in cases where there are delays in 

diagnosis or incomplete infection control data. 

The timing of the research is another limitation. Conducted during the COVID-19 

pandemic, the study was influenced by significant changes in hospital operations, 

patient demographics, and infection control measures. This complicates the ability to 

distinguish the specific effects of A. baumannii infections from the broader impacts of 

the pandemic. 

Another potential issue is the risk of overfitting in the statistical models used. With 

multiple models applied, there is a possibility that the results could be overfitted, 

particularly if the sample size is too small relative to the number of variables being 

analyzed. Finally, the variability of resistance patterns poses a limitation. Since 

resistance patterns can fluctuate over time and vary across locations, the study’s 

findings regarding MDR and XDR A. baumannii may not be applicable to other regions 

or future periods where these patterns could be different 



48 

4.3 Recommendations 

A multifaceted approach is essential to address all risks related to A. baumannii 

infections. Strengthening infection control measures should be a priority, including 

strict hand hygiene, a clean environment, implementation of sterilization techniques, 

and isolation procedures for patients affected by multidrug-resistant organisms. 

Furthermore, promotion of antimicrobial stewardship by judicious use of antibiotics, 

monitoring of prescribing patterns, and adjustment of guidelines according to local 

resistance data is of the essence to contain the development of resistance. A 

multifaceted approach may also include the reduction in device use to minimize 

infection risk by avoiding central lines, catheters, and mechanical ventilation when not 

strictly indicated, as well as timely removal when unnecessary. It includes tailoring 

treatment modality, for example, combination antibiotic therapy based on susceptibility 

profiles and exploring alternative options in the case of XDR strains. This will require 

continuing surveillance of resistance trends at the hospital and regional levels, along 

with exploration into new therapeutic options, to provide information necessary for 

intervention strategies. Equally important is the need for professional education in 

infection control, antimicrobial resistance, and judicious antibiotic use, with increased 

collaborative multidisciplinary efforts to provide more complete and effective care for 

patients. 

4.4 Conclusion 

Infections caused by A. baumannii are a serious problem, especially in intensive care 

units, due to the antibiotic resistance of this pathogen, its environmental tenacity, and 

also due to an association with a high mortality rate. This study shows an alarming 

infection rate of (8.2%) of the ICU patients admitted to three West Bank hospitals, 

which is higher compared to neighboring countries. Principal risk factors, including 

comorbidities, recent ICU hospitalization, invasive procedures, and antibiotic exposure, 

make the management of critically sick patients a quite complex task. Geographic 

variations of resistance patterns therefore indicate the importance of targeted 

epidemiological monitoring in developing customized treatments. 



49 

List of Abbreviat