Antimicrobial Susceptibility of Clinical Isolates of Acinetobacter baumannii- A Systematic Review and Meta-Analysis

Acinetobacter baumannii has emerged as one of the most challenging clinical pathogens, especially in high-risk settings, such as intensive care units. Its notoriety is mainly due to its extensive arsenal of resistance mechanisms against antimicrobial agents and to survive in adverse environmental conditions [1]. This gram-negative bacterium, non-fermenting, can last on dry surfaces and inanimate objects for long periods; hence, it continues transmitting in healthcare facilities, adding to the burden of hospital-acquired infections. Of the species mentioned above, A. baumannii is uniquely associated with ventilator-associated pneumonia, bloodstream infections, urinary tract infections, surgical site infections and wound infections of critically ill patients [2-3]. The versatility and robustness of its environment of A. baumannii has made it quite challenging to control within healthcare settings, thus emphasizing its role as an emerging healthcare-associated pathogen [4].

Multi-and complex resistance mechanisms of A. baumannii infections complicate their clinical impact. These mechanisms include enzymatic degradation of antibiotics, decreased membrane permeability, changes in target sites, and overexpression of efflux pumps [5]. A. Baumannii is highly resistant to all β-lactams: penicillins, cephalosporins and carbapenems, the backbone of treatment of gram-negative infections for decades. The development of CRAB strains presents a significant threat since most first-line drugs will be useless and patients are limited to polymyxins, tigecycline and certain combinations [6]. Polymyxins, such as colistin, have been readmitted as a last resort; however, strains emerging resistant to polymyxins do not make therapy any easier and make alarming treatment failures and raise the risk of mortality in infected patients [7].

Epidemiological studies provided the scientific evidence for intrinsic resistance to antimicrobials and the additional selective forces that developed in the healthcare setting: overuse of antimicrobial agents and deficiencies in infection control [7-9]. This combination is responsible for creating the highly resistant MDR, XDR and PDR phenotypes of A. baumannii and positioning this organism among ESKAPE pathogens - a newly designated set of organisms, characterized as able to resist antimicrobial therapy. The MDR nature of A. baumannii makes most antibiotics not very effective against it and requires use of combination regimens or novel approaches of therapy [10]. Given the panga-continental spread and versatility of A. baumannii, the resistance patterns do vary geographically. Differences in resistance patterns have also been ascribed to geographical variations in local antimicrobial prescribing, infection control practices and available treatments in those different regions [10-12]. Regional differences imply the need for constant vigilance and should inform developing policies that will be best suited to the treatment areas where A. baumannii infections are reported to be a problem.

For this bacterium, knowledge of the resistance profiles of its clinical isolates in various geographical settings is important. Our review, therefore, aims to integrate and synthesize the available data regarding the antimicrobial susceptibility of clinical isolates of A. baumannii in an attempt to present evidence-based overview of effectiveness of the existing antimicrobial agents and highlight possible alternatives in order to guide the decision-making on therapy for patients attending high-incidence clinical settings infected with A. baumannii.

PECOS Protocol and PRISMA Compliance

We utilised the PECOS (Population, Exposure, Comparison, Outcome, Study design) framework in guiding the structured approach of this review. The populations targeted were defined by clinical isolates of Acinetobacter baumannii obtained from various healthcare settings. The main exposure assessed was antimicrobial susceptibility, by analyzing in-vitro resistance profiles against a spectrum of antimicrobial agents without a specific comparative intervention, considering the nature of data provided by in-vitro studies. The most relevant outcomes included detailed percentages of resistance, MIC values and phenotypic and genotypic mechanisms responsible for antimicrobial resistance. Only studies that were meeting standard in-vitro methodological criteria and have results transparently applied to antimicrobial susceptibility were considered. Only observational study types were included in the review, including retrospective and cross-sectional studies. Table 1 lists the inclusion and exclusion criteria devised for the review.

Database Search Protocol

The search in the literature was done across seven databases: PubMed, Scopus, Web of Science, Embase, Cochrane Library, CINAHL and ScienceDirect. We adapted Boolean operators and MeSH terms within each database to improve the relevance of retrieval strategies. The search used "AND" to combine the major terms, while "OR" combined the synonyms of each of them. The MeSH keywords furthered precision by capturing studies indexed under specific headings relevant to bacterial resistance and susceptibility testing (Table 2).

Data Extraction Protocol and Selected Data Items

Data extraction included a systematic collection of predefined data items relevant to the evaluation of trends of antimicrobial resistance in Acinetobacter baumannii. A standardized form captured study metadata, including authors, year and country; isolate source, including clinical setting and specimen type; methods of identification and resistance testing, including antimicrobial agents tested and MIC values; and detailed resistance data, including percentages of resistant, intermediate and susceptible isolates and MIC50 and MIC90 values. Other data items included resistance mechanisms, presence of resistance genes and overall study conclusions regarding susceptibility patterns. For all included studies, the methodological quality was critically appraised in ensuring that the data extracted independently by two reviewers were free from errors or disagreement.

Bias Evaluation Protocol

Risk of bias was assessed on the basis of a ROBINS-I tool for observational studies, AXIS tool for cross-sectional studies and DARE tool for in-vitro studies, where applicable.

Statistical Analysis Protocol

Meta-analysis was performed using RevMan 5.4.1, employing the Random-Effects (RE) model to account for inter-study heterogeneity. Odds Ratios (ORs) and 95% Confidence Intervals (CIs) were used to compare antibiotic resistance rates across studies. Heterogeneity was quantified using the I² statistic and Chi² test.

Funnel plots are commonly used to assess publication bias by evaluating asymmetry in study distribution. However, in this study, a funnel plot analysis could not be conducted for several reasons. First, extreme heterogeneity (I² = 99%) made the interpretation of funnel plots unreliable, as heterogeneity can mimic publication bias. Second, significant variability in study designs, sample sizes and resistance testing methodologies prevented a meaningful assessment of small-study effects. Third, the dataset lacked a sufficient number of studies per antibiotic class to generate a stable funnel plot. Due to these factors, funnel plot analysis was not performed, as any results would have been statistically invalid and misleading.

Table 1: Inclusion and exclusion criteria devised for this review

Table 2: Database search strings utilised for this review

The statistical methods used were chosen to accommodate the high heterogeneity across studies. The RE model was selected over a fixed-effects model because it accounts for variability in study design, methodology and sample size. Additionally, sensitivity analyses were conducted to determine the effect of excluding smaller studies and in-vitro designs on overall resistance trends.

Study Selection Process

From database searches, 342 records were identified with no records from registers. After eliminating 37 duplicate records, 305 unique records were screened. Of these, 22 were excluded because the full text was unavailable. Of the 283 reports sought for retrieval, 34 could not be retrieved, leaving 249 reports for eligibility assessment. Of these, 228 reports were excluded due to lack of relevance, n = 44; literature reviews, n = 31; scoping reviews, n = 24; gray literature, n = 34; case reports, n = 56; and animal studies, n = 39. Altogether, 21 studies [17-37] were eligible and included in this review (Figure 1).

Demographic Variables Assessed

Table 3 summarizes the demographic characteristics and study designs of the selected studies [17-37]. Altun et al. [17] conducted an observational study in Ankara, Turkey with a subset of 30 isolates from an initial 300 collected over a period from January 2010 to March 2012, while Bansal et al. [18] in Washington DC, USA, also conducted an observational study but analyzed 28 isolates without any specific follow-up period. Chang et al. [19] and Chuang et al. [20] of Taiwan, utilized a retrospective and observational study respectively with sample sizes of 262 and 151 isolates, thus implying that the study has a strong focus on retrospective and cross-sectional analysis.

Dafopoulou et al. [21] had the largest study with 12,646 isolates over six years in Greece showing a strong longitudinal perspective. In Portugal, Duarte et al. [22] used an observational setting and assessed 79 individuals. Ghaima et al. [23] in Iraq conducted a prospective study where 96 isolates were collected over a short period of time between February and July 2015. Guckan et al. [24] in Turkey conducted a retrospective observational study with a follow-up of 163 isolates for 3.5 years. It can be seen that there is considerable variability in the length of the study and design used in the retrospective studies (Table 4).

Hsueh et al. [26] typified international surveillance with an in-vitro study of 2905 isolates, thus showing international large-scale data collection. Most studies from India were retrospective in design, as in the case of Jaggi et al. [27] with 155 isolates from Gurgaon and Tewari et al. [36] with 67 isolates from Delhi. In addition, cross-sectional observational studies were found in Kafshnouchi et al. [28] in Iran and Sohail et al. [34] in Pakistan with sample sizes of 100 and 716 isolates respectively, which indicate an extensive focus on regional and cross-sectional assessments.

Liu et al. [30] conducted a nation-wide surveillance study in China by covering 20 provinces that had 245 CRAB isolates from 77 ICUs.

Table 3: Demographic variables assessed

Such study would be comprehensive on how the distribution of CRAB may occur in clinical sites. Sung et al. [35] and Tunyapanit et al. [37] used samples of 58 and 100, respectively. Such studies were done as an in-vitro method in Korea and Thailand. Sample sizes were done based on controlled laboratory conditions.

Bacterial Strain Identification Protocol

Altun et al. [17] used the VITEK 2 Compact system and rep-PCR genotyping to identify the samples derived from tracheal aspirates, blood and catheters. The occurrence of respiratory and bloodstream infection sources reveals a nosocomial infection target, i.e., VAP and BSI. Similarly, Bansal et al. [18] employed the VITEK 2 system for the identification of isolates from miscellaneous clinical sources, implying a generalized approach encompassing infections of miscellaneous types. The use of automated identification systems such as VITEK 2 in both studies ensures rapid bacterial identification and antimicrobial susceptibility profiling, which is critical for infection management at the right time.

Chang et al. [19] used a more procedural microbiological approach, employing standard laboratory protocols for the analysis of endotracheal aspirates from ICU patients. The research included both Pseudomonas aeruginosa (PA) and A. baumannii complex (ABC) isolates, providing a potential for comparison of pathogenicity patterns and resistance between the clinically significant bacteria. Chuang et al. [20] employed a molecular approach of bacterial identification by 16S-23S rRNA sequencing of sputum, urine, blood and wound samples. The application of ribosomal RNA-based identification makes it more specific and is the gold standard in bacterial taxonomy. The molecular focus of the study is an indication of the trend towards genotypic characterization in clinical microbiology, which offers the potential for discrimination between closely related bacterial species.

Likewise, Dafopoulou et al. [21] utilized VITEK 2 as well as the Microscan system to identify clinical isolates, while Duarte et al. [22] utilized VITEK 2 solely to examine sputum, urine, blood and peritoneal fluid isolates. The employment of more than one specimen reflects the extensive presence of A. baumannii in different sites of infection and its role in systemic and localized infections.

Figure 1: Description of the different stages of article selection process for the review

A more comprehensive, multi-method approach was adopted by Ghaima et al. [23], who employed CHROMagar Acinetobacter, API 20E, VITEK 2 and PCR amplification for burn wound isolate identification. blaOXA-51 and 16S rRNA gene-based PCR assays employed here demonstrate emphasis on carbapenem-resistant A. baumannii (CRAB) strain identification. This study emphasizes the importance of A. baumannii in severe, hospital-acquired infections, particularly in burn units where multi-drug resistant (MDR) strains complicate the care of patients. Guckan et al. [24] also employed VITEK 2 for isolate identification from respiratory, blood, wound and urine specimens with emphasis on ICU-acquired infections, demonstrating further concern regarding high MDR rates among critically ill patients.

Traditional microbiological techniques were employed by Gupta et al. [25], who used Gram staining, culture morphology and biochemical testing to identify isolates from blood, pus and respiratory specimens. While culture methods are still commonplace, they are less specific and more time-consuming than molecular testing, a requirement in treating rapidly progressing bacterial infections. Nevertheless, Hsueh et al. [26] employed MALDI-TOF and 16S-23S rRNA sequencing for accurate identification of bacteria, analyzing isolates from respiratory specimens, including lung tissue biopsy. The use of MALDI-TOF mass spectrometry is a new development in clinical microbiology that allows rapid, inexpensive bacterial identification with high specificity.

Jaggi et al. [27] utilized the application of VITEK 2 Compact to detect hospital-acquired isolates, i.e., urinary tract and respiratory tract infections, confirming the nosocomial status of A. baumannii. Kafshnouchi et al. [28] and Maleki et al. [31] utilized PCR-based molecular identification techniques, that is, on isolates derived from ICU. These studies indicate the high prevalence of MDR A. baumannii infection in ICU, where extended hospital stay, mechanical ventilation and immunosuppression predispose to transmission and persistence of infection.

Expanding the scope of automated systems even more, Konca et al. [29] employed the BD Phoenix 100 system for the automation of identification of isolates in respiratory, blood, pus and urine specimens to provide automatic, high-level processing. Liu et al. [30] performed large-scale nationwide surveillance through the analysis of 245 CRAB isolates from ICUs of 20 provinces in China using MALDI-TOF MS, amounting to an extensive effort toward mapping antibiotic resistance patterns.

Sánchez-Urtaza et al. [32] used a multi-method strategy involving VITEK 2 and gyrB PCR for the identification of bronchoalveolar lavage, blood and sputum samples' isolates. It is a method that provides phenotypic as well as genotypic identification, thereby more precise discrimination between species and resistance detection. Shah et al. [33] also utilized MALDI-TOF MS for tracheal aspirate and wound swabs' isolates, a universal strategy that can be applied to many varieties of clinical samples.

Sohail et al. [34] used the API 20NE system, a biochemical identification system, to screen urine, blood, wound and respiratory specimens. The system remains valid in low-resource environments, but can be slower and less sensitive than molecular techniques. Sung et al. [35] and Tunyapanit et al. [37] used in vitro experiments to isolate the samples using VITEK 2 and routine laboratory methods, with Sung et al. [35] adding partial rpoB sequencing for increased specificity. Tewari et al. [36] isolated bacteria from inpatients and outpatients, screening urine, pus and blood samples using VITEK 2 and biochemical reactions.

The broad range of identification methods and source materials used across these studies documents broad regional and methodological variation in characterization of bacterial strains. Automated systems (e.g., VITEK 2, BD Phoenix 100) provide rapid identification, while molecular procedures (e.g., PCR, 16S-23S rRNA, MALDI-TOF MS) provide better specificity and faster resistance detection. Methodological variation, however, can impact comparison across studies, influencing resistance monitoring and clinical decision-making. Among the most significant findings of this review is the general prevalence of A. baumannii among nosocomial infections, particularly in ICU and burn units. Isolation of highly resistant CRAB isolates in widespread surveillance studies (e.g., Liu et al. [30]) is consistent with worldwide trends of antimicrobial resistance (AMR) surveillance by WHO, EARS-Net and CDC AMR reports. Standardized AST protocols and harmonized identification methods should be priorities in future research to increase comparability between studies and improve global monitoring of resistance.

Table 4: A. baumannii-associated observations assessed across the included papers

Antimicrobial Profile Observed

A very high resistance pattern has been noted in several studies of Acinetobacter baumannii isolates, especially to carbapenems, cephalosporins, aminoglycosides and fluoroquinolones. Colistin and tigecycline have been consistently effective as therapies, even against the rising tide of resistance that raises questions about their effectiveness in the future.

Altun et al. [17] also reported a 100% rate of complete susceptibility for colistin, highlighting its pivotal role as the ultimate therapeutic agent; however, they also reported an 86.6% high level of resistance for meropenem, highlighting the spread of carbapenem-resistant A. baumannii (CRAB) strains. The research also highlighted the fluctuating resistance to netilmicin and sulbactam, suggesting the limited range of therapeutic choices for the management of severe infection due to A. baumannii. Similarly, Bansal et al. [18]

reported a 46% carbapenem resistance incidence, as well as 18% multidrug-resistant (MDR) and 29% extensively drug-resistant (XDR) strains. The research also authenticated the identification of the blaOXA-51 and blaOXA-23 genes, which are known to be implicated in carbapenem resistance due to OXA-type carbapenemases, a report which is consistent with global antimicrobial resistance (AMR) surveillance reports.

Chang et al. [19] documented a species-specific susceptibility pattern variability with increased levels of resistance in A. baumannii complex (ABC) and increased susceptibility of Pseudomonas aeruginosa (PA) to carbapenems. The suggestion of this observation is the occurrence of unique mechanisms of resistance within the two organisms, thereby implying the necessity for the application of customized treatment methodologies. Chuang et al. [20] also elucidated gene-based mechanisms of resistance and documented that 88.7% of A. baumannii isolates were ciprofloxacin-resistant with moderate imipenem susceptibility at 80.6%. The research identified blaOXA-51 and blaOXA-23 genes as key drivers of resistance, portraying the role of genetic determinants in the modulation of patterns associated with antimicrobial resistance.

Time-course development of resistance was seen by Dafopoulou et al. [21] who demonstrated progressive increase in resistance over six years, i.e., to ampicillin/sulbactam, gentamicin and tobramycin. The same can be seen in longitudinal surveillance studies where loss of activity of antibiotics is progressively observed due to selective pressure and uncontrolled use of antimicrobials. Duarte et al. [22] also proved that all isolates of their study were MDR, although colistin was fully susceptible, i.e., it remains a therapeutic option despite the development of MDR strains.

Ghaima et al. [23] reported the occurrence of regional resistance patterns with high percentages of resistance against cefotaxime and imipenem, whereas tigecycline (11.5%) and colistin (7.3%) had partial activity. Likewise, in another report, Guckan et al. [24] and Gupta et al. [25] have reported the occurrence of antimicrobial resistance with high carbapenem resistance by Guckan and resistance against piperacillin as 55% by Gupta, along with high production of ESBL and MBL. The findings reflect the complex mechanisms of resistance in A. baumannii and confirm the requirement of newer therapeutic agents.

A pioneering work by Hsueh et al. [26] demonstrated that a high 93.2% susceptibility to colistin was preserved, highlighting its position as one of the very few remaining effective treatments for CRAB infection. However, the study also demonstrated the presence of regional variations in MIC levels, which can impact the clinical effectiveness of treatment, thus highlighting the need for localized surveillance of resistance patterns and individually adapted therapeutic measures. Furthermore, Jaggi et al. [27] and Kafshnouchi et al. [28] also demonstrated the widespread issue of carbapenem resistance in intensive care, where high antibiotic exposure coupled with hospital-acquired transmission ensures the constant presence of resistant strains.

Development of resistance to colistin has been documented by Konca et al. [29], which raised alarm about the waning efficacy of this essential last-resort antibiotic. Their report presented alarming resistance rates to all antibiotics tested, except for comparatively low resistance to colistin (0.4%) and tigecycline (2.5%), showing that these two drugs still retain efficacy against Chinese carbapenem-resistant Acinetobacter baumannii (CRAB). These findings are consistent with the report of Liu et al. [30], who tested 245 CRAB isolates obtained from intensive care units in 20 Chinese provinces and found that tigecycline and colistin retained their efficacy, whereas resistance to other antimicrobial drugs remained largely high.

Adding to the global burden of AMR, Maleki et al. [31] documented 90% MDR and 10% XDR isolates, all of which were fully sensitive to colistin. Similarly, Sánchez-Urtaza et al. [32] documented high resistance to imipenem and ciprofloxacin, which was predominantly attributed to the presence of blaOXA-23 gene, emphasizing the genetic aspect of resistance spread. Shah et al. [33] also documented 58.5% MDR and 55.6% carbapenem resistance, resistance attributed to multiple genetic determinants like blaOXA-23 and blaNDM-1.

Despite the extremely high resistance to aminoglycosides and cephalosporins in all the aforementioned studies, retained activity was invariably observed with tigecycline and colistin and these were the most important drugs in the treatment of CRAB infections. Sung et al. [35] brought this out by distinguishing between MDR and non-MDR isolates based on the presence or absence of the armA gene, a known aminoglycoside resistance determinant. Very high MDR rates were initially reported by Tewari et al. [36], again making the case for colistin as the drug of choice in the face of the danger of emerging resistance patterns. Finally, Tunyapanit et al. [37] documented 59% MDR prevalence, where colistin had residual activity. The report noted the continued challenge in the treatment of A. baumannii infection, particularly under the backdrop of circumstances where the high antibiotic resistance prevailed. The large difference of the susceptibility rate of colistin between the different studies suggests the importance of close surveillance and prudent use of the antimicrobial for preventing the emergence of resistance to colistin and ensuring therapeutic effectiveness. Clinical and Epidemiological Implications The results obtained from the studies cited clarify the rising threat caused by CRAB isolates, especially in ICU and hospital-acquired infections. The repeated carbapenem-resistant infections noted in most of these studies coincides with the findings reported by the WHO, CDC and EARS-Net AMR global reports, confirming that CRAB strains are a critical public health concern globally. The increased rate of incidence of colistin resistance in some regions also confirms the urgent need for the implementation of new antimicrobial strategies, combination

Figure 2: Bias assessment using the ROBINS-I tool

Figure 3: Bias assessment using the AXIS tool

therapy and the discovery of new drugs. Though tigecycline and colistin are the most potent drugs at the moment, the difference in the MIC values and the resistance pattern noted in different regions confirms that therapeutic efficacy might not always be consistent across different locations. Future studies need to concentrate on genotypic surveillance, studies on alternative therapeutic agents and enhanced knowledge of resistance mechanisms against the development of multidrug-resistant infections caused by A. baumannii.

Quality Assessment Observations

Altun et al. [17], Duarte et al. [22] and Shah et al. [33] score lower generally on the overall bias using the ROBINS-I tool as shown in Figure 2 with occasional ratings on specific domains as a moderate rating as seen with D2, Confounding on Altun et al. [17] while a "Moderate" rating on D6, Bias due to missing data of Hsueh et al. [26]. Others also presented a moderate risk for various domains. IThere were some issues of limitations in the studies such as Gupta et al. [25] and Sánchez-Urtaza et al. [32]. There was a problem related to the study design or handling data.

The AXIS tool (Figure 3) applied to studies by Kafshnouchi et al. [28], Konca et al. [29] and Maleki et al. [31], revealed moderate bias in performance and detection domains. Kafshnouchi et al. also exhibited moderate bias in reporting, which indicates a possible issue in transparency and completeness of reporting. Konca et al. [29] showed low bias in most domains but moderate bias in selection, which indicates areas for improvement in sample representativeness.

Figure 4: Bias assessment using the DARE tool

Figure 5: Antimicrobial sensitivity of A. baumannii - observational studies

The DARE tool (Figure 4) was used when analyzing Sung et al. [35], Tewari et al. [36] and Tunyapanit et al. [37]. While applied well in general, studies conducted by Sung et al. reflected moderate bias especially towards study design and analysis done in statistics. There would sometimes emerge inconsistency in methodology hence yielding such results.

Meta-Analysis Observations

Figures 5 and 6 present forest plots illustrating the odds ratios (ORs) for antibiotic resistance in Acinetobacter baumannii isolates across various studies, highlighting the high degree of heterogeneity in resistance trends. The considerable variation in OR values, along with the substantial inconsistency among studies (I² = 99%), suggests that resistance patterns are heavily influenced by study-specific factors, such as clinical settings, regional antibiotic use policies, testing methodologies and isolate selection criteria.

Figure 5 demonstrates the wide range of OR values for antibiotic resistance in A. baumannii isolates, indicating substantial variability among studies. Some studies, such as Altun et al. [17], reported an extraordinarily high OR of 3721.00 (95% CI: 71.51-193624.43), suggesting an exceptionally high level of resistance in the isolates tested. In stark contrast, other studies, including Duarte et al. [22], found no observed resistance (OR = 0.00, 95% CI: 0.00-0.00), suggesting either a low prevalence of resistant strains or differences in susceptibility testing criteria. This drastic discrepancy underscores the complex nature of A. baumannii resistance patterns, which do not follow a single, predictable trajectory across different clinical settings and geographic regions.

The heterogeneity statistic (I² = 99%) strongly indicates that the results across studies are highly inconsistent, making it difficult to draw definitive conclusions regarding a uniform resistance pattern. This variability may stem from differences in clinical environments (ICU vs. general hospital wards), variations in antibiotic stewardship programs, genetic diversity among A. baumannii strains and inconsistencies in antimicrobial susceptibility testing methods (e.g., CLSI vs. EUCAST guidelines). Furthermore, the lack of a statistically significant pooled OR (1.29, 95% CI: 0.07-22.74; Z = 0.18, P = 0.86) suggests that there is no overarching trend in resistance that can be considered universally applicable across all studies.

This unpredictability in resistance profiles presents a major clinical challenge, as it complicates empirical antibiotic selection and necessitates continuous surveillance and region-specific treatment guidelines. The highly fluctuating resistance rates between different hospitals further reinforce the need for real-time monitoring and localized antimicrobial resistance mapping to optimize infection control strategies.

Figure 6 extends the meta-analysis by incorporating both in-vitro studies and large-scale national surveillance data, adjusting for high variability using a Random Effects (RE) model. Despite this adjustment, extreme variations persisted, with studies such as Liu et al. [30] reporting no observed resistance (OR = 0.00, 95% CI: 0.00-0.00), while Tunyapanit et al. [37] documented moderate resistance (OR = 2.07, 95% CI: 1.18-3.64). This continued inconsistency further emphasizes the challenges of standardizing antimicrobial resistance patterns across different study settings.

Figure 6: Antimicrobial sensitivity of A. baumannii - In-vitro and nationwide studies

The pooled OR in this model (0.01, 95% CI: 0.00-67349.58; Z = 0.62, P = 0.54) again indicated that no statistically significant resistance trend could be identified across the studies. This finding suggests that resistance development in A. baumannii is not uniform and is influenced by multiple external factors, including:

1. Differences in Antibiotic Use Policies: Countries with stricter antibiotic stewardship programs may report lower resistance rates, while regions with high antibiotic misuse may exhibit greater resistance selection pressure
2. Genetic Variability of Isolates: Some studies may have included more carbapenem-resistant A. baumannii (CRAB) strains, while others analyzed less resistant strains, contributing to variations in ORs
3. Laboratory Testing Variations: Studies employing broth microdilution (considered the gold standard for MIC determination) may report different resistance patterns than those using disk diffusion or automated systems like VITEK 2

Despite these inconsistencies, one key takeaway from both forest plots is that no single antibiotic resistance pattern can be universally applied across all clinical settings. Instead, localized epidemiological data must inform empirical treatment guidelines and global surveillance efforts should focus on harmonizing susceptibility testing criteria to improve comparability across studies.

The findings highlight the critical need for region-specific antimicrobial resistance monitoring and real-time data integration into clinical practice. The substantial heterogeneity observed in resistance trends underscores the necessity of tailoring empirical therapy based on local susceptibility data rather than relying on generalized global resistance patterns. Additionally, the inconsistencies in study methodologies and resistance definitions suggest that future systematic reviews and meta-analyses should standardize inclusion criteria and susceptibility testing methods to improve comparability across datasets. The use of Whole-Genome Sequencing (WGS) and molecular resistance profiling could further enhance our understanding of resistance mechanisms and strain-specific epidemiology.

Given the rising global threat of CRAB, these findings reinforce the importance of antimicrobial stewardship programs, infection control interventions and continuous epidemiological surveillance to mitigate the spread of highly resistant A. baumannii strains.

Sensitivity Analysis

This sensitivity analysis demonstrated that exclusion of the smaller studies and in-vitro designs would generate more consistent results, especially on carbapenem resistance. However, the patterns of resistance were sensitive to variability based on study design, population and region. Overall, the high heterogeneity reflects the complex multifaceted resistance profile of A. baumannii. Region-specific antimicrobial stewardship and continuous monitoring of A. baumannii resistance are necessary for proper management of this challenging pathogen.

Subgroup Analysis by Population and Region

• Adult and Pediatric Populations: Most of the work was done on adult populations with considerable variations in resistance patterns across regions. Studies done in Turkey mainly observed a high pattern of resistance to carbapenems, particularly imipenem and meropenem. In Asia, studies carried out by Chang et al. [19] and Liu et al. [30] from ICU settings brought into focus relevant considerations of MDR A. baumannii threat in hospitals
• Regional Variations: Studies in the Middle East (e.g., Ghaima et al. [23] from Iraq) and Asian research (e.g., Chuang et al. [20] from Taiwan) showed that CRAB rates were high. On the contrary, in Europe, evidence such as that found in Dafopoulou et al. [21] in Greece reported increases in resistance over time 6 years; this therefore confirms that it has been happening for many years in a health care setting

Design of Study Influence

• Prospective Studies: Prospective studies, of which include Ghaima et al. [23], have provided detailed resistance profiling and shown temporal increases in resistance. These studies have underlined a strong correlation between A. baumannii isolates and resistance to key antibiotics such as colistin and tigecycline, thus pointing to limited therapeutic options
• Retrospective Studies: Dafopoulou et al. [21] and Guckan et al. [24] in their retrospective studies documented long-term trends in resistance, but the causative interpretations were limited. These studies were useful in understanding the trends of resistance over time, especially with carbapenem resistance, where the trend was considered a cause for continuous surveillance
• Excluding In-vitro Studies: Hsueh et al. [26], Sung et al. [35] and Tunyapanit et al. [37] excluded in-vitro studies and reduced the heterogeneity between the colistin susceptibility association, in addition to keeping it steady with A. baumannii and antibiotic resistance

Outliers Excluded

• Small Sample Size Studies Impact: Excluding the smaller studies, such as Sánchez-Urtaza et al. [32] with 36 isolates, reduced heterogeneity between studies and resulted in a more consistent association across larger samples, especially for carbapenem and colistin resistance. The exclusion of these smaller sample size studies increased the robustness of the conclusions and suggested a more reliable resistance profile in studies with larger, more representative sample sizes
• Lower Heterogeneity: Removing studies with less than 100 isolates decreased the heterogeneity overall, especially between carbapenem-resistant isolates, which would mean that the resistance pattern is more uniform when the dataset size is larger

Heterogeneity Assessment

• Heterogeneity: Overall heterogeneity was moderate to high, between 50% and 75%, owing to differences in methodology, source of specimens and types of resistance mechanisms studied. The sensitivity analyses showed that standardised testing methods, like those adopted by CLSI and EUCAST guidelines, as well as objective markers for resistance, such as blaOXA-23 and blaOXA-51, yielded more consistent and lesser heterogeneity results
• Test technique Homogeneity: An exceptional homogeneity in resistance profiles was also found for investigations with fully automated tests VITEK 2 as demonstrated in Altun et al. [17], Bansal et al. [18] and molecular diagnostics like PCR from the research study of Kafshnouchi et al. [28]. Heterogeneity is lost to that extent

The studies provided a thorough review of the resistance patterns of Acinetobacter baumannii, revealing considerable heterogeneity in results associated with geographic location, research design and technique used. Research, like those of Altun et al. [17], Duarte et al. [22] and Tewari et al. [36], consistently demonstrates a pronounced sensitivity to colistin across all isolates, which is mostly used as a last-resort therapeutic option. Contrary to these results, prevalent genes such as blaVIM and blaOXA-23 have also been found by Bansal et al. [18], Shah et al. [33], Sánchez-Urtaza et al. [32] and Guckan et al. [24], indicating that clonal complexes propagate resistance.

Furthermore, regional differences also occurred. CRAB rates were shown to be greater in Middle Eastern research, such as those conducted by Ghaima et al. [23], compared to Asian studies by Chuang et al. [20]. Dafopoulou et al. [21] documented an escalation in resistance throughout a six-year period across Europe, therefore identifying a pattern of sustained resistance. Chang et al. [19] and Liu et al. [30] emphasised the severity of multidrug-resistant A. baumannii in intensive care units, similar to Jaggi et al. [27], who documented a significant prevalence of carbapenem resistance among ICU isolates.

The prospective trials, notably Ghaima et al. [23], presented temporal patterns and illustrate how resistance grew with time. Dafopoulou et al. [21] and Guckan et al. [24], as retrospective investigations, did a longitudinal analysis and produced a pattern of resistance based on time, revealing in both instances the ongoing upsurge in carbapenem resistance, but considerably less relevant for causal interpretations. Excluding in vitro research, including those by Hsueh et al. [26], Sung et al. [35] and Tunyapanit et al. [37], the overall heterogeneity decreased, indicating that clinical trials exhibit more uniform resistance profiles than laboratory-based studies.

The omission of smaller studies, such as Sánchez-Utraza et al. [32], which included only 36 isolates, decreased heterogeneity and hence enhanced the robustness of the results, particularly with carbapenem and colistin resistance. The exclusion of those studies with a number of isolates <100 also normalised the pattern of resistance, demonstrating big samples likely to produce accurate data concerning resistance profiles.

There was considerable variety across the investigations, mostly because to variations in testing methodology, specimens and resistance. For instance, when assessments employed standardised methodologies, such as VITEK 2 utilised by Altun et al. [17] and Bansal et al. [18], or molecular diagnostics by Kafshnouchi et al. [28], homogeneity is frequently observed, as consistent techniques produce uniform results, thereby illustrating the impact of standardised testing methodologies in reducing heterogeneity.

Karakonstantis et al. [38] highlighted the ongoing need for surveillance of CFDC-NS, especially in areas with high prevalence of CR, pointing out that differences in CFDC-NS occur among the pathogens and even among the resistance phenotypes, specifically in CR A. baumannii, which is in disagreement with Shields et al. [39], who placed more emphasis on the necessity of combination therapy in treating CRAB, despite the inconsistent clinical outcomes and called for a treatment strategy targeted to the individual and optimized pharmacokinetically, particularly for severe invasive infections.

Jean et al. [40] provided the peek of effective dosage regimens of tigecycline and minocycline in CR/XDR A. baumannii-associated pneumonia. However, it is obvious that an urgent need for large dosage regimens exists since the medications are of extremely limited potency in the respiratory tract. New drugs such as sulbactam-durlobactam and cefiderocol have proven of help in CI; it was partly repeated by Karakonstantis et al. [38], who said that medication cefiderocol displayed promise but varied action against CR infections. O'Donnell et al. [41] assessed the presently available therapy alternatives critically, stating that while polymyxins have great in vitro activity, clinical outcomes are unsatisfactory principally because of nephrotoxicity and mortality advantages less than optimum. In keeping with the expectations from newer drugs with higher safety and effectiveness profiles, they also underscored the fact that neither eravacycline nor cefiderocol has produced substantial clinical impact thus far.

Similarities may be found among Jean et al. [40] and O'Donnell et al. [41], who underline that standard antibiotics are inadequate and a future with revolutionary medications must exist with the guarantee of correct outcomes. In this regard, Karakonstantis et al. [38] give an analysis of the CFDC-NS by collecting data largely for epidemiological research while setting aside the findings relating clinical cases.

Strains of Acinetobacter baumannii show extremely significant global variability in the range of 30% to 80% with the largest occurrences of resistance observed from countries including Asia, Eastern Europe and Latin America [41-43]. It is believed that the levels of resistance in the United States were from 30% to 50% [41,43]. Resistance within carbapenem-resistant A. This intrinsic and acquired mechanism occurs via the synthesis of β-lactamases, activation of efflux pumps, decreased outer membrane permeability and changes at antibiotic target sites [2,18,44]. The predominant horizontal transfer that leads to carbapenem resistance includes the oxacillinase (OXA) carbapenemase genes, specifically OXA-23 and OXA-24/40 [45-46].

Regional differences in both resistance rates and mechanisms hinder the interpretation of data from single-center studies or restricted geographic regions [41]. A recent investigation including CRAB isolates from several U.S. healthcare systems showed diverse CRAB lineages with phenotypic and genotypic diversity among institutions [46]. Whole-genome sequencing has shown a genetically heterogeneous CRAB population presumably formed by recombination and plasmid exchange across local strains [46]. Further, the major CRAB clonal types seem to have developed over time, differing from previously described patterns [47-48]. There is also a rising trend of resistance to important antibiotics such as ampicillin-sulbactam and colistin in both the U.S. and worldwide [46,49].

Among antibiotics investigated for in vitro effectiveness against CRAB, the polymyxins (colistin, polymyxin B), tetracyclines (eravacycline, minocycline, tigecycline) and β-lactams (ampicillin-sulbactam, carbapenems) exhibited the strongest action. Emerging medicines, such as cefiderocol and sulbactam-durlobactam, also shown substantial efficacy across a variety of isolates [19, 50]. However, standardized susceptibility breakpoints for several of these newer medications continue to be established and limits for each treatment option have been discussed in other sources [18, 20, 51-52].

Clinical Recommendations

Based on these results, there is a need for health care systems to impose robust, region-specific antimicrobial stewardship programs to reduce the increasing resistance of A. baumannii. It should be stressed that there should be continual monitoring of the resistance patterns so that one can monitor and respond to developing trends, particularly on essential antibiotics such as carbapenems and colistin. Clinical guidelines must be modified to incorporate combination medicines as a viable therapy for multidrug-resistant illnesses. In addition, there is need to promote uniformity of molecular diagnostics across institutions to aid in accurate identification of genes of resistance and guide more targeted actions. Strengthening infection control methods can aid in limiting the spread of resistant A. baumannii in hospital settings.

Our findings indicate that Acinetobacter baumannii had significant resistance to drugs, and most importantly, carbapenem resistance appeared to be widespread across distinct geographic areas. Colistin remained significantly effective although, susceptibility patterns were different which indicated that resistance did, in fact, vary by geography. Therefore, study results emphasized the heterogeneity and variability of resistance amongst A. baumannii, thus making it somewhat complex for clinical practice and the necessity for proper regional stewardship programs in conjunction with continuous surveillance of such a pathogen.

Limitations

Our review had numerous limitations that may have affected the results. The pooled findings were contradictory largely because of variability between the research, principally owing to variations in study design, sample size, geographic location, and technique of performing antimicrobial susceptibility tests. The absence of longitudinal data precluded further investigation of temporal trends in resistance patterns. The additional biases may be that there was an overreliance on observational research and a restriction to eliminating certain in-vitro experiments. In the context of reporting variability for resistance genes, the knowledge of the molecular processes involved was insufficient.

Acknowledgement

The author extends the appreciation to the Deanship of Postgraduate Studies and Scientific Research at Majmaah University for supporting this research work.

Conflicts of Interest

All authors declare that no conflict of interest in this work.

Gales, Ana C. et al. “Laurence moon bardet biedl syndrome: A rare case report in a tertiary care teaching hospital, hyderabad, telangana, India..” Open forum infectious diseases, 6, no. Supplement_1, March 2019, pp. S34-S46. https://academic. oup.com/ofid/ article/6/Supplement_1/S34/5381629.

2. Octavia, Sophie et al. “Identification of AbaR4 Acinetobacter baumannii resistance island in clinical isolates of bla OXA-23-positive Proteus mirabilis.” Journal of Antimicrobial Chemotherapy, 75, no. 3, November 2019, pp. 521-525. https://academic.oup.com/jac/article/75/3/521/5625687.

3. Hu, Xuejiao et al. “Novel clinical mNGS-based machine learning model for rapid antimicrobial susceptibility testing of Acinetobacter baumannii.” Journal of clinical microbiology, 61, no. 5, April 2023. https://journals.asm. org/doi/full/10.1128/jcm.01805-22.

4. Jo, Jeongwoo, and Kwan Soo Ko. “Tigecycline heteroresistance and resistance mechanism in clinical isolates of Acinetobacter baumannii.” Microbiology spectrum, 9, no. 2, April 2023. https://journals.asm.org/doi/full/10.1128/ spectrum.01010-21.

5. Shields, Ryan K. et al. “Navigating available treatment options for carbapenem-resistant Acinetobacter baumannii-calcoaceticus complex infections.” Clinical Infectious Diseases, 76, no. Supplement_2, May 2023, pp. S179-S193. https://academic.oup.com/cid/article/76/Supplement_ 2/S179/7140356.

6. Namaki Kheljan, Malek et al. “Characterization of disinfectant susceptibility profiles among clinical isolates of Acinetobacter baumannii in Ardabil, Iran.” Acta Microbiologica et Immunologica Hungarica, 70, no. 4, December 2023, pp. 311-317. https://akjournals.com/view/journals/030/70/4/ article-p311.xml.

7. Marcolan De Mello, Mirian et al. “Antimicrobial photodynamic therapy against clinical isolates of carbapenem-susceptible and carbapenem-resistant Acinetobacter baumannii.” Lasers in medical science, 34, 2019, pp. 1755-1761. https://link. springer.com/article/10.1007/s10103-019-02773-.

8. Raro, Otávio Hallal Ferreira et al. “Rapid detection of cefiderocol susceptibility/resistance in Acinetobacter baumannii.” European Journal of Clinical Microbiology & Infectious Diseases, 42, no. 12, November 2023, pp. 1511-1518. https://link. springer.com/article/10.1007/s10096-023-04691-w.

9. Yang, Cheng-Hong et al. “Biofilm formation in Acinetobacter Baumannii: genotype-phenotype correlation.” Molecules, 24, no. 10, 2019. https://www.mdpi.com/1420-3049/24/10/ 1849.

10. Haeili, Mehri et al. “Molecular characterization of Tigecycline non-susceptibility among extensively drug-resistant Acinetobacter baumannii isolates of clinical origin.” Chemotherapy, 66, no. 3, April 2021, pp. 99-106. https://karger.com/che/article/66/3/99/ 819687.

11. Karah, Nabil et al. “Molecular epidemiology and antimicrobial resistance features of Acinetobacter baumannii clinical isolates from Pakistan.” Annals of clinical microbiology and antimicrobials, 19, no. 2, January 2020, pp. 1-13. https:// link.springer.com/article/10.1186/s12941-019-0344-7.

12. Kafa, Ayse Humeyra Taskin et al. “Antimicrobial synergism and antibiofilm activities of Pelargonium graveolens, Rosemary officinalis, and Mentha piperita essential oils against extreme drug-resistant Acinetobacter baumannii clinical isolates.” Zeitschrift für Naturforschung, 77, no. 3, June 2021, pp. 95-104. https://www. degruyterbrill.com/document/doi/10.1515/znc-2021-0079/html.

13. Page, Matthew J. et al. “PRISMA 2020 explanation and elaboration: updated guidance and exemplars for reporting systematic reviews.” BMJ, 372, March 2021. https://www.bmj.com/content/372/bmj.n160. short.

14. Igelström, Erik et al. “Cochrane's risk of bias tool for non-randomized studies (ROBINS-I) is frequently misapplied: a methodological systematic review.” Journal of clinical epidemiology, 140, December 2021, pp. 22-32. https:// www.sciencedirect.com/science/article/pii/S0895435621002675.

15. Tran, Linh et al. “Quality assessment tools used in systematic reviews of in vitro studies: A systematic review.” BMC Medical Research Methodology, 21, no. 1, May 2021. https://link.springer.com/article/10.1186/s12874-021-01295-w.

16. Downes, Martin J. et al. “Development of a critical appraisal tool to assess the quality of cross-sectional studies (AXIS).” BMJ open, 6, no. 12, 2016. https://bmjopen.bmj. com/content/6/12/e011458.short.

17. Altun, Hatice Uludag et al. “Antimicrobial susceptibilities of clinical Acinetobacter baumannii isolates with different genotypes.” Jundishapur journal of microbiology, 7, no. 12, December 2014. https://pmc.ncbi.nlm.nih.gov/articles/ PMC 4335573/.

18. Bansal, Garima et al. “Antibiotic susceptibility, clonality, and molecular characterization of carbapenem‐resistant clinical isolates of Acinetobacter baumannii from Washington DC.” International Journal of Microbiology, 1, July 2020. https://onlinelibrary.wiley.com/doi/abs/10.1155/2020/2120159.

19. Chang, Ko-Yun et al. “Clinical features and antimicrobial susceptibility of pseudomonas aeruginosa and Acinetobacter baumannii complex isolates in intensive care patients with chronic obstructive pulmonary disease and community-acquired pneumonia in Taiwan.” International journal of chronic obstructive pulmonary disease, 16, June 2021, pp. 1801-1811. https://www.tandfonline.com/doi/full/10.2147/COPD.S311714.

20. Chuang, Yu-Chung et al. “Molecular epidemiology, antimicrobial susceptibility and carbapenemase resistance determinants among Acinetobacter baumannii clinical isolates in Taiwan.” Journal of Microbiology, Immunology and Infection, 47, no. 4, August 2014, pp. 324-332. https://www.sciencedirect.com/science/article/ pii/ S1684118213000613.

21. Dafopoulou, K. et al. “Changes in antimicrobial resistance of clinical isolates of Acinetobacter baumannii group isolated in Greece, 2010–2015.” Journal of Medical Microbiology, 67, no. 4, April 2018, pp. 496-498. https://www.microbiology research.org/content/journal/jmm/10.1099/jmm.0.000708.

22. Duarte, Andreia et al. “Clinical isolates of Acinetobacter baumannii from a Portuguese hospital: PFGE characterization, antibiotic susceptibility and biofilm-forming ability.” Comparative immunology, microbiology and infectious diseases, 45, April 2016, pp. 29-33. https://www.sciencedirect.com/science/article/ abs/pii/S0147957116300157.

23. Ghaima, Kais Kassim et al. “Isolation, molecular identification and antimicrobial susceptibility of Acinetobacter baumannii isolated from Baghdad hospitals.” Burns, 6, no. 5, May 2016.

24. Guckan, R. et al. “Antimicrobial susceptibility of Acinetobacter baumannii complex isolated from different clinical samples in a tertiary care hospital.” Journal of Antibiotics Research, 1, no. 1, April 2017, pp. 1-5.

25. Gupta, N. et al. “Isolation and identification of Acinetobacter species with special reference to antibiotic resistance.” Journal of natural science, biology, and medicine, 6, no. 1, 2015. https://pmc.ncbi.nlm.nih.gov/articles/PMC4367029/.

26. Hsueh, Shun-Chung et al. “In vitro antimicrobial susceptibility data of global meropenem-resistant Acinetobacter baumannii isolates causing pneumonia: Data from the Antimicrobial Testing Leadership and Surveillance Program, 2014–2021, and re-estimations of susceptibility breakpoints and appropriate dosages of important antibiotics for pneumonia treatment.” Journal of Global Antimicrobial Resistance, 36, March 2024, pp. 411-418. https://www.sciencedirect.com/ science/article/pii/S2213716524000249.

27. Jaggi, Namita et al. “Acinetobacter baumannii isolates in a tertiary care hospital: Antimicrobial resistance and clinical significance.” Journal of Microbiology and Infectious Diseases, 2, no. 02, 2012, pp. 57-63. https://dergipark.org. tr/tr/pub/jmid/issue/9955/123151.

28. Kafshnouchi, Maghsoud et al. “Molecular detection of blaOXA-type carbapenemase genes and antimicrobial resistance patterns among clinical isolates of Acinetobacter baumannii.” Global medical genetics, 9, no. 2, 2022, pp. 118-123. https://www.thieme-connect.com/products/ejournals/html/10.1055/s-0041-174 0019.

29. Konca, Capan “Susceptibility patterns of multidrug-resistant Acinetobacter baumannii.” The Indian Journal of Pediatrics, 88, June 2020, pp. 120-126. https:// link.springer.com/article/10.1007/s12098-020-03346-4.

30. Liu, Congcong et al. “Epidemiological and genetic characteristics of clinical carbapenem-resistant Acinetobacter baumannii strains collected countrywide from hospital intensive care units (ICUs) in China.” Emerging microbes & infection, 11, no. 1, February 2022, pp. 1730-1741. https://www. tandfonline.com/doi/full/10.1080/22221751.2022.2093134.

31. Maleki, Abbas et al. “Molecular typing and antibiotic resistance patterns among clinical isolates of Acinetobacter baumannii recovered from burn patients in Tehran, Iran.” Frontiers in microbiology, 13, October 2022. https:// www.frontiersin.org/articles/10.3389/fmicb.2022.994303/full.

32. Sánchez-Urtaza, Sandra et al. “Molecular characterization of multidrug resistant Acinetobacter baumannii clinical isolates from Alexandria, Egypt.” Frontiers in Cellular and Infection Microbiology, 13, July 2023. https://www.frontiersin.org/ journals/cellular-and-infection-microbiology/articles/10.3389/fci mb.2023.1208046/full.

33. Shah, Muhammad Waseem et al. “Antimicrobial susceptibility and molecular characterization of clinical strains of Acinetobacter baumannii in Western Saudi Arabia.” Microbial Drug Resistance, 25, no. 9, October 2019, pp. 1297-1305. https://www.liebertpub. com/doi/abs/10.1089/mdr.2019.0018.

34. Sohail, Muhammad et al. “Antimicrobial susceptibility of Acinetobacter clinical isolates and emerging antibiogram trends for nosocomial infection management.” Revista da Sociedade Brasileira de Medicina Tropical, 49, no. 3, June 2016, pp. 300-304. https://www.scielo.br/j/rsbmt/a/GPD shNFzVDJvwGXGgQZKQ3j/?lang=en.

35. Sung, Ji Youn. “Molecular characterization and antimicrobial susceptibility of biofilm-forming Acinetobacter baumannii clinical isolates from Daejeon, Korea.” Korean Journal of Clinical Laboratory Science, 50, no. 2, June 2018, pp. 100-109. https://www.kjcls.org/journal/view.html?doi=10.15324/ kjcls.2018.50.2.100.

36. Tewari, Rachna et al. “Antimicrobial susceptibility patterns of an emerging multidrug resistant nosocomial pathogen: Acinetobacter baumannii.” The Malaysian journal of medical sciences: MJMS, 25, no. 3, June 2018, pp. 129-134. https://pmc.ncbi.nlm.nih.gov/articles/ PMC6422561/.

37. Tunyapanit, Wanutsanun et al. “Antimicrobial susceptibility of Acinetobacter baumannii isolated from hospital patients.” ScienceAsia, 40, no. 1, October 2013, pp. 28-34. https://www.thaiscience.info/Journals/Article/SCAS/10936869.pdf.

38. Karakonstantis, Stamatis et al. “Global prevalence of cefiderocol non-susceptibility in Enterobacterales, Pseudomonas aeruginosa, Acinetobacter baumannii, and Stenotrophomonas maltophilia: a systematic review and meta-analysis.” Clinical Microbiology and Infection, 30, no. 2, February 2024, pp. 178-188. https://www. sciencedirect.com/science/article/pii/S1198743X23004135.

39. Shields, Ryan K. et al. “Navigating available treatment options for carbapenem-resistant Acinetobacter baumannii-calcoaceticus complex infections.” Clinical Infectious Diseases, 76, no. Supplement_2, May 2023, pp. S179-S193. https://academic.oup.com/cid/article/76/ Supplement_2/S179/7140356.

40. Jean, Shio-Shin et al. “Potentially effective antimicrobial treatment for pneumonia caused by isolates of carbapenem-resistant and extensively drug-resistant Acinetobacter baumannii complex species: what can we expect in the future?.” Expert Review of Anti-infective Therapy, 22, no. 12, May 2024, pp. 1171-1187. https://www.tandfonline.com/ doi/abs/10.1080/14787210.2024.2412637.

41. O'Donnell, J. Nicholas et al. “Treatment of patients with serious infections due to carbapenem‐resistant Acinetobacter baumannii: How viable are the current options?.” Pharmacotherapy: The Journal of Human Pharmacology and Drug Therapy, 41, no. 9, June 2021, pp. 762-780. https://accpjournals.onlinelibrary.wiley.com/doi/ abs/10.1002/phar.2607.

42. Seifert, Harald et al. “Global update on the in vitro activity of tigecycline and comparators against isolates of Acinetobacter baumannii and rates of resistant phenotypes (2016–2018).” Journal of Global Antimicrobial Resistance, 31, December 2022, pp. 82-89. https://www.sciencedirect.com/ science/article/pii/S2213716522001928.

43. Karlowsky, James A. et al. “In vitro activity of sulbactam-durlobactam against global isolates of Acinetobacter baumannii-calcoaceticus complex collected from 2016 to 2021.” Antimicrobial Agents and Chemotherapy, 66, no. 9, August 2022, pp. e00781-e00822. https://journals. asm.org/doi/full/10.1128/aac.00781-22.

44. Abdul-Mutakabbir, Jacinda C. et al. “Contemporary perspective on the treatment of Acinetobacter baumannii infections: insights from the Society of Infectious Diseases Pharmacists.” Infectious diseases and therapy, 10, October 2021, pp. 2177-2202. https://link.springer.com/article/10. 1007/s40121-021-00541-4.

45. Hujer, Andrea M. et al. “A comprehensive and contemporary “snapshot” of β-lactamases in carbapenem resistant Acinetobacter baumannii.” Diagnostic Microbiology and Infectious Disease, 99, no. 2, February 2021. https://www. sciencedirect.com/science/article/pii/S0732889320306192.

46. Iovleva, Alina et al. “Carbapenem-resistant Acinetobacter baumannii in US hospitals: diversification of circulating lineages and antimicrobial resistance.” MBio, 13, no. 2, February 2022, pp. Iovleva, Alina, et al. "Carbapenem-resistant Acinetobacter baumannii in US hospitals: diversification of circulating lineages and antimicrobial resistance." MBio 13.2 (2022): e02759-e02821. https://journals.asm.org/doi/abs/10. 1128/mbio.02759-21.

47. Prim, Núria et al. “Bloodstream infection due to Brachyspira pilosicoli in a patient with multiorgan failure.” Journal of Clinical Microbiology, 49, no. 10, October 2011, pp. 3697-3699. https://journals.asm.org/doi/full/10.1128/jcm.00680-11.

48. Adams, Mark D. et al. “Rapid replacement of Acinetobacter baumannii strains accompanied by changes in lipooligosaccharide loci and resistance gene repertoire.” MBio, 10, no. 2, March 2019. https://journals.asm.org/doi/full/10.1128/mbio.00356-19.

49. Cai, Yun et al. “Colistin resistance of Acinetobacter baumannii: clinical reports, mechanisms and antimicrobial strategies.” Journal of antimicrobial chemotherapy, 67, no. 7, July 2012, pp. 1607-1615. https://academic.oup.com/jac/article-abstract/67/7/1607/ 729433.

50. Hackel, Meredith A. et al. “In vitro activity of the siderophore cephalosporin, cefiderocol, against carbapenem-nonsusceptible and multidrug-resistant isolates of Gram-negative bacilli collected worldwide in 2014 to 2016.” Antimicrobial agents and chemotherapy, 62, no. 2, January 2018. https://journals. asm.org/doi/full/10.1128/aac.01968-17.

51. Butler, David A. et al. “Multidrug resistant Acinetobacter baumannii: resistance by any other name would still be hard to treat.” Current infectious disease reports, 21, November 2019. https://link.springer.com/article/10.1007/s11908-019-0706-5.

52. Viehman, J. et al. “Treatment options for carbapenem-resistant and extensively drug-resistant Acinetobacter baumannii infections.” Drugs, 74, August 2014, pp. 1315-1333. https://link.springer.com/article/10.1007/s40265-014-0267-8.