This study selected clinical data from 137 CRAB patients and 131 AB patients hospitalized at the Affiliated Hospital of Yangzhou University from July 2021 to September 2022. The age range of all patients was 16–94 years, with an average age of 69 years; 174 were male (64.9%) and 94 were female (35.1%); 196 patients (73.1%) were over 60 years old. All strains were isolated from sputum specimens of hospitalized patients.

New or progressive radiographic pulmonary infiltrates appearing 48 h after admission, accompanied by at least two clinical infection symptoms (temperature >38°C or <35.5°C, leukocytosis or leukopenia, purulent secretions). For elderly patients, clinical manifestations may be atypical, such as altered mental status, anorexia, fatigue, or lack of fever. Inclusion criteria also included bacteriological test results.

Patients were excluded if they met any of the following conditions: (i) AB detected within 48 h of admission; (ii) age under 16 years; (iii) incomplete clinical data; (iv) duplicate specimens from the same or different sites of the same patient; (v) bacterial colonization without clinical symptoms, radiographic, or laboratory abnormalities.

We screened for common β-lactamase genes (blaPER, blaSHV, blaIMP, blaVIM, blaSIM, blaNDM-1, blaAmpC, blaOXA-23, blaOXA-24, blaOXA-51, blaOXA-143, blaOXA-58), and sent the primers to Nanjing GenScript Biotechnology Co., Ltd. for synthesis. PCR detection was performed on CRAB 27 and CRAB 49. The final reaction volume for PCR amplification was 20 μL, containing 1 μL of template, 10 μL of Mix, 1 μL each of forward and reverse primers, and 7 μL of ddH₂O. The amplification conditions were as follows: pre-denaturation at 95°C for 5 min, denaturation at 95°C for 30 s, annealing for 30 s, extension at 72°C for 1 min, and 35 cycles. The PCR products were subjected to agarose gel electrophoresis and photographed for documentation. The primer sequences are shown in Table 1.

Bacteria culture and isolation were performed according to the standards of the “National Clinical Laboratory Procedures” (fourth edition). For the same site of the same patient, only the first isolated strain was analyzed. Strain identification was conducted using the Vitek 2 Compact automated microbiology analyzer and Vitek MS automated mass spectrometer from bioMérieux, France. Five strains isolated from clinical samples were inoculated onto LB agar plates and isolated using the four-quadrant streaking method. The inoculated plates were incubated overnight at 37°C, and uniform single colonies were selected using an inoculating loop. The selected single colonies were inoculated into 2 mL of LB medium and cultured overnight at 37°C and 200 rpm in a shaker. The cultured bacterial solution was mixed with 40% glycerol at a 1:1 ratio and stored at −80°C for subsequent experiments. Next, the stored bacterial solution was inoculated into 1 mL of MH broth at a 1:100 ratio and cultured until the OD600 nm value reached 0.5. Then, 10 µL of the bacterial solution with an OD600 nm of 0.5 was added to 990 µL of MH broth for a 1:100 dilution. The diluted bacterial solution (200 µL) was added to a 96-well microplate, with three replicates for each strain and 200 µL of MH medium as a blank control. The microplate was incubated statically at 37°C for 18 h. After incubation, the supernatant was discarded, and the wells were washed twice with PBS, stained with 0.4% crystal violet for 30 min, and washed twice again with PBS. Finally, the dye was eluted with 95% ethanol, and the absorbance was measured at OD590 nm using a microplate reader.

Single colonies were inoculated into 10 mL of MH broth and cultured at 37°C and 200 rpm until the OD600 nm reached 0.5. The bacterial solution was then diluted 100-fold to a final concentration of 1 × 10⁶ CFU/mL. Different concentrations of antibiotic solutions (0.5, 1, 2, 4, 8, 16, 32, 64 μg/mL) were prepared, and 100 µL of the corresponding concentration of antibiotic was added to each test well. Another 100 µL of the diluted bacterial solution was added to each well, with a negative control well containing 200 µL of MH medium. After thorough mixing, the plates were incubated at 37°C for 18 h. After incubation, the OD600 nm value was measured using a microplate reader. The minimum inhibitory concentration (MIC) of the antibiotic was defined as the lowest concentration that inhibited bacterial growth. Antibiotic susceptibility results were determined according to the 2021 Clinical and Laboratory Standards Institute (CLSI) guidelines and classified as resistant, intermediate, or sensitive. Quality control strains included Escherichia coli ATCC 25922 and Pseudomonas aeruginosa ATCC 27853. All experimental operations strictly followed standard operating procedures. All experiments were repeated three times to ensure reproducibility and reliability of the results.

Complete medical records of CRAB patients meeting the inclusion criteria were collected, including the following: (i) Basic patient information: gender, age, history of diabetes, history of invasive procedures, glucocorticoid use, length of hospital stay, length of ICU stay, duration of mechanical ventilation, antibiotic use; (ii) Clinical indicators and antibiotic susceptibility results; (iii) Hospital outcomes: outcomes were classified as improvement (including cure and clinical and/or laboratory and radiographic improvement) and poor prognosis (including death and non-recovery).

PCR amplification of seven housekeeping genes (gltA, gyrB, gdhB, recA, cpn60, gpi, rpoD) was performed using internationally recognized primers. The amplification products were sequenced by Nanjing GenScript Biotechnology Co., Ltd. According to the Oxford MLST scheme, MLST typing analysis was performed on strains AB1, AB2, and AB3. The obtained gene sequences were uploaded to the Acinetobacter baumannii PubMLST database (http://pubmlst.org/abaumannii/) to determine their sequence type (ST) by comparison with existing sequences in the database. The total volume of the PCR reaction system was 25 μL, including 1 µL of primer, 1 µL of template, 12.5 µL of PCR Mix, and 9.5 µL of ddH2O. The PCR amplification conditions were set as follows: pre-denaturation at 94°C for 5 min; denaturation at 94°C for 30 s, annealing for 30 s, extension at 72°C for 1 min, for a total of 30 cycles; final extension at 72°C for 10 min. Primer sequences and annealing temperatures are detailed in Supplementary Table S1. All experiments were performed with three technical replicates to ensure the reproducibility and reliability of the results.

To evaluate the in vitro antibacterial activity of antibiotic combinations against CRAB, a checkerboard broth microdilution assay was first performed. Based on the results of the previous antibiotic susceptibility tests, the MICs of the two antibiotics were determined, and their concentrations were further diluted to one to two times the MIC. According to the two-fold dilution method, the two antibiotics were diluted along the vertical and horizontal coordinates of the 96-well plate, with 50 µL of the corresponding concentration of antibiotic MH broth added to each well. After overnight culture at 37°C, single colonies were transferred to 5 mL of fresh MH broth and cultured until the OD590 nm reached 0.5 (1 × 10⁸ CFU/mL), then diluted 100-fold to a final concentration of 1 × 10⁶ CFU/mL. Each well was inoculated with 100 µL of the diluted bacterial solution and mixed with 100 µL of MH broth containing antibiotics, with negative and blank controls set. After mixing, the plates were incubated at 37°C for 18 h, and the optical density (OD) at 600 nm was measured using a microplate reader. The bacterial synergistic inhibition rate was calculated using the following formula: Bacterial synergistic inhibition rate (%) = [(OD positive control group − OD negative control group) − (OD sample - OD negative control group)]/(OD positive control group − OD negative control group) × 100%. The fractional inhibitory concentration index (FICI) of the combined drugs was calculated using the following formula: FICI = MIC meropenem (combination)/MIC meropenem (single) + MIC ciprofloxacin (combination)/MIC ciprofloxacin (single). FICI ≤0.5, synergy; 0.5≤ FICI ≤4, no interaction; FICI >4, antagonism (Vidaillac et al., 2012). Additionally, the growth curve assay was performed on the CRAB2 strain. After streaking single colonies using the quadrant method, they were cultured overnight at 37°C and transferred to 5 mL of fresh MH broth at a 1:100 ratio, cultured until the OD590 nm was 0.5 (1 × 10⁸ CFU/mL), and then diluted to 1 × 10⁶ CFU/mL. Subsequently, the two antibiotics were added to the bacterial solution alone or in combination, with no antibiotics added as the control group. The cultures were incubated at 37°C and 200 rpm, and 200 µL samples were taken every 2 h, and the OD600 nm was measured using a microplate reader to monitor bacterial growth (YOURASSOWSKY et al., 1985).

Analysis of CRAB resistance, intermediate resistance, and sensitivity was performed using SPSS 26.0 software for statistical analysis. For quantitative data that follow a normal distribution or an approximate normal distribution, they are presented as mean ± standard deviation. Independent-samples t-test or one-way ANOVA (depending on the number of groups) is used for analysis. If the data do not meet the requirements of normal distribution or homogeneity of variance, they are described by median/interquartile range. The Mann-Whitney U test for two independent samples or the Kruskal–Wallis test for multiple independent samples is used for result analysis. Categorical variables or ranked data are described using percentages. The chi-square test or Fisher’s exact probability method is used for between-group difference analysis.

The AB-27 strain of CRAB was selected for transcriptomic analysis. After culturing AB-27 in Mueller-Hinton broth (MHB) to the early logarithmic growth phase, the strain was treated with a combination of ciprofloxacin and meropenem. Following RNA quality assessment, cDNA synthesis, fragmentation, and sequencing were performed. The cDNA library was then amplified and sequenced using the Illumina HiSeq™ platform. Technical services and subsequent bioinformatics analysis were provided by Guangzhou Gene Denovo Biotechnology Co., Ltd. Differentially expressed genes (DEGs) were identified through comparative analysis of RNA-Seq data. Differential expression analysis was conducted using the DESeq2 package, and genes with significant differences were selected based on the predefined thresholds (P < 0.05, |log₂FoldChange| > 1). For the selected DEGs, Gene Ontology (GO) enrichment analysis was performed using the R package clusterProfiler. The GO annotations were categorized into three main terms: Biological Process (BP), Molecular Function (MF), and Cellular Component (CC). The significance of each GO term was calculated using Fisher’s exact test, followed by multiple testing correction using the Benjamini–Hochberg method. GO terms with a P-value <0.05 were considered significantly enriched. KEGG pathway enrichment analysis was performed on the DEGs using the KEGG database. KEGG enrichment analysis was conducted using the R package clusterProfiler, which calculates the enrichment of DEGs across various KEGG pathways using Fisher’s exact test, followed by False Discovery Rate (FDR) correction for P-values. Pathways with a P-value <0.05 were considered significantly enriched.

In this study, 137 carbapenem-resistant A. baumannii (CRAB) strains were isolated from sputum samples. These strains were primarily obtained from the Intensive Care Unit (ICU) (103 cases), the Department of Respiratory Medicine (12 cases), and the Department of Neurosurgery (11 cases), accounting for 75.2%, 8.8%, and 8.0% of the total, respectively (Supplementary Table S2).

In this study, carbapenem resistant Acinetobacter baumannii strains were selected for research. Clinical antimicrobial susceptibility tests on 137 CRAB strains against 24 antimicrobial agents were conducted. Results showed that among β-lactam antibiotics, resistance rate to carbapenem antibiotics was the highest, reaching 99.27%. This was followed by third-generation cephalosporins, with a resistance rate exceeding 96%. The resistance rates to broad-spectrum penicillins (such as piperacillin and ticarcillin) were also high, exceeding 95%.CRAB exhibited moderate resistance to tigecycline, with a resistance rate of 42.33%, while minocycline still showed good antibacterial activity against CRAB, with a resistance rate of 25.54% (Supplementary Table S3).

To compare the clinical characteristics between the CRAB group and the AB group, we conducted a comparative analysis of multiple parameters in both groups of patients. These parameters included the number of invasive procedures, length of hospital stay, length of ICU stay, duration of mechanical ventilation, types and duration of antibiotic use, conjugated bilirubin levels, alanine aminotransferase (ALT) levels, aspartate aminotransferase (AST) levels, urea levels, and the number of unresolved cases. We found that the CRAB group had higher values in all these indicators compared to the AB group, with statistically significant differences between the two groups (P < 0.05). The number of patients undergoing invasive procedures was 129 in the CRAB group and 111 in the AB group, accounting for 94.16% and 84.73%, respectively. The length of hospital stay was 21 ± 11 days for the CRAB group and 17 ± 9 days for the AB group. The length of ICU stay was 13 ± 9 days for the CRAB group and 5 ± 7 days for the AB group. The duration of mechanical ventilation was 9 ± 8 days for the CRAB group and 3 ± 6 days for the AB group. The number of types of antibiotics used was 4 ± 7 in the CRAB group and 2 ± 1 in the AB group. The duration of antibiotic use was 18 ± 9 days in the CRAB group and 13 ± 7 days in the AB group. The conjugated bilirubin level was 6.32 ± 22.40 μmol/L in the CRAB group and 3.69 ± 14.68 μmol/L in the AB group. The average alanine aminotransferase (ALT) level was 36.00 U/L in the CRAB group and 23.00 U/L in the AB group. The average aspartate aminotransferase (AST) level was 28.00 U/L in the CRAB group and 41.00 U/L in the AB group. The urea level was 11.19 ± 7.00 mmol/L in the CRAB group and 8.41 ± 5.67 mmol/L in the AB group. The number of unresolved cases was 53 in the CRAB group and 25 in the AB group, accounting for 38.69% and 19.08%, respectively (Supplementary Table S4).

To elucidate the prognosis of patients with CRAB infection, we compared multiple parameters between the non-improvement group and the improvement group. The results showed that the non-improvement group had higher values in ICU stay days, mechanical ventilation days, white blood cell count, neutrophil count, C-reactive protein level, procalcitonin level, conjugated bilirubin, unconjugated bilirubin, aspartate aminotransferase, urea, creatinine, and creatine kinase-MB compared to the improvement group, with statistically significant differences between the two groups (P < 0.05). Specifically, the ICU stay days were 15 ± 11 days in the non-improvement group and 11 ± 8 days in the improvement group; mechanical ventilation days were 11 ± 9 days in the non-improvement group and 8 ± 8 days in the improvement group; white blood cell count was 12.17 ± 5.19 × 10⁹/L in the non-improvement group and 9.82 ± 3.53 × 10⁹/L in the improvement group; neutrophil count was 11.37 ± 7.00 × 10⁹/L in the non-improvement group and 8.05 ± 3.20 × 10⁹/L in the improvement group; C-reactive protein level was 102.19 ± 77.72 mg/L in the non-improvement group and 50.09 ± 39.31 mg/L in the improvement group; procalcitonin level was 4.49 ± 8.57 ng/L in the non-improvement group and 0.68 ± 1.64 ng/L in the improvement group; conjugated bilirubin was 14.60 ± 34.45 μmol/L in the non-improvement group and 1.10 ± 2.67 μmol/L in the improvement group; unconjugated bilirubin was 16.06 ± 12.83 μmol/L in the non-improvement group and 10.24 ± 6.61 μmol/L in the improvement group; aspartate aminotransferase average level was 53.00 U/L in the non-improvement group and 36.00 U/L in the improvement group; urea was 13.84 ± 8.40 mmol/L in the non-improvement group and 9.52 ± 5.36 mmol/L in the improvement group; creatinine was 126.59 ± 94.72 μmol/L in the non-improvement group and 83.21 ± 75.63 μmol/L in the improvement group; creatine kinase-MB was 19.45 ± 26.03 U/L in the non-improvement group and 7.64 ± 6.40 U/L in the improvement group. These data indicate that the clinical manifestations and laboratory indicators of patients in the non-improvement group were significantly worse than those in the improvement group (Supplementary Table S5).

The results of the logistic multivariate regression analysis showed that the duration of antibiotic use, application of glucocorticoids, C-reactive protein levels, and creatine kinase isoenzyme levels were independent risk factors for poor prognosis in CRAB patients (P < 0.05) (Table 2). Additionally, C-reactive protein and creatine kinase isoenzyme levels have certain predictive value in forecasting poor prognosis in CRAB patients, with an AUC of 0.69 for C-reactive protein and an AUC of 0.64 for creatine kinase isoenzyme (Figure 1).

We tested the biofilm formation ability of five CRAB strains, with results reflected by OD590 nm values. The results indicated that strain 2 had the strongest biofilm formation ability, followed by strains 1 and 3 (Figure 2). Therefore, we selected these three strong biofilm-forming CRAB strains for further antibiotic susceptibility testing. We found that all strains exhibited high levels of resistance (MIC ≥64 μg/mL) to meropenem (MEM), imipenem (IPM), gentamicin (GEN), ticarcillin (TIC), ceftriaxone (CRO), tetracycline (TET), and vancomycin (VCA). Notably, CRAB1 had minimum inhibitory concentrations (MICs) of 4 μg/mL for ciprofloxacin (CIP), while CRAB2 and CRAB3 had MICs of 8 μg/mL for this drugs, respectively. These results indicate that these strains exhibit intermediate sensitivity to levofloxacin and ciprofloxacin. Additionally, all strains were highly sensitive to tigecycline (TGC) (MIC <1 μg/mL) (Table 3).

Two CRAB strains (CRAB27 and CRAB49) were collected for analyzing the synergistic effects of carbapenem antibiotics and ciprofloxacin in bactericidal experiments. Prior to the experiments, the β-lactamase gene carriage status of both CRAB strains was analyzed, with the results shown in Figure 3. PCR detection revealed that both CRAB strains carried the blaVIM, blaOXA-51, blaAmpC, and blaOXA-23 genes. The blaVIM gene encodes a Class B metallo-β-lactamase (MBL), while blaOXA-51 and blaOXA-23 encode Class D β-lactamases (OXA enzymes), and blaAmpC encodes a Class C β-lactamase (AmpC enzyme). There were no significant differences in the β-lactamase genotypes between the two CRAB strains, indicating that they likely share similar resistance mechanisms. By confirming the absence of significant genotypic differences between the strains, potential interference from genotypic variations in subsequent resistance mechanism studies can be ruled out. This facilitates a more accurate assessment of the impact of combination therapy on CRAB resistance.

The checkerboard assay results indicated that the combination of carbapenem antibiotics and ciprofloxacin exhibited significant synergistic effects against AB27 and AB49. In contrast, the combination of carbapenem antibiotics and colistin showed no interaction against AB27 and AB49. We further used the growth curve of bacteria to describe the bactericidal effects of different antibiotics on CRAB. According to the curve analysis, meropenem alone (blue line) had poor inhibitory effects on the strains, with the OD600 nm value gradually increasing over 24 h, indicating continuous bacterial growth. Ciprofloxacin alone (green line) showed some inhibitory effects compared to meropenem but still failed to completely inhibit bacterial growth. When meropenem was combined with ciprofloxacin (orange line), the OD600 nm value was significantly lower than in other groups within 24 h, indicating a significant bactericidal effect of this combination. In summary, The FICI of MEM combined with CIP is 0.375, indicating that the combination of the two has a synergistic antibacterial effect against CRAB. The FICI of MEM combined with COL is greater than 0.5, indicating that the combination of the two does not have a synergistic effect (Table 4). This demonstrates the potential advantage of the combination therapy of MEM and CIP against CRAB strains (Figure 4).

To further elucidate the effects of various antibiotics on CRAB, we examined the impact of different antibiotic treatments on the outer membrane permeability, membrane integrity, membrane potential, and reactive oxygen species (ROS) production of CRAB. The results showed that the combination of meropenem and ciprofloxacin significantly increased the outer membrane permeability of CRAB (P < 0.001), with the combined use of meropenem and ciprofloxacin being more effective than either drug alone. Membrane integrity analysis indicated that the meropenem group had significantly higher membrane integrity than other groups (P < 0.01). In contrast, the ciprofloxacin alone and combination groups did not show significant differences in membrane integrity but were both lower than the meropenem group. Membrane potential measurements showed that the membrane potential of the meropenem and ciprofloxacin combination group was significantly higher than that of the control group (P < 0.05), indicating a significant synergistic effect of this combination in altering membrane potential. In terms of ROS production, both the ciprofloxacin alone and combination groups significantly increased ROS levels, although the combination group showed higher ROS production, the difference was not statistically significant compared to the control group (Figure 5).

In this study, multilocus sequence typing (MLST) was used to analyze the sequence types of carbapenem-resistant A. baumannii strains. The results revealed that strains 8, 27, and 49 al l belonged to Sequence Type 208 (ST208). These strains shared identical allele numbers across seven loci: gltA, cnp60, gdhB, gpi, gyrB, rpoD, and recA. This finding suggests that the strains have a common genetic background and likely originated from the same clonal group (Table 5).

To investigate the synergistic bactericidal mechanism of ciprofloxacin combined with meropenem against CRAB, the transcriptome of the AB-27 strain was analyzed after ciprofloxacin-meropenem combination treatment. The results showed that, compared to the control group, 480 genes were upregulated and 112 genes were downregulated in CRAB following the combined treatment (Figure 6A). GO enrichment analysis revealed that the differentially expressed genes (DEGs) under meropenem combined with ciprofloxacin treatment (Figure 6B) are primarily associated with the following biological functional categories: Biological Processes: The DEGs are significantly enriched in cellular processes, metabolic processes, localization, and response to stimuli, indicating that the drug treatment has a substantial impact on the basic biological activities of the bacteria. Molecular Functions: The DEGs are predominantly enriched in catalytic activity, binding activity, and transport activity, which are closely related to the bacteria’s metabolic and adaptive responses. Cellular Components: The DEGs are mainly concentrated in cellular components and protein complexes, suggesting that the treatment may induce alterations in the structure and function of the bacterial cell. KEGG pathway analysis further revealed the biological pathways associated with these differentially expressed genes (DEGs) (Figure 6C). The most significantly enriched pathways include: Metabolic Pathways, with the Global and Overview Maps (132 genes) and Amino Acid Metabolism (48 genes) being the most prominently enriched. Other important pathways include Energy Metabolism, Carbohydrate Metabolism, and Cofactor and Vitamin Metabolism, suggesting that the drug treatment may affect bacterial growth and antibiotic resistance by altering metabolic pathways. Genetic Information Processing pathways, such as Folding, Sorting, and Degradation (12 genes) and Transcription (11 genes), indicating that the combined treatment may impact gene expression and protein synthesis in the bacteria. Cellular Processes and Environmental Information Processing pathways were also enriched, highlighting the bacterial adaptive response to environmental changes.