Distribution and antibiotic resistance analysis of clinically isolated bacteria at a national regional cardiovascular medical center, 2018–2024

Abstract

Objective

This study aims to analyze the distribution and drug resistance patterns of clinically isolated bacteria at the National Center for Cardiovascular Diseases (Fuwai Central China Cardiovascular Hospital) from 2018 to 2024, thereby guiding the precise selection of effective antimicrobial agents in clinical practice and providing data support for the development of individualized anti-infective treatment strategies.

Methods

Clinical isolates from outpatients and inpatients were collected at the Fuwai Central China Cardiovascular Hospital, excluding duplicate strains. Statistical analysis was conducted using descriptive epidemiological methods with SPSS 22.0, Prism 10.0, and WHONET 5.6 software.

Results

Over seven years, 9,180 clinical isolates were identified, with a positive detection rate of 7.78%–8.41%. Gram-negative bacteria accounted for 67% (6,151 strains), while Gram-positive bacteria comprised 23% (2,139 strains). The top three isolated pathogens were Klebsiella pneumoniae, Acinetobacter baumannii, and Pseudomonas aeruginosa. Resistance analysis revealed high overall resistance rates among common pathogens. Carbapenem resistance rates during this period were as follows: K. pneumoniae (25.0%–35.8%), A. baumannii (76.5%–94.7%), P. aeruginosa (13.3%–25.3%), and Escherichia coli (2.0%–10.7%). Methicillin-resistant Staphylococcus aureus (MRSA) detection rates ranged from 35.3% to 60%, while methicillin-resistant coagulase-negative Staphylococci (MRCNS) detection rates were 80.4%–91.2%.

Conclusion

Investigating bacterial distribution and resistance patterns in a regional cardiovascular center provides critical evidence for optimizing antibiotic use and infection control. Importantly, multidrug-resistant infections, particularly CRABA and MRSA, were associated with significantly increased mortality and prolonged hospitalization, underscoring the urgent need for enhanced stewardship and surveillance to mitigate this clinical burden.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12879-025-12367-3.

Introduction

The escalating global crisis of antimicrobial resistance (AMR), driven by widespread clinical antibiotic use, represents one of the most urgent public health threats. Projections from The Lancet estimate that AMR-associated infections will cause over 39 million deaths globally between 2025 and 2050, with direct AMR fatalities already reaching 1.14 million in 2021 and projected to rise to 1.91 million by 2050 [1]. The global AMR landscape exhibits significant geographical disparities: prevalence growth rates are slower in high-income countries (0.3%), while accelerating in low- and middle-income nations, particularly in South Asia and sub-Saharan Africa [2]. Variations in resistance mechanisms also exist globally; for instance, CTX-M-14 is the predominant extended-spectrum β-lactamase-producing Enterobacterales (ESBL-E) subtype in China, contrasting with the CTX-M-15 dominance in Europe and North America [3]. Socioeconomic and climatic factors synergistically exacerbate AMR dissemination, including inadequate healthcare resources, antibiotic overuse, and environmental transmission of resistance genes [2]. China faces particularly acute AMR challenges due to its large population, high antibiotic consumption, and regional disparities in healthcare quality, resulting in persistently high resistance rates [4]. Recent 2024 data indicate a strikingly high detection rate of third-generation cephalosporin-resistant Enterobacterales in China, substantially exceeding the global average [3].

In cardiovascular patients, particularly those undergoing invasive procedures or device implantation, AMR is associated with increased postoperative infection rates, prolonged hospitalization, higher mortality, and elevated healthcare costs. Patients with comorbidities such as heart failure, diabetes, or renal impairment are at heightened risk for resistant infections, which can compromise the success of cardiovascular interventions such as valve replacement, coronary artery bypass grafting, or left ventricular assist device implantation. Japanese research has linked Streptococcus anginosus group carriage to a 20% increased stroke risk [5], suggesting resistant bacteria may exacerbate cardiovascular events through microbiome dysbiosis; altered gut microbiota patterns (e.g., increased Lactobacillus abundance) in coronary heart disease patients have also been associated with recurrent cardiovascular events [6]. Despite the establishment of international AMR surveillance networks [7] and China’s national China Antimicrobial Resistance Surveillance System (CARSS) system, the pathogen distribution and resistance profiles within specialized cardiovascular hospitals remain insufficiently characterized at regional and specialty levels. This retrospective study analyzes clinical isolates from Fuwai Central China Cardiovascular Hospital, a national pilot regional medical center, between 2018 and 2024. It aims to elucidate the unique bacterial distribution and temporal resistance trends within this high-volume cardiovascular care setting. It seeks to evaluate the effectiveness of current antimicrobial strategies and provide specialized data to optimize antibiotic stewardship, enhance infection control, support global AMR mitigation efforts, and ultimately reduce infection risk in this vulnerable patient population.

Materials and methods

Data collection

This study retrospectively analyzed clinically significant bacterial strains isolated from outpatients and inpatients at Fuwai Central China Cardiovascular Hospital between January 1, 2018, and December 31, 2024. Clinically significant isolates were defined as those obtained from normally sterile sites (e.g., blood, cerebrospinal fluid) or from non-sterile sites (e.g., sputum, urine) in the presence of clinical signs and symptoms of infection, according to Centers for Disease Control and Prevention (CDC)/National Healthcare Safety Network (NHSN) criteria. Isolates considered probable contaminants (e.g., a single set of blood cultures positive for coagulase-negative staphylococci in the absence of clinical signs) were excluded.

Duplicate isolates from the same patient with the same species and identical antibiogram within a 30-day period were excluded according to established laboratory protocols. Patient identification was based on a unique hospital admission number. The selected specimen types encompassed sputum, blood, urine, bronchoalveolar lavage fluid (BALF), tissue, catheter tips/specimens, and wound exudates/samples from patients. The dataset encompassed critical clinical parameters including specimen collection dates, microbial identification results, antimicrobial susceptibility testing (AST) profiles, patient demographics (age and sex), and clinical care locations (hospital wards).

Pathogen identification and antimicrobial susceptibility testing

Bacterial identification was performed consistently using MALDI-TOF MS (Bruker Biotyper system) throughout the study period. Antimicrobial susceptibility testing followed the latest Clinical & Laboratory Standards Institute (CLSI) guidelines [8]. The BD Phoenix M50 system was used for the majority of isolates and antimicrobials. Disk Diffusion (OXOID, UK) was used as a supplementary method for specific agents and for quality control. The E-test (Zhejiang Wenzhou Kangtai Biotechnology Co., Ltd) was employed for special antibiotics that lack a breakpoint by the disk diffusion method. The BD Phoenix M50 system uses dedicated susceptibility panels. Species-specific media was applied to different pathogens: Mueller–Hinton (MH) agar (Zhengzhou Antu Bioengineering Co., Ltd.) for Moraxella catarrhalis, 5% defibrinated sheep blood-supplemented MH agar for Streptococcus pneumoniae susceptibility confirmation, and Haemophilus Test Medium (HTM) for Haemophilus influenzae. Interpretive criteria followed the latest CLSI M100 guidelines (Editions 28–34, corresponding to years 2018–2024) [8], with the European Committee on Antimicrobial Susceptibility Testing (EUCAST) breakpoints applied to selected antimicrobials (Versions 8.0–14.0, 2018–2024) [9]. Tigecycline and colistin susceptibility was assessed according to the Chinese Expert Consensus on Polymyxins, Tigecycline and Ceftazidime/Avibactam Susceptibility Testing [10]. Quality control strains included E. coli ATCC 25,922 and ATCC 35,218, Pseudomonas aeruginosa ATCC 27,853, Enterococcus faecalis ATCC 29,212, S. aureus ATCC 25,923 and ATCC 29,213, S. pneumoniae ATCC 49,619, and H. influenzae ATCC 49,247, ensuring standardized testing procedures throughout the study.

Definitions of antimicrobial resistance phenotypes

The following definitions were used for key resistance phenotypes:

MRSA/MRCNS: resistance to oxacillin (minimum inhibitory concentration (MIC) ≥ 4 µg/mL) or cefoxitin disk diffusion zone diameter ≤ 21 mm for S. aureus and ≤ 24 mm for CoNS; ESBL Production: phenotypically confirmed using combination disk test (cefotaxime and ceftazidime with/without clavulanic acid; ≥5 mm increase confirmed ESBL); carbapenem-resistant Enterobacterales (CRE): resistance to imipenem (MIC ≥ 4 µg/mL), meropenem (MIC ≥ 4 µg/mL), or ertapenem (MIC ≥ 2 µg/mL); carbapenem-resistant A. baumannii (CRABA)/carbapenem-resistant K. pneumoniae (CRKPN)/carbapenem-resistant P. aeruginosa (CRPAE): resistance to imipenem or meropenem (MIC ≥ 8 µg/mL for A. baumannii and P. aeruginosa, ≥ 4 µg/mL for K. pneumoniae); vancomycin-resistant Enterococci (VRE): vancomycin MIC ≥ 32 µg/mL; high-level gentamicin resistance (HLGR) in Enterococci: gentamicin MIC > 500 µg/mL.

Statistical analysis

Descriptive epidemiological analysis was conducted on the data, with numerical data processed and analyzed using SPSS 22.0 and GraphPad Prism software (version 10.0). Temporal trends in resistance rates for key pathogen-antibiotic combinations were analyzed using binomial regression with a logit link function, with models weighted by the annual number of isolates; this trend analysis described population-level changes and was not adjusted for patient-level covariates such as age, gender, or ward type. Results are reported as the annual change in the log odds of resistance (β) with 95% confidence intervals (CI) and p-values, considering p < 0.05 statistically significant. The distribution of isolates by gender, age group, and ward type was analyzed using descriptive statistics, with categorical variables compared via the chi-square (χ²) test. In contrast, the analysis of clinical outcomes associated with resistant phenotypes (e.g., mortality, length of stay) presented in Supplementary Table S4 employed multivariable models adjusted for age, gender, and ward type where data were available to control for potential confounding. WHONET 5.6 software was used to manage and analyze antimicrobial susceptibility data; to complement its outputs and account for temporal autocorrelation in formal analyses, regression modeling was performed in SPSS as described.

Results

Isolation of pathogenic bacteria

A total of 9,180 non-duplicate clinical isolates (excluding repeat isolates from the same patient) were obtained from specimens cultured at Fuwai Central China Cardiovascular Hospital between January 1, 2018, and December 31, 2024. The annual specimen culture volumes were 10,451 (2018), 14,639 (2019), 16,030 (2020), 15,780 (2021), 14,552 (2022), 22,335 (2023), and 21,529 (2024). Corresponding annual isolate counts were 844, 1,231, 1,311, 1,264, 1,143, 1,810, and 1,674, yielding positivity rates of 8.07%, 8.41%, 8.18%, 8.01%, 7.85%, 8.10%, and 7.78%, respectively. The progressive increase in specimen culture volumes and corresponding isolate counts coincided with the expansion of hospital service capacity, with specimen culture density ranging from 56.4 to 74.5 per 1000 patient-days over the study period, indicating that the isolation rates were stable relative to healthcare exposure (Supplementary Table S1).

Among the positive isolates, Gram-negative bacteria predominated (n = 6,151; 67.0%). The top five Gram-negative pathogens were Klebsiella pneumoniae (n = 1,102; 17.9%), Acinetobacter baumannii (n = 991; 16.1%), P. aeruginosa (n = 848; 13.8%), E. coli (n = 654; 10.6%), and Stenotrophomonas maltophilia (n = 462; 7.5%). Gram-positive bacteria accounted for 2,139 isolates (23.3%), with the most prevalent species being S. aureus (n = 389; 18.2%), S. epidermidis (n = 345; 16.1%), Corynebacterium striatum (n = 218; 10.2%), Enterococcus faecium (n = 200; 9.3%), and S. pneumoniae (n = 178; 8.3%). The ten most frequently isolated pathogens over the seven-year period were K. pneumoniae, A. baumannii, P. aeruginosa, E. coli, S. maltophilia, S. aureus, Enterobacter cloacae, S. epidermidis, Burkholderia cepacia, and H. influenzae. Detailed distribution data are presented in Table 1. The number of unique patients and isolate-per-patient distribution are summarized in the Supplementary Table S2.

Table 1.

Species distribution of the clinical isolates in 2018 − 2024

Organism	2018 (n = 844)	2019 (n = 1231)	2020 (n = 1311)	2021 (n = 1264)	2022 (n = 1143)	2023 (n = 1810)	2024 (n = 1674)	Total No. of Strains	Proportion (%)
K. pneumoniae	120	177	186	147	114	217	141	1102	12.0
A. baumannii	103	128	133	125	132	234	136	991	10.8
P. aeruginosa	112	98	137	112	100	156	133	848	9.2
E. coli	68	75	90	68	104	110	139	654	7.1
S. maltophilia	45	62	71	58	52	102	72	462	5.0
S. aureus	52	60	50	60	50	55	62	389	4.2
E. cloacae	40	48	55	45	42	75	46	351	3.8
S. epidermidis	57	79	103	93	95	161	155	345	3.8
B. cepacia	38	45	52	48	43	76	35	337	3.7
H. influenzae	35	40	42	38	32	45	28	260	2.8
C. albicans	31	40	45	35	28	45	25	249	2.7
C. striatum	25	32	40	35	30	36	20	218	2.4
E. faecium	28	35	42	33	25	22	15	200	2.2
S. marcescens	22	28	35	30	25	35	12	187	2.0
S. pneumoniae	25	30	35	28	20	25	15	178	1.9
C. tropicalis	18	22	28	25	18	28	10	149	1.6
S. haemolyticus	20	25	30	25	20	20	8	148	1.6
E. aerogenes	15	20	25	22	18	20	9	129	1.4
S. hominis	18	20	22	20	15	18	8	121	1.3
C. glabrata	12	15	20	18	15	18	5	103	1.1
Others	180	232	290	240	225	352	240	1759	19.2
Total Isolates	844	1231	1311	1264	1143	1810	1674	9180	100.0
Total Specimens Cultured	10,451	14,639	16,030	15,780	14,552	22,335	21,529	-	-
Positivity Rate (%)	8.07	8.41	8.18	8.01	7.85	8.1	7.78	-	-

Distribution characteristics of isolated strains

Among the 9,180 pathogenic isolates, the five most common specimen sources were sputum (n = 4,615; 50.3%), whole blood (n = 1,055; 11.5%), BALF (n = 934; 10.2%), urine (n = 915; 10.0%), and wound exudates/specimens (n = 772; 8.4%), as illustrated in Fig. 1.

Fig. 1.

Sample source of the clinical isolates

A strong male predominance was observed (n = 5,814; 63.3% vs. female: n = 3,366; 36.7%). The age distribution showed that older adults (≥ 60 years) constituted the largest patient group (n = 4,644; 50.6%), followed by middle-aged patients (18–59 years: n = 3,659; 39.9%) and pediatric patients (0–17 years: n = 877; 9.6%). The elderly cohort’s predominance aligns with established age-related susceptibility to healthcare-associated infections (Fig. 2).

Fig. 2.

Baseline Data of Patients. (A) Gender; (B) Age

Pathogens were predominantly isolated from the Adult Cardiac Surgery Intensive Care Unit (ICU) (32.2%, n = 2,956), General ICU (13.3%, n = 1,221), Coronary Heart Disease ICU (11.5%, n = 1,056), Pediatric Cardiac Center ICU (8.4%, n = 771), and the Respiratory Medicine Department (6.5%, n = 597). The isolation rate in the Adult Cardiac Surgery ICU was significantly higher than in all other departments (χ²=1842.7, p < 0.001) (Fig. 3).

Fig. 3.

Distribution of clinical isolates by hospital department

Antimicrobial resistance of Klebsiella pneumoniae, 2018–2024

K. pneumoniae exhibited persistently high resistance to β-lactams, with carbapenem resistance remaining a major clinical concern. Carbapenem resistance (imipenem) showed a non-significant upward trend (β = 0.012 per year, 95% CI: −0.042 to 0.066, p = 0.661), peaking at 35.8% in 2023. Resistance to β-lactam/β-lactamase inhibitor combinations increased significantly, notably for piperacillin-tazobactam (β = 0.055 per year, 95% CI: 0.008 to 0.102, p = 0.022) and ampicillin-sulbactam (reaching 61.9% in 2023). Fluoroquinolone resistance remained substantial but stable, while gentamicin resistance declined significantly (β = −0.048 per year, 95% CI: −0.089 to − 0.007, p = 0.023). Crucially, tigecycline (resistance ≤ 2.9%) and polymyxin B (≤ 4.0%) retained near-complete susceptibility, affirming their role as last-line options. These trends underscore escalating multidrug resistance, particularly to carbapenems and β-lactamase inhibitor combinations. Detailed annual resistance data are provided in Table 2.

Table 2.

Resistance rates of K. pneumoniae to antimicrobial agents in 2018–2024 (%)

Antimicrobial agent	2018	2019	2020	2021	2022	2023	2024
	n = 120	n = 177	n = 186	n = 147	n = 114	n = 217	n = 155
Piperacillin	58.5	51.7	61.7	57.7	56.5	68.0	59.6
Cefoperazone	57.0	52.3	66.2	60.6	54.2	66.0	56.6
Cefuroxime	56.5	53.7	59.7	56.2	57.0	62.4	60.6
Ciprofloxacin	54.3	52.3	56.5	52.2	51.9	62.4	52.3
Cefotaxime	53.4	51.1	58.3	54.7	51.9	63.2	55.6
Ampicillin-sulbactam	50.4	48.9	56.5	48.9	49.5	61.9	44.4
Cefepime	44.9	47.1	56.0	52.6	50.9	59.3	53.6
Trimethoprim-sulfamethoxazole	44.1	37.9	53.6	35.0	32.4	44.8	45.1
Aztreonam	43.6	45.4	48.8	44.5	49.1	56.7	48.4
Tetracycline	41.9	42.5	45.8	46.0	45.4	49.5	45.4
Gentamicin	41.5	39.7	39.3	34.3	34.3	42.8	30.9
Nitrofurantoin	40.0	80.0	90.9	76.9	57.1	54.5	38.5
Levofloxacin	39.0	37.0	41.7	32.1	35.5	45.9	32.7
Amoxicillin-clavulanic acid	37.6	36.8	39.9	31.4	35.2	43.8	30.7
Ceftazidime	36.4	40.8	43.5	38.0	42.6	53.1	41.8
Piperacillin-tazobactam	34.2	33.9	36.3	32.1	34.3	45.4	28.5
Cefoperazone-sulbactam	33.0	34.5	36.6	29.5	35.1	41.5	28.4
Meropenem	30.2	29.9	30.4	26.3	24.1	34.7	26.1
Imipenem	29.6	29.9	31.1	27.7	25.0	35.8	25.7
Chloramphenicol	26.5	27.0	28.6	21.1	41.2	32.4	26.3
Amikacin	25.4	24.7	25.6	20.4	18.5	26.4	15.0
Tigecycline	1.6	0.6	2.7	2.9	0.0	0.5	0.7
Polymyxin B	0.0	0.6	2.7	0.7	0.9	3.4	4.0

Antimicrobial resistance of Acinetobacter baumannii, 2018–2024

A. baumannii demonstrated alarmingly high and sustained resistance to nearly all antimicrobial classes, with a critical surge observed in 2022. Carbapenem resistance (imipenem) remained hyperendemic, showing a concerning though non-significant upward trend (β = 0.070 per year, 95% CI: −0.010 to 0.150, p = 0.085), exceeding 94% for most agents in 2022. Similarly, resistance to third- and fourth-generation cephalosporins (e.g., ceftazidime), fluoroquinolones, and aminoglycosides consistently exceeded 75%. In stark contrast, tigecycline (resistance ≤ 2.6%) and polymyxin B (≤ 0.7%) maintained near-complete susceptibility, confirming their indispensable role as last-resort therapies. These data reveal a predominance of extensively drug-resistant (XDR) strains, highlighting an urgent need for enhanced infection control and stewardship targeting this pathogen. Full resistance profiles are presented in Table 3.

Table 3.

Resistance rates of A. baumannii to antimicrobial agents in 2018–2024 (%)

Antimicrobial agent	2018	2019	2020	2021	2022	2023	2024
	n = 103	n = 128	n = 133	n = 125	n = 132	n = 234	n = 152
Ceftazidime	80.6	77.8	77.3	82.4	94.7	87.3	80.8
Piperacillin	80.6	79.4	75.8	83.1	95.5	88.4	79.6
Ciprofloxacin	80.6	77.8	75.8	81.6	94.7	86.8	79.6
Levofloxacin	80.6	77.0	75.0	81.6	94.7	86.4	78.9
Cefotaxime	79.6	79.4	77.3	80.8	94.7	88.1	78.8
Cefepime	79.6	79.4	77.3	85.6	95.5	86.8	78.3
Piperacillin-tazobactam	80.6	79.4	75.8	81.6	95.5	86.8	78.3
Imipenem	79.2	79.4	76.5	84.0	94.7	86.8	78.3
Meropenem	79.2	80.2	76.5	84.0	94.7	86.8	78.3
Gentamicin	82.5	81.0	76.3	83.2	94.7	86.4	75.7
Ampicillin-sulbactam	79.6	78.6	75.0	81.6	95.5	86.4	74.3
Cefoperazone-sulbactam	71.6	73.2	74.2	75.2	92.4	84.2	70.4
Amikacin	68.0	75.4	70.5	72.8	88.6	82.8	63.2
Trimethoprim-sulfamethoxazole	80.6	71.4	72.7	76.0	87.1	78.3	52.4
Minocycline	/	/	/	69.5	51.1	18.4	43.3
Tigecycline	2.4	0.0	0.0	2.4	0.0	0.0	2.6
Polymyxin B	0.0	0.0	0.0	0.0	0.0	0.4	0.7

Antimicrobial resistance of Pseudomonas aeruginosa, 2018–2024

P. aeruginosa exhibited generally low and declining resistance to most agents, representing a relative bright spot in the resistance landscape. A significant decreasing trend was observed for key antipseudomonal agents: meropenem (β = −0.063 per year, 95% CI: −0.112 to − 0.014, p = 0.013), ciprofloxacin (β = −0.105 per year, 95% CI: −0.173 to − 0.037, p = 0.003), and piperacillin-tazobactam. Aminoglycosides maintained exceptionally low resistance, with gentamicin and amikacin reaching 0.0% in multiple years. Polymyxin B demonstrated 100% susceptibility throughout the study period. These improving susceptibility profiles suggest that, unlike A. baumannii, therapeutic options for P. aeruginosa remain more viable in this setting. Complete annual resistance rates are shown in Table 4.

Table 4.

Resistance rates of P. aeruginosa to antimicrobial agents in 2018–2024 (%)

Antimicrobial agent	2018	2019	2020	2021	2022	2023	2024
	n = 112	n = 98	n = 137	n = 112	n = 100	n = 156	n = 151
Aztreonam	26.0	21.2	20.8	20.8	11.2	16.9	20.0
Imipenem	25.3	18.8	16.8	16.2	13.3	16.2	16.5
Meropenem	19.2	14.1	12.8	10.9	9.2	12.3	16.4
Piperacillin	16.0	15.3	5.6	6.9	10.3	7.0	12.1
Ceftazidime	15.0	11.8	11.2	9.0	9.2	8.5	12.1
Levofloxacin	23.0	22.4	19.2	17.8	14.3	13.8	10.7
Piperacillin-tazobactam	12.0	14.1	4.8	7.0	9.2	5.4	10.7
Cefoperazone-sulbactam	20.0	15.6	9.0	7.5	9.1	4.5	10.0
Cefepime	14.1	14.1	7.2	10.9	12.4	5.4	10.0
Ciprofloxacin	16.0	15.3	10.4	5.0	8.2	3.9	4.3
Gentamicin	11.0	3.5	4.8	1.0	4.1	4.5	0.0
Amikacin	5.1	4.7	0.0	1.0	2.0	0.0	0.0
Polymyxin B	0.0	0.0	0.0	0.0	0.0	0.0	0.0

Antimicrobial resistance of Escherichia coli, 2018–2024

E. coli maintained characteristically high resistance to ampicillin (83.8–97.1%) and fluoroquinolones (ciprofloxacin 57.4–73.3%), consistent with widespread ESBL production. Notably, resistance to β-lactam/β-lactamase inhibitor combinations showed marked and significant improvement: piperacillin-tazobactam resistance decreased by 66% (β = −0.142 per year, 95% CI: −0.226 to − 0.058, p = 0.001). Carbapenems retained exceptional activity, with meropenem resistance remaining low (1.3–9.5%) and showing a declining trend (β = −0.093 per year, 95% CI: −0.203 to 0.017, p = 0.098). Last-resort agents including tigecycline, polymyxin B, and amikacin demonstrated minimal resistance (≤ 6.7%). These findings highlight a successful stewardship outcome in preserving carbapenem efficacy and restoring the utility of certain β-lactamase inhibitor combinations. All resistance data are detailed in Table 5.

Table 5.

Resistance rates of E. coli to antimicrobial agents in 2018 − 2024 (%)

Antimicrobial agent	2018	2019	2020	2021	2022	2023	2024
	n = 68	n = 75	n = 90	n = 68	n = 104	n = 110	n = 151
Ampicillin	97.1	92.0	92.2	83.8	89.4	87.3	86.1
Piperacillin	92.6	89.3	88.9	80.9	86.5	81.2	80.8
Tetracycline	70.6	63.5	60.0	70.1	56.7	64.7	68.9
Ciprofloxacin	57.4	68.0	73.3	63.2	67.3	68.3	64.9
Cefoperazone	77.6	73.3	70.6	61.2	64.4	70.6	64.7
Levofloxacin	52.9	65.3	62.1	57.4	64.4	63.0	58.3
Cefuroxime	72.9	66.7	59.6	58.2	61.5	66.0	56.3
Cefotaxime	75.0	66.7	58.9	57.4	58.7	61.8	53.0
Trimethoprim-sulfamethoxazole	61.8	58.7	64.4	54.4	53.8	59.8	53.0
Cefepime	60.3	62.7	45.6	50.0	50.5	50.0	46.4
Gentamicin	47.1	40.0	52.2	35.3	36.5	33.3	37.7
Aztreonam	39.7	50.7	32.2	33.8	41.3	27.5	27.8
Ampicillin-sulbactam	33.8	48.0	34.4	30.9	25.0	26.5	22.5
Chloramphenicol	15.7	37.5	20.8	32.5	17.2	29.2	21.0
Ceftazidime	26.5	41.3	21.1	22.1	21.2	14.7	15.9
Cefoperazone-sulbactam	9.8	28.0	6.7	9.0	8.7	7.3	9.3
Amoxicillin-clavulanic acid	16.2	33.3	10.0	10.3	5.8	13.7	8.6
Piperacillin-tazobactam	10.3	20.0	5.6	5.9	5.8	4.9	4.0
Nitrofurantoin	6.2	4.0	7.9	4.0	11.1	1.7	2.5
Imipenem	5.9	10.7	4.4	4.4	3.8	5.9	2.0
Meropenem	5.9	9.5	4.4	2.9	3.8	3.9	1.3
Amikacin	2.9	6.7	3.3	0.0	1.9	1.0	1.3
Polymyxin B	2.8	1.3	1.3	0.0	0.0	0.0	0.0
Tigecycline	0.0	0.0	0.0	0.0	0.0	0.0	0.0

Antimicrobial resistance of Staphylococcus aureus, 2018–2024

S. aureus displayed encouraging trends, with a significant and sustained decline in methicillin resistance. Oxacillin resistance (indicative of MRSA) decreased from 59.6% to 35.3% (β = −0.112 per year, 95% CI: −0.185 to − 0.039, p = 0.003). Parallel reductions were observed in macrolide (erythromycin: β = −0.089 per year, 95% CI: −0.151 to − 0.027, p = 0.005) and lincosamide (clindamycin) resistance. Penicillin resistance remained near-ubiquitous (> 92%). Critically, last-line agents including vancomycin, teicoplanin, and linezolid maintained 100% susceptibility, reinforcing their role as therapeutic cornerstones. These trends suggest positive impacts of local antimicrobial stewardship and infection control measures targeting MRSA. Annual resistance rates are provided in Table 6.

Table 6.

Resistance rates of S. aureus to antimicrobial agents in 2018–2024 (%)

Antimicrobial agent	2018	2019	2020	2021	2022	2023	2024
	n = 52	n = 60	n = 50	n = 60	n = 50	n = 55	n = 68
Penicillin G	94.2	96.7	94.0	100.0	96.0	92.3	100.0
Erythromycin	80.8	75.0	74.0	66.1	60.0	56.9	60.3
Clindamycin	62.5	50.0	44.0	39.0	40.0	37.3	36.8
Oxacillin	59.6	60.0	50.0	38.3	42.0	40.7	35.3
Tobramycin	25.0	21.7	18.4	60.0	53.8	48.0	13.6
Trimethoprim-sulfamethoxazole	13.5	13.3	18.0	18.6	16.0	25.0	13.2
Tetracycline	42.3	26.7	24.0	15.3	8.2	19.2	13.2
Levofloxacin	22.7	13.3	10.0	13.3	16.3	11.1	12.3
Ciprofloxacin	21.2	11.7	12.0	13.6	12.2	11.5	11.9
Gentamicin	21.2	20.0	16.0	16.9	10.2	17.3	10.3
Amikacin	9.6	5.0	4.0	1.9	6.0	9.6	9.0
Rifampicin	13.5	6.7	4.0	1.7	0.0	5.8	1.5
Mupirocin	1.9	1.7	4.8	2.7	4.1	0.0	0.0
Linezolid	0.0	0.0	0.0	0.0	0.0	0.0	0.0
Vancomycin	0.0	0.0	0.0	0.0	0.0	0.0	0.0
Teicoplanin	0.0	0.0	0.0	0.0	0.0	0.0	0.0

Antimicrobial resistance of coagulase-negative Staphylococci (CoNS), 2018–2024

CoNS exhibited persistently high and concerning resistance patterns, distinct from MRCNS remained hyperendemic (80.6–91.2%) without a significant trend (β = −0.023 per year, 95% CI: −0.098 to 0.052, p = 0.544). Fluoroquinolone resistance demonstrated a significant and worrying upward trajectory (levofloxacin: β = 0.071 per year, 95% CI: 0.020 to 0.122, p = 0.007). Aminoglycoside resistance varied drastically, with tobramycin resistance spiking to 95.1% in 2022, suggesting potential clonal outbreaks or selective pressure. Vancomycin retained complete activity (0.0% resistance), though sporadic non-susceptibility to teicoplanin (up to 4.7%) emerged, warranting vigilance. These findings underscore CoNS as a reservoir of multidrug resistance in the healthcare setting. Complete data are available in Table 7.

Table 7.

Resistance rates of Coagulase-negative Staphylococci (CoNS) to antimicrobial agents in 2018–2024 (%)

Antimicrobial agent	2018	2019	2020	2021	2022	2023	2024
	n = 57	n = 79	n = 103	n = 93	n = 95	n = 161	n = 155
Penicillin G	91.2	93.2	95.8	97.6	98.8	97.8	94.6
Erythromycin	78.9	86.1	81.4	73.9	76.1	81.2	81.0
Clindamycin	36.8	45.6	47.1	38.0	33.3	48.1	44.2
Oxacillin	87.7	87.3	91.2	87.1	80.4	90.3	80.6
Tobramycin	50.9	53.2	58.0	70.6	95.1	69.4	54.2
Trimethoprim-sulfamethoxazole	69.6	54.4	65.7	56.0	65.6	70.6	56.9
Tetracycline	17.5	12.7	20.8	20.7	17.4	11.2	18.0
Levofloxacin	43.8	40.8	53.5	58.4	54.8	65.6	61.3
Ciprofloxacin	43.8	47.4	59.8	53.3	56.5	60.7	52.3
Gentamicin	48.9	45.6	52.0	39.1	37.6	45.4	41.8
Amikacin	5.3	7.6	9.8	2.3	5.4	6.5	3.9
Rifampicin	24.6	12.7	15.7	13.0	16.1	22.7	13.1
Linezolid	1.8	1.3	2.9	1.1	4.3	2.6	0.6
Vancomycin	0.0	0.0	0.0	0.0	0.0	0.0	0.0
Teicoplanin	1.8	0.0	1.0	4.3	1.1	3.9	4.7

Antimicrobial resistance of Enterococcus, 2018–2024

Enterococcus spp. presented a mixed resistance profile with both alarming and encouraging trends. Fluoroquinolone resistance escalated to extremely high levels (levofloxacin: β = 0.039 per year, 95% CI: 0.002 to 0.076, p = 0.039), severely compromising oral therapeutic options. In contrast, high-level gentamicin resistance (HLGR) declined dramatically from 76.5% to 2.6% (β = −0.342 per year, 95% CI: −0.446 to − 0.238, p < 0.001), which may restore the utility of aminoglycoside synergy in severe infections. Glycopeptide resistance remained rare but showed sporadic, non-significant emergence (vancomycin up to 5.7%; teicoplanin up to 6.2%), necessitating ongoing monitoring. Linezolid preserved 100% susceptibility. These divergent trends highlight the complex and evolving resistance ecology of enterococci. Full resistance trends are summarized in Table 8.

Table 8.

Resistance rates of Enterococcus to antimicrobial agents in 2018–2024 (%)

Antimicrobial agent	2018	2019	2020	2021	2022	2023	2024
	n = 19	n = 29	n = 34	n = 31	n = 55	n = 63	n = 73
Levofloxacin	91.2	83.3	83.3	92.3	88.9	91.7	96.6
Ciprofloxacin	78.6	81.5	86.7	90.9	90.9	87.5	93.5
Nitrofurantoin	88.9	64.7	86.7	88.9	86.4	86.4	77.4
Penicillin G	94.1	53.6	56.5	63.0	87.0	90.2	75.4
Erythromycin	93.8	84.0	42.9	50.0	65.6	72.6	70.6
Ampicillin	88.2	72.4	55.9	78.6	73.6	67.2	64.8
High-concentration gentamicin	76.5	55.2	55.9	19.0	10.8	5.9	2.6
Tetracycline	57.1	74.1	33.3	27.3	56.5	48.0	61.8
Amikacin	70.6	75.0	65.4	68.2	51.0	59.0	59.7
Vancomycin	0.0	0.0	0.0	0.0	5.7	0.0	4.3
Teicoplanin	6.2	0.0	0.0	0.0	0.0	1.6	4.3
Linezolid	0.0	0.0	0.0	0.0	0.0	0.0	0.0

Surveillance of multidrug-resistant organisms from 2018 to 2024

Based on the longitudinal data, several key resistance trends are evident. A. baumannii exhibited the most severe and persistent multidrug resistance, particularly peaking in 2022. In contrast, S. aureus showed a clear and encouraging decline in resistance to key agents like oxacillin and clindamycin. Meanwhile, K. pneumoniae and E. coli presented concerning but distinct profiles, with the former showing high carbapenem resistance and the latter maintaining high fluoroquinolone resistance but greater carbapenem susceptibility (Fig. 4).

Fig. 4.

Heatmap of resistance trends in 2018–2024

Surveillance data revealed distinct temporal patterns across 11 critical resistance phenotypes: CRKPN exhibited fluctuating resistance (25.0–35.8%) without significant linear trend (β = 0.28%/year, p = 0.698, 95%CI: -1.52 to 2.08), while CRABA demonstrated persistently hyperendemic resistance (76.5–94.7%) peaking in 2022 and showing a non-significant upward trajectory (β = 1.38%/year, p = 0.121, 95%CI: -0.36 to 3.12). CRPAE significantly declined from 25.3% to 16.5% (β=-1.42%/year, p = 0.013, 95%CI: -2.54 to -0.30), contrasting with carbapenem-resistant E. coli (CRECO) which decreased from 5.9% to 2.0% albeit non-significantly (β=-0.63%/year, p = 0.173, 95%CI: -1.54 to 0.28). MRSA resistance substantially decreased from 59.6% to 35.3% (β=-3.21%/year, p = 0.017, 95%CI: -5.78 to -0.64), whereas ESBL-producing E. coli (CTX-CRO-R-ECO) declined non-significantly from 75.0% to 53.0% (β=-3.07%/year, p = 0.056, 95%CI: -6.22 to 0.08). Fluoroquinolone-resistant E. coli (QNR-ECO) remained consistently high (57.4–73.3%) without temporal trend (β = 1.23%/year, p = 0.478, 95%CI: -2.17 to 4.63), while ESBL-producing K. pneumoniae (CTX-CRO-R-KPN) fluctuated between 51.1 and 63.2% (p = 0.842). MRCNS maintained hyperendemic resistance (80.6–91.2%, β=-0.92%/year, p = 0.189, 95%CI: -2.29 to 0.45). Vancomycin-resistant enterococci remained rare with E. faecalis (VREF) at 0.0% throughout, though E. faecium (VREM) showed sporadic non-susceptibility (2022:5.0%; 2024:6.5%) without significant trend (Fisher’s exact p = 0.172). Collectively, these data demonstrate divergent evolutionary trajectories, with CRPAE and MRSA exhibiting encouraging declines, CRABA representing a critical persistent threat, and emerging VREM isolates warranting vigilant monitoring despite low absolute prevalence. Detailed data are presented in Fig. 5.

Fig. 5.

Trend of detection rate of multidrug-resistant bacteria in 2018–2024

Clinical impact and epidemiological characteristics of resistant infections

Stratified analysis of resistance rates for key pathogen-drug combinations by specimen type and patient location revealed notable disparities (Supplementary Table S3). Overall, isolates from ICU patients and respiratory specimens (sputum) exhibited the highest resistance rates. For instance, meropenem resistance in A. baumannii was substantially higher in the ICU (90.2%) compared to non-ICU settings (76.8%). Similarly, meropenem resistance in K. pneumoniae from sputum specimens (37.2%) exceeded that from blood (18.5%) or urine (15.8%). These findings underscore the importance of considering both the site of infection and the patient care environment when formulating empiric antibiotic therapy.

Analysis of clinical outcomes associated with resistant phenotypes revealed that patients infected with MDROs faced significantly increased mortality risks and prolonged hospitalization (Supplementary Table S4). For instance, patients with CRABA infection had significantly higher 30-day all-cause mortality (40.0% vs. 20.6%) and median hospital length of stay (38.5 days vs. 25.0 days) compared to those with susceptible strains. Furthermore, infection with MRSA and CRKPN was also associated with significantly elevated mortality risk (adjusted HR 1.72–2.51) and longer ICU stays. These data underscore the substantial clinical burden imposed by multidrug-resistant infections.

Discussion

Dynamic surveillance of clinical bacterial distribution and AMR constitutes a critical research domain in clinical microbiology and infection control, providing essential scientific evidence for developing precision anti-infective strategies through elucidating pathogen distribution dynamics and resistance phenotype evolution. Our study analyzing clinical isolates from Fuwai Central China Cardiovascular Hospital (2018–2024) revealed a progressive increase in specimen culture volumes, with pathogen positivity rates remaining stable (7.78–8.41%). Gram-negative bacteria consistently predominated (> 65% of isolates), with K. pneumoniae, A. baumannii, and P. aeruginosa representing the top three pathogens—aligning with CARSS surveillance data [11] but contrasting with global patterns: Ugandan surveillance (2018–2021) [12] reported S. aureus and Salmonella spp. as most prevalent, while other African studies identified S. aureus, Klebsiella spp., and Proteus spp. as dominant [13]. The significantly higher pathogen isolation rate observed in the Adult Cardiac Surgery ICU, compared to all other departments, underscores this unit as the epicenter of healthcare-associated infections within our cardiovascular specialty hospital. This finding is multifactorial, reflecting the unique vulnerability of its patient population. These patients typically undergo the most complex and invasive surgical procedures (e.g., open-heart surgery with cardiopulmonary bypass), which are associated with prolonged operative times, extensive tissue trauma, and frequent use of invasive devices such as endotracheal tubes, central venous catheters, and chest drains. The confluence of these factors creates multiple portals of entry for pathogens and induces a state of systemic immunosuppression, substantially elevating the risk of postoperative infections like ventilator-associated pneumonia, surgical site infections, and bloodstream infections.

Analysis of antimicrobial resistance surveillance data from this medical center during 2018–2024 revealed a severe bacterial resistance profile: the detection rates of CRKPN, CRABA, CRECO, MRSA, MRCNS, QNR-ECO, CTX-CRO-R-KPN, and CTX-CRO-R-ECO all exceeded the 2023 national levels reported by the CARSS, underscoring regional variability and the need for localized surveillance [11]. This elevated resistance may be attributed to the unique characteristics of cardiovascular specialty hospitals, where patients often present with multiple comorbidities, compromised immune function, and require advanced invasive interventions. Contributing factors include the concentration of critically ill patients, frequent invasive procedures, extensive use of broad-spectrum antibiotics, and high transmission risks of drug-resistant pathogens in ICU environments.

The persistently high and often increasing resistance rates among K. pneumoniae and A. baumannii, particularly to carbapenems (Meropenem resistance in K. pneumoniae peaked at 34.7% in 2023; A. baumannii exceeded 94% for Imipenem/Meropenem in 2022) and broad-spectrum β-lactams/β-lactamase inhibitors, strongly suggest the endemic presence and possible clonal dissemination of multidrug-resistant (MDR) and XDR strains. The high-level, cross-class resistance profile observed in A. baumannii is characteristic of successful epidemic clones that accumulate multiple resistance determinants, rather than arising solely from sporadic acquisition of single carbapenemase genes. Although molecular characterization was not performed in this study, the high carbapenem resistance observed in K. pneumoniae and A. baumannii may be attributable to the dissemination of carbapenemase genes such as blaKPC, blaNDM, and blaOXA-48-like, which are prevalent in China [14–19]. The potential co-carriage of these genes with others conferring resistance to fluoroquinolones and aminoglycosides, as suggested by the co-resistance patterns in our data, would severely limit therapeutic options and complicate outbreak management. Future molecular studies are warranted to identify the prevailing clones and resistance gene platforms to inform targeted infection control interventions. Alarmingly, A. baumannii demonstrated near-pan-resistance (> 75% for most agents) by 2022–2023, indicating severely limited therapeutic options; even traditionally reliable agents like Amikacin exhibited rising resistance (peaking at 88.6% in 2022), likely due to the combined effects of aminoglycoside-modifying enzymes and 16 S rRNA methylases [20, 21]. The clinical implication is that empirical therapy for suspected A. baumannii infections in high-risk units must assume extensive drug resistance, often necessitating immediate consideration of combination regimens with last-resort agents like polymyxins or tigecycline, which are themselves associated with higher toxicity and require therapeutic drug monitoring. While P. aeruginosa generally showed lower and often decreasing resistance (e.g., Meropenem resistance declined from 19.2% to 16.4%), the rebound in Aztreonam resistance to 20.0% in 2024 warrants vigilance against mechanisms like AmpC derepression or the emergence of metallo-β-lactamases that can hydrolyze this agent [22]; however, the preserved susceptibility to agents like Polymyxin B, Amikacin, and Gentamicin suggests that membrane permeability and target site alterations remain uncommon locally [23].

The sporadic but concerning emergence of resistance to last-resort agents like Polymyxin B (rising to 4.0% in K. pneumoniae and 0.7% in A. baumannii by 2024) and Tigecycline (peaking at 2.9% in K. pneumoniae) warrants serious concern and pre-emptive action. Although molecular characterization was not performed in this study, the emerging resistance to polymyxin B, particularly if coupled with carbapenem resistance, could be hypothetically linked to the acquisition of mobile mcr genes or mutations in chromosomal regulatory systems (pmrAB, phoPQ), which have been reported in China and pose a grave threat due to their potential for horizontal transfer [24, 25]. If confirmed locally, this would necessitate enhanced surveillance for colistin resistance and strict isolation precautions for carriers. A review of the literature indicates that the tigecycline resistance documented in this study is likely mediated mainly by the overexpression of efflux pumps (e.g., mediated by ramR mutations) [26], a mechanism that can contribute to reduced susceptibility across multiple antibiotic classes. Future molecular studies are needed to clarify local mechanisms and their genetic context (plasmid-borne vs. chromosomal), which directly impacts the risk of further spread.

E. coli maintained high resistance to Ampicillin, Piperacillin, and Fluoroquinolones (Ciprofloxacin ~ 65–73%), consistent with widespread plasmid-mediated ESBLs (e.g., CTX-M) and qnr genes [27], but crucially retained susceptibility to carbapenems (< 6% for Meropenem), Piperacillin-Tazobactam (< 5.6%), and Amikacin (< 6.7%), highlighting the effectiveness of stewardship in preserving these agents. Clinically, these data mandate rigorous antimicrobial stewardship: avoiding carbapenems for suspected ESBL E. coli where Piperacillin-Tazobactam or Cefoperazone-Sulbactam remain viable (resistance < 9.3%), using Amikacin or Polymyxin B judiciously for P. aeruginosa and A. baumannii MDR infections, and reserving Tigecycline/Polymyxins for confirmed XDR K. pneumoniae. It is worth noting that empiric therapy should be guided by local antibiogram and patient severity. Hospitals must implement enhanced infection control: strict contact precautions for carriers of MDR A. baumannii or CRE, active surveillance cultures in high-risk units, robust environmental decontamination given A. baumannii’s persistence, and real-time resistance data dissemination to guide empirical therapy [28]. Continuous resistance monitoring and molecular epidemiology studies are essential to detect emerging resistance mechanisms early and adapt treatment protocols accordingly. Integrating molecular surveillance (e.g., PCR for carbapenemases, whole-genome sequencing) into routine practice could enhance early detection of resistance outbreaks and guide targeted interventions. While resource-intensive, such approaches are increasingly feasible in specialized centers and could be prioritized in high-risk units.

Among Gram-positive pathogens, distinct evolutionary and epidemiological patterns were observed. The persistently high penicillin resistance in staphylococci (> 90%) is a baseline expectation, but the diverging trends in methicillin resistance are noteworthy. The significant decline in MRSA suggests possible success of targeted infection control measures against this classic pathogen, potentially reducing the prevalence of healthcare-associated clones carrying the mecA gene encoding PBP2a [29]. Erythromycin and Clindamycin resistance, often linked to erm genes conferring MLS < sub > B</sub > resistance [30], shows a promising decreasing trend in S. aureus but remains high and stable in CoNS and Enterococcus. Enterococcus exhibits alarmingly high resistance to fluoroquinolones (Levofloxacin > 83%, Ciprofloxacin > 78%) and Penicillin G/Ampicillin, likely due to alterations in drug targets (gyrases, topoisomerases) and low-affinity PBPs, respectively [31, 32]; however, a significant and encouraging decline in High-concentration Gentamicin resistance (76.5% to 2.6%) suggests reduced prevalence of high-level aminoglycoside resistance (HLAR) mechanisms like aminoglycoside-modifying enzymes [33]. The volatile Tobramycin resistance in both Staphylococci, notably the sharp spike in CoNS (95.1% in 2022), warrants investigation into local aminoglycoside usage or potential clonal outbreaks. Critically, resistance to last-line agents Vancomycin, Teicoplanin, and Linezolid remains exceptionally low or absent in S. aureus and minimal in CoNS/Enterococcus, though the sporadic low-level resistance to Glycopeptides in Enterococcus (Vancomycin up to 5.7%, Teicoplanin up to 4.7%) and CoNS (Teicoplanin up to 4.7%) necessitates vigilant monitoring for emerging van gene-mediated resistance [34, 35].

Consequently, empiric therapy in this cardiovascular center must be highly nuanced. For suspected staphylococcal infections, vancomycin remains the cornerstone, especially in ICU settings or post-surgery, given the high MRCNS burden and fluctuating MRSA rates. The dramatic decline in HLGR in enterococci is practice-changing; for serious infections like prosthetic valve endocarditis, ampicillin plus gentamicin can again be considered a viable synergistic regimen once susceptibility is confirmed, reducing reliance on last-line agents. However, the extreme fluoroquinolone resistance in enterococci mandates that this class not be used empirically for conditions like urinary tract infections in patients with recent healthcare exposure. This longitudinal dataset provides more than a warning; it offers a blueprint for a precision infection management strategy. The persistence of high resistance, particularly the XDR A. baumannii and the divergent trends in gram-positive pathogens, underscores that a one-size-fits-all approach to infection control and stewardship is insufficient. For example, the response to a surge in A. baumannii resistance (as observed in 2022) or a spike in CoNS tobramycin resistance must differ. The former likely requires enhanced environmental decontamination and review of device utilization, while the latter might prompt an investigation into a specific drug usage pattern or a point-source outbreak. The instance where enhanced measures coincided with a decline in resistance rates is a compelling, real-world example of the effectiveness of data-driven intervention. Therefore, our findings argue for a dynamic, pathogen-specific, and even unit-specific approach. Antimicrobial stewardship programs must evolve beyond broad restriction policies to incorporate real-time resistance data and rapid diagnostic results to guide therapy. Infection control protocols should be tiered based on the transmission dynamics and environmental persistence of the target MDRO (e.g., more stringent measures for A. baumannii versus S. aureus).

The findings of this study provide critical evidence for formulating targeted anti-infection strategies, optimizing antimicrobial stewardship (AMS), and enhancing infection prevention and control (IPC) in cardiovascular specialty hospitals. We propose establishing a multidimensional prevention system comprising the following key components: establishing an antimicrobial stewardship team to rigorously enforce prescribing criteria for carbapenems and glycopeptides; implementing pre-emptive isolation for patients colonized with multidrug-resistant organisms (MDROs) combined with weekly environmental surveillance in high-risk departments; developing machine learning-based predictive models for resistance risk to guide empirical therapy; deploying rapid point-of-care molecular diagnostics to shorten time-to-diagnosis; and creating an early-warning scoring system for post-cardiovascular intervention infections. By integrating microbiological data with clinical characteristics, a closed-loop management model encompassing “rapid diagnosis – precision treatment – dynamic monitoring – timely adjustment” can be established. This integrated approach is expected to significantly improve the accuracy of empirical anti-infective therapy and effectively curb the transmission of resistant pathogens.

Limitations of this study

It should be noted that our evaluation and discussion are limited to the cohort from our single institution, which may not be fully representative of broader patterns. The retrospective design may introduce selection bias. Furthermore, the absence of molecular characterization of resistance mechanisms means that statements regarding the genetic basis of resistance (e.g., carbapenemase genes) remain speculative. Future multicenter studies incorporating molecular epidemiology and clinical outcomes are warranted to obtain a more comprehensive understanding. Additionally, prospective evaluation of the proposed multidimensional prevention system is warranted.

Conclusion

In summary, our data demonstrate alarming antimicrobial resistance rates among major clinical isolates to first-line antibiotics. The persistent emergence of novel MDROs poses substantial therapeutic challenges for the management of infectious diseases. These findings underscore the critical need to sustain antimicrobial resistance surveillance programs, optimize antibiotic stewardship protocols, and reinforce hospital infection control measures to curb the spread and transmission of resistant pathogens.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

AMR

Antimicrobial Resistance

AMS

Antimicrobial Stewardship

AST

Antimicrobial Susceptibility Testing

BALF

Bronchoalveolar Lavage Fluid

CARSS

China Antimicrobial Resistance Surveillance System

CLSI

Clinical and Laboratory Standards Institute

CoNS

Coagulase-Negative Staphylococci

CRE

Carbapenem-Resistant Enterobacterales

CRABA

Carbapenem-Resistant Acinetobacter baumannii

CRKPN

Carbapenem-Resistant Klebsiella pneumoniae

CRPAE

Carbapenem-Resistant Pseudomonas aeruginosa

CTX-CRO-R-KPN

ESBL-producing K. pneumoniae

CTX-CRO-R-ECO

ESBL-producing E. coli

ESBL

Extended-Spectrum β-Lactamase

EUCAST

European Committee on Antimicrobial Susceptibility Testing

HLGR

High-Level Gentamicin Resistance

ICU

Intensive Care Unit

MDR

Multidrug-Resistant

MDRO

Multidrug-Resistant Organism

MRSA

Methicillin-Resistant Staphylococcus aureus

MRCNS

Methicillin-Resistant Coagulase-Negative Staphylococci

QNR-ECO

Fluoroquinolone-resistant E. coli

VRE

Vancomycin-Resistant Enterococci

XDR

Extensively Drug-Resistant

Author contributions

Q. Wang participated in the conception, design, analysis, interpretation of the study, and wrote the draft manuscript; F.Y. Wang, P.S. Xu, and H.Z. Liu critically reviewed the report and manuscript drafts for intellectual content, and made substantive revisions; Q.S. Zhang, Y.Y. Dong, Y.l. Qiao, and T. Li reviewed the final manuscript for intellectual content and scientific integrity. All authors read and approved the final manuscript.

Funding

The authors declare that they received financial support for the research, authorship, and/or publication of this article. This work was supported by the General Program of the Natural Science Foundation of Henan Province (242300421291), Provincial-Ministerial Co-Construction Key Project of Henan Medical Science and Technology Research Program (SBGJ202402005), Young and Middle-aged Academic Leaders Training Program in Health System of Henan Province (HNSWJW-2022010), and Excellent Young Talents Program of Scientific and Technological Innovation in Health Care for Young Researchers in Henan Province (YXKC2022043).

Data availability

All data used to analyze and generate the results of this study are included in this article.

Declarations

Ethics approval and consent to participate

This retrospective study was conducted in accordance with the principles of the Declaration of Helsinki. It was approved by the Ethics Committee of Fuwai Central China Cardiovascular Hospital (approval number: 2023-LS-84). The requirement for both consent to participate and individual informed consent was waived by the Ethics Committee of Fuwai Central China Cardiovascular Hospital due to the use of de-identified data and the retrospective nature of the analysis.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.