The relationship between catheter-related bloodstream infection and multi-drug resistant bacteria: a five-year retrospective study

Abstract

Background

Catheter-related bloodstream infection (CRBSI) is a severe hospital-acquired infection primarily associated with using central venous catheters (CVCs). With the rise of antibiotic resistance and the emergence of multidrug-resistant organisms (MDROs), there is a growing concern about the relationship between MDROs and CRBSI.

Objective

To explore the association between MDROs and CRBSI.

Methods

A retrospective study was conducted on hospitalized patients with implanted CVCs over 54 months (January 2019 to June 2023).

Results

We included 50,718 patients with indwelling central venous catheters; 122 were diagnosed with central venous CRBSI) The incidence of CRBSI was 0.24% (122/50,718 patients). The rate of MDRO infections among non-CRBSI patients was only 3.86%, while the rate among CRBSI patients was 62.30% (χ² = 1065.80, p < 0.001). After excluding factors related to age, gender, medical history, environmental factors (ward of residence, length of hospitalization), catheter factors (catheter placement site, duration of catheterization, number of insertions), and other factors (duration of fever, abnormal white blood cell counts, frequency of mechanical ventilation) associated with CRBSI infection, the odds ratio for MDROs was 4.63 (95% CI: 2.86–7.50; P < 0.001; Fully adjusted model).

Conclusion

MDROs are independently associated with a higher incidence of CRBSI. These findings highlight the critical need for targeted infection control strategies against MDRO-associated CRBSI and formulate specific management strategies for these infections while promoting interdisciplinary collaboration to improve patient outcomes.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12879-025-11367-7.

Introduction

Catheter-related bloodstream infections (CRBSI) are a severe form of hospital-acquired infection, primarily associated with using central venous catheters (CVSs) [1]. Central line catheters are a significant risk factor for bloodstream infections, with more than 250,000 cases of hospital-acquired central line-associated bloodstream infections (CRBSI) reported annually in the United States [2]. The incidence rates in developing nations are significantly higher than in developed countries. Studies have shown that in developing countries, the incidence can range from 1.7 to 44.6 per 1,000 catheter days [3].

CRBSI is associated with increased mortality, prolonged hospitalization, and elevated healthcare costs. In a prospective nationwide surveillance study of 2,563 nosocomial bloodstream infections (nBSIs) across Brazil (where 70.3% of patients had CVCs present), the overall crude mortality rate reached 40% [4]. This exceeds the 27% crude mortality rate reported for nosocomial bloodstream infections in the United States [5]. Treating these infections can be prohibitively expensive, with costs soaring to as much as $39,000 in the United States and approximately $90,000 in developing countries [6, 7].

In recent years, a concerning trend has emerged: The emergence of antibiotic resistance among the ESKAPE pathogens (including Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter species) poses significant challenges to the prevention and treatment of these infections [8, 9]. In the United States, infections caused by multidrug-resistant organisms (MDROs) result in over 2 million infections and 23,000 deaths annually, leading to a two- to three-fold increase in hospitalization costs and a 35–60% increase in 1-year mortality rates [10, 11]. In CRBSI, MDROs account for 20–67% of cases [12], highlighting the urgent need for healthcare professionals to understand whether MDROs are associated with the rates of CRBSI.

Although there are many studies on CRBSI, the literature specifically addressing CRBSI related to MDROs is relatively scarce [13, 14]. Some studies may focus on the overall pathogen types associated with CRBSI without delving into the impact of specific multidrug-resistant bacteria on the occurrence of CRBSI [15, 16]. Specific MDROs, such as methicillin-resistant Staphylococcus aureus (MRSA) and multidrug-resistant Escherichia coli, may play a significant role in the pathogenesis of CRBSI, yet related research remains insufficient [17, 18]. Research on the specific role of the hospital environment in CRBSI and its relationship with MDROs is relatively scarce [19, 20].

This study hypothesizes an association between MDROs and CRBSI, aiming to confirm the role of MDROs in CRBSI through data analysis. Additionally, clinical factors such as patients’ underlying conditions, types of catheters used, hospital departments, and duration of hospitalization are investigated for their association with CRBSI. By exploring the relationship between MDROs and CRBSI, this study seeks to provide insights into how these resistant pathogens, along with patient conditions, catheter types, and hospital settings, contribute to the incidence of CRBSI.

Materials and methods

A retrospective cohort study was conducted at LiHuili Hospital, a tertiary medical center in Ningbo, China, a large tertiary hospital in Ningbo city, Zhejiang province, covering 54 months from January 2019 to June 2023. The observational study was approved by the Institutional Review Board of Li Huili Hospital of Ningbo Medical Center approved the study (KY2024SL105).

Data source

This structured data collection aims to provide a comprehensive overview of the patient population and the factors associated with catheter usage and infection control. The collected data includes the following categories:

1. Demographic characteristics: Age, gender, department, surgical history, diabetes and hypertension status, history of malignant disease, and length of hospital stay.

2. Catheter factors: Catheter insertion site, duration of catheterization, and the number of central venous catheterizations.

3. Factors associated with infection: Maximum consecutive fever days, number of abnormal white blood cell counts, mechanical ventilation, and presence of multi-drug resistant organisms.

We retrospectively collected study data from two different sources:

The resident database contains demographic and clinical data from the electronic medical record system. The data were initially collected during patient admission and clinical courses as part of routine hospital operations, and we retrospectively accessed these records for analysis.

Hospital infection control database: includes infection-related factors, such as blood test results and multi-drug resistant organisms, collected by the hospital infection control team through standardized surveillance protocols.

This retrospective study included patients who met the following criteria:

Inclusion criteria: (1) Age ≥ 18 years; (2)Hospitalized with an indwelling central venous catheter (CVC) during the study period (from January 2019 to June 2023), including central venous catheters such as subclavian vein catheters, internal jugular vein catheters, axillary vein catheters, femoral vein catheters, and peripherally inserted central catheters (PICC); (3) Complete clinical and laboratory data available in both the electronic medical record and hospital infection control databases; (4) Catheter indwelling time ≥ 48 h.

Exclusion criteria: (1) Incomplete or missing key variables (e.g., blood culture results, catheter insertion site); (2) Loss to follow-up due to transfer or death during the study period; (3) Unclear diagnosis of catheter-related infection or failure to meet CRBSI diagnostic criteria.

Catheter protocol

Our hospital uses double-lumen and single-lumen catheters (LPPCVC model, provided by Lepu Medical Technology (Beijing) Co., Ltd.). Double-lumen catheters, which allow for simultaneous infusion of medications and parenteral nutrition solutions, are predominantly used, while single-lumen catheters are used in a smaller percentage (approximately 20% of cases).

Infection control protocol

All catheter operations follow CDC-based hospital protocols. Maximal sterile barriers, chlorhexidine/ethanol skin disinfection, and trained personnel (Physicians for CVCs, Specialist Nurses for PICCs) are used for catheter insertion. Dressings are changed regularly/as needed. Connection ports are disinfected for > 15 s. Heparinized lumen maintenance is performed. Monthly compliance audits are conducted [21]. Supplementary File 1.

Definitions

The definitions for catheter infection and colonization are derived from the guidelines established by the Centers for Disease Control and Prevention (CDC) for bloodstream infections and the semi-quantitative culture technique developed by Maki et al. [22].

CRBSI is diagnosed based on [12]: Concordant positive blood cultures from a peripheral vein and the central venous catheter (CVC) with the same organism; Higher colony count or shorter time-to-positivity in the CVC sample; and Exclusion of other infection sources(e.g., pneumonia, urinary tract infection). This study uses CRBSI as the primary outcome due to its direct link between infection and catheter use.

MDROs refer to microorganisms demonstrating antimicrobial resistance to at least one drug in three or more different antimicrobial categories. Catheter colonization: The presence of ≥ 15CFU of a single organism per catheter if not accompanied by a laboratory-confirmed BSI of the patient [23].

Evaluation of CVC infection

The diagnostic criteria for CRBSI typically involve a comprehensive assessment of clinical manifestations, laboratory tests, and microbiological evidence [22].

1. Clinical manifestations: Does the patient exhibit symptoms of infection such as fever or chills.

2. Catheter site examination: Check whether the catheter insertion site shows signs of redness, exudate, or other indications of infection.

3. Blood cultures: Simultaneously collect blood samples from the peripheral vein and the CVC for culture. If the culture from the CVC is positive and the peripheral blood culture is also positive, it is generally considered a primary bloodstream infection.

4. Assessment of infection sources: Other potential sources of infection, such as pneumonia or urinary tract infections, should be ruled out.

Actual bloodstream infection requires at least two blood cultures yielding the same organism for common skin commensals (e.g., coagulase-negative staphylococci and Corynebacterium sp.). A differential time-to-positivity (DTP) of ≥ 2 h for the CVC versus the peripheral sample supports catheter-related infection. Other infection sources (e.g., pneumonia, urinary tract infection) must be excluded to confirm catheter-related bloodstream infection and minimize contamination misclassification.

Evaluation of MDROs

1. Laboratory-confirmed infection meeting both criteria: Microbiological evidence: Positive culture from a sterile site (blood, BAL, etc.) or significant growth (≥ 10⁵ CFU/mL) from the non-sterile site with compatible symptoms.

2. Clinical correlation: Documented signs/symptoms of infection; Physician diagnosis of active infection; Targeted antimicrobial therapy initiated; Absence of alternative explanation for findings.

Data collection and outcome identification

Potential CRBSI cases were initially screened through electronic health records using predefined criteria. Trained personnel then performed manual chart review applying strict CDC criteria. This involved extracting clinical documentation of symptoms and catheter site findings, reviewing microbiology reports (peripheral/CVC culture concordance, colony counts, time-to-positivity), and verifying alternative infection sources through progress notes and diagnostic reports. Cases meeting CRBSI criteria but with confirmed non-catheter sources of MDRO infection (e.g., pulmonary infection with MRSA, urinary tract infection with carbapenem-resistant Enterobacteriaceae) were systematically excluded from the CRBSI cohort.

Statistical analysis

We analyzed the data using R language 4.2.3. Descriptive statistics were employed to characterize the patient population, reporting counts and mean values (+/− standard deviation) for qualitative and quantitative variables, and the median (M) with interquartile range (Q₁, Q₃) for skewed variables. As appropriate, comparisons between groups were made using the Chi-square test or the Z-test (Wilcoxon rank-sum test). We employed a multi-model logistic regression strategy to assess the association between MDROs and CRBSI. Variables were selected a priori based on clinical relevance and entered into the model using a stepwise approach. Model validation was performed using 10-fold cross-validation to ensure robustness, focusing on the association between MDROs and central line-associated bloodstream infections (CRBSI). A p-value of<0.05 was considered statistically significant.

Results

Descriptive statistics

Among 50,718 patients with indwelling CVCs, 122 (0.24%) were diagnosed with CRBSI. Among the 1,420 Intensive Care Unit (ICU) patients, 70 individuals (4.9%) were diagnosed with CRBSI. The cohort’s mean age was 58.68 ± 15.34 years, with CRBSI patients significantly older than non-CRBSI patients (69.67 ± 14.62 vs. 58.66 ± 15.33, p < 0.001). Males represented 55.29% of the total population but accounted for 71.31% of CRBSI cases (p < 0.001).

Medical history analysis revealed a higher prevalence of diabetes (33.61% vs. 12.54%, p < 0.001) and hypertension (50.00% vs. [non-infected group percentage], p < 0.001) in CRBSI patients compared to non-CRBSI patients. (Table 1)

Table 1.

Demographic and clinical characteristics of the study population

Variables	Total (n = 50718)	Non-CRBSI (n = 50596)	CRBSI (n = 122)	P
Age, Mean ± SD (years)	58.68 ± 15.34	58.66 ± 15.33	69.67 ± 14.62	< 0.001
Sex, n (%)				< 0.001
Male	28,040 (55.29)	27,953 (55.25)	87 (71.31)
Female	22,678 (44.71)	22,643 (44.75)	35 (28.69)
Clinical characteristics
Surgical history, n (%)				0.077
No	10,435 (20.57)	10,402 (20.56)	33 (27.05)
Yes	40,283 (79.43)	40,194 (79.44)	89 (72.95)
Diabetes mellitus, n (%)				< 0.001
No	44,331 (87.41)	44,250 (87.46)	81 (66.39)
Yes	6387 (12.59)	6346 (12.54)	41 (33.61)
Hypertension, n (%)				< 0.001
No	34,274 (67.58)	34,213 (67.62)	61 (50.00)
Yes	16,444 (32.42)	16,383 (32.38)	61 (50.00)
Oncologic disease, n (%)				< 0.001
No	30,569 (60.27)	30,477 (60.24)	92 (75.41)
Yes	20,149 (39.73)	20,119 (39.76)	30 (24.59)
Hospital length of stay, median (Q₁, Q₃) (days)	9.00 (5.00, 14.00)	9.00 (5.00, 14.00)	40.00 (25.00, 73.75)	< 0.001
Hospital department, n (%)				< 0.001
ICU	1420 (2.80)	1350 (2.67)	70 (57.38)
Surgical department	31,916 (62.93)	31,887 (63.02)	29 (23.77)
Medical department	17,382 (34.27)	17,359 (34.31)	23 (18.85)
Catheter insertion site, n (%)				< 0.001
Internal jugular vein	8770 (17.29)	8731 (17.26)	39 (31.97)
Subclavian vein	11,797 (23.26)	11,778 (23.28)	19 (15.57)
PICC	4550 (8.97)	4503 (8.90)	47 (38.52)
Femoral vein	1694 (3.34)	1687 (3.33)	7 (5.74)
Axillary vein	23,907 (47.14)	23,897 (47.23)	10 (8.20)
Catheter indwelling time				< 0.001
M (Q₁, Q₃) (days)	8.00 (4.25, 13.00)	8.00 (4.13, 13.00)	28.33 (10.00, 52.00)
Number of CVC insertions, n (%)				< 0.001
One time	41,886 (82.59)	41,835 (82.68)	51 (41.80)
Two times	5080 (10.02)	5046 (9.97)	34 (27.87)
Three times	3752 (7.40)	3715 (7.34)	37 (30.33)
Additional clinical factors
Longest duration of consecutive fever, median (Q₁, Q₃) (days)	0.00 (0.00, 1.00)	0.00 (0.00, 1.00)	4.00 (2.00, 7.00)	< 0.001
Number of abnormal white blood cell counts, median (Q₁, Q₃)	0.00 (0.00, 1.00)	0.00 (0.00, 1.00)	2.00 (1.00, 2.00)	< 0.001
Number of days on mechanical ventilation, median (Q₁, Q₃)	0.00 (0.00, 0.00)	0.00 (0.00, 0.00)	1.00 (0.00, 1.75)	< 0.001
Multidrug-resistant organisms (MDROs), n (%)				< 0.001
No	48,687 (96.00)	48,641 (96.14)	46 (37.70)
Yes	2031 (4.00)	1955 (3.86)	76 (62.30)

t: t-test, Z: Mann-Whitney test, χ²: Chi-square test

SD: standard deviation, M: Median, Q₁: 1 st Quartile, Q₃: 3 st Quartile

Factors associated with CRBSI

Environmental factors analysis revealed that CRBSI patients had a significantly more extended median hospital stay than non-CRBSI patients (p < 0.001). The majority of CRBSI cases (57.38%) occurred in the ICU (p < 0.001).

Catheter-related factors showed that insertion site was significantly associated with CRBSI risk (χ² = 178.85, p < 0.001), with higher rates observed for internal jugular (31.97%) and PICC (38.52%) insertions. CRBSI patients also had longer catheter indwelling times (Z = −11.05, p < 0.001) and a higher proportion of multiple catheter insertions (χ² = 150.46, p < 0.001).

Other clinical factors included a longer duration of fever in CRBSI patients (Z = −20.74, p < 0.001). CRBSI patients also exhibited more abnormal white blood cell counts (Z = −13.26, p < 0.001) and a greater need for mechanical ventilation (Z = −23.36, p < 0.001).

Among 50,718 patients with CVCs, 2,031 (4.00%) had multidrug-resistant organism (MDRO) infections. CRBSI was strongly associated with MDRO prevalence, with 62.30% of CRBSI patients having MDRO infections compared to 3.86% of non-CRBSI patients (χ² = 1065.80, p < 0.001). (Table 1)

MDROs and CRBSI: a multimodel analysis

The multiple regression analyses demonstrated a strong and consistent association between MDROs and CRBSI across all models (Table 2). In the unadjusted model, MDRO-positive status was associated with a 41-fold increased odds of CRBSI (OR = 41.11, 95% CI: 28.43–59.45, p < 0.001). After adjusting for demographic and fundamental clinical factors (Model 1: age, sex, surgical history, diabetes mellitus, hypertension, and oncologic disease), the association remained highly significant (OR = 30.36, 95% CI: 20.82–44.26, p < 0.001). Further adjustment for comprehensive clinical and catheter-related variables (Fully adjusted model) attenuated but maintained a statistically significant association (OR = 4.63, 95% CI: 2.86–7.50, p < 0.001), suggesting MDROs are an independent risk factor for CRBSI even after accounting for potential confounders.

Table 2.

Association between multidrug-resistant organisms (MDROs) and CRBSI in multiple regression models

Variables	Unadjusted model	P	Model 1	P	Fully adjusted model	P
	OR (95%CI)		OR (95%CI)		OR (95%CI)
MDROs
No	1.00 (Reference)		1.00 (Reference)		1.00 (Reference)
Yes	41.11 (28.43 ~ 59.45)	< 0.001	30.36 (20.82 ~ 44.26)	< 0.001	4.63 (2.86 ~ 7.50)	< 0.001

OR: Odds Ratio, CI: Confidence Interval

Model 1: Adjusted for age, sex, surgical history, diabetes mellitus, hypertension, and oncologic disease

Fully adjusted model: Adjusted for age, sex, surgical history, diabetes mellitus, hypertension, oncologic disease, hospital length of stay, hospital department, catheter insertion site, catheter indwelling time, number of CVC insertions, longest duration of consecutive fever, number of abnormal white blood cell counts, and number of days on mechanical ventilation

Distribution and co-infection of MDROs

Among single-strain infections, Carbapenem-resistant Enterobacteriaceae (CRE) had the highest incidence at 509 cases, followed by MRSA with 483 cases. In two-strain infections, significant co-infections included Carbapenem-resistant Acinetobacter baumannii (CRAB) and Multidrug-resistant Acinetobacter baumannii (MDR-AB) at 301 cases. For three-strain infections, notable combinations included CRAB, MDR-AB, and Pandrug-resistant Pseudomonas aeruginosa (PDR-AB) (79 cases). CRAB, MRSA, MDR-AB, and PDR-AB co-infected 62 cases in four-strain infections. The five-strain combination of CRAB, Carbapenem-resistant Pseudomonas aeruginosa (CRPA), MDR-AB, PDR-AB, and Pandrug-resistant Acinetobacter baumannii (PDR-PA) accounted for 11 cases. (Fig. 1)

Fig. 1.

Distribution of multidrug-resistant organisms (MDROs) in Catheter-related bloodstream infections (CRBSI)

Discussion

Factors related to CRBSI infections

In this study, we evaluated the demographic and clinical characteristics of 50,718 patients and compared the differences between patients with central venous CRBSI and those with non-central venous non-CRBSI.

Firstly, age and sex were significant differential factors between CRBSI patients and non-CRBSI patients. This is consistent with previous studies, which have shown that older patients are more susceptible to CRBSI [24, 25]. This may be attributed to age-related immune dysfunction, higher comorbidity burden, and prolonged hospitalization. Our studies have shown that male patients are more likely to develop CRBSI compared to females. This may be due to a higher frequency of invasive procedures in male patients and physiological differences (such as immune responses) [12].

The presence of comorbidities significantly influenced CRBSI risk in this study. Diabetes was associated with a higher incidence of CRBSI (33.61% vs. 12.54%), consistent with its known immunosuppressive effects and heightened infection susceptibility [26]. These findings emphasize the necessity for rigorous surveillance protocols and optimized preventive strategies in this high-risk population.

Clinical studies [27, 28] indicate that the incidence of CRBSI varies across regions and healthcare institutions. ICU patients demonstrated 21.5-fold higher CRBSI risk (57.38% of cases; incidence rate = 4.9%, p < 0.001), consistent with global reports, which is consistent with the reported rates and further confirms the significant risk of this infection in the ICU setting.

CRBSI patients had a significantly more extended hospital stay compared to non-CRBSI patients. This indicates that CRBSI may lead to extended treatment durations, affecting overall hospital resource management. For example, a prospective study found that patients with CRBSI had an average length of stay extended by up to 10 days, which places an additional burden on hospital resource management [29].

Furthermore, the data suggests that a significant proportion of CRBSIs occur in patients who have undergone multiple catheter insertions. This highlights the importance of minimizing the number of catheter insertions and using alternative routes for vascular access when possible. Educational programs for healthcare workers to reinforce best practices in catheter care and insertion techniques can also play a crucial role in reducing CRBSI rates. The residence time is significantly correlated with the occurrence of CRBSI [30]. It is suggested that in clinical practice, the residence time of the catheter should be reduced as much as possible to lower the risk of infection.

It is worth mentioning, although the presence of more cancer patients in the non-CRBSI group may initially seem counterintuitive (24.59% vs. 39.76% in the non-CRBSI group), it is worth noting that the lower proportion of cancer patients in the CRBSI group in this study may reflect confounding factors. First, ICU exposure, a significant risk factor for CRBSI, was significantly lower in oncology patients (16.55% vs. 83.45% in non-oncology patients; χ² = 2799.00, p = 0.001), freeing them from the highest risk setting. Second, oncology patients received fewer catheter insertions (68.39% vs. 31.61% for non-tumor patients; 61.73% vs. 38.27% for non-tumor patients), consistent with evidence that reducing the number of insertions reduces the risk of CRBSI by minimizing skin barrier disruption [31]. Supplementary File 2.

Additionally, factors such as the number of days with persistent fever, the use of antimicrobial agents, and abnormal white blood cell counts were significantly higher in CRBSI patients compared to the non-infected group. Research has shown that persistent fever is common in patients with CVC-BSI [32]. Fever is typically considered an important indicator of infection, and CRBSI patients often exhibit a longer duration of fever [33]. This is related to the systemic inflammatory response caused by pathogens. Patients with CRBSI usually require more frequent use of antimicrobial agents to control the infection. These patients often face more complex infection scenarios and may develop resistance.

The association between MDRO and CRBSI

Our study indicates that in clinical practice, the risk of CRBSI is significantly increased in patients with MDRO (Multidrug-Resistant Organism) infections, establishing a particular connection between the two. Multiple studies have shown that This may be attributed to age-related immune patients with MDRO infections requiring CVCs face a significantly elevated risk of developing CRBSI [34, 35]. This elevated risk may be related to the use of catheters and the characteristics of resistant bacteria.

MDROs typically spread in hospital environments, and their antibiotic resistance renders traditional antibiotic therapies ineffective. According to research by Wattal C et al., MDROs cause bloodstream infections and colonize the catheter surface by forming biofilms, leading to CRBSI [36]. Bacteria within biofilms can resist host immune responses and antibiotic treatment, further exacerbating the infection. Patients with MDRO infections often exhibit higher clinical complexity, such as multiple comorbidities and immunosuppressive states, which may make these patients more susceptible to CRBSI [37].

The high incidence of MDROs in CRBSI cases merits careful consideration. In our study, the CRBSI incidence rate in the ICU was significantly higher than that in general wards, reaching 4.9% compared to only 0.24%. This disparity underscores the ICU as a critical area for controlling infection. The unique characteristics of the ICU environment likely contribute to this elevated incidence. ICU patients often receive broad-spectrum antibiotics, which can exert selective pressure, favoring the emergence and proliferation of resistant strains [38]. The high volume of invasive procedures in the ICU, such as CVC insertions, can breach skin barriers, providing pathogens with direct access to the bloodstream [39]. Moreover, most ICU patients are immunocompromised due to underlying diseases, illness, or immunosuppressive treatments, making them more vulnerable to infections [40]. Additionally, the proximity of patients in the ICU and the high frequency of patient care activities can facilitate the transmission of pathogens [41].

Notably, our study reported the MDRO infection rate among CRBSI patients in a single Chinese center, which was 62.30%, significantly higher than the 3.86% observed in the non-CRBSI population (p < 0.001). Multivariate analysis further confirmed MDROs as an independent risk factor for CRBSI (OR = 3.79). This finding highlights the extreme importance of MDRO control in the local ICU environment for CRBSI prevention. Additionally, we conducted a thorough outbreak investigation during the study period and found no evidence of clustered cases or transmission chains indicative of an outbreak. We have ruled out the interference of COVID-19, as our country implemented a designated policy for treating COVID-19 patients during the study period, ensuring that the patients included in our study were not affected by the pandemic. Given that the CRBSI incidence rate in the ICU is over 20 times higher than that in general wards, our study’s findings underscore the importance of infection control and MDRO prevention in the ICU setting and contribute valuable local data for potential guideline development.

Distribution of single bacterial infections

The study shows that the most prevalent infections are CRE and MRSA. According to a study by Krupanandan et al., the incidence of CRE infections has significantly increased in hospital settings, particularly in intensive care units [42]. MRSA is recognized as one of the major pathogens responsible for hospital-acquired infections, leading to severe clinical consequences [43]. MRSA prevalence is declining in North America/Western Europe due to infection control, consistent with our data.

Frequency and combinations of multibacterial infections

Among patients with infections caused by two types of bacteria, the combination of CRAB and multidrug-resistant Acinetobacter baumannii (MDR-AB) is the most common. CRAB, MDR-AB, and PDR-PA co-occurred most frequently (79 cases) in tripartite infections, reflecting severe multidrug resistance. Guidance (2022) shows that such polymicrobial MDRO infections significantly worsen clinical prognosis and complicate therapeutic decisions [44]. Critically, A. baumannii (not MRSA) was the cornerstone organism in multistrain infections, underscoring its pivotal role in MDRO dissemination. The high MDRO burden extends beyond infected patients: empiric carbapenem use for such infections may inadvertently amplify CRE colonization in non-CRBSI cases. Therefore, CRBSI protocols must dynamically integrate international standards and local real-time data to balance coverage and resistance prevention.

Conclusion

This five-year retrospective study conducted at a tertiary hospital in Ningbo, China, demonstrates a significant association between CRBSI and the isolation of MDROs within our specific patient. These findings highlight the substantial burden of MDROs complicating CRBSI in our setting and underscore the critical need for enhanced infection prevention and control strategies within our institution. Based on our observations, rigorous monitoring and targeted management protocols for patients with CRBSI, particularly those involving MDROs, appear warranted. Furthermore, our data reinforces the importance of clinical decision-making being guided by local pathogen distribution and susceptibility patterns. Future research, particularly prospective multi-center studies within China, is needed to confirm these associations in broader populations and to develop robust predictive models.

Limitations

This study has several important limitations that should be considered when interpreting the results: First, the cross-sectional design inherently limits causal interpretation of the observed associations. While we identified a significant correlation between CRBSI and MDRO colonization, this study design cannot establish predictive relationships or determine temporal sequences between exposure and outcome. Second, the single-center retrospective design may affect generalizability. All data were collected from one tertiary hospital in Ningbo, China, which may not reflect patterns in community hospitals, private institutions, or regions with different socioeconomic or demographic characteristics. This limitation particularly affects the external validity of our findings across diverse healthcare settings. Third, the retrospective methodology introduces potential selection bias and data completeness issues. Case identification depended on medical record documentation, which may have led to underreporting or misclassification of CRBSI cases and MDRO status. Finally, although we found a statistically robust association, our binary classification of MDROs (present/absent) precluded more nuanced analysis. This approach cannot characterize risk differences among specific resistant pathogens (e.g., MRSA, VRE, ESBL-producing Enterobacterales, or carbapenem-resistant Gram-negative bacteria) within the CRBSI population. Future research should aim to address these limitations by launching prospective cohort studies to verify causality.

Abbreviations

CRE

Carbapenem-resistant Enterobacteriaceae

MRSA

Methicillin-resistant Staphylococcus aureus

CRAB

Carbapenem-resistant Acinetobacter baumannii

MDR-AB

Multidrug-resistant Acinetobacter baumannii

PDR-PA

Pandrug-resistant Pseudomonas aeruginosa

CRPA

Carbapenem-resistant Pseudomonas aeruginosa

PDR-AB

Pandrug-resistant Acinetobacter baumannii

Author contributions

All authors made substantial contributions in this manuscript that every author responsibility is specified in the following: Authors JL contributed in conception and design of the study. YZ and JM wrote the main manuscript text. YZ analyzed the data and interpreted the results. HD and LC prepared figures. SL and LS collect data. YZ and JM are corresponding authors. All authors reviewed the manuscript.

Funding

Zhejiang Provincial Medical and Health Technology Project (2025KY220).

Zhejiang Provincial Science and Technology Plan for Traditional Chinese Medicine (2024ZL156).

Zhejiang Yangtze River Delta Health Research Fund Project (2023CSJ-3-D006).

Data availability

The datasets generated and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was conducted per the Declaration of Helsinki. All methods were carried out following relevant guidelines and regulations. This retrospective study was approved by the Ethics Committee of Ningbo Medical Center LiHuili Hospital (Approval No: KY2022SL246). As this was a retrospective analysis of anonymized patient data, the committee granted an exemption from obtaining individual informed consent.

Competing interests

The authors declare no competing interests.