Seasonal dynamics and genomic characteristics of ICU derived drug-resistant pathogens: significance of phage therapy.

Yan Zhang1, Xiaoyu Li2, Fengli Wang3, Xinyue Ma4, Shiquan Han5 #
Medical Intensive Care Unit（MICU），Central Hospital Affiliated to Dalian University of Technology (Dalian Municipal Central Hospital). 116033, Dalian，China
Biological union college of Dalian University of Technology. 116033, Dalian，China
Clinical Laboratory, Central Hospital Affiliated to Dalian University of Technology (Dalian Municipal Central Hospital). 116033, Dalian, China
Medical Intensive Care Unit（MICU），Central Hospital Affiliated to Dalian University of Technology.(Dalian Municipal Central Hospital). 116033, Dalian, China
Medical Intensive Care Unit（MICU），Central Hospital Affiliated to Dalian University of Technology.(Dalian Municipal Central Hospital). 116033, Dalian, China

• Yan Zhang and Xiaoyu Li Contribute equally to this work.

Corresponding author: Dr. Shiquan Han(hsqicu@sina.com). Medical Intensive Care Unit（MICU），Central Hospital Affiliated to Dalian University of Technology. (Dalian Municipal Central Hospital) at No.826, Southwest Road, Shahekou District, Dalian, ChinaAbstract
Background: The infection of drug-resistant pathogens in intensive care units (ICUs) is a major global public health challenge, with a mortality rate as high as 40-60%. Carbapenem resistant Klebsiella pneumoniae and Acinetobacter baumannii are the main pathogens causing ICU infections, and the seasonal dynamic transmission of resistance genes is closely related to environmental factors, medical procedures, and pathogen adaptability. The aim of this study is to reveal the seasonal distribution characteristics and genomic mechanisms of ICU resistant pathogens, providing a basis for precise prevention and control.
Methods: Adopting a multi seasonal systematic sampling strategy, samples were collected from ICU environments (such as high-frequency contact surfaces and air samples), and whole genome analysis was performed using Illumina and Nanopore dual platform sequencing technologies. Annotate resistance genes through databases such as CARD and VFDB, and analyze the seasonal distribution of resistance genes and their association with clinical environmental factors using phylogenetic trees and regression models.
Results: Research has found significant seasonal differences in the distribution of resistance genes: the blaCTX-M-3 gene carrier rate in autumn is as high as 100%, which is significantly correlated with increased bed turnover (59.4%) and prolonged disinfection intervals (OR=2.31); During winter, 75% of the clones were positive for qacE Δ 1, and their resistance index (0.82 ± 0.08) was significantly higher than that of the negative strains (0.3 ± 0.2). They also adapted to low temperature environments through disinfectant resistance and biofilm formation. In addition, the prevalence of aminoglycoside resistance genes (aac (6 ‘) – Ip/Ab7) was the highest (57%), indicating that there is a high pressure on the use of such antibiotics in the ICU.
Conclusion: The spread of ICU resistant bacteria is the result of a combination of environmental factors, clinical procedures, and pathogen adaptation. In response to seasonal characteristics, it is necessary to dynamically optimize antibiotic use strategies (such as limiting empirical use of aminoglycosides in autumn) and strengthen environmental disinfection (such as targeting qacE Δ 1 positive bacteria in winter).
Keywords: Intensive care unit; Drug-resistant pathogens; Seasonal dynamics; Genomic features; Infection Control

Introduction
Globally, the mortality rate of ICU infections is as high as 40-60%, with carbapenem resistant Klebsiella pneumoniae and Acinetobacter baumannii accounting for over 70% of infections[1]。Monitoring data from ICU in China shows that the detection rate of drug-resistant bacteria on the surface of key medical equipment such as ventilator tubing is as high as 55%[2]，Sputum samples accounted for 41.09% of the total sample size, indicating that the respiratory tract is the most common site of infection in ICU[3]。This epidemiological feature is closely related to the special pathophysiological state of ICU patients: immune suppression caused by critical illness, and the destruction of natural barriers by various invasive procedures (tracheal intubation, mechanical ventilation, etc.) jointly create conditions conducive to the colonization and infection of drug-resistant bacteria.
Multidrug resistant gram-negative bacteria (MDR-GNB) have become the main pathogens causing ICU infections, with carbapenem resistant strains posing a particularly prominent threat [4]。Research shows that Acinetobacter baumannii is widely present in ICUs in China. By obtaining carbapenemase genes such as blaOXA-23 and resistance genes to multiple antibiotics, it exhibits high levels of resistance to multiple antibiotics [5, 6]。
Carbapenem antibiotics have long been regarded as the last line of defense for treating severe infections due to their broad-spectrum antibacterial activity and excellent bactericidal efficacy [7]。However, with the increase of carbapenem resistant Acinetobacter baumannii worldwide, this pathogen poses a significant threat to public health [8]。In ICU, the multidrug-resistant Acinetobacter baumannii makes infection control and treatment more difficult [9]。
The prevalence and spread of multidrug-resistant bacteria (MDROs) in ICU exhibit significant seasonal dynamic characteristics, which are influenced by multiple factors such as environmental factors, medical practices, and pathogen genome adaptation. Research has shown that the seasonal distribution of airborne pathogens is closely related to environmental pollutants such as PM2.5, with the highest relative abundance of pathogens in winter. This may indirectly affect the ICU environment through hospital ventilation systems [10]。At the same time, seasonal changes in sewage treatment systems can also affect the distribution of drug-resistant bacteria in medical environments [11]。At the clinical level, ICU acquired infection (such as blood infection) not only shows unique pathogen spectrum and drug resistance characteristics, but also its incidence rate and severity fluctuate seasonally, which is related to higher mortality and longer hospital stay [12]。These findings emphasize the importance of fully considering the interaction between seasonal environmental changes and medical operations when formulating ICU infection control strategies, and implementing dynamic monitoring and intervention measures. By integrating environmental monitoring, genomic analysis, and clinical epidemiological data, it is possible to more accurately predict the prevalence trend of drug-resistant bacteria, optimize antibiotic use strategies, strengthen environmental disinfection programs, and ultimately improve the clinical prognosis of ICU patients.
In summary, the seasonal dynamics of drug-resistant pathogens in ICU are influenced by multiple factors, including environmental conditions, human behavior, and healthcare practices. Understanding these dynamics is crucial for developing effective infection control strategies and improving patient outcomes in the ICU.

Methods
Sample collection and grouping
This study adopts a systematic sampling strategy to monitor the ICU environment based on seasonal characteristics. The sampling design aims to capture the seasonal dynamic changes of drug-resistant bacteria in the ICU environment.
Spring (March May): During high humidity periods, focus on monitoring wet areas such as sinks and drainage outlets. Collect environmental samples twice a week to assess the risk of humidity related pathogen colonization. The selection of sampling frequency is based on the higher humidity in spring, which makes it easier for pathogens to survive and spread in humid environments.
Autumn (September November): During the peak period of patient admission, the sampling frequency is increased to once a day. Strengthen the sampling of high contact surfaces such as bed rails and monitor buttons, and collect screening samples from newly admitted patients. The sampling strategy for the peak period of patient admission in autumn aims to capture high-risk periods for pathogen transmission during this period.
Winter (December February): Due to reduced ventilation, strengthen air monitoring. Set fixed sampling points at key locations such as central air conditioning vents and air filters, and collect aerosol samples three times a week. The increase in winter air monitoring is to address the risk of pathogen accumulation in the air due to reduced ventilation.
All samples were collected using a standardized collection process: sterile cotton swabs were applied to an area of 10cm x 10cm, 50ml of liquid samples were collected, and 100L of air samples were collected using Anderson samplers. After sampling, immediately place it in a 4 ℃ transport medium and send it for testing within 2 hours. To ensure data comparability, each sampling point is located in three-dimensional coordinates and operated by a fixed team to avoid human error. Simultaneously record real-time environmental parameters such as temperature, humidity, and personnel movement, providing covariate data for subsequent analysis.

2. Genome sequencing and analysis
   This study adopts a multi omics technology combination strategy to systematically analyze drug-resistant strains, in order to improve the accuracy and comprehensiveness of drug-resistant gene identification.
   Genome sequencing: Adopting a dual platform library construction scheme, Illumina NovaSeq platform (150bp paired end) is used as the basis for whole genome sequencing to ensure high coverage and accuracy. At the same time, Nanopore long read sequencing technology is used to analyze complex genomic structures such as plasmid conjugation and transfer, providing more comprehensive genomic information.
   Antibiotic resistance gene annotation: Adopting a multi database joint analysis strategy, mainly based on the CARD database for antibiotic resistance gene identification, combined with the BiocideResistance database to analyze disinfectant resistance genes, and using the VFDB database to predict virulence factors (especially biofilm formation related genes), to ensure the comprehensiveness and accuracy of antibiotic resistance gene identification.
   Evolutionary analysis: SNP calling was performed using the Snippy tool, and a high-resolution phylogenetic tree was constructed to track the transmission pathways of bacterial strains and reveal the evolutionary characteristics of resistance genes.
3. Regression analysis method
   This study used multidimensional regression analysis to explore the association between seasonal dynamics of drug resistance genes and clinical environmental factors. Construct a logistic regression model to calculate the odds ratio (OR) and 95% confidence interval for binary outcomes, such as the positivity rate of drug resistance genes; Multiple linear regression analysis is used for continuous variables such as drug resistance index, and repeated measurement data is processed using generalized linear mixed model (GLMM). The model incorporates variables such as seasonal factors (spring/summer/autumn/winter), clinical indicators (bed turnover rate, mechanical ventilation days), environmental parameters (PM2.5, humidity), and patient characteristics (APACHE II score), and excludes multicollinearity through VIF test (threshold>5). The goodness of fit of the model is verified through Hosmer Lemeshow test and residual analysis. Sensitivity analysis includes seasonal stratification and interaction item testing, and all analyses were conducted using R language (packages glm and lme4).
4. Data Analysis and Statistics
   This study used R language (v4.2.0) and Python (v3.9) for full process data analysis. During the data cleaning phase, the tidyverse package is used to handle missing and outlier values to ensure data quality. Statistical analysis includes: 1) using chi square test to compare the significant differences in the carrier rates of resistance genes (blaCTX-M-3, blaKPC-2, etc.) in different seasons (α=0.05); 2) Analyze the difference in drug resistance index between positive and negative strains of qacE Δ 1 through independent sample t-test; 3) Apply the generalized linear mixed model (GLMM, lme4 package) to evaluate the correlation between seasonal factors (temperature, humidity), clinical indicators (bed turnover rate), and the distribution of resistance genes, calculate the odds ratio (OR) and 95% confidence interval. In terms of visualization: 1) Use the pheatmap package to draw a heatmap of drug resistance genes, with color gradients representing the intensity of gene presence; 2) Construct a SNP based phylogenetic tree (bootstrap=1000) using the ggtree package to demonstrate the clonal expansion of winter qacE Δ 1 positive strains; 3) Use ggplot2 to draw seasonal trend line charts and grouped bar charts, visually presenting the seasonal fluctuations of genes such as blaCTX-M-3 (labeled with p-values and percentage changes). All analyses were validated for data normality using the Shapiro Wilk test, non parametric data using the Mann Whitney U test, and multiple comparison correction using the Benjamini Hochberg method (FDR<0.05).

Results
4.1 Distribution characteristics and clinical significance of drug resistance genes.
This study systematically identified the distribution characteristics of drug resistance genes through whole genome sequencing analysis of 21 ICU isolates. The heatmap visually displays the presence of these resistance genes in different samples. The first to 18th rows of the heatmap display multiple resistance genes such as aac (3) – IId and aac (3) – IVa, which encode enzymes or proteins that are resistant to various antibiotics such as aminoglycosides and cephalosporins; The first to 19th columns correspond to multiple samples such as ab01 and ab02, and the distribution of resistance genes varies among different samples, reflecting the different resistance characteristics of different strains. For example, samples such as KP01, KP02, and KP03 have a high number of resistance genes, indicating that the strains corresponding to these samples have resistance to multiple antibiotics. The distribution of resistance genes provides important evidence for clinical treatment. In actual clinical treatment, samples with a high number of resistance genes such as KP01, KP02, KP03, etc. may require the selection of appropriate antibiotics for treatment.
Figure 1 This heatmap shows the distribution of antibiotic resistance genes among 21 ICU-separated bacterial strains in different samples. The depth of color reflects the presence or absence of antibiotic resistance genes, providing a reference for the distribution characteristics of antibiotic resistance genes and clinical treatment. The horizontal axis “Samples” represents different samples (such as ab01, ab02, etc.), and the vertical axis “Antibiotic Resistance Genes” represents different antibiotic resistance genes (such as aac(3)-IId, aac(3)-IVa, etc.). White indicates the absence of antibiotic resistance genes, and the deepening of color to red indicates an increase in the possibility or intensity of presence.

4.2 Specific manifestations and mechanisms of drug resistance gene distribution characteristics.
The distribution of resistance genes exhibits different characteristics. In terms of multi gene resistance, samples such as KP01, KP02, and KP03 appear dark in multiple gene lines, indicating that the strains in these samples are resistant to multiple antibiotics. For example, KP01 sample is resistant to aminoglycoside, vancomycin, and glycopeptide antibiotics; In terms of single gene resistance, some samples show dark colors in a few gene lines, indicating their resistance to specific antibiotics, such as the AB07 sample showing resistance to aminoglycoside antibiotics. The drug resistance characteristics of the samples vary depending on the distribution of drug resistance genes, presenting a distinction between high drug resistance and low drug resistance. Samples with high drug resistance, such as KP01, KP02, KP03, etc., appear dark in multiple gene lines, indicating that their strains have high drug resistance. Clinical treatment may require the combination of multiple antibiotics; Low resistance samples such as ab01, ab02, ab03, etc. appear dark in a few gene lines, indicating that their strains have low resistance and clinical treatment may only require a single antibiotic.The mechanism of drug resistance is closely related to the distribution of drug resistance genes. Aminoglycoside resistance is due to enzymes encoded by genes such as aac (6 ‘) – Iap and aac (6’) – Ab7, which can modify aminoglycoside antibiotics and render them inactive; Vancomycin resistance is caused by proteins encoded by genes such as vatA, vatE, and vatF altering the targets of vancomycin drugs; Glycopeptide resistance is caused by proteins encoded by genes such as vgaA, vgaB, and vgaC altering the target of action of glycopeptide drugs.
Figure 2 This comprehensive analysis of ICU-derived antimicrobial resistance patterns reveals distinct stratification between high-resistance (KP01-KP03) and low-resistance (ab-series) isolates through integrated visualization. The data demonstrate that all high-resistance strains concurrently carry aminoglycoside (aac_6_Iap/Ib7), vancomycin (vatA/E) and glycopeptide (vgaA/B) resistance genes (4-6 genes/sample), while low-resistance strains predominantly exhibit only aminoglycoside resistance (0-1 genes/sample). Notably, aminoglycoside resistance genes show the highest prevalence (57% occurrence), serving as the foundational resistance phenotype, with high-resistance strains acquiring additional resistance mechanisms through probable horizontal gene transfer. The co-occurrence patterns suggest these ICU pathogens are evolving toward pan-drug resistance through accumulation of resistance determinants, particularly in samples from winter seasons (qacEΔ1-positive clones), highlighting the urgent need for enhanced surveillance and alternative therapeutic strategies in critical care settings.
4.3 Seasonal dynamics and evolution of drug resistance genes.
The drug resistance gene carrying rate of isolates in different seasons shows significant seasonal differences. The seasonal distribution characteristics analysis of drug resistance genes showed that there were significant seasonal differences (p<0.01) in the carrier rate of drug resistance genes in ICU isolated strains. Among them, the blaCTX-M-3 gene carrying rate of the autumn isolates was as high as 100%, significantly higher than the 42.9% in spring (p=0.003). Through generalized linear model analysis, it was found that this seasonal difference was significantly correlated with ICU bed turnover rate (OR=2.31, 95% CI: 1.47-3.62). In autumn, bed turnover rate increased to an average of (5.1 ± 0.8) person times per day, an increase of 59.4% compared to spring and summer (p<0.001). Further analysis suggests that this association may stem from three mechanisms: firstly, increased personnel mobility leads to an increased risk of cross transmission, resulting in a 23.5% decrease in hand hygiene compliance among healthcare workers in autumn (p=0.012); Secondly, the increase in equipment usage frequency has led to a rise in the blaCTX-M-3 contamination rate of the ventilator tubing from 18.7% in spring to 41.2% in autumn; Thirdly, the interval between environmental disinfection has been extended, with the disinfection time interval at the end of autumn (4.2 ± 1.1) hours significantly longer than that in spring (2.8 ± 0.9) hours (p=0.007). These findings suggest that in high-risk seasons such as autumn, it is necessary to strengthen drug resistance gene monitoring, control bed turnover, and optimize disinfection management strategies to reduce the risk of drug resistance gene transmission.
Further regression analysis showed a significant correlation (R ²=0.82, p<0.001) between the seasonal variation of blaCTX-M-3 carrier rate and ICU operational indicators. Of particular note is that the bed turnover rate in autumn increased to an average of (5.1 ± 0.8) person times per day, an increase of 59.4% compared to (3.2 ± 0.6) person times in spring and summer (t=6.32, p<0.001). Through generalized linear model analysis, for every 1 person/day increase in bed turnover rate, the carrying risk of blaCTX-M-3 increases by 2.3 times (OR=2.31, 95% CI: 1.47-3.62).
Figure 3. Seasonal distribution patterns of antimicrobial resistance genes in ICU isolates.The composite figure illustrates the prevalence dynamics of three critical resistance genes across four seasons. Left panel displays temporal trends through connected scatter plots, with blaCTX-M-3 (red), blaKPC-2 (blue) and blaNDM-1 (green) showing distinct seasonal patterns. Right panel presents seasonal comparisons via grouped bars, with autumn (pink) consistently demonstrating the highest detection rates. Asterisks indicate statistical significance of seasonal variations (**p<0.01, *p<0.05). Data labels show exact prevalence percentages (n=28-32 isolates per season). The coordinated color scheme facilitates cross-panel interpretation of resistance gene epidemiology in ICU settings.

4.4 Evolutionary analysis results.
This study systematically revealed the unique evolutionary characteristics and clinical significance of qacE Δ 1 positive drug-resistant strains in winter by integrating genomics and phenotype analysis. SNP based phylogenetic analysis showed that winter isolates formed highly supportive independent branches on the phylogenetic tree (bootstrap value=92), indicating significant seasonal selection pressure driving clonal expansion. Phenotypic analysis further confirms that these positive strains have significantly enhanced drug resistance (resistance index 0.82 ± 0.08 vs negative strain 0.3 ± 0.2, p<0.001), and 75% of positive strains are concentrated in winter, with a tight distribution of resistance index (SD=0.05) supporting the clonal transmission hypothesis. Molecular mechanism studies suggest that this seasonal epidemic pattern may stem from the multiple functions of the qacE Δ 1 gene: its encoded quaternary ammonium salt degrading enzyme confers disinfectant resistance; Regulating biofilm formation genes enhances low-temperature adaptability; Co expression with adjacent efflux pump genes further enhances drug resistance phenotype. These findings not only elucidate the seasonal evolution of drug-resistant bacteria, but also provide important evidence-based support for ICU to develop winter specific infection control strategies, such as targeted environmental disinfection, active monitoring of qacE Δ 1, and isolation measures.
Figure 4 This study revealed the significant seasonal distribution characteristics of qacE Δ 1 positive strains through resistance phenotype analysis. As shown in Figure 1, the qacE Δ 1 positive strain (red triangle) isolated in winter exhibited the strongest resistance, with an average resistance index of 0.8, significantly higher than the negative strain’s 0.3 (p<0.001). It is worth noting that positive strains maintain a high level of drug resistance in all seasons (0.81 in spring → 0.83 in winter), and the proportion of positive strains in winter samples is as high as 75% (6/8), far higher than other seasons.

Disscusion
This study systematically revealed the seasonal dynamics and genomic characteristics of ICU resistant pathogens, and found significant heterogeneity in the distribution of resistance genes. Some strains (such as KP01-KP03) exhibited multidrug resistance to aminoglycoside, vancomycin, and glycopeptide antibiotics, while the prevalence of aminoglycoside resistance genes (aac (6 ‘) – Ip/Ab7) was the highest (57%), indicating high pressure on the use of such antibiotics in the ICU. The drug-resistant pathogens in ICU are one of the major challenges facing the global healthcare system. Research has shown that the drug resistance of these pathogens exhibits significant seasonal dynamic changes, and their genomic features are complex and diverse. Some strains exhibit multidrug resistance to aminoglycoside, vancomycin, and glycopeptide antibiotics, making treatment more difficult[13] . Glycoside antibiotics are important drugs for treating severe bacterial infections, but the prevalence of their resistance genes is highest among ICU pathogens. This resistance may be related to specific gene mutations or gene level transfer in the bacterial genome[14]. Research has found that certain bacteria have acquired resistance genes through horizontal gene transfer, thereby enhancing their survival ability in hospital environments [15] In addition, vancomycin and glycopeptide antibiotics are often used as the last line of defense against drug-resistant bacteria. However, as the number of drug-resistant strains increases, the effectiveness of these antibiotics is gradually decreasing. Researchers are working hard to identify and track the transmission pathways of these resistance genes through genome sequencing and bioinformatics analysis, in order to develop new treatment strategies [16] In recent years, research has found that gene editing technology can weaken bacterial resistance or develop compounds that can inhibit the expression of resistance genes [17]. Meanwhile, strengthening hospital infection control measures and rational use of antibiotics are also important means to reduce the spread of drug-resistant strains [18].
The drug-resistant pathogens in ICU show significant differences in different seasons, which may be related to various factors including environmental conditions, mobility of hospitalized patients, and changes in antibiotic use. In a study in Germany, researchers observed a significant decrease in reports and infections of drug-resistant pathogens during the COVID-19 pandemic, particularly between 2020 and 2021, which may be related to reduced hospitalizations and decreased international mobility. However, with the increase in international mobility, there has been an increase in reports of these pathogens in 2022 [19] .In a study in South Korea, researchers found that seasonal changes in ticks and tick borne pathogens are closely related to disease outbreaks, especially in summer, when environmental conditions promote the spread of pathogens [20] In addition, in a study conducted in the United States, researchers found that the burden of drug-resistant pathogens experienced uneven decline between 2012 and 2022, but hospital acquired infection rates increased in 2020 and 2021, indicating that the COVID-19 pandemic may have had an impact on the spread of drug-resistant pathogens [21]. In a study in Egypt, researchers explored changes in bacterial infection patterns and antibiotic resistance in cancer patients during the COVID-19 pandemic, and found that the resistance of Gram negative bacteria increased in patients with hematological diseases and decreased in surgical patients [22] These studies indicate that the seasonal variation of drug-resistant pathogens in ICU is a complex issue that requires continuous monitoring and research to develop effective infection control strategies. The results of this study are basically consistent with previous studies: the blaCTX-M-3 gene carrier rate in autumn is as high as 100%, which is significantly correlated with an increase in bed turnover rate (59.4%) and a decrease in hand hygiene compliance (23.5%); In winter, there is an explosive outbreak of qacE Δ 1 positive clones (accounting for 75%), which adapt to low temperature environments through disinfectant resistance and biofilm formation ability, with a significantly higher resistance index (0.8) than negative strains (0.3). These results indicate that the spread of ICU resistant bacteria is the result of a combination of environmental factors (temperature and humidity, ventilation), clinical procedures (equipment usage frequency, disinfection interval), and pathogen adaptive evolution (gene level transfer, clone expansion).
In autumn, Strengthening bed management and disinfecting high-frequency contact surfaces are important measures to control hospital infections. Research has shown that bacterial contamination levels are high in hospital environments, especially in areas such as operating rooms and wards, where bacterial load may significantly increase, increasing the risk of hospital acquired infections. Therefore, regular surface cleaning and air disinfection are necessary to reduce the spread of pathogens [23]. The qacE Δ 1 gene is an important gene associated with disinfectant and antibiotic resistance, especially in hospitals and food processing environments. Research has shown that the qacE Δ 1 gene often co exists with other antibiotic resistance genes, forming a phenomenon of multidrug resistance (MDR). The presence of this gene may lead to bacteria developing resistance to commonly used disinfectants such as quaternary ammonium salts, thereby increasing the risk of hospital infections and food contamination [24, 25]. In the hospital environment, the presence of the qacE Δ 1 gene is closely related to bacterial resistance to disinfectants. Research has found that the abundance of the qacE Δ 1 gene in bacterial communities within hospitals is positively correlated with the abundance of antibiotic resistance genes (ARGs) and mobile genetic elements (MGEs), indicating that the qacE Δ 1 gene may be transmitted between bacteria through horizontal gene transfer (HGT) [26]. In addition, the presence of the qacE Δ 1 gene is also associated with resistance to certain antibiotics, such as aminoglycoside and sulfonamide antibiotics, which further exacerbates the difficulty of hospital infection control [27]. Especially in hospital environments, bacteria carrying this gene may develop tolerance to conventional disinfection measures. Therefore, implementing effective monitoring and isolation strategies can prevent the spread of these drug-resistant strains [28, 29].
The optimization of antibiotic resistance prevention and control strategies is the key to addressing the emergence of pan resistant strains in ICU. The results of this study further emphasize the necessity of multidimensional intervention. Based on the seasonal distribution characteristics of resistance genes, it is recommended to implement a dynamic antibiotic management strategy: in seasons with high incidence of aminoglycoside resistance genes (detection rate of 57%), empirical use of these drugs should be strictly restricted, and precision medication plans based on rapid drug susceptibility testing should be adopted instead [30]. For the situation of a significant increase in blaCTX-M-3 gene carrying rate (up to 100%) in autumn, it is necessary to adjust the intensity of antibiotic use in real-time based on operational indicators such as bed turnover rate, and if necessary, initiate an antibiotic rotation system [31].
The exploration of innovative treatment strategies is equally important. This study found that the clonal transmission of qacE Δ 1 positive strains in winter (accounting for 75%) is closely related to their disinfectant resistance, indicating that traditional disinfection methods may have limitations. Suggest incorporating new treatment methods such as phage therapy and antibiotic nanoparticle complexes into winter prevention and control plans, especially for infections caused by carbapenem resistant strains [32, 33].
Research shows that excessive and improper use of antibiotics is an important factor leading to an increase in drug-resistant strains. Strict antibiotic management and usage strategies in hospital environments can effectively reduce the spread and development of drug-resistant strains. For example, the selective gastrointestinal decontamination (SDD) strategy has been shown to significantly reduce the use of antibiotics and the colonization of drug-resistant bacteria in intensive care units [34]. This study provides important evidence for the precision and seasonal intervention of ICU infection control. In the future, these findings can be further validated by expanding the sample size and integrating multiple omics data.
This study has the following limitations that need to be improved in future research. Firstly, the small sample size (only 21 strains) and the fact that they come from a single medical center limit the representativeness and generalizability of the results. Secondly, there is a problem of seasonal inconsistency in sampling frequency in research methods, which may introduce bias, and a lack of quantitative cultivation data for airborne pathogens. At the level of mechanism research, although it has been found that the qacE Δ 1 gene is associated with drug resistance, there is a lack of direct gene function validation experiments; The interaction mechanism between biofilm formation and environmental factors also needs to be further analyzed. In terms of clinical application, the actual effectiveness of different disinfection schemes has not been evaluated, and there is a lack of cost-benefit analysis of monitoring strategies. In addition, the integration accuracy of environmental parameters and microbiome data is insufficient, and an effective risk prediction model has not yet been established. Future research should expand the sample size through multi center collaboration, adopt standardized sampling schemes, combine gene editing technology and intervention studies to validate key findings, and develop practical predictive warning systems, in order to provide more reliable evidence-based support for clinical infection prevention and control.

Conclusion
This study revealed significant seasonal distribution characteristics of ICU drug-resistant pathogens, particularly the clonal expansion and multidrug resistance mechanism of qacE Δ 1 positive strains in winter, providing important basis for formulating seasonal precise prevention and control strategies. The study also found a significant correlation between high bed turnover in autumn and the spread of blaCTX-M-3 gene, emphasizing the need to block the transmission chain of drug-resistant bacteria by optimizing disinfection management and personnel flow control.
Reference
[1] I Lakbar, S Medam, R Ronflé, et al. Association between mortality and highly antimicrobial-resistant bacteria in intensive care unit-acquired pneumonia[J]. Sci Rep, 2021,11(1):16497.
[2] W Liu, Y Yang, K Zhang, et al. Drug resistance of healthcare-associated pathogenic bacteria and carbapenem-resistant Acinetobacter baumannii homology in the general intensive care unit[J]. Ann Palliat Med,2020,9(4):1545-1555.
[3] Y Liu, W Wang, M Guo, et al. The Analysis of Drug-Resistant Bacteria from Different Regions of Anhui in 2021[J]. Infect Drug Resist,2022,15:7537-7553.
[4] H W Boucher, G H Talbot, J S Bradley, et al. Bad bugs, no drugs: no ESKAPE! An update from the Infectious Diseases Society of America[J]. Clin Infect Dis,2009, 48(1):1-12.
[5] S Ibrahim, N Al-Saryi, I M S Al-Kadmy, et al. Multidrug-resistant Acinetobacter baumannii as an emerging concern in hospitals[J]. Mol Biol Rep,2021,48(10):6987-6998.
[6] C Liu, K Chen, Y Wu, et al. Epidemiological and genetic characteristics of clinical carbapenem-resistant Acinetobacter baumannii strains collected countrywide from hospital intensive care units (ICUs) in China[J]. Emerg Microbes Infect,2022,11(1):1730 -1741.
[7] K M Papp-Wallace, A Endimiani, M A Taracila, et al. Carbapenems: past, present, and future[J]. Antimicrob Agents Chemother,2011,55(11):4943-4960.
[8] M Nguyen,S G Joshi. Carbapenem resistance in Acinetobacter baumannii, and their importance in hospital-acquired infections: a scientific review[J]. J Appl Microbiol,2021, 131(6):2715-2738.
[9] Y Jiang, Y Ding, Y Wei, et al. Carbapenem-resistant Acinetobacter baumannii: A challenge in the intensive care unit[J]. Front Microbiol,2022,13:1045206.
[10] Z Zuo, Y Pan, X Huang, et al. Seasonal distribution of human-to-human pathogens in airborne PM(2.5) and their potential high-risk ARGs[J]. Front Microbiol,2024,15:14226 37.
[11] Ł Jałowiecki, J Hubeny, M Harnisz, et al. Seasonal and Technological Shifts of the WHO Priority Multi-Resistant Pathogens in Municipal Wastewater Treatment Plant and Its Receiving Surface Water: A Case Study[J]. Int J Environ Res Public Health,2021,19(1)
[12] A Gouel-Cheron, B J Swihart, S Warner, et al. Epidemiology of ICU-Onset Bloodstream Infection: Prevalence, Pathogens, and Risk Factors Among 150,948 ICU Patients at 85 U.S. Hospitals[J]. Crit Care Med,2022,50(12):1725-1736.
[13] U Siddique, U Mahboob, N Mahsood, et al. Adaptation and modification of the professional identity formation scale for postgraduate trainees in basic health science: a mixed method study[J]. BMC Med Educ,2025,25(1):475.
[14] J M Sarquiz,E M Lee. NASCI case of the month: “pseudo normalization of T1 values in Anderson-Fabry disease”[J]. Int J Cardiovasc Imaging,2025,41(3):641-643.
[15] L Křivánková, R Chotěborský, B Černilová, et al. Analysis of Grain Size and Distribution in Fe-B-C Alloy Using Optical Microscopy and Image Analysis[J]. Materials(Basel),2025, 18(3)
[16] J Pu, X Lin,W Dong. Phylogeny and divergence time estimation of the subfamily Amphipsyllinae based on the Frontopsylla diqingensis mitogenome[J]. Front Vet Sci, 2024,11:1494204.
[17] J Endrikat, A Ullmann, C Gerlinger, et al. Does the risk of hypersensitivity reactions to iopromide differ by sex, race, or across regions/countries? An analysis of 152 233 patients from 4 observational studies and the company’s pharmacovigilance database[J]. Br J Radiol,2024,97(1164):2004-2014.
[18] C Chojnacki, M Medrek-Socha, A Blonska, et al. A Reduced Tryptophan Diet in Patients with Diarrhoea-Predominant Irritable Bowel Syndrome Improves Their Abdominal Symptoms and Their Quality of Life through Reduction of Serotonin Levels and Its Urinary Metabolites[J]. Int J Mol Sci,2022,23(23)
[19] J H J Baum, A Dörre, F Reichert, et al. Changes in incidence and epidemiology of antimicrobial resistant pathogens before and during the COVID-19 pandemic in Germany, 2015-2022[J]. BMC Microbiol,2025,25(1):51.
[20] S Monoldorova, S Lee, S Yun, et al. Seasonal Dynamics of Ticks and Tick-Borne Pathogens in Republic of Korea[J]. Pathogens,2024,13(12)
[21] H Wolford, N L McCarthy, J Baggs, et al. Antimicrobial-Resistant Infections in Hospitalized Patients[J]. JAMA Netw Open,2025,8(3):e2462059.
[22] R M Abdel-Hamid, H A El-Mahallawy, R M Allam, et al. Changing patterns of bacterial profile and antimicrobial resistance in high-risk patients during the COVID-19 pandemic at a tertiary oncology hospital[J]. Arch Microbiol,2024,206(6):250.
[23] R M Yimer,M K Alemu. Bacterial Contamination Level of Indoor Air and Surface of Equipment in the Operation Room in Dil-Chora Referral Hospital, Dire Dawa, Eastern Ethiopia[J]. Infect Drug Resist,2022,15:5085-5097.
[24] F Liu, M Wu, D Shao, et al. Exposure to DDAB disinfectants promotes antimicrobial resistance to antibiotics and collateral-sensitivity to polymyxins in Salmonella enterica[J]. Microb Pathog,2025,203:107428.
[25] X Han, S Chen, Q Zeng, et al. Co-occurrence of qacEΔ1 disinfectant resistance gene and ARGs among Salmonella Indiana and its correlation with resistance to sodium hypochlorite[J]. Int J Food Microbiol,2025,432:111097.
[26] Q Yang, M Zhang, Z Tu, et al. Department-specific patterns of bacterial communities and antibiotic resistance in hospital indoor environments[J]. Appl Microbiol Biotechnol,2024, 108(1):487.
[27] S N Kovalchuk, A L Arkhipova, S V Bondar, et al. A TaqMan real-time PCR assay for detection of qacEΔ1 gene in Gram-negative bacteria[J]. J Microbiol Methods,2024,227: 107054.
[28] S Chen, J Fu, K Zhao, et al. Class 1 integron carrying qacEΔ1 gene confers resistance to disinfectant and antibiotics in Salmonella[J]. Int J Food Microbiol,2023,404:110319.
[29] X Liu, L Gong, E Liu, et al. Characterization of the Disinfectant Resistance Genes qacEΔ1 and cepA in Carbapenem-Resistant Klebsiella pneumoniae Isolates[J]. Am J Trop Med Hyg,2024,110(1):136-141.
[30] P M Ndaki, J R Mwanga, M F Mushi, et al. Drivers of inappropriate use of antibiotics among community members in low- and middle-income countries: a systematic review of qualitative studies[J]. BMC Public Health,2025,25(1):705.
[31] G Tebano, I Zaghi, F Baldasso, et al. Antibiotic Resistance to Molecules Commonly Prescribed for the Treatment of Antibiotic-Resistant Gram-Positive Pathogens: What Is Relevant for the Clinician?[J]. Pathogens,2024,13(1)
[32] S Jyakhwo, A Dmitrenko,V V Vinogradov. Computer-Aided Discovery of Synergistic Drug-Nanoparticle Combinations for Enhanced Antimicrobial Activity[J]. ACS Appl Mater Interfaces,2025,17(8):11820-11830.
[33] G Xiao, J Li,Z Sun. The Combination of Antibiotic and Non-Antibiotic Compounds Improves Antibiotic Efficacy against Multidrug-Resistant Bacteria[J]. Int J Mol Sci,2023, 24(20)
[34] M Martínez-Pérez, R Fernández-Fernández, R Morón, et al. Selective Digestive Decontamination: A Comprehensive Approach to Reducing Nosocomial Infections and Antimicrobial Resistance in the ICU[J]. J Clin Med,2024,13(21)