Rising Drug Resistance Among Gram-Negative Pathogens in Bloodstream Infections: A Multicenter Study in Ulanhot, Inner Mongolia (2017–2021)

Background

Bloodstream infection is a serious illness that occurs when pathogenic microorganisms invade the human bloodstream and it poses a significant threat to life and health due to its high mortality rate and potential for severe disruption of bodily functions [1,2]. Worldwide, bloodstream infection is emerging as an increasing menace to public health [3]. Every year, North America and Europe witness around 2 million instances of bloodstream infections, a statistic that correlates with 250 000 fatalities. Consequently, bloodstream infection stands as the primary cause of death resulting from infections in these regions [4]. Mortality induced by central line-associated bloodstream infections could be as high as 28% to 30% [5]. Furthermore, bloodstream infections have the potential to trigger sepsis, an intense systemic reaction to an infection. Sepsis is linked to elevated mortality rates, prolonged hospital stays, and increased healthcare expenses [6]. The timely diagnosis and treatment of bloodstream infections have been a major hurdle for laboratory technicians and medical professionals alike, given the complex nature of the disease and the critical need for prompt and accurate identification of causative microorganisms [7]. Nonetheless, the clinical and microbiological data concerning bloodstream infections have recently exhibited ongoing fluctuations [8]. The medical community has recently witnessed the introduction and utilization of cutting-edge technology, such as blood flow infection diagnosis and treatment technology, and a standardized set of guidelines in the form of the code of practice for blood culture in the clinical microbiology laboratory [4].

In 2016, a study conducted in 10 major educational hospitals spanning 7 regions in China, under the China Antimicrobial Surveillance Network (CHINET), highlighted a high occurrence of bloodstream infections, of 97.3% in 2773 instances [9]. Another nationwide prospective cohort study was conducted from 2007 until 2016 in 16 teaching hospitals across China [10]. The study revealed a rising mortality trend for patients infected with Pseudomonas aeruginosa or Acinetobacter baumannii, in contrast to a decline in the death rate linked to Staphylococcus aureus-related bloodstream infections [10]. These advancements have equipped a majority of medical professionals with robust theoretical support. This is of paramount significance due to the hazardous, urgent, and grave consequences that are associated with bloodstream infections [11–13]. Despite the availability of modern medical technology, many hospital clinicians continue to rely on the conventional practice of using anti-infective and clinical empirical drugs for the treatment of bloodstream infections [14]. Unfortunately, this practice has contributed to the steady rise in drug-resistant pathogenic microorganisms, posing great challenges to the effective management of infections [15,16]. Furthermore, community-acquired bloodstream infections exhibit traits such as swift onset, serious circumstances, and an elevated fatality rate [17]. Thus, it becomes imperative to administer prompt and suitable antimicrobial therapy to lower the mortality rate, particularly among individuals with sepsis or septic shock. To address this issue, several hospitals were selected as research bases in our study, in which we aimed to determine the distribution and drug resistance patterns of blood-borne pathogens, as well as the prevalence of pathogenic bacteria in bloodstream infections. This was achieved through the monitoring and statistical analysis of the pathogens obtained from blood cultures, coupled with antibiotic susceptibility testing. In this study, we aimed to collect theoretical data that would aid in the empirical treatment of clinical blood flow infections, while simultaneously reducing the prevalence of multidrug-resistant bacteria and enhancing the accuracy and effectiveness of drug use. This study characterized the prevalence of microbial pathogens and antimicrobial resistance in 2498 blood samples from patients in Ulanhot, Inner Mongolia, treated between 2017 and 2021. Specifically, we investigated the distribution of blood culture-positive strains in several microbiological laboratories. We also analyzed the resistance rates for the main pathogenic bacteria against specific antibacterial drugs.

Material and Methods

ETHICS STATEMENT:

This study was approved by the Ethical Review Committee of Xing’an League People’s Hospital (approval number 2017-021) and complied with the requirements of the Declaration of Helsinki. It was also approved by the participating institutions. Participant data were kept confidential and used for academic research only. Due to the retrospective design of this study, the Ethics Review Committee waived patient informed consent.

EXPERIMENT AND MATERIALS:

Columbia blood agar, Muller Hinton agar, MacConkey agar, chocolate agar, and Sabouraud glucose agar were purchased from Zhengzhou Antu Biological Engineering Co., Ltd. The antibacterial drugs used in this study included penicillin, ampicillin, ceftriaxone, vancomycin, linezolid, clindamycin, erythromycin, levofloxacin, meropenem, imipenem, cefepime, cefoperazone sulbactam, gentamycin, amikacin, ampicillin sulbactam, cefuroxime, aztreonam, and trimethoprim/sulfamethoxazole, purchased from Oxoid UK. In addition, minimum inhibitory concentration (MIC) test strips for meropenem, imipenem, vancomycin, penicillin, and ceftriaxone were purchased from Wenzhou Kangtai Biotechnology Co., Ltd.

The BacT/ALERT 3D system and its accompanying blood culture equipment, used for the culture and detection of bacterial and mycobacterial infections, were purchased from BioMerieux in France. The system includes culture bottles for different types of microbes, comprising aerobic and facultative anaerobic microbes (BacT/ALERT FA PLUS), anaerobic and facultative anaerobic microbes (BacT/ALERT FN PLUS), and aerobic and facultative anaerobic microbes (BacT/ALERT PF). The VITEK 2 Compact system, also manufactured by BioMerieux, was used for bacterial identification and drug sensitivity analysis, and is accompanied by drug sensitivity cards for different types of bacteria, including gram-positive bacteria (AST-GP67 and AST-P639), gram-negative bacteria (AST-GN13 and AST-N335), and gram-negative bacteria (AST-N334).

BLOOD CULTURE:

The guidelines concerning the collection and reporting of blood culture results were established following the expert consensus on the clinical practice of blood culture technology for the accurate diagnosis of bloodstream infections [18] and the Chinese expert consensus on standardized sample collection of blood culture for children [19].

DRUG SENSITIVITY TEST:

The breakpoint of the drug sensitivity test recommended by CLSI M100-S31 was used to analyze drug sensitivity test results [20]. Currently, tigecycline does not have a breakpoint in CLSI M100-S31. Therefore, the criteria outlined in the Expert Consensus on the Operating Procedures for In Vitro Drug Sensitivity Test of Tigecycline were adopted [21]. Moreover, cefoperazone/sulbactam has no break point established by the FDA, CLSI, or EUCAST. However, the break point of cefoperazone against relevant gram-negative bacteria is currently being used [22].

The drug sensitivity of Staphylococcus and Enterococcus was analyzed using the VITEK 2 Compact full-automatic bacterial identification and drug sensitivity analysis system, (France BioMerier Company), and the results were identified by the AST-GP67 or AST-P639 card (instrument method). The drug sensitivity tests for Enterobacteriaceae bacteria were performed using the VITEK 2 Compact full-automatic bacterial identification and drug sensitivity analysis system along with the supporting AST-GN13 or AST-N334 card from France BioMerier Co. For other non-fermenting sugar-negative bacteria, except Stenotrophomonas maltophilia, the VITEK 2 Compact full-automatic bacterial identification and drug sensitivity analysis system with the supporting AST-N335 card from France BioMerier were applied to conduct drug sensitivity testing and result interpretation. The quality control strains were S. aureus ATCC 25923 (disk diffusion method), ATCC 29213 (instrument method), E. faecalis ATCC 29212, E. coli ATCC 25922, P. aeruginosa ATCC 27853, K. pneumoniae (ATCC700603), and S. aureus (ATCC43300). Strains of Staphylococcus or Enterococcus that are resistant to vancomycin or linezolid, as determined by instrumental methods, needed to be re-identified. Once confirmed, the MIC values of the strains were re-verified using the appropriate Epsilometer test (E-test) strip. Furthermore, in the case of all Enterobacteriaceae strains displaying resistance to meropenem and imipenem, it was crucial to conduct re-identification and re-evaluate MIC values using the corresponding E-test strip as subsequent steps.

STATISTICAL ANALYSIS:

The data obtained from both the paper diffusion method and the automated instrument method MIC for drug sensitivity results were processed and analyzed using WHONET 5.6 software. Demographic data were presented as mean±SD for age, and the rate was expressed as a percentage. SPSS 19.0 statistical software was used for data analysis. The 4-table chi-square test, multiple independent sample contingency table chi-square tests, and Fisher’s exact tests were used to compare differences among groups. P<0.05 was considered statistically significant.

Results

DEMOGRAPHIC INFORMATION OF PATIENTS:

Patients’ data were collected from 2017 to 2021 from hospitals in Ulanhot, Inner Mongolia. The demographic characteristics of patients stratified by year are presented in Table 1.

PROFILING COMPOSITION AND DISTRIBUTION OF BACTERIAL SPECIES IN THE TARGET POPULATION:

Between 2017 and 2021, we collected a sample of 2498 pathogenic bacteria known to cause bloodstream infections. Among these bacteria, 35.83% (895/2498) were gram-positive, 62.45% (1560/2498) were gram-negative, and 1.72% (43/2498) were fungi. Figure 1 and Table 2 show the distribution of the main pathogenic bacteria.

ANALYSIS OF ANTIBIOTIC RESISTANCE RATES OF MAJOR GRAM-POSITIVE BACTERIA:

In S. aureus and coagulase-negative staphylococci, the rates of detecting methicillin-resistant S. aureus (MRSA) and methicillin-resistant coagulase-negative staphylococci (MRCNS) were 15.62% and 70.13%, respectively. Further analysis revealed that the rates of drug resistance were higher for MRSA and MRCNS relative to levels in methicillin-sensitive S. aureus (MSSA) and methicillin-sensitive coagulase-negative staphylococci (MSCNS; P<0.01). However, no strains that were resistant to vancomycin and linezolid were detected in the Staphylococcus group. In terms of resistance rates, MRSA, MSSA, MRCNS, and MSCNS showed resistance against penicillin G at rates of 100.00%, 80.87%, 100.00%, and 86.25%, respectively. Similarly, the resistance rates of these bacteria against clindamycin were 79.26%, 56.24%, 72.78%, and 60.32%, respectively. Furthermore, the percentage of resistance shown by these bacteria toward erythromycin were 83.57%, 65.83%, 90.75%, and 73.34%, respectively. It should be noted that these resistance rates were higher than 56.24%. Details of the resistance rates of MRSA, MSSA, MRCNS and MSCNS against antibacterial drugs are presented in Table 3.

The predominant strain of enterococcus identified was E. faecium, comprising 1.48% (37/2498) of all strains tested. None of the E. faecium isolates were resistant to linezolid, quinupristin/dalfopristin, or tigecycline. The drug-resistant strains were correctly identified and confirmed by reviewing the drug sensitivity results using the corresponding E-test strip. Analysis of data from 2017 to 2021 indicated that, apart from high-concentration gentamicin, levofloxacin, and tetracycline, the resistance rates of E. faecium to other antibiotics did not show significant differences across the 5 years (P>0.05). Further details on the antibiotic resistance rates of E. faecium are shown in Table 4.

E. coli and K. pneumoniae were the most prevalent bacteria among the Enterobacteriaceae strains, representing 35.15% (878/2498) and 15.73% (393/2498) of the total bacterial isolates, respectively (Figure 1, Table 2). The monitoring results among the 5 years indicated that E. coli had a significantly higher drug resistance than did K. pneumoniae in the common antibiotics, particularly for piperacillin (χ2=54.52, P<0.001), ceftriaxone (χ2=32.22, P<0.001), ciprofloxacin (χ2=129.07, P<0.001), levofloxacin (χ2=89.01, P<0.001), and compound sulfamethoxazole (χ2=56.27, P<0.001). However, there was no significant difference in cefazolin (χ2=3.30, P=0.069). The analysis of monitoring results indicated notable variations in drug resistance rates of E. coli for the following antibiotics between 2017 and 2021: piperacillin (χ2=11.68, P=0.002), amoxicillin/clavulanic acid (χ2=27.806, P<0.001), ampicillin/sulbactam (χ2=18.201, P=0.001), and cefepime (χ2=33.004, P<0.001). While in K. pneumoniae, only amoxicillin/clavulanic acid (χ2=35.994, P<0.001) showed a significant difference in drug resistance rate changes among the 5 years. No significant changes in resistance rates were observed over 5 years for the rest of the antibiotics in both E. coli and K. pneumoniae (all P>0.05). Further details on the antibiotic resistance rates of E. coli and K. pneumoniae are shown in Tables 5 and 6.

The non-fermenting gram-negative bacteria that were most commonly identified were P. aeruginosa and A. baumannii, constituting 2.88% (72/2498) and 1.64% (41/2498) of the total strains, respectively. There were no changes in the resistance rates among the 5 years in P. aeruginosa and A. baumannii (all P>0.05). The drug resistance rates of A. baumannii to antibiotics, including ceftazidime, ciprofloxacin cefatriaxone, cefepime, amikacin, levofloxacin, piperacilin/tazobactam, amikacin/sulbactam, gentamicin, meropenem, imipenem, minocycline, piperacillin, and compound sulfamethoxazole, were all above 30%. Conversely, the resistance rate of P. aeruginosa to tested antibiotics, including aztreonam, imipenem, ceftazidime, levofloxacin, piperacillin, ciprofloxacin, meropenem, cefepime, piperacillin/tazobactam, gentamicin, and amifloxacin, from 2017 to 2021 remained below 30%. The antibiotic resistance rates of P. aeruginosa and A. baumannii from 2017 to 2021 are shown in Tables 7 and 8.

Discussion

In this study, conducted between 2017 and 2021 in Ulanhot, Inner Mongolia, we investigated the composition, distribution, and antibiotic resistance rates of pathogenic bacteria from hospital patients. Of the 2498 bacteria collected, 35.83% were gram-positive, 62.45% were gram-negative, and 1.72% were fungi. S. aureus and coagulase-negative staphylococci exhibited considerable resistance rates to antibiotics, especially to methicillin, while no resistance to vancomycin and linezolid was found. E. faecium, the predominant Enterococcus strain, exhibited stable resistance rates over the years, with no resistance to linezolid, quinupristin/dalfopristin, and tigecycline. Among the Enterococcus strains, E. coli and K. pneumoniae were the most prevalent and exhibited varying antibiotic resistance patterns. A. baumannii displayed high resistance to several antibiotics, although resistance to minocycline decreased annually. In contrast, P. aeruginosa showed resistance rates below 30% to several antibiotics during the study period. These findings shed light on the regional prevalence of bacterial strains and their resistance patterns, which is essential for improving antibiotic use in healthcare settings.

Previous studies indicated an increasing trend in mortality among patients with bloodstream infections due to P. aeruginosa or A. baumannii, and decreasing mortality with S. aureus-related bloodstream infections [10]. Our study contributes to the understanding of these observations by providing in-depth insights into the prevalence and antibiotic resistance patterns of these bacteria. We found that P. aeruginosa showed a resistance rate below 30% to several antibiotics, while A. baumannii demonstrated high resistance to multiple antibiotics, potentially explaining the increase in mortality rates in patients infected with these bacteria. Overall, although previous studies have established the seriousness of bloodstream infections in China, our research provides more recent and detailed information on the bacterial profile and antibiotic resistance patterns in Inner Mongolia. Our findings are expected to guide the application of targeted antimicrobial therapies, improve patient outcomes, and contribute to the overall management of bloodstream infections.

Septicemia and bacteremia are severe infectious diseases that affect the bloodstream [17]. They are characterized by serious clinical symptoms, such as high fever, sudden chills, rapid heartbeat, shortness of breath, and rash, and in some cases, altered mental status [23]. Previous research indicates that shock is a common complication in patients with severe bloodstream infections, and the infections can lead to disseminated intravascular coagulation and multiple organ failure [24]. The use of various diagnostic and treatment technologies, along with drugs and other factors, has led to an increase in cases of S. aureus and coagulase-negative staphylococci in recent years [25]. These factors include iatrogenic factors, like the use of anti-infectives, hormones, chemotherapy, and immunosuppressants, as well as non-iatrogenic factors, such as malnutrition, HIV, old age, and coma. The number of bloodstream infections caused by enterococci and fungi has been increasing every year, and the pathogenic bacteria are becoming more resistant to drugs as the incidence of these infections rises [26]. As an illustration, the extensive resistance displayed by K. pneumoniae limits the availability of viable antibiotic options, thereby posing a significant danger to human life. K. pneumoniae has the ability to generate carbapenemase enzymes, giving it resistance against carbapenem antibiotics. This particular strain also exhibits resistance to penicillin and cephalosporin antibiotics. Epidemiological studies have shown that bloodstream infections can result in high mortality rates, prolonged hospital stays, expensive medical bills, and significant harm [27]. Thus, strategies that can prevent and control bloodstream infections should be explored, although there have been no reported cases in Inner Mongolia [28]. In the present study, we discovered that out of 2498 strains, MRSA accounted for 15.62% of S. aureus, while coagulase-negative staphylococci made up 70.13% of MRCNS. The reason for the higher prevalence of coagulase-negative staphylococci is because they are commonly found in human mucosal tissues and the microflora of skin tissue environments, although they are not entirely considered normal. Additionally, their prevalence is also attributed to various invasive procedures that are performed. Inserting medical devices such as indwelling gastric tubes and urinary catheters can create a pathway for skin flora to invade the body and cause bloodstream infections. Notably, the presence of coagulase-negative staphylococci in blood cultures can also compromise their quality and safety [29]. Therefore, it is necessary to analyze the specific circumstances of each patient to determine whether contamination or infection is caused by an increase in the proportion of coagulase-negative staphylococci.

Strict adherence to disinfection procedures and using a single set of bottles are crucial for accurate collection of blood samples from patients. Collecting blood from various body parts can enhance the precision of the sample, ultimately improving the accuracy of the test. Furthermore, by incorporating the patient’s symptom performance, procalcitonin test results, and blood culture results, a comprehensive analysis of the test results can be conducted to further enhance the detection of pathogenic bacteria. MRSA and MRCNS showed significantly higher resistance rates to aminoglycosides, lactams, fluoroquinolones, and macrolides than did MSSA and MSCNS. Nevertheless, Staphylococcus strains did not show any resistance to vancomycin, linezolid, and tigecycline, and E. faecium was not resistant to vancomycin, teicoplanin, tigecycline, and linezolid. However, its resistance rate to ampicillin, penicillin G, levofloxacin, and erythromycin exceeded 80%. Although the resistance rate of K. pneumoniae to carbapenems slightly increased among Enterobacteriaceae bacteria, that of E. coli decreased slightly to reduce the incidence of carbapenem-resistant Enterobacteriaceae infection [30–32]. The resistance level of E. coli to commonly used fluoroquinolones, namely ciprofloxacin and levofloxacin, exceeded 60%. Particularly, the resistance rate for ciprofloxacin has shown a significant annual increase. Research has indicated that the efficacy of these 2 drugs in treating E. coli-related urinary tract infections (UTIs) among hospitalized patients is approximately 30% only, which aligns with our findings. In cases of community-acquired UTIs, the emergence of the BLA CTX-M-14 gene within E. coli has contributed to heightened drug resistance. As such, there is a pressing need to restrict the empirical usage of cephalosporins and fluoroquinolones for treating UTIs in China. It is crucial to be vigilant for the emergence of staphylococcal strains that contain S. aureus that is resistant to vancomycin, linezolid, and tigecycline (vancomycin-resistant S. aureus [VRSA]), or S. aureus mediated by vancomycin, linezolid, and tigecycline (vancomycin-intermediate S. aureus [VISA]), as the clinical use of vancomycin, linezolid, and tigecycline continues to increase [33].

The use of vancomycin, linezolid, and tigecycline is considered an effective strategy for MRSA and MRCNS therapy. Due to the high resistance of carbapenem-resistant Enterobacteriaceae to commonly used antibacterial drugs and the limited availability of alternative antibacterial drugs, treatment options are severely limited [34]. It is crucial to monitor the emergence of carbapenem-resistant Enterobacteriaceae infections during routine tasks in the future. Strategies to identify potential hospital-acquired infection outbreaks rapidly, inform relevant departments promptly, and investigate the causes through scientific antibacterial drug management should be developed through collaborative efforts from different departments to reduce the frequency of carbapenem-resistant Enterobacteriaceae infections [35]. Carbapenem drug resistance occurs and develops primarily due to the presence of carbapenemase, which includes drug-resistant bacteria, class D carbapenemase, New Delhi metallo-beta lactamase, vasoactive intestinal polypeptide, and hypoxanthine nucleotide. The Clinical Laboratory Standards Association’s recommended method can be used to detect carbapenem resistance, and the quality of susceptibility testing can be enhanced by incorporating drugs with the lowest inhibitory concentration, such as fosfomycin, tigecycline, cefoperazone, and polymyxin [36].

P. aeruginosa was another significant opportunistic pathogen responsible for hospital infections, with a resistance rate of less than 20% to carbapenems and other antibacterial drugs [37]. Since the occurrence of such circumstances is linked to the limited application options for certain drugs, the co-administration of aminoglycosides with other medications can also lower the resistance rate of antibiotics such as carbapenem. According to the national drug usage survey, the application of P. aeruginosa and the resulting drug resistance vary across different regions. Blindly administering drugs without comprehensively analyzing the basic situation of each region can promote the development of bacterial resistance and increase the incidence of infections. Hence, while using antibacterial drugs, it is crucial to properly match them with other drugs, considering the region’s drug resistance. To combat infections caused by K. pneumoniae and A. baumannii, stringent prevention and control measures must be implemented, and the monitoring of drug resistance and research into drug resistance mechanisms must be intensified [38]. Following a comprehensive assessment of the infection site, bacteria involved, and factors influencing drug resistance, healthcare professionals must consider the local distribution of pathogens and promptly initiate appropriate empiric antibacterial therapy. Additionally, they should regularly investigate the epidemiological data of patients with bloodstream infections. It is critical to monitor the changes in drug resistance of infectious pathogens, use antibiotics judiciously, and control the spread and prevalence of drug-resistant strains in hospitals.

Although this study has provided valuable insights into the prevalence and antibiotic resistance patterns of bacteria causing bloodstream infections in Ulanhot, Inner Mongolia, it is important to acknowledge certain limitations. First, the study was conducted in a specific geographic region, which can limit the generalizability of our findings to other areas of China or globally. Second, although we collected a sizable sample of bacterial isolates, the actual diversity and prevalence of bacterial species causing bloodstream infections could be underrepresented. Third, our study focused primarily on bacterial pathogens and does not provide insights into fungal or viral causes of bloodstream infections, which are also clinically relevant. Fourth, changes in the hospital environment, patient population, and local antibiotic stewardship practices over the study period may have influenced the trends observed in our data. Finally, owing to the retrospective nature of the study, we could not control for potential confounding factors such as patient comorbidities, variations in clinical practice, or differences in laboratory methods across the participating hospitals. Despite these limitations, we believe our findings contribute significantly to the understanding of bloodstream infections and antibiotic resistance patterns in this region and provide a strong foundation for future prospective and multicentric studies.

Conclusions

Our study provides critical insights into the prevalence, distribution, and antibiotic resistance patterns of bacteria causing bloodstream infections in Ulanhot, Inner Mongolia, from 2017 to 2021. These results demonstrate that gram-negative bacteria were the primary type of pathogenic bacteria identified from bloodstream infections, Among the tested bacteria, A. baumannii, P. aeruginosa, K. pneumoniae, and E. coli showed an increasing resistance to antibiotics. Alarming rates of resistance to commonly used antibiotics were observed, indicating the pressing need for vigilant monitoring and judicious use of antibiotics. Our findings underscore the importance of local surveillance in informing empirical antibiotic therapy and guiding interventions to control the spread of antibiotic resistance.