This study investigated the prevalence of CRE in fecal samples of hospitalized children in Shandong Province and found a CRE detection rate of 4.6% in the children's feces. This rate was comparable to previous report, such as a 4.5% CRE positivity rate in rectal swabs from children in the United Kingdom in 2012[15], and a 5.2% detection rate in the United States [16]. In China, a previous study revealed the CRE positivity rate for pediatric inpatients in Southern China is 3.92%[7], while in Shanghai, the carriage rate for outpatient children was 3.6% in 2016[8], and 8.6% for hospitalized children in 2019[17]. Our results indicated that the carriage rate of CRE in children in Shandong, northern China was not significantly higher than in these regions,reflecting a moderate CRE burden in our pediatric population comparable to both Western and other regional Chinese settings.
Regarding age grouping, there was no significant difference was found in the detection rate of CRE in children in this study. The overall prevalence of CRE in children in China was reported is 6.4%, with 8.8% in the neonatal period, 7.3% in infancy, 3.8% in toddlerhood, 4.0% in the preschool period, 4.7% in school age, and 7.4% in adolescence[18]. Notably, our study found that the detection rate in neonatology and neonatal surgery was significantly higher than in other departments. This finding was consistent with a study from the Asian region in 2023, which indicated that infants under one month of age had a higher rate of infection than other children[19]. This suggests that neonates, particularly those requiring surgical intervention, are at increased susceptibility to CRE infections, underscoring the need for stringent infection control measures in these high-risk areas.
This study conducted a detailed analysis of the antimicrobial drug resistance profiles of CRE strains. The findings of this study were consistent with previous reports, confirming that CRE strains were commonly resistant to many classes of antibiotics [20], which provided an important basis for future prevention and treatment strategies and guidance on antibiotic use. The higher multidrug resistance rate observed in this study further highlighted the major challenges in clinical treatment. It was particularly noteworthy that the resistance rate to β-lactam antibiotics was as high as 90%-100%, which reflected the widespread clinical use of this class of drugs and the corresponding resistance pressure. In addition, the resistance rates to SXT and TET were as high as 95% and 85% respectively, highlighting the prevalence of CRE isolates resistant to these commonly used antimicrobials. Resistance rates to aminoglycoside antibiotics in particular were also detected to be higher, posing a serious challenge to treatment options for pediatric patients, as some key antibiotics had limited use in children. For instance, fluoroquinolones were unsuitable for children and adolescents because of potential damage to joint cartilage and effects on bone growth[21]; and aminoglycosides, although commonly used, could cause ototoxicity and nephrotoxicity if overused[22];Tetracycline antibiotics may affect bone and tooth development in children[23]. At the same time, there were some differences compared to the previous report [24], which found higher susceptibility rates to CZA and CT. In this study, most CRE isolates showed intermediate resistance to CT, which showed that CT had limited clinical efficacy.
Our results showed that CRE strains generally carried multiple drug resistance genes. In particular, β-lactamase genes such as blaNDM−1, blaNDM−5, and OXA-1 are ubiquitous among strains, and the high prevalence of these genes was consistent with CRE reported in many countries worldwide [25]. In addition, a variety of types of resistance genes were discovered in our study, including the sulfa antibiotic resistance genes sul1, sul2 and sul3, and the TET resistance gene tetA. The presence of these genes reflected the bacterial response to the environment. Adaptability to antibiotic selection pressures. The prevalence of macrolide antibiotic resistance genes such as mphA and ermB also indicated the difficulties that these antibiotics might encounter in clinical practice[26]. For fluoroquinolones and other broad-spectrum antibiotics, we observed that 32% of strains carried quinolone resistance genes, a finding consistent with the global trend of increasing quinolone resistance[27]. The increase in such resistance was associated with a high frequency of use, inappropriate use and widespread use in agriculture[28]. It was worth noting that the ubiquity of multidrug resistance efflux pump genes such as mdfA illustrates the ability of CRE strains to enhance their drug resistance through multiple mechanisms[29]. This type of efflux pump could pump a variety of antibiotics out of the cell, effectively reducing the internal antibiotic concentration, thereby reducing the bactericidal or bacteriostatic effect of the antibiotics.
The cluster analysis results of this study showed that three patients carried carbapenem-resistant E. coli isolates with ST167, suggesting that dominant strains of this ST type might exist in children's hospitals. It was worth noting that the ST167 strain had not been reported before 2017, but the latest research showed that this strain was not only found in hospital patients but also in healthy individuals, indicating its possible spread in the population wider than expected[30]. Globally, ST167 strains were widely distributed and found in 164 strains in 25 countries, 95 of which carry the blaNDM gene, and blaNDM−5 was the most common. In China, ST167 strains have the highest prevalence, accounting for 37%[31]. These data indicated that the prevalence of ST167 carbapenem-resistant E. coli was a public health issue requiring global attention. Moreover, Klebslella pasteurii strain isolated was with unique or incompletely defined sequence type ST*ff57 and the type of Raoultella ornithinolytica strain cannot be determined. Raoultella ornithinolytica was reclassified from the genus Klebsiella in 2001, so data on its epidemiology and clinical relevance are not yet complete, and more studies are needed to determine its ST pattern[32].
In plasmid analysis, we observed that IncFIB and IncFII series plasmids were extremely prevalent in CRE strains, suggesting that these plasmids might be the main vectors of horizontal transmission of resistance genes. Especially for the IncFIB family, its frequent detection in multiple strains highlighted its central role in the spread of antibiotic resistance. According to existing studies, the high prevalence of plasmids such as IncFIB, IncFII, IncR and IncFIA in CRE further indicated their importance in the spread of multidrug resistance[33]. These findings highlighted the role of specific plasmids in antibiotic resistance and provided important information for further research and countermeasures. Although the IncX3 plasmid showed high prevalence in environmental samples from Shandong and children's samples from Shanghai[17, 34], the detection rate of IncX3 was lower in children's samples from Shandong, while the IncFIB plasmid appeared to be particularly common. This difference might stem from the different compositions of the microbial communities. In our study, the main CRE strain isolated was carbapenem-resistant E. coli. A study on E. coli isolates contained a variety of plasmids, mainly IncFIB (AP001918). This plasmid has been confirmed to be associated with resistance to a variety of antibacterial drugs, including β-lactas, aminoglycosides, sulfonamides, tetracyclines, etc[29]. These observations highlighted the importance of considering sample type and geography when studying the spread of resistance genes.
This study filled a research gap on the carriage of CRE among hospitalized children in Shandong Province, northern China, providing preliminary data for understanding the CRE infection status in this population. However, this present study has limitations, mainly in sample representativeness and the lack of detailed clinical information. First, only stool sample was collected and detected from each child, and collecting and incorporating data from multiple clinical specimens (blood, urine, and sputum) of hospitalized children may provide additional power to our study. Second, due to a lack of adequate clinical information about hospitalize children, stool samples may be obtained from the children after antibiotic treatment; this may have affected the results. Future research should further delineate population characteristics, especially with more exhaustive observation of the neonatal group, to better understand CRE transmission and infection in different populations. Furthermore, studies with larger sample sizes and more techniques are warranted to confirm these findings and investigate the mechanisms of antimicrobial drug resistance underlying the observed associations.