Introduction: Carbapenem-resistant Enterobacteriaceae (CRE) poses a significant challenge in treating infections, leading to increased mortality and healthcare burden. Early and accurate identification of CRE is crucial for appropriate antibiotic selection. This study aimed to evaluate the prevalence of CRE isolated from ICU patients and carbapenemase gene profiles among Klebsiella pneumoniae isolates. Methods: A cross-sectional study was conducted on Enterobacteriaceae isolates from 53 ICU patients from March to May 2023. Among 125 initial isolates, 89 non-duplicated Enterobacteriaceae strains were included. Carbapenem resistance was determined from Kirby-Bauer disk diffusion susceptibility testing results. Carbapenemase-producing phenotype and carbapenemase classification were classified from BD Phoenix M50 results using the NMIC-500 CPO panel. Carbapenemase-encoding gene were identified from amplification results of carbapenemase-encoding genes in K. pneumoniae strains using multiplex real-time PCR. Results: Of 89 isolates, 51 (57.3%) were carbapenem-resistant, predominantly K. pneumoniae (34/48, 70.8%) and S. marcescens (12/13, 92.3%). All isolates were carbapenem-resistant due to carbapenemase production. Class D carbapenemase was the most prevalent (39/51, 76.5%). Among K. pneumoniae isolates, 12/34 harboured single carbapenemase genes: blaOXA-48-like (17.7%), blaKPC (14.7%), and blaNDM-1 (2.9%); Sixteen isolates carried two genes: blaNDM-1+blaOXA-48-like and blaKPC+blaOXA-48-like; Three isolates possessed three genes: blaKPC+blaNDM-1+blaOXA-48-like. Multiple gene carriage was significantly associated with increased antibiotic resistance (p<0.01). Conclusion: This study revealed a high prevalence of carbapenemase-producing, carbapenem-resistant Enterobacteriaceae. This significantly increased the risk of spreading multidrug-resistant pathogens. Therefore, stronger infection control measures are in demand. Additionally, automated susceptibility testing and carbapenemase class identification are crucial for guiding treatment decisions.
Lê et al. (Tue,) studied this question.