High-throughput clinical antimicrobial susceptibility testing and drug-resistant subpopulation detection in...

The optimized Easy and Zooming MTT (EZMTT)-based antimicrobial susceptibility testing (AST) method for Gram-negative bacteria demonstrated 5- to 20-fold enhanced sensitivity and improved the detection limit to approximately 1.13% ± 0.30% growth. This advancement enables the detection of drug-resistant subpopulations that were previously undetectable by the gold standard broth microdilution (BMD) method, which has a growth limitation of approximately 10.15% ± 4.70%. Consistent results were achieved through AST of 205 clinical bacterial samples using both the EZMTT and BMD methods. At 450 nm, the overall positive and negative compliance rates were 98.94% and 99.58%, respectively, with a kappa value of 0.981 and a 95% CI of 0.974-0.988. For each of the 23 antibiotics tested, compliance rates exceeded 90.00%, with kappa values above 0.794 ~ 1.000 for 21 of them. Notably, the AST screening identified a clinical K. pneumoniae strain (KP007) containing drug-resistant subpopulations. Further analysis using the K-B method identified 13 drug-resistant subpopulations in KP007, with detection rates of 100% by the EZMTT method, 77% by the BMD method, and less than 44% by the VITEK method. A series of heteroresistant strains was generated by mixing the parent-sensitive strain (DH10B) with fivefold dilutions of its resistant strain transformed with an ampicillin resistance gene (DH10B-Amp^R). Results indicated that the EZMTT method is more than 3,000 times more effective at detecting the resistant subpopulation (DH10B-Amp^R) compared to VITEK. In conclusion, this study demonstrates that the EZMTT method exhibits good consistency and stability and is a highly sensitive AST method for detecting drug-resistant subpopulations in Gram-negative bacteria.IMPORTANCEBacterial drug resistance represents a significant global health concern. Currently, the minimum inhibitory concentration (MIC) is widely accepted as a primary criterion for determining bacterial sensitivity or resistance. However, pathogenic bacteria classified as sensitive based on MIC criteria may still exhibit resistance. One crucial factor contributing to this phenomenon is the heterogeneity of bacterial resistance. Due to the small number of drug-resistant bacteria in these subpopulations, their corresponding drug resistance is often challenging to detect in MIC data. AST remains a critical method for guiding the use of antibacterial agents. In this study, we employed the EZMTT method to accurately detect drug-resistant bacterial subpopulations. This method significantly improved assay consistency, stability, and sensitivity in clinical bacterial strains when compared with other methods. Consequently, the application of EZMTT-modified BMD assay in clinical settings may potentially prevent the development of multidrug-resistant bacteria and enhance the treatment of infectious diseases in the future.