TDLAS-Based Antimicrobial Susceptibility Testing Using Microbial CO2 Evolution and Validation in Gram-Positive and Gram-Negative Bacteria

Introduction Rapid and accurate antimicrobial susceptibility testing (AST) is pivotal for guiding clinical therapy and curbing antimicrobial resistance (AMR). Traditional AST methods despite being the gold standard, lack real-time growth kinetic information and often involve invasive or cumbersome procedures. This study aims to develop a universal, non-invasive AST method based on Tunable Diode Laser Absorption Spectroscopy (TDLAS). Methods Through the monitoring of microbial CO2 evolution in the headspace of sealed culture bottles, established a calibration curve relating threshold time (TT) to total viable count (TVC) and constructed a dose-response model to quantitatively determine the minimum inhibitory concentration (MIC). Results Validation using the Staphylococcus sciuri standard strain demonstrated that the MIC of ampicillin (0.16–0.18 mg∙mL−1) determined by this method was highly concordant with the CLSI M07 standard broth microdilution method. Four independent replicates exhibited excellent stability, with a coefficient of variation (CV) of only 2.75%. The method was further successfully extended to clinical isolates, validating its applicability across Gram-positive bacteria (Staphylococcus aureus) and Gram-negative bacteria (Escherichia coli, Enterobacter cloacae). Different from traditional methods, the continuous monitoring capability of this system visually captures detailed growth dynamics under antibiotic stress (such as growth delays and partial inhibition) providing a richer dimension of information for resistance evaluation. Conclusion As a universal, non-invasive, and low-cost approach, this AST method provides MIC values consistent with standard protocols while providing richer detailed growth dynamics information than traditional methods. Consequently, this method shows significant potential for deployment in clinical on-site settings and resource-limited environments.