Diagnostic Approaches for Detecting Metallo-β-Lactamase–Producing Gram-Negative Bacteria: A Review of MALDI-TOF Mass Spectrometry, Culture Methods, and Emerging Machine Learning Applications

Abstract

Metallo-β-lactamase (MBL)–producing Gram-negative bacteria represent a major challenge to global healthcare due to their ability to confer resistance to carbapenems, often leaving limited therapeutic options. Early and accurate detection of these organisms is essential for appropriate antimicrobial therapy, infection control, and antimicrobial stewardship. Conventional culture-based and phenotypic methods remain widely used but are limited by prolonged turnaround times, subjective interpretation, and variable diagnostic performance. Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry (MALDI-TOF MS) has emerged as a rapid tool for bacterial identification and has recently been adapted for functional detection of antimicrobial resistance through modified carbapenem hydrolysis assays. In parallel, machine learning approaches applied to MALDI-TOF spectral data have shown promising results in automating and enhancing resistance detection. This review critically summarizes current diagnostic approaches for MBL detection, compares their diagnostic performance, and discusses the evolving role of machine learning–assisted MALDI-TOF MS in routine clinical microbiology. Key challenges, standardization issues, and future research directions are also highlighted to support clinical translation.