Simplified bacterial identification and antimicrobial resistance detection in bacteraemia by on-chip multiplexed LAMP from blood cultures

Abstract

Bacteraemia is a life-threatening infection associated with substantial morbidity and mortality. Current diagnosis relies on blood culture followed by culture-based identification and antimicrobial susceptibility testing, which remain the clinical gold standard but require prolonged turnaround times. Isothermal amplification technologies such as loop-mediated isothermal amplification (LAMP) offer a rapid, cost-effective, and accessible alternative for pathogen identification and antimicrobial resistance detection directly from positive blood cultures, particularly in resource-limited settings. However, their widespread implementation remains constrained by sample preparation requirements and limited multiplexing capabilities. To address these challenges, we developed ALADIM (Accessible and integrated LAMP Analysis for multiplex Detection of Infectious agents in Microfluidics), an integrated microfluidic platform combining bacterial DNA extraction, multiplex on-chip LAMP amplification, and instrument-free amplicon detection by lateral flow immunoassay (LFIA). The complete workflow can be performed in less than two hours, compared with the 24-48 h typically required for conventional diagnostic methods. As a proof of concept, the system was designed to detect E. coli, the most frequently reported pathogen in bacteraemia, together with two clinically relevant antimicrobial resistance determinants, blaCTX-M-15 and blaOXA-48, as well as an internal positive control. The multiplex assay was first evaluated using 14 bacterial strains cultured in LB medium. Performance of the on-chip workflow was comparable to that of the reference method consisting of conventional tube-based LAMP followed by LFIA detection, achieving 100% sensitivity and >84% specificity for malF and blaOXA-48, and 68% sensitivity with 83% specificity for blaCTX-M-15. The complete ALADIM workflow was subsequently assessed using overnight positive blood cultures generated from 15 PCR-characterised strains, corresponding to 61 independent experiments encompassing sample preparation, DNA amplification, and LFIA readout. In this more challenging matrix, specificity remained above 77% for all targets. Sensitivity exceeded 94% for malF and blaOXA-48, whereas detection of blaCTX-M-15 showed lower sensitivity (56%). A comparable reduction in sensitivity was also observed with the reference in-tube method (>68%), underscoring the analytical challenges associated with direct analysis of blood culture-derived samples. Overall, this study demonstrates the feasibility of integrating sample preparation, multiplex nucleic acid amplification, microfluidic handling, and lateral-flow-based detection within a single user-friendly platform. While further optimization is required to improve performance in complex matrices, increase analytical sensitivity, and expand multiplexing capabilities, ALADIM provides a promising foundation for the development of accessible molecular diagnostics for bloodstream infections in low- and middle-income countries.