From bulk samples to single cells: measurement strategies in Raman-based antibiotic susceptibility testing.

Area of Science:

• Microbiology
• Analytical Chemistry
• Spectroscopy

Background:

• Antibiotic susceptibility testing (AST) is crucial for guiding treatment decisions.
• Conventional AST methods can be slow, hindering timely clinical intervention.
• Raman spectroscopy presents a rapid, phenotypic approach to bacterial AST.

Purpose of the Study:

• To systematically compare three distinct Raman-based AST measurement strategies.
• To evaluate their performance for detecting bacterial response to antibiotics.
• To identify optimal methods for diverse sample types, including low-cell-number samples.

Main Methods:

• Comparison of dried bulk, dielectrophoresis (DEP)-chip cell cloud, and filter-based single-cell Raman measurements.
• Exposure of *Escherichia coli* to amoxicillin and levofloxacin at varying concentrations.
• Analysis of Raman spectra using PCA-LDA for classification, validated against broth microdilution (BMD).

Main Results:

• Dried bulk measurements demonstrated the highest agreement with BMD and robust classification.
• DEP-chip measurements offered reduced handling in a closed system with good performance.
• Single-cell measurements provided the simplest workflow for low bacterial concentrations but had reduced robustness.

Conclusions:

• Methodological choices in Raman-based AST involve trade-offs between robustness, complexity, and suitability for low-cell-number samples.
• Dried bulk measurements are most robust but less suitable for low cell counts.
• DEP-chip and single-cell methods offer alternatives with specific advantages and limitations for different applications.