Detection of colistin resistance from Escherichia colicolonies grown on agar cultures and bacterial pellets obtained from positively flagged blood cultures by MALDI-TOF MS-based rapid assay: aproof-of-concept study.

Purpose: Polymyxins (polymyxin B and colistin) are one of the last-resort antibiotics, whose use is limited in the alternative treatment of certain infections due to carbapenem-resistant Enterobacterales, difficult-to-treat Pseudomonas aeruginosa, and carbapenem-resistant Acinetobacter baumannii. With the increased usage of colistin in clinical settings, the colistin resistance among Enterobacterales gradually increased and became one of the most important public health threats. Therefore, the rapid and accurate detection of colistin resistance have a significant effect on the mortality rate of Gram-negative bacterial infections. For this purpose, new methodologies are being developed, and matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF) mass spectrometry (MS)-based tools are one of these promising techniques.

Methods: A total of 103 Escherichia coli isolates were included in the study (57 colistin-susceptible and 46 colistin-resistant). For the determination of colistin resistance, strains were tested with the MALDI Biotyper sirius System (Research Use Only [RUO]; Bruker Daltonics GmbH & Co. KG, Germany) equipped with the MBT HT LipidART Module (RUO; Bruker Daltonics GmbH & Co. KG, Germany) using MALDI-TOF MS in negative-ion mode. In these processes, two starting materials were used: (I) Bacteria grown on agar culture (AC), (II) bacterial pellets obtained from positively flagged blood cultures (BC) purified by the MALDI Sepsityper Kit 50 (Bruker Daltonics GmbH & Co. KG, Germany). The research use only MBT Lipid Xtract™ Kit (RUO; Bruker Daltonics GmbH & Co. KG; Germany) was used for the extraction of Lipid A molecules from bacteria grown on AC and from bacterial pellets of BCs. The performance of the MBT HT LipidART Module was determined for both sample types. The colistin susceptibility was tested by a broth microdilution based commercial kit (UMIC Colistin; Bruker Daltonics GmbH & Co. KG, Germany), as the reference method.

Results: Using the MBT HT LipidART Module in Compass HT, the sensitivity and specificity of the AC-derived samples were determined to be 97.7%, and 98.2%, respectively. For BC-derived samples, the corresponding values were 93.3%, and 94.7%, respectively. The AC-grown E. coli strains (both colistin-susceptible and colistin-resistant), and the BC-grown, colistin-susceptible E. coli strains had a dominating, native Lipid A peak at m/z 1796. In contrast, in BC-grown, colistin-resistant E. coli isolates, three dominant native Lipid A peaks (m/z 1796, m/z 1824, and m/z 1840) were observed. In addition, modified Lipid A peaks at m/z 1919 and m/z 2043 were observed in the mass spectra.

Conclusion: The MBT Lipid Xtract™ Kit (RUO) in combination with the MBT HT LipidART Module (RUO) proves to be a promising tool for an effective detection of colistin resistance from AC-grown bacterial isolates and BC-derived samples, after appropriate development and regulatory approval.