Abstract INTRODUCTION: The World Health Organization recently highlighted the serious worldwide problem of the emergence of antibiotic-resistant or antibiotic multidrug-resistant bacteria. Carbapenem-resistant Enterobacterales, including carbapenemase-producing Enterobacterales (CPE), are major antibiotic-resistant bacteria that can be identified by various methods, including antibiotic susceptibility testing, PCR, and immunologic assays. However, there is a need for a faster, more accurate, low-cost, and easy method to detect CPE strains.
OBJECTIVES: We previously developed an osmotic shock matrix-assisted laser desorption/ionization mass spectrometry (OS-MALDI MS) method for directly detecting intact Klebsiella pneumoniae carbapenemase (KPC) using osmotic shock cell lysis. In this study, we evaluated the OS-MALDI MS method and compared it with two other methods (octyl-glucoside-aided direct KPC detection method [OG-MALDI MS] and Bruker's MBT subtyping module indirect method [MBT-SM MALDI MS]).
METHODS: To direct detect KPC-2 protein on MALDI-TOF MS, all Gram-negative bacterial isolates were lysed by the osmotic shock method. We completed an analytical performance evaluation of the OS-MALDI MS method with 24 clinical isolates (12 KPC-2 positive & 12 KPC-negatives) according to Clinical and Laboratory Standards Institute guidelines. Upon the criteria (e.g., intensity threshold and mass window), clinical testing was performed with 437 clinical isolates, including 292 KPC-producing bacteria and 145 non-KPC-producing bacteria. For OS, OG and MBT-SM MALDI MS, MALDI MS spectrometry was performed in linear and positive mode (m/z range: 12,000 ~ 32,000). Pairwise compassion studies between two methods (OS versus OG, OS versus MBT-SM, and OG versus MBT-SM) were performed by Mann-Whitney test (independent samples) using the MedCalc program (version 20.106).
RESULTS: The OS-MALDI MS and comparator methods (OG-MALDI MS and MBT-SM MALDI MS) were analyzed as a three-way comparison between the candidate method, the comparison method, and diagnostic accuracy criteria according to the CLSI EP12-A2 guideline. OS-MALDI MS showed significantly higher sensitivity than OG and MBT-SM MADLI MS (P-value, 0.05) while all the three methods showed 100% specificity. The difference in sensitivity between the OS-MALDI MS and OG-MALDI MS methods was 62.0% and the difference between the OS-MALDI MS and MBT-SM MALDI MS methods was 70.6%. Accuracy of the OS-MALDI MS, OG-MALDI MS, and MBT-SM MALDI MS methods was 97.3%, 55.9%, and 50.2%, respectively. Based on a disease prevalence of 66.74% (292/437) in the evaluated samples, the OS-MALDI MS method exhibited 95.9% sensitivity (TP / TP +FN), 100.0% specificity (TN / FP + TN), and 100.0% precision (TP / TP + FP) for detecting KPC. The OS-MALDI MS method had a positive predictive value of 99.29% (95%score confidence limit = 97.24% to 99.82%) and a negative predictive value of 91.82% (95% score confidence limit: 86.84% to 95.03%).
CONCLUSION: The OS-MALDI MS method clearly outperformed the other methods, exhibiting the highest accuracy and sensitivity of the three methods. We propose the OS-MALDI MS method as a practical, useful method for clinic environments, which may help guide appropriate antibiotic treatment and contribute to the prevention of the spread of CPE.
|