ABSTRACT
We compared the performance of 2 automated systems for detection of carbapenemase-producing Enterobacteriaceae, BD MAX Check-Points CPO (CPO assay) and Xpert Carba-R assay, with culture confirmed by polymerase chain reaction as the reference method. Using 867 samples from 627 patients, the overall sensitivity, specificity, total positive predictive value, and negative predictive value of the CPO assay were 95.7%, 96.5%, 60.8% and 99.8% and for the Xpert assay were 97.9%, 99.8%, 95.8%, and 99.9%, respectively. The cycle threshold values (Ct value) of the false-positive CPO assay results were significantly higher than those of true-positive cases (P < 0.001). By applying a modified cut-off Ct value of 37.3 for klebsiella pneumoniae carbapenemases (KPC), the positive predictive value for KPC was improved from 52.9% to 89.5%. The CPO assay will be useful when handling many specimens, as tests are conducted in batches. However, positive cases showing high Ct values should be confirmed by another assay to rule out false positivity.
Subject(s)
Carbapenem-Resistant Enterobacteriaceae , Enterobacteriaceae Infections , beta-Lactamases , Bacterial Proteins/genetics , Carbapenem-Resistant Enterobacteriaceae/genetics , Enterobacteriaceae Infections/diagnosis , Humans , Klebsiella pneumoniae/genetics , Polymerase Chain Reaction , Sensitivity and Specificity , beta-Lactamases/geneticsABSTRACT
BACKGROUND: Rapid and accurate microbial identification and antimicrobial susceptibility testing (AST) are essential for timely use of appropriate antimicrobial agents for bloodstream infection. To shorten the time for isolating colonies from the positive blood culture, various preparation methods for direct identification using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) system were developed. Here, we evaluated the SepsiPrep kit (ASTA Corp.) for direct identification of microorganisms and AST from positive blood cultures using MicroIDSys Elite MALDI-TOF MS system (ASTA Corp.) and VITEK-2 system (bioMérieux). METHODS: For direct identification, a total of 124 prospective monomicrobial positive blood culture bottles were included. For direct identification, the pellet was prepared by centrifugation and washing twice. For direct AST, the pellet was suspended in 0.45% saline and adjusted to McFarland 0.5. The results from the direct identification and AST using MicroIDSys Elite and VITEK-2 system were compared to those from the conventional method performed with pure colony subcultured on agar plate. RESULTS: Compared to the conventional method using pure colony, correct direct identification rate was 96.5% and 98.5% for 57 gram-positive isolates and 67 gram-negative isolates, respectively. For direct AST, among the 55 gram-positive isolates, the categorical agreement (CA) for staphylococci, streptococci, and enterococci was 96.7%, 98.4%, and 94.1%, respectively. For 66 gram-negative isolates, the CA for Enterobacterales and non-fermentative gram-negative rods was 99.0% and 96.6%, respectively. CONCLUSIONS: The SepsiPrep kit was easy to use combined with MicroIDSys Elite and VITEK-2 system and also, the correct identification and AST rate were very high.
Subject(s)
Bacteremia/microbiology , Bacterial Typing Techniques/methods , Microbial Sensitivity Tests/methods , Anti-Bacterial Agents/pharmacology , Blood Culture/instrumentation , Blood Culture/methods , Gram-Negative Bacteria/classification , Gram-Negative Bacteria/drug effects , Gram-Positive Bacteria/classification , Gram-Positive Bacteria/drug effects , Humans , Microbial Sensitivity Tests/instrumentation , Prospective Studies , Spectrometry, Mass, Matrix-Assisted Laser Desorption-Ionization/methodsABSTRACT
Rapid detection of carbapenemases and accurate reporting of carbapenem MICs is critical for appropriate treatment and infection control. We evaluated the BD Phoenix NMIC-500 panel for detection and classification of carbapenemases and antimicrobial susceptibility testing (AST) for carbapenems. A total of 235 isolates were tested; 47 carbapenemase-producing Enterobacterales, 52 non-carbapenemase-producing carbapenem-resistant Enterobacterales (non-CP-CRE), 136 carbapenem-susceptible Enterobacterales (CSE). The sensitivity of carbapenemase-producing organism (CPO) detection was 97.9%, the specificity was 100% for CSE but 32.7% for non-CP-CREs. All the 35 false-positive cases were non-CP-CREs; 23 out of the 35 were determined as untyped carbapenemase producer (CP), nine were mistyped as class B, and three were as class A. The detection rate/correct classification rate for class A, B, and D carbapenemase was 100%/78.6%, 100%/100%, and 80%/60%, respectively. To supplement the low specificity, it is suggested to report carbapenemase-producer (CP) positive results as "strongly suspicious for carbapenem resistance but carbapenemase production needs to be confirmed" and perform the confirmatory test. The EA and CA for ertapenem, imipenem, and meropenem was 99.1%/99.6%, 89.4%/90.6%, and 95.3%/95.7%. In conclusion, the BD Phoenix CPO detect panel provides advantage in that the carbapenemase test is automated and the results can be obtained within 6 h but the low specificity in CREs needs to be improved. In addition, accurate reporting of meropenem MICs will be helpful for clinicians to choose treatment options.
