Evaluation of the updated MS-2 Bacterial Identification system in comparison with the API 20E system

Autor: J R DiPersio, R D Vannest, J W Dyke
Rok vydání: 1983
Předmět:
Zdroj: Journal of Clinical Microbiology. 18:128-135
ISSN: 1098-660X
0095-1137
DOI: 10.1128/jcm.18.1.128-135.1983
Popis: The recently updated MS-2 Bacterial Identification system software (Abbott Laboratories, Diagnostic Division, Irving, Tex.) was compared with the original MS-2 Bacterial Identification software and the API 20E, using 968 strains of Enterobacteriaceae. The updated MS-2 software correctly identified 94.4% of the isolates tested. API 20E and the original MS-2 software correctly identified 91 and 85.3% of the strains, respectively. MS-2 responses were considered to be equivocal (needing additional tests for verification) if the percent likelihood values were less than 80%. The percentage of equivocal responses was reduced from 6.5% with the original software to 2.2% with the updated software, and the percentage of incorrect identifications was reduced from 8.2 to 3.4% with the original and updated software, respectively. Organisms belonging to 25 taxonomic groups were tested. Direct comparison of the two MS-2 programs showed that the updated MS-2 software increased the identification accuracy of Salmonella spp., Enterobacter cloacae, Providencia stuartii, Escherichia coli, Shigella spp., Klebsiella pneumoniae, Serratia marcescens, Proteus mirabilis, and Acinetobacter calcoaceticus. A decrease in accuracy was seen with Citrobacter freundii, Hafnia alvei, Enterobacter agglomerans, and Yersinia pseudotuberculosis when the updated software was used. The remaining 12 taxonomic groups were not affected by the software changes. The updated MS-2 software appears to significantly improve the identification accuracy of the MS-2 Bacterial Identification system.
Databáze: OpenAIRE