Research article

Comparison of the MicroScan, VITEK 2, and Crystal GP with 16S rRNA sequencing and MicroSeq 500 v2.0 analysis for coagulase-negative Staphylococci

Miyoung Kim^1* , Se Ran Heo² , Soon Hee Choi² , Hyelin Kwon² , Jeong Su Park¹ , Moon-Woo Seong³ , Do-Hoon Lee³ , Kyoung Un Park^1,2,4 , Junghan Song^1,2,4 and Eui-Chong Kim^1,4

¹  Department of Laboratory Medicine, Seoul National University Hospital, 101 Daehang-no, Jongno-gu, Seoul, South Korea

²  Department of Laboratory Medicine, Seoul National University Bundang Hospital, 300 Gumi-dong, Bundang-gu, Seongnam-si, Gyeonggi-do, South Korea

³  Department of Laboratory Medicine, National Cancer Center, 809 Madu-1-dong, Ilsandong-gu, Goyang, Gyeonggi-do, South Korea

⁴  Department of Laboratory Medicine, Seoul National University College of Medicine, 101 Daehang-no, Jongno-gu, Seoul, South Korea

author email corresponding author email* Contributed equally

BMC Microbiology 2008, 8:233doi:10.1186/1471-2180-8-233

Published:	23 December 2008

Abstract

Background

Three phenotypic identification systems (MicroScan, VITEK 2, and Crystal GP) were evaluated for their accuracy to identify coagulase-negative staphylococci (CNS). A total of 120 clinical isolates confirmed to be CNS via 16S rRNA sequencing and analysis with the MicroSeq 500 v2.0 database were assessed.

Results

The MicroScan, VITEK 2, and Crystal GP systems correctly identified 82.5%, 87.5%, and 67.5% of the isolates, respectively. Misidentification was the main problem in MicroScan (10.8%) and Crystal GP (23.3%) systems, whereas the main problem of VITEK 2 was low-level discrimination (7.5%).

Conclusion

None of the 3 phenotypic systems tested could accurately and reliably identify CNS at the species level. Further verifications such as biochemical testing or 16S rRNA sequencing together with analysis using a comparable database might be helpful in this regard.