Article

Original Article

Korean J Lab Med 2007; 27(3): 162-168

Published online June 1, 2007 https://doi.org/10.3343/kjlm.2007.27.3.162

Copyright © Korean Society for Laboratory Medicine.

Evaluation of the Abbott Cell-Dyn Sapphire Hematology Analyzer

Younhee Park, M.D., Jaewoo Song, M.D., Sungwook Song, M.D., Kyung Soon Song, M.D., Mee Suk Ahn, M.T., Mi-Sook Yang, M.T., Il Kim, M.T., and Jong Rak Choi, M.D.

Department of Laboratory Medicine, Yonsei University College of Medicine, Seoul, Korea

Correspondence to: 최 종 락
우 120-752 서울시 서대문구 신촌동 134 연세대학교 의과대학 진단검사의학교실
전화: 02-2228-2445, Fax : 02-364-1583
E-mail: cjr0606@yumc.yonsei.ac.kr

Received: January 4, 2007; Revised: April 5, 2007; Accepted: April 25, 2007

Abstract

Background : The performance of Cell-Dyn Sapphire (Abbott Diagnostic, USA) was compared to the Bayer Advia 2120 (Bayer Diagnostics, USA), Sysmex XE-2100 (Sysmex Corporation, Japan), and reference microscopy.
Methods : Three hundred samples for routine CBC and WBC differentials were randomly chosen for a comparison analysis. The Cell-Dyn Sapphire system was evaluated according to the linearity, imprecision, inter-instrument correlations, and white blood cell differential.
Results : The CBC parameters (WBC, RBC, hemoglobin and platelet) showed a significant linearity with correlation coefficients greater than 0.99 (P<0.0001). Coefficients of variation (CV) for withinrun and differential count of WBC were less than 5% except for Total CV for monocytes, eosinophils, and basophils and within-run CV for low valued eosinophils. The correlation coefficients with manual count were lower in monocytes, eosinophils, and basophils than in neutrophils and lymphocytes. The correlation with other hematology anlayzers was significant exclusive of basophils.
Conclusions : These results demonstrate that the Cell-Dyn Sapphire has a good linearity, an acceptable reproducibility, a minimal carryover, and a comparable performance with the sysmex XE- 2100 and Advia 2120.

Keywords: Cell-Dyn Sapphire, Performance, Correlation