Performance Evaluation of an Automated Microbiological Stainer, KS-S100

Lee, Sun Min; Lee, Hyun Ji; Chang, Chulhun L

Abstract

BACKGROUND
There have been few evaluations of the performance of automatic stainers used instead of manual staining in clinical laboratories. We evaluated the performance of an automated stainer, KS-S100 (KS Co. Ltd., Korea), in Gram and acid-fast fluorescence staining of clinical specimens.
METHODS
Repeatability was evaluated by comparing the results of Gram stains performed 10 times over 5 days using two standard strains and duplicated acid-fast bacilli (AFB) fluorescence stains using 40 samples. Comparisons to manual staining and to another automatic stainer, AT-2000 (Dagatron, Korea), were performed using 40 remnant respiratory specimens or liquid AFB culture media. The results were interpreted by two experienced technicians, and total test time was measured during the staining of 1, 10, and 20 slides.
RESULTS
All 10 pairs of Gram staining results obtained with KS-S100 were identical, and the concordance rate of the duplicate AFB fluorescence stains was 92.5%. Among the results for 20 culture-positive liquid media samples, three AFB fluorescent staining results were not concordant. The weighted kappa values between KS-S100 and manual staining for Gram and AFB fluorescence staining were both 0.94, with concordance within the ±1-grade range. For the staining of a single slide, manual staining was faster. KS-S100 was more efficient than AT-2000 at staining 20 slides.
CONCLUSIONS
The automated stainer KS-S100 showed reproducible results comparable to those of manual staining in two major microbiological staining methods, Gram and AFB fluorescence staining, used in clinical applications. Performance evaluations should be conducted before implementation of the automated stainer in microbiological staining.