Comparative evaluation of diagnostic and analytical performance of DIRUI FUS-200 and MUS-3600 fully automated urine...

The aim was to assess and compare the analytical performance of the FUS-200 and its new upgrade, the MUS-3600 urine autoanalyzers, and to compare performance with manual microscopy. First morning, void urine samples were randomly collected, and suitable samples were analyzed on both autoanalyzers. In addition, concurrent manual microscopic examinations were performed for all suitable urine samples. Carry-over, linearity and imprecision analysis were performed to assess analytical and diagnostic performance of both urine autoanalyzers according to Clinical and Laboratory Standards Institute (CLSI) EP15-A2, and the study was conducted in accordance with CLSI GP16-A3 Urinalysis; Approved Guideline. A total of 518 samples were collected, and of these, 494 (95.4%) were suitable for analysis and were included in the study. Sensitivity and positive likelihood ratio (LR+) values for WBC and squamous epithelium (SqEC) counts for both autoanalyzers were >90% and >10%, respectively. Specificity and LR+ values obtained from MUS-3600 were better compared to the FUS-200 (respectively 92.8 vs. 84.7; 5.5 vs. 10.1). One group agreement between the MUS-3600 and manual microscopy was >90%. Both analyzers displayed comparable performance for RBC and WBC counts, which were moderately correlated with each other. These results show the diagnostic performance of the MUS-3600 and FUS-200 was satisfactory for urine sediment analysis and was compatible with manual microscopy findings. However, random urine samples are sub-optimal for evaluating diagnostic performance, particularly for bacterial, cast, crystal and yeast analysis. Thus, we recommend using more appropriate urine samples for this comparison, such as from patients with suspected urinary tract infections.