Àá½Ã¸¸ ±â´Ù·Á ÁÖ¼¼¿ä. ·ÎµùÁßÀÔ´Ï´Ù.
KMID : 1100720190390050464
Annals of Laboratory Medicine
2019 Volume.39 No. 5 p.464 ~ p.469
Performance of the cobas u 701 Analyzer in Urinary Tract Infection Screening
de la Tabla Victoria Ortiz

Gazquez Gregoria
Infante Ana
Martin Coral
Bunuel Fernando
Gutierrez Felix
Abstract
Background: Negative urine cultures to rule out urinary tract infections (UTI) generate a considerable laboratory workload; thus, a rapid screening test is desirable. We evaluated the performance of a new automated microscopy analyzer, cobas u 701 (Roche Diagnostics International, Rotkreuz, Switzerland) for the screening of UTI, and developed a rule-out strategy to reduce the number of samples requiring culture. We also assessed squamous epithelial cell (SEC) count as a predictor of culture contamination.

Methods: In total, 1,604 urine samples from outpatients were analyzed with cobas u 701 and culture. Bacterial (BAC) and white blood cell (WBC) counts were used for sample interpretation. To determine a useful cut-off point to predict negative cultures, we selected the highest sensitivity and specificity values obtained from ROC curves. Diagnostic accuracy by age and gender was evaluated.

Results: Urine culture showed growth of ¡Ã104 colony forming units (CFU)/mL in 256 samples (16.0%). The highest sensitivity (91.8%) and specificity (68.4%) were obtained for cut-off points of 119 BAC/¥ìL and 22 WBC/¥ìL. The combination of BAC and WBC improved the performance of the rule-out strategy with a low rate of false-negative results (1.5%) and a high negative predictive value (NPV, 97.3%). Fifty-seven percent of the samples would not have required culture. SEC count was a poor predictor of culture contamination.

Conclusions: cobas u 701 can substantially reduce the number of urine samples requiring culture, with a low false-negative rate and a high NPV.
KEYWORD
Automated microscopy analyzer, cobas u 701, Screening urine samples, Urinary tract infection, Urine culture
FullTexts / Linksout information
 
Listed journal information
SCI(E) MEDLINE ÇмúÁøÈïÀç´Ü(KCI) KoreaMed