BACKGROUND: Cultures for urinary tract infections (UTI) constitute a large workload in the clinical microbiology laboratory, although up to 80% are usually negative. Several automated methods are available to screen urines for UTI, one being the flow cytometry-based Sysmex((R)) UF-100. METHODS: The performance of the UF-100 was evaluated over a 16-month period using urine culture as the reference method. RESULTS: During this period, a total of 5356 urine samples were studied (469 children; 3229 women and 1658 men), of which 706 were culture positive (593 grew Gram negative bacilli). Receiver operating characteristics (ROC) curve analysis showed an area under the curve (AUC) of 0.83 for leukocytes and 0.85 for bacterial count. Applying cut-off values reported in the literature gave sensitivities ranging from 75% to 90%, resulting in 73-174 false negatives (FN). Using a logical combination (leukocytes >or=15x10(6)/L OR bacteria >or=500x10(6)/L) gave a sensitivity of 98%. However, the specificity dropped to 25%, resulting in 15 FN. CONCLUSIONS: Screening urine samples for UTI detects a large number of culture positive samples. However, the rather large number of FN observed precludes the use of the UF-100 as a routine screening method to exclude urine samples from culture.
BACKGROUND: Cultures for urinary tract infections (UTI) constitute a large workload in the clinical microbiology laboratory, although up to 80% are usually negative. Several automated methods are available to screen urines for UTI, one being the flow cytometry-based Sysmex((R)) UF-100. METHODS: The performance of the UF-100 was evaluated over a 16-month period using urine culture as the reference method. RESULTS: During this period, a total of 5356 urine samples were studied (469 children; 3229 women and 1658 men), of which 706 were culture positive (593 grew Gram negative bacilli). Receiver operating characteristics (ROC) curve analysis showed an area under the curve (AUC) of 0.83 for leukocytes and 0.85 for bacterial count. Applying cut-off values reported in the literature gave sensitivities ranging from 75% to 90%, resulting in 73-174 false negatives (FN). Using a logical combination (leukocytes >or=15x10(6)/L OR bacteria >or=500x10(6)/L) gave a sensitivity of 98%. However, the specificity dropped to 25%, resulting in 15 FN. CONCLUSIONS: Screening urine samples for UTI detects a large number of culture positive samples. However, the rather large number of FN observed precludes the use of the UF-100 as a routine screening method to exclude urine samples from culture.
Authors: Gernot Bonkat; Olivier Braissant; Malte Rieken; Anna Solokhina; Andreas F Widmer; Reno Frei; Andre van der Merwe; Stephen Wyler; Thomas C Gasser; Alexander Bachmann Journal: World J Urol Date: 2012-07-26 Impact factor: 4.226
Authors: Hong Phuoc Duong; Karl Martin Wissing; Nathalie Tram; Georges Mascart; Philippe Lepage; Khalid Ismaili Journal: J Clin Microbiol Date: 2016-09-28 Impact factor: 5.948