Age-Specific Cutoffs of the Sysmex UF-1000i Automated Urine Analyzer for Rapid Screening of Urinary Tract Infections in Outpatients

Abstract

We investigated the usefulness of age-specific cutoffs for screening of urinary tract infections (UTIs) in Korean outpatients, using the automated urine analyzer UF-1000i (Sysmex, Kobe, Japan). We retrospectively reviewed outpatient medical records. Urine samples of 7,443 outpatients from January 2010 to December 2017 were analyzed using urine culture and UF-1000i. ROC curves were calculated for each UF-1000i parameter based on the culture results. There were 1,398 culture positive samples, 5,774 culture negative samples, and 271 contaminated samples. UF-1000i had an area under the curve of ≥0.9 in outpatients >15 years. The appropriate cutoffs, which are the sum of bacterial (B-A-C) and white blood cell (WBC) counts, were 297.10/μL (15-24 years), 395.65/μL (25-44 years), 135.65/μL (45-64 years), 67.95/μL (65-74 years), and 96.5/μL (≥75 years). B-A-C and WBC counts differed among the three most frequently identified bacteria (Escherichia coli, Klebsiella pneumoniae, and Enterococcus faecalis). The UF-1000i system is useful for applying age-specific cutoffs to screen for UTIs, thereby preventing antibiotic abuse and reducing antibiotic resistance. Future studies can explore how its B-A-C and WBC counts can facilitate selection of empirical antibiotics by distinguishing between gram-positive and gram-negative bacteria.

Keywords: Age-specific cutoff; Outpatients; UF-1000i; Urinary tract infection.

Fig. 1. UF-1000i (Sysmex, Kobe, Japan) ROC curves by age. The sum of B-A-C and WBC graphs showed the highest AUC, so the cutoff of this graph is the most suitable for screening. The optimal ROC curve values (AUC, cutoff, sensitivity and specificity) for each age group were as follows: <4 years (0.790, 74.75/µL, 77.2% and 65.7%), 4–14 years (0.759, 42.9/µL, 81.0% and 65.1%), 15–24 years (0.908, 297.10/µL, 95.1% and 71.7%), 25–44 years (0.902, 395.65/µL, 90.2% and 74.9%), 45–64 years (0.908, 135.65/µL, 90.2% and 72%), 65–74 years (0.916, 67.95/µL, 94.1% and 67.8%), and ≥75 years (0.912, 96.5/µL, 92.8% and 71.1%).

Abbreviations: WBC, white blood cell; B-A-C, bacteria; AUC, area under the curve.

Fig. 2. Graph of median WBC vs B-A-C counts determined by UF-1000i (Sysmex, Kobe, Japan) in outpatients with a urinary tract infection confirmed to be caused by Escherichia coli (937 samples), Klebsiella pneumoniae (86 samples), and Enterococcus faecalis (72 samples). The distribution range (95% confidence interval, minimum, maximum, interquartile range) is as follows: E. coli B-A-C (11,765.0–14,186.4/µL, 2.0/µL, 99,086.0/µL, 17,826.1/µL), E. coli WBC (1,073.3–1,464.4/µL, 0.0/µL, 32,723.0/µL, 1,007.4/µL); E. faecalis B-A-C (1,836.6–5,964.5/µL, 0.0/µL, 53,597.0/µL, 3,016.4/µL), E. faecalis WBC (285.4–847.1/µL, 1.0/µL, 6,779.0/µL, 332.6/µL); K. pneumoniae B-A-C (10,532.2–18,205.2/µL, 2.0/µL, 80,533.0/µL, 20,834.0/µL), and K. pneumoniae WBC (582.3–2,037.0/µL, 1.0/µL, 31,392.0/µL, 1,081.3/µL).

Abbreviations: WBC, white blood cell; B-A-C, bacteria.