Clin Chem Lab Med 2019; aop

Matthijs Oyaert*, Marijn Speeckaert, Jerina Boelens and Joris R. Delanghe

Renal tubular epithelial cells add value in the

diagnosis of upper urinary tract pathology
https://doi.org/10.1515/cclm-2019-1068
Received October 15, 2019; accepted November 23, 2019
Introduction
Abstract Differentiation between lower and upper urinary tract
infections (UTI) is of paramount clinical importance, for
Background: Diagnosis of upper urinary tract infections
which clinicians often have to rely on clinical symptoms.
(UTI) is challenging. We evaluated the analytical and
However, in adults and children, clinical criteria do not
diagnostic performance characteristics of renal tubular
allow a 100% discrimination between both pathologies
epithelial cells (RTECs) and transitional epithelial cells
[1, 2]. Also, current laboratory methods do not always
(TECs) on the Sysmex UF-5000 urine sediment analyzer.
allow a clear-cut distinction between upper UTI and
Methods: Urinary samples from 506 patients presenting
lower UTI (LUTI). Microbiological culture is not helpful in
with symptoms of a UTI were collected. Only samples for
this respect as it does not provide information regarding
which a urinary culture was available were included. Ana-
the localization of the infection. Therefore, differentiat-
lytical (imprecision, accuracy, stability and correlation
ing LUTI from upper UTI remains a diagnostic challenge
with manual microscopy) and diagnostic performance
in modern urinalysis. Urine sediment analysis can be
(sensitivity and specificity) were evaluated.
helpful: the presence of leukocyte casts in the urine sedi-
Results: The Sysmex UF-5000 demonstrated a good ana-
ment is suggestive of an upper UTI, but its diagnostic
lytical performance. Depending on the storage time,
sensitivity remains low [3]. Specific urinary proteins (e.g.
storage conditions (2–8 °C or 20–25 °C) and urinary pH,
α1-microglobulin) are established markers of tubular func-
RTECs and TECs were stable in urine for at least 4 h. Using
tion, which have been shown to be helpful to some extent
Passing-Bablok and Bland-Altman analysis, an acceptable
[4]. However, these markers may not be available in every
agreement was observed between the manual and auto-
routine clinical laboratory.
mated methods. Compared to TECs, RTECs demonstrated
The Sysmex UF-5000 (Sysmex Corporation, Kobe,
an acceptable diagnostic performance for the diagnosis of
Japan) represents the newest generation of urinary par-
upper UTI.
ticle analyzers, designed for automated urine sediment
Conclusions: While TECs do not seem to serve as a helpful
and body fluid analysis. This third-generation fluores-
marker, increased urinary levels of RTECs add value in the
cence flow cytometry analyzer for urinalysis offers the
diagnosis of upper UTI and may be helpful in the discrimi-
enhanced possibility to accurately count and differen-
nation between upper and lower UTIs.
tiate a broad variety of urinary cells. The instrument is
Keywords: automated urinary sediment analyzer; renal able to differentiate epithelial cells into squamous (SEC),
tubular epithelial cells; urinary tract infections. transitional (TECs) and renal tubular epithelial cells
(RTECs). RTECs along with TECs line the urinary tract,
originate from the proximal or distal urine segments and
therefore have a diagnostic potential [5], whereas high
SEC counts might point toward contaminated urine and
add only little diagnostic value. RTECs are specific for
*Corresponding author: Matthijs Oyaert, Department of
the presence of tubular damage and their detection in
Laboratory Medicine, Clinical Chemistry, Ghent University Hospital,
C. Heymanslaan 10, 9000 Ghent, Belgium, Phone: 09/332 35 12, a urinary sample could allow early recognition of renal
Fax: 09/332 49 85, E-mail: matthijs.oyaert@uzgent.be damage when other kidney function parameters are
Marijn Speeckaert: Department of Nephrology, Ghent University still normal [5, 6]. On the contrary, TECs are markers of
Hospital, Ghent, Belgium; and Research Foundation Flanders, ureteral damage related to the presence of an infection,
Brussels, Brussels, Belgium
kidney stones or invasive procedures [5]. As these types
Jerina Boelens: Department of Laboratory Medicine, Microbiology,
Ghent University Hospital, Ghent, Belgium
of cells may differ in morphological presentation, they
Joris R. Delanghe: Department of Laboratory Medicine, Clinical are a diagnostic challenge in modern clinical urinalysis
Chemistry, Ghent University Hospital, Ghent, Belgium laboratories [6].

Brought to you by | provisional account

Unauthenticated
Download Date | 12/26/19 4:45 AM
2      Oyaert et al.: Added value of RTECs in UTI
The aim of the present study was to explore the analyt-
ical performance and diagnostic value of RTECs and TECs
in a cohort of well-documented patients presenting with
symptoms of a UTI. In particular, we aimed to determine
the value of RTECs and TECs for the diagnosis of upper
UTI. Furthermore, we wanted to compare the diagnostic
performance of RTECs and TECs with that of some estab-
lished upper UTI markers (α1-microglobulin, pathological
casts, γ-glutamyl transpeptidase [GGT] activity).
Materials and methods
Patient population
The total population consisted of 506 patients (median age: 58 years
[range: 18–97 years]; males: n = 237; females: n = 269) with a suspicion
of a UTI and for whom a midstream urine sample for test strip and
sediment analysis was sent to the clinical laboratory of the Ghent
University Hospital. Based on the final diagnosis made by the clini-
cian, patients were retrospectively categorized into upper UTI (pyelo-
nephritis [n = 73], urosepsis [n = 32] and renal cystic infection [n = 2]),
LUTI (cystitis [n = 35] and prostatitis [n = 12]) and 352 patients without
UTI were categorized as disease controls. The main reasons for per-
forming urine sediment analysis were suspected clinical symptoms
of a UTI (pollakisuria, dysuria, abdominal or flank pain, fever chills,
etc.). Only patients with a proven microbiological UTI were included.
A summary of the patient characteristics along with the RTEC and TEC
count is presented in Table 1. The exclusion criteria were an unclear
diagnosis, the inability to collect enough urine and an age <18 years.
Urine samples were collected by the aspiration technique using
Sarstedt Monovette urinary collection tubes (Sarstedt, Numbrecht,
Germany) and were processed within 2–4 h after arrival in the labo-
ratory. Test strip analysis on a Sysmex UC-3500 analyzer (Sysmex,
Kobe, Japan) [7] was carried out before flow cytometry analysis on the
Sysmex UF-5000 (Sysmex, Kobe, Japan). After microscopic analysis,
urinary chemistry tests (see further) were determined.
The study was performed with the full respect for individuals’
rights to confidentiality and in accordance with procedures super-
vised by local authorities responsible for ethical research (Belgian
registration number of ethical approval: B670201837110).
Table 1: Overview of the patient characteristics and RTEC and TEC counts (/μL) of the different patient groups.
Final diagnosis n Men/women, n/n
Upper urinary tract infection
 Urosepsis 32
 Pyelonephritis 73
Lower urinary tract infection
 Cystitis 35
 Prostatitis 12
Non-urological/non-nephrological patients 354 181/173
a
Age is presented as years (range). bThe value within brackets indicates the interquartile range (IQR). RTECs: renal tubular epithelial cells;
TECs: transitional epithelial cells.
Urine particle analysis using UF-5000 and microscopic
analysis
Measurement of RTECs and TECs was performed using the Sysmex
UF-5000 (Sysmex, Kobe, Japan). The UF-5000 is able to recognize,
count and classify cells by analyzing forward scatter light, side scatter
light, side fluorescent light and the depolarized side scattered light of
stained particles. Depolarized side scattered light was introduced to
improve the sensitivity of crystals and to better discriminate red blood
cells (RBCs) from crystals. The principle is based on a 488-nm blue
laser flow cytometry. The UF-5000 measures urinary conductivity and
categorizes the particles based on their size, intracellular structure and
staining characteristics. The signals are displayed in scattergrams and
histograms and the results are given as counts per μL as well as counts
per high power field (HPF) [8]. The UF-5000 automatically detects and
counts RBCs, non-lysed RBCs, white blood cells (WBCs), WBC clumps,
bacteria, yeast-like cells, crystals, RTECs, TECs, sperm cells and (hya-
line and pathological) casts. Urinary particles that cannot be classified
into one of the former categories are counted as “other cells”.
Following the analysis on the automated urine sediment ana-
lyzer, microscopic identification and counting were performed by
phase-contrast microscopy on uncentrifuged urine samples using
disposable Uriglass counting chambers (A. Menarini Diagnostics,
Florence, Italy) by two expert laboratory technicians blinded to the
results of the Sysmex UF-5000. Each 1-μL chamber is divided into 10
large squares. Each square corresponds to a volume of 0.1 μL and is
subdivided into 16 small squares. RTECs and TECs were counted in 5
x 16 squares, and the results are expressed as the number of particles
per μL of urine.
Biochemical investigations
Following the urinary sediment analysis, biochemical investigations
were also performed. Total protein was determined on all samples by
a pyrogallol red-molybdate method (Instruchemie BV, Delfzijl, The
Netherlands) [9]. Urinary α1-microglobulin was assayed immunotur-
bidimetrically using Roche reagents (Mannheim, Germany). Urinary
GGT activity was measured according to the method of Szasz [10]
using commercial reagents from Abbott. Urinary creatinine and uri-
nary urea were also determined using Abbott commercial reagents.
All analyses were performed on the Abbott Alinity C analyzer (Abbott
Diagnostics, Wiesbaden, Germany).
Age, years (range)a RTECs, /μLb TECs, /μLb
19/13 71 (27–93) 5.5 (3.3–10.6) 0.6 (0.1–1.3)
19/54 59 (18–97) 9.8 (6.3–13.6) 0.6 (0.3–1.3)
6/29 58 (18–95) 1.4 (0.3–4.3) 0.3 (0.1–0.9)
12/0 69 (40–85) 1.5 (0.7–2.1) 0.7 (0.0–2.1)
55 (18–94) 1.3 (0.4–2.6) 0.1 (0.0–0.3)
Brought to you by | provisional account
Unauthenticated
Download Date | 12/26/19 4:45 AM
 Oyaert et al.: Added value of RTECs in UTI      3

Microbiological culture considered statistically significant. Receiver operating characteris-

tic (ROC) curve analysis and areas under the ROC curve (AUC) were
calculated to evaluate the diagnostic accuracy of the different para-
Biplates with Tryptic Soy Agar (TSA) + 5% sheep blood and MacCo-
meters measured. Sensitivities and specificities were determined
nkey agar (Becton Dickinson, Cockeysville, MD, USA) were used for
with the clinical diagnosis. Optimal cut-offs were defined as cut-off
chairside inoculation of 1 μL of freshly voided urine, and incubated
values with the highest sum of sensitivity and specificity, based on
upon arrival at 35 °C ambient atmosphere overnight and for another
ROC curve analysis.
24 h. Interpretation is done after 24 and 48 h of incubation. All sus-
Statistical analysis was performed using the MedCalc software
pect colony types were identified using matrix-assisted laser desorp-
(version 15.6.1., Mariakerke, Belgium).
tion/ionization time-of-flight (MALDI-TOF) mass spectrometry (MS)
(Microflex LT, Bruker, Billerica, MA, USA). Criteria for positivity of a
urine culture were in accordance with the European Federation for
Urinalysis guidelines [11].

Results
Statistical analysis

Imprecision of RTECs and TECs on the UF-5000 was assessed using

three patient samples (low, medium and high level). Intra-run impre-
cision was determined by measuring each sample 10 times in the
same run. As the stability of RTECs is limited, only within-run impre-
cision was assessed. Bias was determined by calculating the % dif-
ference between the mean RTEC value of the intra-run imprecision
experiment and the target value. This target value was determined
by manually counting the number of RTECs and TECs. The total error
(TE) was calculated using the formula: TE = bias + 1.65 * (intra-run
imprecision [%]).
The stability of RTECs and TECs at two different storage condi-
tions (room temperature [20–25 °C, 24  h] and refrigerator [2–8 °C,
24  h]) on 10 different samples was determined (RTEC: mean con-
centration: 21.7/μL, range: 9.3–38.9/μL; TEC: mean concentration:
17.4/μL, range: 5.8–55.4/μL). Samples were determined 2, 4, 6, 8, 10,
12, 16 and 24 h after the initial measurement. Allowable TE internal
laboratory criteria for SECs (i.e. 30%) were used as criterion for sig-
nificant deviation with respect to the initial value. RTEC and TEC con-
centrations were determined using the UF-5000.
For analytical method comparison, Bland-Altman plots, Pass-
ing-Bablok regression analysis and Spearman’s rank correlation coef-
ficients were determined for both assays.
Differences in RTEC count between (sub)groups were assessed
for significance using the Mann-Whitney U test. p-Values <0.05 were
Analytical performance
Imprecision, bias, total error and stability
Within-run imprecision ranged from 6.0 to 10.6% and 5.0
to 12.9% for RTECs and TECs, respectively. The bias rela-
tive to the manual count ranged from −2.6 to 6.5% and
−5.9 to 15.2% for RTECs and TECs, respectively. The calcu-
lated TE was acceptable and ranged from 12.5 to 24.0% for
RTECs and 9.8 to 36.5% for TECs. A summary is presented
in Table 2.
We determined the stability of RTECs and TECs at
both room temperature (20–25 °C) and in the refrigerator
(2–8 °C) and found that the stability of both parameters
depended on the pH of the urinary sample. RTEC and TEC
counts in samples with a pH ≥7.5 were stable for maximum
4 h at both room and refrigerator temperature. RTEC and
TEC counts in samples with a pH <7.5 were stable for 8 h
at room temperature. In the refrigerator, TEC counts were
stable for 12  h and RTECs for 16  h. The results are pre-
sented in Figure 1.

Table 2: Analytical performance (imprecision, bias and total error) of RTECs and TECs on the Sysmex UF-5000.

Imprecision (within-run) Bias Total error

Mean, /μL SD, /μL CV, % Target % Deviation Results, % Criteria, %a

RTECs
 Low 5.6 0.6 10.6 6.0 6.5 24.0 50
 Medium 16.4 1.4 8.7 16.0 −2.4 16.8 30
 High 39.0 2.3 6.0 38.0 −2.6 12.5 30
TECs
 Low 2.5 0.3 12.9 3.0 15.2 36.5 50
 Medium 20.6 1.4 7.1 19.0 −5.9 17.7 30
 High 49.8 2.5 5.0 49.0 −1.6 9.8 30

SD, standard deviation; CV, coefficient of variation; RTECs, renal tubular epithelial cells; TECs, transitional epithelial cells. aSource criteria:
internal laboratory criteria.

Brought to you by | provisional account

Unauthenticated
Download Date | 12/26/19 4:45 AM
 4      Oyaert et al.: Added value of RTECs in UTI

RTEC TEC
20 pH: 5.5–7.0; 2–8 °C 20 pH: 5.5–7.0; 2–8 °C
pH: 5.5–7.0; 20–25 °C pH: 5.5–7.0; 20–25 °C
pH: 7.5–9.0; 2–8 °C pH: 7.5–9.0; 2–8 °C

% Difference to the initial value

% Difference to the initial value

0 pH: 7.5–9.0; 20–25 °C 0 pH: 7.5–9.0; 20–25 °C

–20 –20

–40 –40

–60 –60

–80 –80

0 4 8 12 16 20 24 0 4 8 12 16 20 24
Time, h after initial measurement Time, h after initial measurement

Figure 1: Overview of the stability of RTEC and TEC counts.

Mean difference ranges (%) of renal tubular epithelial cell (RTEC) (A) and tubular epithelial cell (TEC) (B) counts stored for 24 h at 2–8 °C and
at 20–25 °C. The mean difference to the initial value for the different RTEC and TEC counts is presented. The level of acceptance (30%) is
shown as a dashed line. The error bars represent the 95% confidence interval.

Method comparison proportional difference of approximately 4% (slope:

0.96 [95% CI: 0.91–1.00]; intercept: −0.42 [95% CI: −0.50
Method comparison was performed on 504  samples. to −0.36]; R2: 0.972 [95% CI: 0.966–0.976]) (Figure 2A1).
Passing-Bablok regression analysis between manual Using Bland-Altman analysis, a mean difference of −10.1
microscopic and automated RTEC counts revealed a RTEC cells/μL (95% CI: −16.0 to −4.2/μL) was obtained

A1 B1
(RTEC manual count - RTEC UF-5000, /µL)/average %

80 200
Renal tubular epithelial cells, manual count, /µL

70 150

60 100 +95% LOA

84.5
50 50

40 0 Mean
–10.1
30 –50

–100 –95% LOA

20
–104.7
10 –150

0 –200
0 20 40 60 80 0 20 40 60 80
Renal tubular epithelial cells, UF-5000, /µL Mean of RTEC manual count and RTEC UF-5000, /µL

A2 B2
(TEC manual count - TEC UF-5000, /µL)/average %

25 80
Transitional epithelial cells, UF-5000, /µL

60
+95% LOA
20
40 46.5

20
15
0 Mean
–2.3
10 –20

–40
–95% LOA
5
–51.1
–60

0 –80
0 5 10 15 20 25 0 5 10 15 20 25
Transitional epithelial cells, manual count, /µL Mean of TEC manual count TEC UF-5000, /µL

Figure 2: Method comparison of renal tubular epithelial cell (RTEC) and transitional epithelial cell (TEC) counts.
Method comparison of RTEC and TEC counts with Passing-Bablok (A) and Bland-Altman (B) analysis of manual microscopy vs. Sysmex UF-5000.

Brought to you by | provisional account

Unauthenticated
Download Date | 12/26/19 4:45 AM

with limits of agreement (LOA) of −104.7 cells/μL (95%
CI: −114.8 to −94.6) and 84.5 cells/μL (95% CI: 74.4–94.6)
(Figure 2B1).
For TEC, no significant proportional or systemic dif-
ferences were obtained between manual and automated
counts (slope: 1.00 [95% CI: 0.90–1.30]; intercept: 0.30
[95% CI: 0.30–0.30]; R2: 0.980 [95% CI: 0.977–0.984])
(Figure 2A2). Bland-Altman analysis revealed a mean dif-
ference of −2.3 TEC cells/μL (95% CI: −8.1 to 3.6) with LOA
of −51.1 (95% CI: −61.1 to −41.0) and 46.5 (95% CI: 36.5–
56.6) (Figure 2B2).
Correlation study
Multiple linear mixed model analysis revealed signifi-
cant correlations between RTECs and log-transformed
GGT (β = 3.1865, standard error [SE] = 0.648, p < 0.0001),
completed by log-transformed bacteria (β = 1.1091,
SE = 0.207, p < 0.0001) and α1-microglobulin (β = 0.0269,
SE = 0.007, p < 0.0001). The r2 value of the final model
was poor (0.167). No significant correlations were
obtained between TECs and the other measured urinary
parameters.
Diagnostic performance
Among the study population, the prevalence of upper
UTI was 21.1% (107/506). Quantitative data distribu-
tion showed that RTEC counts were significantly higher
(p < 0.05) in patients with a final diagnosis of urosepsis
and pyelonephritis compared to counts in patients diag-
nosed with cystitis and prostatitis (see Figure 3). On the
p < 0.01
p < 0.01
70
60
50
RTEC, /µL
40
30
20
10
0
Cystitis Non-urological/ Prostatitis
nephrological patients
Figure 3: Box-whisker plots representing renal tubular epithelial
cell (RTEC) counts among the different clinical conditions.
Values falling outside the box-whisker plot are outliers.
Oyaert et al.: Added value of RTECs in UTI      5
contrary, no significant difference in TECs between the
different subgroups was observed.
In our population, the diagnostic accuracy
(expressed as AUC) for the parameters α1-microglobulin,
bacteria, GGT, WBCs, RTECs, TECs and pathological casts
on the Sysmex UF-5000  was found to be 0.735 (95% CI:
0.694–0.773), 0.787 (95% CI: 0.748–0.821), 0.586 (95%
CI: 0.541–0.629), 0.816 (95% CI: 0.779–0.849), 0.923 (95%
CI: 0.897–0.945), 0.790 (95% CI: 0.752–0.825) and 0.751
(95% CI: 0.711–0.788), respectively (see Figure  4). The
AUCs were statistically significantly different (p < 0.05)
between RTECs and α1-microglobulin, bacteria, GGT,
pathological casts, TECs and WBCs. Statistically signifi-
cant differences in AUCs were also observed between
α1-microglobulin and WBCs and pathological casts and
WBCs (p-value <0.01). A summary of the diagnostic per-
formance of these parameters measured on the Sysmex
UF-5000 in our study population is presented in Table 3.
The same AUCs were demonstrated for RTECs and TECs
counted by manual microscopy (data not shown).
The median conductivity and urinary creatinine con-
centrations of the samples in the study were 13.2 mS/cm
(interquartile range [IQR]: 9.2–18.6 mS/cm) and 91.0 mg/
dL (IQR: 48.7–140.9  mg/dL), respectively. The obtained
diagnostic performance (expressed as AUC) for RTECs
corrected for conductivity and RTECs corrected for
urinary creatinine (0.900 [95% CI: 0.867–0.927] and 0.905
p < 0.01
Pyelonephritis Urosepsis
Figure 4: Presentation of the receiver operating characteristic (ROC)
curve comparison for the parameters α1-microglobulin, urinary
bacterial count, white blood cells (WBCs), renal tubular epithelial
cells (RTECs), tubular epithelial cells (TECs) and pathological casts.
Brought to you by | provisional account
Unauthenticated
Download Date | 12/26/19 4:45 AM
 6      Oyaert et al.: Added value of RTECs in UTI

[95% CI: 0.872–0.931], respectively) was not significantly

28.2/μL
WBCs

  0.816a (0.779–0.849)
0.0195
90.7 (83.5–95.4)
67.3 (62.5–71.9

Indicates statistically different AUC compared to RTECs (p < 0.05); bIndicates statistically different AUC with WBC (p < 0.05); AUC, area under the curve; ROC, receiver operating characteristic;
different compared to the AUC of RTECs (AUC: 0.911
[95% CI: 0.880–0.936]), demonstrating that there is no
added value when the results are corrected for dilution
parameters.
TECs  

0.2/μL  

 
 
 
  0.790a (0.752–0.825)
0.0247
74.53 (65.1–82.5)
72.29 (67.6–76.6) Discussion
While SECs and TECs offer little useful information, we
have demonstrated for the first time that RTECs can be
regarded as an interesting supplemental parameter in
RTECs  

3.1/μL  

 
 
 

the discrimination between upper UTI and LUTI. In line

  0.923 (0.897–0.945)
0.0139
93.5 (87.0–97.3)
82.2 (78.0–85.8)

with previous studies [8, 12], our study showed that the
Sysmex UF-5000 urine sediment analyzer is able to count
the RTECs and TECs with an acceptable analytical perfor-
mance. These parameters were not available on earlier
generations of urinary flow cytometers [3, 13, 14], catego-
Pathological casts  

0.13/μL  

 
 
 

rizing RTECs as “small round cells”. Compared to manual

Table 3: Sensitivities and specificities of the different parameters at the cut-off with maximal sensitivity and specificity.

  0.751a,b (0.711–0.788)
0.0269
62.62 (52.7–71.8)
84.67 (80.8–88.1)

microscopy, a sensitivity and a specificity of 95% and

RTECs, renal tubular epithelial cells; SE, standard error; TECs, transitional epithelial cells; WBCs, white blood cells.

75% for RTECs and 71% and 94% for TECs at a cut-off of
3 and 5 cells/μL were reported, respectively [8]. Due to
the improved categorization of urinary particles of the
UF-5000, the discriminatory power of this new-generation
urinary flow cytometer with respect to localization of UTI
GGT  

50.2 U/L  

 
 
 
  0.586a,b (0.541–0.629)
0.0336
56.4 (44.2–64.4)
63.6 (58.6–68.4)

has much improved.

Multiple regression analysis revealed that RTEC values
strongly correlated with bacterial and WBC counts, which
are indicators of UTI. Similarly, a strong correlation was
found between RTEC count and urinary α1-microglobulin
concentration and GGT activity in urine. The latter two
parameters have been previously described as indicators
Bacterial count  

964.9/μL  

 
 
 
  0.787a (0.748–0.821)
0.0215
76.6 (67.5–84.3)
73.1 (68.5–77.4)

of upper UTI [3, 15].

In the past, different potential markers to discriminate
between upper UTI and LUTI have been studied [16, 17].
Although urinary neutrophil gelatinase-associated lipoca-
lin (NGAL) concentrations were increased in patients with
upper UTI and LUTI compared to the control group, a
α1-Microglobulin  

 
 
 
18.8 mg/L

  0.735a,b (0.694–0.773)
0.0281
68.6 (58.7–77.5)
68.7 (63.9–73.3)

differentiation between upper UTI and LUTI using this

marker was not possible [17]. As the proximal tubule is
not mainly affected in upper UTI [18] and as kidney injury
molecule-1 (KIM-1) represents especially an injury marker
of the proximal tubule, urinary KIM-1 can also not be con-
sidered as a suitable marker for discriminating upper UTI
from LUTI [16].
 

 
Sensitivity (95% CI) 
Specificity (95% CI) 
maximal sensitivity

As expected [5], increased RTEC counts were observed

ROC’s cut-off at

and specificity

in urine specimens from patients with upper urinary tract

AUC (95% CI)

pathology. Inflammation is known to target TECs. Tubular

cells have been implicated in the response to inflamma-
tory mediators in ischemic and septic renal damage [19].
SE

Brought to you by | provisional account

Unauthenticated
Download Date | 12/26/19 4:45 AM

We demonstrated that RTECs may add value in the dis-
crimination between upper UTI and LUTI. At a cut-off
level of 3.1 cells/μL, which corresponds with the manufac-
turer’s set upper reference limit (0–3 cells/μL), we found
a sensitivity and a specificity of 93.5% and 82.2%, respec-
tively. The diagnostic performance of RTECs in the differ-
entiation between upper UTI and LUTI exceeded by far the
one of TECs.
In our previous study, α1-microglobulin proved to be a
useful discriminator between upper UTI and LUTI [3, 20].
However in this study, comparative ROC analysis dem-
onstrated that the diagnostic utility of RTECs for detect-
ing upper UTI outperforms that of α1-microglobulin, GGT
activity, leukocyturia and pathological casts. In accord-
ance with the findings of Penders et al. [3], we found that
pathological casts showed a low specificity for localizing
UTI. These findings could be attributed to the low number
of casts usually found in urinary specimens and the dif-
ficult quantification of these urinary structures in urinary
particle analysis [13, 21, 22].
Time between sampling and performance of the
examination procedure is critical for the reliability of
urinary test results [23]. Stability data are known for most
usual urine sediment parameters [24], but lack for RTECs
and TECs. We found that the stability of RTECs and TECs at
room temperature is limited to 4 h. As most clinical labo-
ratories are able to process urinary samples for urinalysis
immediately after arrival in the laboratory, this limited sta-
bility can be overcome. However, special attention should
be paid to the pH dependent stability of RTECs and TECs,
as a higher pH (pH > 7.5) decreases the RTEC and TEC sta-
bility with approximately 8 h. The same dependency has
been observed for urinary casts and WBCs [25], which are
also lost in alkaline urines. High urinary pH values may
be observed in infections with urease-producing bacteria
[25]. Cooling of the specimen at 2–8 °C improves the stabil-
ity of both parameters.
Although the inbuilt UF-5000 osmolality parameters
may allow to correct for urinary dilution, the diagnos-
tic power of RTECs cannot further be improved by cor-
recting results using urinary conductivity and urinary
creatinine. This could be explained by the fact that the
stability of both parameters is, as urinary WBCs, affected
by extremely diluted or concentrated specimens [25].
In conclusion, our study adds interesting information
on the performance of some new parameters that recently
have become available on a routine new generation urine
sediment analyzer [26]. Along with the results of urinary
microbiological culture and clinical evaluation, RTECs
may be considered as an interesting supplemental analyte
for differentiating upper UTI from LUTI.
Oyaert et al.: Added value of RTECs in UTI      7
Author contributions: All the authors have accepted
responsibility for the entire content of this submitted
manuscript and approved submission.
Research funding: None declared.
Employment or leadership: None declared.
Honorarium: None declared.
Competing interests: The funding organization(s) played
no role in the study design; in the collection, analysis, and
interpretation of data; in the writing of the report; or in the
decision to submit the report for publication.
References
1. Stein R, Dogan HS, Hoebeke P, Kocvar R, Nijman RJ, Tekgüul S.
Urinary tract infections in children: EAU/ESPU guidelines. Eur
Urol 2015;67:546–58.
2. Grabe M, Bartoletti R, Bjerklund-Johansen TE, Bartoletti R,
Cek M, Naber KG, et al. Guidelines on urological infections.
European Association of Urology 2015. https://uroweb.org/wp-
content/uploads/19-Urological-infections_LR2.pdf. Accessed:
12 Aug 2019.
3. Penders J, Fiers T, Everaert K, Barth J, Dhondt A, Delanghe J.
Diagnostic performance of combined specific urinary proteins
and urinary flow cytometry in urinary tract pathology. Clin Chem
Lab Med 2007;45:499–504.
4. Penders J, Delanghe J. Alpha 1 microglobulin: laboratory and
clinical aspects. Clin Chim Acta 2004;346:107–18.
5. Fogazzi GB. The urinary sediment: an integrated view. p57.
Milano: Elsevier, 2010.
6. Becker GJ, Garigali G, Fogazzi GB. Advance in urine microscopy.
Am J Kidney Dis 2016;67:954–64.
7. Oyaert M, Delanghe JR. Semiquantitative, fully automated urine
test strip analysis. J Clin Lab Anal 2019;33:e22870.
8. Previtali G, Ravasio R, Seghezzi M, Buoro S, Alessio MG. Per-
formance evaluation of the new fully automated urine particle
analyser UF-5000 compared to the reference method of the
Fuchs-Rosenthal chamber. Clin Chim Acta 2017;472:123–30.
9. Orsonneau HL, Douet P, MAssoubre C, Lustenberger P, Bernard
S. An improved pyrogallol red-molybdate method for determin-
ing total urinary protein. Clin Chem 1989;35:2233–6.
10. Korsten CB, Persijn JP, Van der Slik W. The application of the
serum gamma-glutamyl transpeptidase and the 5’ nucleotide
assay in cancer patients: a comparative study. Z Klin Chem Klin
Biochem 1974;12:116–20.
11. Aspevall O, Hallander H, Gant V, Kouri T. European guidelines
for urinalysis: a collaborative document produced by
European clinical microbiologists and clinical chemists under
ECLM in collaboration with ESCMID. Clin Microbiol Infect
2001;7:173–8.
12. Enko D, Stelzer I, Böckl M, Derler B, Schnedl WJ, Anderssohn P,
et al. Comparison of the diagnostic performance of two auto-
mated urine sediment analyzers with manual phase-contrast
microscopy. Clin Chem Lab Med 2020;58:268–73.
13. Delanghe JR, Kouri TT, Huber AR, Hanneman-Pohl K, Guder WG,
Lun A, et al. The role of automated urine particle flow cytometry
in clinical practice. Clin Chim Acta 2000;301:1–18.
Brought to you by | provisional account
Unauthenticated
Download Date | 12/26/19 4:45 AM
8      Oyaert et al.: Added value of RTECs in UTI
14. Langlois MR, Delanghe JR, Stefaert SR, Everaert KV, De Buyzere
ML. Automated flow cytometry compared with an automated
dipstick reader for urinalysis. Clin Chem 1999;45:118–22.
15. Yesel EE, Paker N, Yesel A, Kayatas K, Laleli Y, Ari E. Urinary gamma-
glutamyl transferase to creatinine ratio as an indicator of tubular
function in bence jones proteinuria. Ren Fail 2014;36:390–2.
16. Urbschat A, Obermüller N, Paulus P, Reissig M, Hadji P, Hofman
R, et al. Upper and lower urinary tract infections can be detected
early but not be discriminated by urinary NGAL in adults. Int Urol
Nephrol 2014;46:2243–9.
17. Yilmaz A, Sevketogly E, Gedikbasi A, Karyagar S, Kiyak A,
Mulazimoglu M, et al. Early prediction of urinary tract infection
with urinary neutrophil gelatinase associated lipocalin. Pediatr
Nephrol 2009;24:2387–92.
18. Han WK, Bailly V, Abichandani R, Thadhani R, Bonventre JV.
Kidney injuring molecule 1: a novel biomarker for human renal
proximal tubule injury. Kidney Int 2002;62:237–44.
19. Cantaluppi V, Quercia AD, Dellepiane S, Ferrario S, Camussi G,
Biancone L. Interaction between systemic inflammation and renal
tubular epithelial cells. Nephrol Dial Transplant 2014;29:2004–11.
20. Mantur M, Kemona H, Dabrowska M, Dabrowska J, Sobolewski
S, Prokopowicz J. Alpha1-microglobulin as a marker of proximal
tubular damage in urinary tract infection in children. Clin Neph-
rol 2000;53:283–7.
21. Fenili D, Pirovano B. The automation of sediment analysis
using a new urine flow cytometer (UF-100). Clin Chem Lab Med
1998;36:909–17.
22. Hanneman-Pohl K, Kampf SK. Automation of urine sediment
examination: a comparison of the Sysmex UF-100 automated
flow cytometer with routine manual diagnosis (microscopy, test
strips, and bacterial culture). Clin Chem Lab Med 1999;37:
753–64.
23. Delanghe J, Speeckaert M. Preanalytics in urinalysis. Clin
­Biochem 2016;49:1346–50.
24. Kouri T, Fogazzi G, Gant V, Hallander H, Hofmann W,
Guder WG. European Confederation of Laboratory Medi-
cine. ­European u
­ rinalysis guidelines. Scand J Clin Lab Invest
2000;231(Suppl):1–86.
25. Kouri T, Fogazzi G, Gant V, Hallander H, Hofmann W, Guder
WG. European Urinalysis guidelines. Scand J Clin Lab Invest
2000;60:1–96.
26. Oyaert M, Speeckaert MM, Delanghe JR. Estimated urinary
osmolality based on combined urinalysis parameters: a critical
evaluation. Clin Chem Lab Med 2019;57:1169–76.
Brought to you by | provisional account
Unauthenticated
Download Date | 12/26/19 4:45 AM