Key words
CEUS - upper urinary tract - urothelial cell carcinoma - ultrasound - CT urography
MR urography
Introduction
Upper urinary tract malignancies are relatively rare and account for only 5 % of urothelial
cell carcinomas (UCCs) [1]
[2]. Most upper urinary tract UCCs (75 %) are located in the renal pelvis and calyces,
while the remaining 25 % are located in the ureter: 70 % in the distal ureter, 25 %
in the mid-ureter and 5 % in the proximal ureter [3].
Multidetector computed tomographic urography (MDCTU) and magnetic resonance urography
(MRU) are widely considered to be the best available modalities for the diagnosis
and staging of urothelial carcinoma [4]
[5]
[6]
[7]
[8]
[9]
[10]. The sensitivity (MDCTU 92.9 – 100 %; MRU 62.9 – 94.1 %) and specificity (MDCTU
93 – 100 %; MRU 81 – 100 %) of these methods are high but diagnosis is often based
on the presence of indirect signs, such as wall thickening and hydroureteronephrosis,
rather than on the actual recognition of the lesion.
Ultrasound (US) is the first-line imaging modality in patients with hematuria [2], as color Doppler US is useful in urological imaging and power Doppler US improves
the sensitivity in the detection of slow blood flow [11].
The introduction of contrast-enhanced US (CEUS) has made it possible to detect and
image blood flow in small vessels using microbubbles to assess regional microvascular
perfusion in tissues such as the liver, kidney and myocardium [12]. The advantage of this method compared to MDCTU and MRU lies in the ability to provide
real-time images of blood flow within the lesions.
The main objective of this study was to assess the effectiveness of CEUS in the diagnosis
of upper urinary tract neoplasms compared to MDCTU and MRU which are considered to
be the gold standard. Secondary objectives were to assess the usefulness of CEUS in
distinguishing high-grade tumors from low-grade ones on the basis of time-intensity
curves and also to compare the tumor size measured with CEUS, MDCTU and MRU.
Materials and methods
Patient population
From January 2009 to September 2011, 18 patients, 10 males and 8 females (mean age:
74; range: 62 – 85), were referred to diagnostic investigation with suspicion of upper
urinary tract carcinoma due to persistent gross hematuria (n = 11) or microscopic
hematuria (n = 7), positive or atypical urine cytology results but no abnormalities
at cystoscopy ([Tab. 1]).
Tab. 1
Patient population.[1]
|
age
|
sex
|
location
|
symptoms
|
specimen length
(cm)
|
urine cytology
|
grading
|
|
1
|
81
|
M
|
DU
|
Gross
Hematuria
|
2.3
|
P
|
UCC-HG
|
|
2
|
72
|
M
|
DU
|
Gross
Hematuria
|
2.7
|
P
|
UCC-HG
|
|
3
|
69
|
M
|
PU
|
Gross
Hematuria,
Flank Pain
|
3.2
|
P
|
UCC-LG
|
|
4
|
63
|
F
|
C
|
Microscopic
Hematuria
|
1.2
|
P
|
UCC-HG
|
|
5
|
75
|
F
|
DU
|
Microscopic
Hematuria
|
2.4
|
A
|
UCC-LG
|
|
6
|
74
|
M
|
MU
|
Gross
Hematuria
|
1.9
|
P
|
UCC-LG
|
|
7
|
81
|
M
|
DU
|
Microscopic
Hematuria
|
2.5
|
P
|
UCC-HG
|
|
8
|
79
|
F
|
DU
|
Gross
Hematuria
|
3.1
|
P
|
UCC-LG
|
|
9
|
84
|
F
|
PU
|
Gross
Hematuria
|
3.8
|
P
|
UCC-LG
|
|
10
|
70
|
M
|
MU
|
Microscopic
Hematuria
|
2.1
|
A
|
UCC-LG
|
|
11
|
68
|
F
|
DU
|
Microscopic
Hematuria
|
2.5
|
P
|
UCC-HG
|
|
12
|
72
|
M
|
PU
|
Gross
Hematuria
|
4.1
|
A
|
UCC-LG
|
|
13
|
70
|
M
|
DU
|
Gross
Hematuria
Flank Pain
|
3.9
|
P
|
UCC-LG
|
|
14
|
73
|
F
|
MU
|
Gross
Hematuria
|
2.5
|
A
|
UCC-LG
|
|
15
|
85
|
M
|
MU
|
Microscopic
Hematuria
|
7.1
|
P
|
UCC-HG
|
|
16
|
62
|
M
|
DU
|
Microscopic
Hematuria
|
3.5
|
P
|
UCC-LG
|
|
17
|
79
|
F
|
DU
|
Gross
Hematuria
|
4.3
|
P
|
UCC-HG
|
|
18
|
73
|
F
|
C
|
Gross
Hematuria
|
3.3
|
A
|
UCC-LG
|
1 M: male, F: female, DU: distal ureter, MU: mid ureter, PU: proximal ureter, C: pielo-calyceal
system, P: positive, A: atypical, UCC: urothelial cell carcinoma, HG: high grade,
LG: low grade.
Written informed consent was obtained from all patients, and the procedures were in
accordance with the ethical standards of the Committee on Human Experimentation of
this institution.
Imaging techniques
Investigational protocol
All patients underwent MDCTU or MRU to assess the presence of a neoplastic lesion.
They subsequently underwent grayscale US, color Doppler (CDUS) and CEUS during the
same session. The same radiologists performed US, color Doppler and CEUS; they were
aware of the MDCTU and MRU findings.
All patients underwent surgery (nephroureterectomy with bladder cuff excision). The
dimensions of the surgical specimen were recorded before it was submitted to histological
examination which confirmed UCC in all cases.
MDCTU
Thirteen patients underwent MDCTU, which was performed on 64-MDCT scanners (Siemens,
Erlangen, Germany) using a three-phase protocol. 400 – 500 ml of water were administered
orally to the patients 20 minutes before the examination. Unenhanced CT scans of the
abdomen and pelvis were obtained. Nephrographic phase images were then obtained 80
sec after bolus injection (2 ml/kg body weight) of iohexol (Omnipaque 300, GE Healthcare,
USA) or iomeprolo (Iomeron 350, Bracco, Italy). Excretory phase images of the abdomen
and pelvis were obtained 8 minutes after injection of the contrast agents. Unenhanced
and nephrographic phase scans were reconstructed with sections with a thickness of
2 – 5 mm. The excretory phase scan was reconstructed as sections with a thickness
of 1.25 mm.
MRU
Five patients underwent MRU, which was performed on 1.5 T (Siemens) scanners. Multisequence
MRI was carried out including T1-weighted and T2-weighted sequences (slice thickness:
6.6 – 7.7 mm). Diffusion-weighted imaging (DWI) was carried out with single-shot spin-echo
planar imaging sequences (slice thickness: 6 – 7 mm). Spectral attenuated inversion
recovery (SPAIR) fat suppression was performed under free-breathing acquisition. In
the dynamic contrast-enhanced (DCE) MRI sequences, 24 slices were sequentially acquired
30, 80, and 180 sec after administration of meglumine gadopentetate (0.1 ml/kg body
weight).
US and CDUS
All patients underwent grayscale US and CDUS. US examination was performed on Aplio
(Toshiba, Tokyo, Japan), MyLab and MyLab-Twice (Esaote, Genoa, Italy) using a 3.5 – 5 MHz
multi-frequency broadband convex transducer. Examination was performed when the urinary
bladder was adequately full to visualize the possible presence of lesions in the ureteral-bladder
junction. The location and size of the lesions detected during grayscale US were recorded.
CDUS was performed using scanning parameters set for maximum sensitivity for slow
flow while the power output was increased to maximum. Color gain was increased to
the point just before the appearance of random noise. The pulse repetition frequency
was set at the lowest possible level. The images were assigned the following scores:
0 = no color signal, 1 = low color signal, 2 = intense color signal.
CEUS
In the same session all patients underwent real-time continuous examination using
a low mechanical index (range: 0.04 – 0.1) after the administration of SonoVue (Bracco™).
SonoVue is a blood pool US contrast agent comprising microbubbles. A total of 4.8 ml
was administered in 2 intravenous bolus doses of 2.4 ml; the second dose was injected
15 minutes after the first and both were followed by 5 ml of a saline flush. The second
injection was performed to evaluate the contralateral urinary tract.
Still images of the entire examination were digitally recorded on magnetic optical
disk. Recording was initiated at the start of the first contrast injection and at
the start of the second injection. All time measurements started at these points,
which were “time-0” in all recorded video clips.
The images were analyzed off-site at consensus conferences to determine the size of
the lesion, baseline and CDUS appearance as well as the signal intensity of the enhancement
after SonoVue administration.
The images were interpreted by the two radiologists who performed the examination.
Both were skilled in urologic US and had more than 5 years of CEUS examination experience.
Perfusion software was then used to estimate the perfusion pattern (Qontrast v 3.0,
Bracco, Milan, Italy) [13]. This quantitative software analyzes the temporal sequence of the images. Time-intensity
curves (TIC) were extracted from regions of interest in the lesion, and TIC quantitative
analysis was performed. The following parameters were considered: wash-in time, time-to-peak
(TTP), maximum signal intensity (SI) and wash-out time. The signal intensity was calculated
for every pixel for every second, thus generating maps of perfusion parameters.
Statistical analysis
The paired T-test was performed to evaluate the statistical significance of the measurements
obtained using the four imaging techniques. The measurements were compared to the
actual length of the lesion measured on the surgical specimen.
Results
MRU and MDCTU identified upper urinary tract masses in all 18 patients, who subsequently
underwent CEUS. Histological examination after surgery confirmed UCC in all patients.
The lesions were located as follows: 9 (50 %) in the distal ureter; 4 (22.2 %) in
the mid-ureter; 3 (16.6 %) in the proximal ureter; 2 (11.2 %) in the pelvicalyceal
system. The mean dimension was 3.1 cm (range 1.2 – 7.1 cm).
In MDCTU and MRU, 18 lesions exhibited enhancement. In the urographic phase, the lesions
did not exhibit direct enhancement, and the diagnosis of upper urinary tract neoplasm
was made on the basis of indirect indicators. The lesion size measured on MDCTU and
MRU images was found to be 15 – 20 % greater than the actual size of the surgical
specimen.
Grayscale US identified 15 out of 18 lesions (83.3 %). One lesion in the upper pelvicalyceal
system and 2 in the mid-ureter were not identified. The lesion size measured on grayscale
US images was found to be about 25 % greater than the actual size of the surgical
specimen.
In CDUS the 15 lesions presented different color signals: 7 lesions exhibited no flow
signal (score 0); 7 presented low color signal (score 1); 1 exhibited intense color
signal (score 2).
In CEUS 17 patients were found positive for upper urinary tract neoplasm, while 1
was found negative. The undetected lesion was small (1.2 cm) and located in the upper
pelvicalyceal system of the kidney. It was not identified with grayscale US. CEUS
examination therefore resulted in 17 true-positive cases and 1 false-negative case,
yielding a sensitivity of 94.4 %.
The lesion size measured on the CEUS images came close to the actual dimensions of
the surgical specimen ([Tab. 2]). In CEUS all detected lesions exhibited homogeneous enhancement in every part of
the lesion.
Tab. 2
Urothelial lesions; length (cm).
|
US
|
MDCTU\MRU
|
CEUS
|
specimen
|
|
2.8
|
2.7
|
2.4
|
2.3
|
|
3.3
|
3.2
|
2.8
|
2.7
|
|
4.0
|
3.7
|
3.4
|
3.2
|
|
unidentified
|
1.4
|
unidentified
|
1.2
|
|
3.0
|
2.8
|
2.5
|
2.4
|
|
2.4
|
2.3
|
2.0
|
1.9
|
|
3.1
|
2.8
|
2.6
|
2.5
|
|
3.9
|
3.5
|
3.2
|
3.1
|
|
4.8
|
4.4
|
4.0
|
3.8
|
|
unidentified
|
2.4
|
2.2
|
2.1
|
|
3.1
|
3.0
|
2.6
|
2.5
|
|
5.0
|
4.8
|
4.3
|
4.1
|
|
4.8
|
4.6
|
4.1
|
3.9
|
|
unidentified
|
2.9
|
2.6
|
2.5
|
|
8.1
|
7.9
|
7.3
|
7.1
|
|
4.4
|
4.0
|
3.7
|
3.5
|
|
5.2
|
5.0
|
4.5
|
4.3
|
|
4.3
|
3.8
|
3.5
|
3.3
|
The paired T-test results were: US vs. specimen P > 0.50; MDCTU/MRU vs. specimen P > 0.50;
CEUS vs. specimen P < 0.05.
TIC and semi-quantitative analysis produced the following results: high-grade UCCs:
wash-in after 13 – 18 sec, TTP > 30 sec, SI > 55 %, wash-out time > 80 secs; low-grade
UCCs: wash-in time 10 sec, TTP < 25 sec, SI < 50 %, wash-out time < 50 sec ([Tab. 3]) [2].
Tab. 3
HG-UCC e LG-UCC in bladder and upper urinary tract (UUT).
|
HG-UCC
|
LG-UCC
|
|
Bladder
|
UUT
|
Bladder
|
UUT
|
|
wash-in time
|
13 sec
|
13 – 18 sec
|
10 sec
|
10 sec
|
|
TTP
|
> 28 sec
|
> 30 sec
|
< 26 sec
|
< 25 sec
|
|
SI
|
> 50 %
|
> 55 %
|
< 50 %
|
< 50 %
|
|
wash-out time
|
58 sec
|
> 80 sec
|
< 50 sec
|
< 50 sec
|
Discussion
MDCTU and MRU are the gold standard for imaging of the upper urinary tract and have
replaced intravenous excretory urography [14]
[15]
[16]. Analysis of the data available in the literature shows that the sensitivity and
specificity of MDCTU are close to 100 % with a greater sensitivity and specificity
in lesions located in the renal pelvis (100 %) than in the ureter (92.9 %) [1]
[16]
[17]
[18].
In the literature, the two most common sequences in MRU are T2-weighted hydrographic
sequences without contrast agent and T1-spoiled gradient-recalled echo (GRE) sequences
during the excretory phase after gadolinium administration [19]
[20] with a sensitivity of 62.9 – 76.5 % and a specificity of 89 – 100 % [20]
[21]. The use of DWI sequences substantially increases the sensitivity to up to 94.1 %,
as compared to T1 and T2 sequences [9]
[10].
In the present study, 18/18 lesions exhibited enhancement in MDCTU and MRU. During
the urographic phase the lesions were diagnosed on the basis of other indicators such
as thickening of the ureter wall or pelvicalyceal system as well as filling defects
which were particularly evident in the 7 cases (38.8 %) affected by hydroureteronephrosis
[17]
[20]
[22].
In grayscale US, UCC of the renal pelvis typically appears as a solid mass which is
difficult to distinguish from the renal sinus, particularly in the absence of hydronephrosis
[23]
[24]
[25]. US has a limited role in the evaluation of UCC of the ureter as the ureter is rarely
visualized in its entirety, even if dilated. When visualized, these tumors are typically
intraluminal soft-tissue masses with proximal distention of the ureter.
In the present study, grayscale US identified 15 lesions, while 3 were not identified.
Two of the unidentified lesions were located in the mid-ureter and there was mild
dilation of the ureter. The third undetected lesion was the smallest (1.2 cm) and
was located in the upper pelvicalyceal system.
The use of color Doppler did not significantly increase US sensitivity owing to the
frequent absence of a vascular signal within the neoplastic lesion: 7 lesions out
of 18 (38.8 %) were assigned a score of 0.
In CEUS, 17 lesions out of 18 (94.4 %) exhibited enhancement offering direct visualization
of the lesion. The lesion that was not detected in CEUS was located in the upper pelvicalyceal
system of the kidney and was not identified in grayscale US ([Fig. 1]). It probably went undetected because of the intense enhancement of the renal parenchyma
which masked the enhancement of the lesion. TTP of the renal parenchyma occurred about
18 sec after the first injection of contrast agent and wash-out took more than 5 min
thereby completely masking the lesion. In the present patient population, CEUS identified
another lesion located in the calyceal cavity as it was larger and caused swelling
of the renal pelvis. This may indicate a greater sensitivity of CEUS in masses located
in the ureter (16 out of 16 lesions were identified; sensitivity 100 %) than in smaller
lesions located in the calyceal system (1 out of 2 lesions was identified; sensitivity
50 %). The overall sensitivity of CEUS was 94.4 %. Because of the lack of false-positive
and true-negative patients, the sensitivity was the only statistical datum that could
be extrapolated about CEUS in the detection of upper urinary tract UCCs.
Fig. 1 A 63-year-old woman with high-grade UCC located in the upper pelvicalyceal group
of the left kidney: a MDCTU shows a small mass in the calyceal cavity of the upper pole (red arrow); b MRU confirms the presence of the lesion (yellow arrow); c CDUS does not show the lesion, normal vascularity of the renal parenchyma; d US (left: gray-scale; right: CEUS): the lesion is not identified; e Surgical specimen: lesion length 12 mm (black arrow).
Abb. 1 Eine 63-jährige Frau mit hochgradigem UCC, welches im oberen Nierenbeckenkelchsystems
der linken Niere lokalisiert ist: a MDCTU zeigt eine kleine Raumforderung im Kelchraum des oberen Pols (roter Pfeil);
b MRU bestätigt den Herdbefund (gelber Pfeil); c CDUS zeigt keine Läsion, normale Vaskularität des Nierenparenchyms; d US (links: B-Bild, rechts: CEUS): die Läsion wird nicht erkannt; e Operationspräparat: Länge des Herdes 12 mm (schwarzer Pfeil).
MDCTU and MRU overestimated tumor size by 15 – 20 %. This phenomenon is probably due
to the fact that it is difficult to differentiate between the lesion and the ureteral
walls or the calyceal walls because of concurrent enhancement of the lesion and the
surrounding wall with no neoplasm [26].
Grayscale US also overestimated tumor size by about 25 %. In most cases this phenomenon
is linked to the presence of intraluminal echoes due to artifacts, cellular debris
and/or blood clots.
CEUS yielded tumor sizes close to the actual dimensions of the surgical specimens
as amplitude subtraction permitted the suppression of intraluminal echoes caused by
artifacts ([Fig. 2], [3]). Agreement between the actual tumor size of the surgical specimen and the size
measured in CEUS shows that areas exhibiting enhancement in CEUS are actually neoplastic
lesions.
Fig. 2 An 81-year-old man with high-grade UCC located in the left distal ureter: a MDCTU shows moderate hydroureteronephrosis of the urinary tract, the lesion is moderately
enhanced (red arrow); b CDUS: the lesion is poorly vascularized (score 1) and the margins are blurred due
to the presence of intraluminal artifacts (yellow arrow) (length 38 mm); c CEUS: average late-phase enhancement of the lesion (length 25 mm) (green arrows);
d TIC showing the following values: wash-in time 13 sec, TTP 29 sec, SI > 60 and wash-out
time > 80 sec suggesting high-grade UCC.
Abb. 2 Ein 81-jähriger Mann mit hochgradigem UCC in der linken distalen Harnröhre: a MDCTU zeigt eine moderate Hydroureteronephrose des Harntraktes; die Läsion ist mäßig
verstärkt (roter Pfeil); b CDUS: die Läsion ist schlecht vaskularisiert (Grad 1) und die Ränder sind aufgrund
intraluminaler Artefakte verschwommen (gelber Pfeil) (Länge 38 mm); c CEUS: durchschnittliche Signalverstärkung der Läsion in der Spätphase (Länge 25 mm)
(grüne Pfeile); d TIC zeigt folgende Werte: Wash-in-Zeit 13 s, TTP 29 s, SI > 60 und Wash-out-Zeit
> 80 s, was für ein hochgradiges UCC spricht.
Fig. 3 A 79-year-old woman with high-grade UCC located in the left distal ureter: a MDCTU shows left-sided hydroureteronephrosis and thickening of the distal ureter
with poor enhancement (red arrow); b US (left: CEUS; right: gray-scale) intense enhancement of the lesion (green arrows)
(length 43 mm), in the right image, intraluminal artifacts (yellow arrow) do not permit
an accurate measurement; c Surgical specimen, lesion length measured on the specimen corresponds to measurement
made at CEUS; d TIC showing the following values: wash-in time 14 seconds, TTP 35 sec, SI > 64 and
wash-out time > 80 sec suggesting high-grade UCC.
Abb. 3 Eine 79-jährige Frau mit hochgradigem UCC in der linken distalen Harnröhre: a MDCTU zeigt eine linksseitige Hydrourethernephrose und eine Verdickung des distalen
Urethers mit geringer Signalverstärkung (roter Pfeil); b im US (links: CEUS, rechts: B-Bild) hochgradige Signalverstärkung der Läsion (grüne
Pfeile) (Länge 43 mm), in der rechten Abbildung, die intraluminalen Artefakte (gelber
Pfeil) lassen keine genaue Messung zu; c Operationspräparat: die am Präparat gemessene Größe der Läsion entspricht der Messung
mittels CEUS; d TIC zeigt folgende Werte: Wash-in-Zeit 14 s, TTP 35 s, SI > 64 und Wash-out-Zeit
> 80 s, was für ein hochgradiges UCC spricht.
The fact that CEUS showed a vascular signal from the lesion in an elevated percentage
of cases, i. e., higher than CDUS and close to MDCTU/MRU (CEUS 94 %; CDUS 61 %; MDCTU/MRU
100 % of cases), may be attributed to the ability of the US blood pool contrast agent
to provide real-time depiction of the lesion vascularity.
The clips obtained in CEUS were used for the construction of TICs for the comparison
between enhancement patterns of high-grade and low-grade UCCs. This semi-quantitative
analysis was compared to data obtained from studies of the bladder [27]
[28]. Semi-quantitative analysis of high-grade and low-grade UCCs of the urinary tract
yielded results similar to those related to bladder cancer.
The only contrasting data with regard to the signal intensity of high-grade UCCs was
a case in which the lesion was located in the mid-ureter in close proximity to the
iliac artery. In this case the SI was lower than the SI observed in the other cases.
The close proximity of the iliac artery, where there were high concentrations of contrast
agent, may have altered the SI measurement of the neoplastic lesion ([Fig. 4]).
Fig. 4 An 85-year-old man with high-grade UCC located in the right mid ureter: a Retrograde pyelography shows filling defect of the ureter (red arrow); b Gray-scale US shows a solid mass in the lumen of the right ureter (yellow arrows),
the iliac artery and vein are visible below; c CDUS shows poor vascular signal within the lesion (score 1); d CEUS shows homogeneous enhancement of the neoplastic lesion (green arrows) (length
70 mm); e Surgical specimen: lesion length is 71 mm; f TIC showing the following values: wash-in time 13 sec, TTP 31 sec, SI 44 and washout
time > 80 sec suggesting high-grade UCC.
Abb. 4 Ein 85-jähriger Mann mit hochgradigem UCC in der rechten mittleren Harnröhre: a Die retrograde Pyelografie zeigt die Füllung des Harnleiters (roter Pfeil); b das B-Bild zeigt eine kompakte Raumforderung im Lumen des rechten Urethers (gelbe
Pfeile), die Arterie und Vene des Beckens sind darunter sichtbar; c CDUS zeigt ein schlechtes vaskuläres Signal innerhalb der Läsion (Grad 1); d CEUS zeigt eine homogene Signalverstärkung der neoplastischen Läsion (grüne Pfeile)
(Länge 70 mm); e Operationspräparat: Die Länge der Läsion beträgt 71 mm; f TIC zeigt die folgenden Werte: Wash-in-Zeit 13 s, TTP 31 s, SI 44 und Wash-out-Zeit
> 80 s spricht für ein hochgradiges UCC.
The limitations of this study are related to the patient population. The first limitation
is the low number of patients which is partly due to the relative rarity of upper
urinary tract UCCs. Only 5 – 10 % of UCCs affect the upper urinary tract, while the
remaining 90 – 95 % are located in the bladder [2]. The second limitation is related to the selection criteria, as all the recruited
patients already had a MDCTU and/or MRU report positive for the presence of an upper
urinary tract lesion and urinary cytology positive for malignant cells. However, this
was needed to assess the reliability of CEUS in the detection of upper urinary tract
UCCs.
Conclusions
The results of this study show that CEUS is a useful tool for evaluating upper urinary
tract tumors as nearly all lesions were detected. CEUS allows a semi-quantitative
analysis for differentiating the degree of malignancy, an accurate measurement of
the mass by image subtraction, thus avoiding artifacts, and distinction of the tumor
from adjacent structures based on enhancement timing.
CEUS may become a first-line examination in the diagnosis of upper urinary tract masses
because of the short execution time and relatively low cost and thereby become an
alternative to MRU and MDCTU in patients who are allergic to contrast agents or are
affected by renal failure. However, further studies involving larger patient populations
are mandatory to confirm these encouraging preliminary results.
