Introduction
Introduction
In modern uroradiology, unenhanded helical CT meanwhile became established as the
best imaging modality for the diagnosis of urolithiasis [1]. In contrast, urographic visualization of the urinary tract anatomy by means of
contrast material-enhanced CT in the excretory phase could not really compete with
intravenous pyelography (IVP), although several studies demonstrated a general feasibility
of CT urography (CTU) [2]
[3]. In comparison with IVP, the conventional CTU examination technique still revealed
a deficiency in the accurate delineation of pelvicaliceal details [3]. Furthermore, complete opacification of all ureteral segments seemed to be an inconsistent
finding in previous CT urograms which necessitated the use of a compression device
even in CT [3]. In the present context, we assume that two aspects can help to improve the examination
technique for CT urography: Firstly, it makes sense to utilize all the technical advantages
of the new multi-slice technology [4] also for the performance of CTU. Secondly, derived from the knowledge of the positive
interaction of low-dose furosemide and contrast material for optimizing the morphologic
accuracy in excretory MR urography [5], it would be interesting to know whether this modification can also be successfully
applied to CT urography. Therefore, the purpose of the preliminary study was to obtain
initial results as to whether diuretic-enhanced multi-slice CT urography (MS-CTU)
can improve the quality of routine CT to visualize the urinary tract in the excretory
phase.
Patients and methods
Patients and methods
Multi-slice CT urography was performed in 21 patients (12 men, 9 women), ranging in
age from 38 to 72 years. The serum creatinine level was normal in 18 patients. Two
patients had an increased serum creatinine of 1.5 mg/dl (133 µmol/l) and one patient
of 1.7 mg/dl (150 µmol/l). Twenty patients were referred to our department for a multiphase
CT examination of the kidneys and urinary tracts because of suspected extrinsic or
intrinsic tumor disease by sonography (n = 9), by intravenous pyelography (n = 1),
or by CT from an external department (n = 1) and, moreover, because of sonographically
diagnosed hydronephrosis (n = 1), hematuria and/or flank pain of uncertain cause (n
= 7), and hematuria after trauma (n = 1). In another patient with a carcinoma of the
cervix, MS-CTU was carried out to provide a preoperative assessment of the ureters.
Informed consent was obtained from each patient after the procedure of the examination
had been explained carefully, including the additional injection of low-dose furosemide
(Lasix; Hoechst, Germany). No extra CT scans were scheduled for the multiphase examination
protocol, which included the excretory phase.
All CT examinations were performed using a multi-slice scanner (Somatom Volume Zoom;
Siemens, Forchheim, Germany). The patients were placed in supine position. For contrast-enhancement,
all patients received a single intravenous bolus injection of 100 ml of Iopromide
(Ultravist 370; Schering, Berlin, Germany) at a flow rate of 3 ml/sec. In 16 of 21
patients, 10 mg of furosemide were intravenously administered 3 - 5 minutes before
contrast material injection, followed by a bolus of 30 ml of physiologic saline solution
immediately after application of contrast material. In 5 of 21 patients, MS-CTU was
performed without use of furosemide, but with intravenous bolus injection of 250 ml
of physiologic saline solution (150 ml 5 min before, 100 ml immediately after contrast
material injection). The contrast agent and the saline solution were administered
via a standard double power-injector (Liebel-Flarsheim, Cincinatti, U.S.A.). Oral
contrast material opacification of bowel loops was applied in only one patient.
After completion of the standard CT scans in the corticomedullary and nephrographic
phases, a low-dose single CT slice „test-image” (20mAs) was obtained from the kidneys
within 10 minutes after contrast material injection in order to make sure that the
contrast agent had definitely reached the pelvicaliceal system on either side. With
the information of this „test-image”, it was possible to start the subsequent CT urography
scan exactly in the excretory phase. In each case, the CTU scan was planned to extend
from the upper pole of the left kidney down to the symphysis. An abdominal compression
device was not used. The following scan parameters were employed for MS-CTU: 4 x 2.5
mm collimation; 12.5 mm table feed per rotation; pitch 1.25 ; 120 kV; 100 - 150 mAs;
0.5 sec rotation time; standard abdominal kernel. All data were acquired within a
short breath-hold lasting between 12 and 16 sec encompassing a z-coverage of up to
40 cm. For the diagnostic reading, axial images were reconstructed with a slice thickness
and increment of 5 mm. For postprocesing procedures, we reconstructed source images
with a slice thickness of 3 mm and an increment of 1 mm generating an overlap of 67
%. From these source images, maximum intensity projection (MIP) images in multiple
different planes were reconstructed using the standard software (Wizard, Siemens)
of the CT scanner. In addition, superimposing bones and calcified arteries were eliminated
from the MIP images using the volume of interest (VOI) software tool also implemented
on our system.
The MS-CT urograms were analyzed by two radiologists (author 1 and 2) in a consensus
assessment concerning the feasibilty of the new technique, the completeness of urinary
tract visualization, and pathologic findings. The ability of MS-CTU to visualize the
upper urinary tratcs was evaluated separately for the ureters and pelvicaliceal systems
on either side using a scale from 1 to 3, with
-
fragmented opacification of the ureter/pelvicalices with distinct gaps in the contrast
material column.
-
near complete visualization of the ureter/pelvicalices with only one or two short
unopacified segments; (a) with limited caliceal detail and (b) with good caliceal detail.
-
complete opacification of the ureter/pelvicalices; (a) with limited caliceal detail and (b) with good caliceal detail.
The assessment comprised a total of 42 pelvicaliceal systems and ureters. The urinary
bladder was not included in this preliminary analysis. For a quantitative assessment,
the Hounsfield units (HU) of each contrast material-enhanced renal pelvis were measured
in the axial CTU source images using the region-of-interest (ROI) method.
Results
Results
In the five patients (10 pelvicalices and ureters) in whom physiologic saline solution
was administered instead of low-dose furosemide, complete opacification of the collecting
system was possible, however, the scores varied markedly between the different patients
(Tab. [1]). With use of saline solution, density values between 947 and 2544 HU (mean: 1514
HU) were measured inside the 10 opacified pelvicaliceal systems after 10 min of contrast
material injection. A mean standard deviation of 352 HU (23 %) per ROI was indicative
of a relatively inhomogeneous bright enhancement. The demonstration of pelvicaliceal
details was assessed as limited in all cases (Tab. [1]). In MS-CTU performed using saline solution, the calices frequently appeared to
be blunted and blurred (Fig. 1). Dense contrast material inside the papillae and medullary collecting tubules partly
also affected the sharp delineation of the caliceal fornices.
Table 1 Consensus assessment of two reviewers: scores for the visualization of the upper
urinary tract in MS-CTU.
|
MS-CTU with NaCl 0.9 % |
MS-CTU with furosemide |
| score |
ureter
(n = 10) |
pelvicalices
(n = 10) |
ureter
(n = 32) |
pelvicalices
(n = 32) |
| 1 |
4 (40 %) |
1 (10 %) |
2 (6 %) |
0 |
| 2 |
3 (30 %) |
a 2 (20 %)
b 0 |
8 (25 %) |
a 2 (6 %)
b 2 (6 %) |
| 3 |
3 (30 %) |
a 7 (70 %)
b 0 |
22 (69 %) |
a 1 (3 %)
b 27 (85 %) |
Fig. 1 MIP image from MS-CTU performed after injection of 250 ml physiologic saline solution
instead of furosemide in a 53-year old man. Although the pelvicaliceal systems are
brightly enhanced, the calices appear blunted. Dense contrast material is also seen
in many papillae (arrows).
In the 16 patients (32 pelvicalices and ureters) who underwent diuretic-enhanced MS-CTU,
the injection of 10 mg of furosemide led to uniform distribution of the subsequently
excreted contrast material inside the urinary tracts. In each of our patients, only
one test-scan and one breath-hold MS-CTU scan were necessary to provide a sufficient
urographic view after 10 minutes following contrast material injection. A good urographic
effect was possible even in those three patients with an increased serum creatinine
of up to 1.7 mg/dl (150 µmol/l). Near complete or complete opacification was constantly
achieved with use of low-dose furosemide (Tab. [1]). In two cases, the left ureter was insufficiently visualized. In one of these patients,
the left ureter was superimposed by an inserted radiopaque double-J stent. Artefacts
caused by very dense endoluminal accumulation of contrast material were not observed.
Morphologic details such as the caliceal fornices were frequently visualized (Fig.
2). After injection of furosemide, the density values measured inside the 32 opacified
pelvicalices were markedly lower than obtained without a diuretic agent, now ranging
between 181 and 492 HU (mean: 354 HU). A low mean standard deviation of 21 HU (6 %)
per ROI indicated a very uniform endoluminal opacification.
Fig. 2 MIP images from furosemide-enhanced MS-CTU in a 45-year old male patient. Complete
opacification of the tortuously running ureters and sharp delineation of the pelvicaliceal
systems are achieved 10 min after injection of contrast material. The anterior (a) and lateral (b) MIP images demonstrate a calculus inside a lower pole calix (arrow) of the left
kidney, whereas phleboliths are clearly separated from the distal ureteral segments.
The postprocessing time per patient lasted up to 20 minutes. Most of this time had
to be spent for the elimination of skeletal structures using the VOI method. Multiplanar
MIP images enabled to assess each calyx from any preferred view without superposition
of neighbouring calices. The distal segments of the ureters were best seperated from
the bladder in the near lateral MIP images (Fig. 2 b, 3). Orally administered contrast material inside of bowel loops partially affected
the visibility of the ureters on MIP images. Calcifications of the iliac arteries
were present in some patients but did not simulate ureteral calculi at the site of
the uretero-iliac crossover (Fig. 4).
Fig. 3 Forty-three-year old man with uncharacteristic left-sided flank pain and hematuria.
The near lateral MIP image from furosemide-enhanced MS-CTU shows a calculus obstructing
the left ureterovesical junction. A mild dilatation of the ipsilateral urinary tract
is also visible. Multiplanar MIP images confirm that the calculus cannot be misdiagnosed
as a phlebolith.
Fig. 4 Sixty-seven-year old woman who sonographically presented with bilateral hydronephrosis.
The axial source image from furosemide-enhanced MS-CTU (a) discloses the true pathology caused by multiple parapelvic cysts (arrows). The MIP
image (b) most obviously demonstrates the pelvicaliceal compression and displacement on either
side (large arrows). Also note that the ureters are partly superimposed by calcified
iliac arteries.
MS-CTU was normal in 10 patients. A total of 16 pathologic findings involving the
upper urinary tract were diagnosed in 11 patients: dilatation (n = 4; Fig. 3), bifid ureter (n = 2; Fig. 5); caliceal deformities caused by chronic pyelonephritis (n = 2; Fig. 5); multiple parapelvic cysts with displacement of the pelvicaliceal systems (n = 2;
Fig. 4); nephrolithiasis without obstruction (n = 5; Fig. 2,5), and ureterolithiasis with mild obstruction (n = 1; Fig. 3). On MIP images from furosemide-enhanced MS-CTU, calculi could be well differentiated
from the adjacent contrast material-enhanced urine (Fig. 2,3,5). Phleboliths were also easily identified and could not be misdiagnosed as ureteral
calculi (Fig. 2,3). A transitional cell carcinoma was not found. Extrinsic affection of the ureters
was ruled out prior to surgery in the patient with a cervical carcinoma.
Fig. 5 Marked chronic pyelonephritic changes in a 43-year old female patient with a pelvicaliceal
duplex system of the right urinary tract. Furosemide-enhanced MS-CTU shows distinct
bilateral pelvicaliceal deformities including two, previously unknown calculi in the
right lower calices (thin arrows) and severe parenchymal atrophy of the left upper
pole (thick arrow). The right bifid ureter (arrowheads) is not completely opacified
(score = 2).
Discussion
Discussion
With the development of the multi-slice technology, helical computed tomography has
experienced several important advances, which also form the basis for innovations
in CT urography. With use of MS-CT, the scan time is markedly shortened by the simultaneous
acquisition of, currently, up to four slices, which enables us to carry out a thin-collimation
scan protocol of the entire urinary tract in a breath-hold of only 12 to 16 seconds.
This aspect does not only help to improve the patients’ cooperation during the breath-hold
data acquisition. Potentially more important is that MS-CT increases the resolution
of the CT data in the axial direction (z-resolution) [4], which especially improves the postprocessing capabilities of the CT data in the
manner that we are now able to reconstruct high-quality multiplanar urographic MIP
images.
Furthermore, we could demonstrate for MS-CTU that, in analogy to contrast-enhanced
MR urography [5], the additional injection of a low-dose diuretic proved to be beneficial for obtaining
accurate depiction of the urinary tract morphology. In patients with normal or moderately
impaired renal excretory function, the injection of only 10 mg of furosemide constantly
achieved a rapid and uniform distribution of the radiopaque contrast material inside
the entire upper urinary tract. In contrast to the conventional CTU technique used
by McNicholas et al. [3], no additional contrast material injection was necessary using the furosemide-enhanced
method. In most of our patients, the ureters were completely visualized although an
inconvenient compression device was not used. Low-dose furosemide only led to a mild
distension of the nonobstructed urinary tracts. Even fine anatomic details such as
caliceal fornices could be visualized on diuretic-enhanced MS-CT urograms. McNicholas
et al. reported in their series a partial loss of caliceal detail at conventional
window settings due to artefacts from dense contrast material [3]. Such a caliceal „pseudo-blunting”, probably caused by too high papillary and forniceal
concentration of contrast material, was not observed with use of low-dose furosemide
in our series. Loss of caliceal detail was, however, found on our MS-CT urograms obtained
using exclusively physiologic saline solution. We could demonstrate that the average
radiopacity inside the enhanced pelvicaliceal systems was 4 - 5 times higher and more
inhomogeneous when physiologic saline solution was used instead of furosemide. On
the other hand, the furosemide induced dilution of the excreted amount of contrast
material and also the increase in urine flow did not destroy the contrast enhancement
of the collecting systems, as it may occur in conventional intravenous pyelography.
The dilution of the contrast medium also explains for the finding that on MIP images,
calculi could still be differentiated inside the opacified urine.
Based on our initial findings obtained from a limited number of patients, we do not
claim that the use of low-dose furosemide is the only way to provide high-quality
CT urograms. Comparative studies are necessary to find out, if the sole infusion of
physiologic saline solution may be as effective as furosemide in achieving the same
accurate enhancement of the urinary collecting system. The use of only 10 mg of furosemide
is, at least, as simple and harmless as an infusion of 250 - 500 ml of physiologic
saline solution. From our experience, it is advantageous that in patients with normal
or moderately impaired excretory function, the injection of low-dose furosemide allows
good control over the extent and the timing of the hydration of the urinary tract
which cannot be achieved with use of saline solution alone.
In the average patient, an mAs-value of 100, or probably even less, should provide
a sufficient contrast between the opacified urinary tract and the retroperitoneum,
which also helps to keep down the radiation dose. Compared to a standard abdominal
CT scan, a dose-reduction in CTU is possible because the enhanced urine as a ‘high-contrast
region-of-interest’ allows to reconstruct excellent MIP images even if the source
data were acquired at low mAs-values. Moreover, in order to avoid additional radiation
exposure to the patient, it is important for the practicability of CTU that the completely
opacified urinary tract be imaged by only one acquisition, as far as possible. For
this purpose, it proved useful to check the bilateral contrast enhancement of the
renal pelvis with a low-dose ‘single-slice test image’ before starting the definite
CTU scan. In our patients, the diuretic-enhanced technique allowed us to complete
the examination after a single MS-CTU scan within 10 minutes of contrast material
injection. Delayed scans were not necessary in our series. However, delayed scans
have to be considered in patients with asymmetrical renal excretion, in whom it will
be impossible to get a sufficient urographic overview by acquisition of only a single
CT scan.
Another limitation may occur in those applications in which additional oral or rectal
administration of contrast material is required. Opacified bowel loops can superimpose
parts of the urinary tract and, thus, may interfere with the MIP reconstruction algorithm.
Anyway, postprocessing of our large MS-CT data sets, including up to 400 source images,
was still time consuming, which was almost exclusively attributed to the limited computing
rate of the VOI tool used to eliminate bones, calcified vessels, and bowel loops from
MIP images.
Nevertheless, our preliminary results demonstrate that the multi-slice scanning technology
and the combined use of contrast material and low-dose furosemide are effective means
to improve the feasibility of CT urography in patients who tolerate the application
of radiopaque contrast agents. Diuretic-enhanced MS-CTU allows us to obtain detailed
views of the upper urinary tract morphology and together with the diagnostic impact
of the axial CT source images, it now appears hard to find any aspect that would favor
the preference of standard intravenous pyelography. In the clinical routine, CTU can
be easily integrated as part of a multiphase CT protocol of the kidneys, especially,
in patients presenting with hematuria of uncertain cause. Because multislice CT provides
excellent image resolution, it is allowed to assume that MS-CTU suggests a clinical
potential for the detection of early-stage intrinsic tumors inside the non-dilated
upper urinary tract. Furthermore, MS-CTU can also be combined with the examination
of pelvic organs in patients with gynecologic tumor disease prior to surgical resection.
There is no doubt that in acute urolithiasis, unenhanced CT represents the examination
of primary choice. MS-CTU can be applied to those diagnostic problem cases, in which
unenhanced CT cannot safely answer the question whether a calcified structure is located
endoluminally or not. MS-CTU now has reached the state, in which the technique rather
competes with MR urography than with conventional pyelography.
Acknowledgement
Acknowledgement
The authors thank Roman Iwa for his support in the postprocessing of the CTU data.
