-
EDUCATION EXHIBIT 1609
Transitional Cell Carci-noma of the Upper Uri-nary Tract:
Spectrumof Imaging Findings1
LEARNINGOBJECTIVESFOR TEST 5After reading thisarticle and
takingthe test, the reader
will be able to:
� Identify the typicaland atypical appear-ances of upper
uri-nary tract TCC withconventional imagingmodalities (EU, RP,US)
and emergingCT and MR imagingtechniques.
� Describe the com-prehensive, mul-tiphasic CT and MRimaging
techniquesfor one-stop evalua-tion of upper urinarytract TCC.
� Discuss the com-bined radiologic andendourologic ap-proach to
staging andfollow-up of upperurinary tract TCC.
Ronan F. J. Browne, MB, BCh, MSc, FFRRCSI ● Conor P. Meehan,MB,
BCh, MRCPI ● Jane Colville, MB, BCh ● Raymond Power, MB,BCh, FRCSI
● William C. Torreggiani, MB, BCh, MRCPI, FRCR,FFRRCSI
Transitional cell carcinoma (TCC) accounts for up to 10% of
neo-plasms of the upper urinary tract and usually manifests as
hematuria.Imaging plays an important role in assessment of upper
tract disease,unlike in bladder TCC, diagnosis of which is usually
made at cystos-copy. Traditional imaging modalities, such as
excretory urography,retrograde pyelography, and ultrasonography,
still play pivotal roles indiagnosis of upper tract TCC, in
combination with endourologic tech-niques. The multicentric nature
of TCC makes assessment of the en-tire urothelium essential before
treatment. The advent of minimallyinvasive surgery, which allows
renal preservation in selected patients,makes accurate tumor
staging mandatory to determine the appropriatetherapy; staging is
usually performed with computed tomography (CT)or magnetic
resonance (MR) imaging. Vigilant urologic and radiologicfollow-up
is warranted to assess for metachronous lesions and recur-rence.
The emerging technique of CT urography allows detection ofurinary
tract tumors and calculi, assessment of perirenal tissues,
andstaging of lesions; it may offer the opportunity for one-stop
evaluationin the initial assessment of hematuria and in follow-up
of TCC. SimilarMR imaging protocols can be used in patients who are
not candidatesfor CT urography, although detection of urinary tract
calcificationsmay be suboptimal.©RSNA, 2005
Abbreviations: EU � excretory urography, MIP � maximum intensity
projection, RP � retrograde pyelography, TCC � transitional cell
carci-noma, 3D � three-dimensional
RadioGraphics 2005; 25:1609–1627 ● Published online
10.1148/rg.256045517 ● Content Code:
1From the Departments of Radiology (R.F.J.B., C.P.M., J.C.,
W.C.T.) and Urology (R.P.), Adelaide and Meath Hospital, Tallaght,
Dublin 24, Ire-land. Presented as an education exhibit at the 2003
RSNA Annual Meeting. Received March 30, 2004; revision requested
May 12; final revision re-ceived April 21, 2005; accepted April 22.
All authors have no financial relationships to disclose. Address
correspondence to W.C.T. (e-mail:[email protected]).
©RSNA, 2005
Radio
Gra
phic
s
CME FEATURESee accompanying
test at http://www.rsna.org
/education/rg_cme.html
-
IntroductionTransitional cell carcinoma (TCC) is
commonlyencountered in the urinary bladder and is usuallydiagnosed
at cystoscopy. Five percent of urothe-lial tumors arise from the
ureter or the renal pelvisor calices, accounting for approximately
10% ofupper tract neoplasms (1,2). Patients with TCCtypically
present with hematuria, which may befrank or microscopic. Up to
one-third of patientspresent with flank pain or acute renal colic,
symp-toms more typically associated with calculi. Occa-sionally,
tumors may manifest with distant metas-tases or be discovered
incidentally at radiologicexamination.
Renal TCC most frequently arises in the extra-renal part of the
pelvis, followed by the infun-dibulocaliceal region (3). The
distribution isequal between the left and right kidneys, with2%–4%
of cases occurring bilaterally. Twenty-five percent of upper tract
tumors occur in theureter, where 60%–75% of cases are found in
thelower third, with no side predominance (2). Tu-mor spread occurs
by mucosal extension or local,hematogenous, or lymphatic invasion.
The mostcommon sites for metastases are the liver, bone,and lungs.
The tumor stage at diagnosis influ-ences the development of local
recurrence andmetastases and hence overall survival (1,4).
Mul-ticentric TCC is common and is associated withpoor survival
(4). Synchronous or metachronoustumor of the ipsilateral or
contralateral collectingsystem is also common, necessitating
vigilant uro-logic and radiologic follow-up.
This article reviews the characteristic imagingfeatures of upper
tract TCC and outlines the roleof imaging in diagnosis,
preoperative staging, andfollow-up. The appearances at excretory
urogra-phy (EU), retrograde pyelography (RP), and re-nal
ultrasonography (US) are reviewed. Emphasisis placed on
cross-sectional imaging modalitiessuch as computed tomographic (CT)
urography,which is fast becoming the investigation of choicein the
assessment of patients with suspectedTCC. The emerging role of
magnetic resonance
(MR) imaging techniques such as gadolinium-enhanced
three-dimensional (3D) MR angiogra-phy and urography is also
discussed.
PathogenesisUpper tract TCC typically occurs in the 6th and7th
decades of life, with males affected threetimes more often than
females (3). Besides in-creasing age and male gender, the most
importantrisk factor is smoking, with smokers being two tothree
times more likely to develop TCC thannonsmokers (3). Chemical
carcinogens (ani-line, benzidine, aromatic amine, azo dyes),
cyclo-phosphamide therapy, and heavy caffeine con-sumption are also
associated with TCC, and allpredispose to synchronous and
metachronoustumor development (2). These substances aremetabolized
and excreted in the urine as carcino-genic substances that act
locally on the urothe-lium. Stasis of urine and structural
abnormalitiessuch as horseshoe kidney are also associated
withincreased prevalence (5).
Upper tract TCC is common in families af-fected with “Balkan
endemic nephropathy.” Fa-milial metabolic abnormalities in these
patientslead to tubulointerstitial nephritis, renal
failure,carcinogenic glomerulotubular toxins, and mul-tiple tumors
(2). Analgesic abuse, particularlylong-term use of phenacetin,
produces capillo-sclerosis and predisposes to a highly invasive
typeof TCC that preferentially involves the renal pel-vis (5).
Human papilloma virus and hereditarynonpolyposis colon cancer have
also been sug-gested as risk factors for TCC of the upper tract,and
the prevalence is significantly higher in areaswhere endemic
“blackfoot disease” is seen (2).
Pathologic FeaturesUpper tract TCC is histologically and
cytologi-cally similar to bladder TCC (6). Eighty-five per-cent of
upper tract TCCs are low-stage, superfi-cial, papillary neoplasms
with a broad base andfrondlike morphologic structure (7). These
tu-mors are usually small at diagnosis, grow slowly,and follow a
relatively benign course (8). Pedun-culated or diffusely
infiltrating tumor is less com-mon, accounting for approximately
15% of upper
1610 November-December 2005 RG f Volume 25 ● Number 6
Radio
Gra
phic
s
-
tract TCCs, but tends to behave more aggres-sively and be more
advanced at diagnosis (9). In-filtrating tumors are characterized
by thickeningand induration of the ureteric or renal pelvic wall.If
the renal pelvis is involved, there is often inva-sion into the
renal parenchyma. However, thisinfiltrative growth pattern
preserves renal contourand differs from renal cell carcinoma, which
istypically expansile.
Synchronous bilateral TCC has been reportedto occur in 1%–2% of
cases of renal lesions and2%–9% of cases of ureteric lesions.
Eleven per-cent to 13% of patients with upper tract TCCsubsequently
develop metachronous upper tracttumors (3). Furthermore, up to 50%
of patientsinitially presenting with upper tract TCC will de-velop
metachronous tumors in the bladder, typi-cally developing within 2
years of surgical treat-ment and seen more commonly with ureteric
tu-mors than with renal tumors (2,10). Two percentof patients with
bladder TCC also have synchro-nous upper tract tumors at
presentation, and 6%will develop metachronous upper tract
disease(11).
DiagnosisThe evaluation of hematuria requires assessmentof the
entire urothelium and the renal paren-chyma for tumor and of the
urinary tract for cal-culi. The standard work-up for these patients
asrecommended by the American Urological Asso-ciation consists of
urinalysis and cytologic analy-sis, cystoscopy, and EU (12,13). The
initial diag-nosis of TCC is usually made on the basis of find-ings
from urine cytology; the diagnostic yield isimproved with selective
lavage and collection andwith brush biopsies performed at
cystoscopy orRP (14,15). However, these techniques are inva-sive
and technically demanding. The limitationsof EU in assessing the
renal parenchyma usuallyrequire the supplemental use of US, CT, or
MRimaging to evaluate the kidneys for masses (16–18). Furthermore,
additional imaging is oftenrequired to clarify indeterminate
findings at EU.
Recently, the technique of multiphasic CTurography has emerged
as an alternative methodof assessing patients with hematuria,
offering su-
perior detection of urinary calculi and renal pa-renchymal
masses, and in some studies, improveddetection of urothelial
lesions. Because surround-ing structures can also be assessed, CT
urographyis rapidly replacing EU as the definitive study forthese
patients, potentially shortening the durationof diagnostic
evaluation. MR imaging, includingthe newer techniques of MR
angiography andMR urography, is also being used, particularly
inpatients who cannot tolerate iodinated contrastmaterial and in
whom multiplanar, vascular, andcollecting system imaging is
required. Because ofthe multifocal and metachronous nature of
TCC,thorough assessment of the entire urothelium isrequired before
treatment. Therefore, evaluationof the upper tract with EU (or CT
urography ifequivalent) is indicated in those with newly diag-nosed
bladder TCC; conversely, patients withupper tract TCC should
undergo cystoscopicevaluation.
Excretory UrographyThe diagnosis of upper tract TCC is most
fre-quently made at EU in patients undergoing inves-tigation for
hematuria. EU remains the noninva-sive method of choice for imaging
the detailedanatomy of the pelvicaliceal system and
ureters(15,17–19), although this is likely to change asCT urography
becomes more refined and ac-cepted as a primary diagnostic
investigation. Theappearances of upper tract lesions at EU are
welldescribed. Calcification may be visualized on con-trol
radiographs but is uncommon, occurring in2%–7% of tumors, and, when
present, maymimic urinary tract calculi (3). Enlargement ofthe
kidney may be seen with a large infiltratingtumor or a ureteric
tumor causing prolonged ob-struction.
Renal TCC usually manifests as a filling defectwithin the
contrast-enhanced collecting system,which may be single or multiple
and smooth, ir-regular (Fig 1), or stippled. The stipple sign
refersto tracking of contrast material into the intersticesof a
papillary lesion (Fig 2) (2,3). However, this
RG f Volume 25 ● Number 6 Browne et al 1611
Radio
Gra
phic
s
-
sign may also be seen with blood clot and fungusballs and should
be interpreted with caution.Stricture-like lesions of the
pelvicaliceal systemmay be evident and, if multiple, may mimic
renaltuberculosis (5). Filling defects within dilatedcalices may
occur secondary to tumor obstructionof the infundibulum and may
lead to caliceal“amputation” (Fig 3). Tumor-filled,
distendedcalices have been called “oncocalices.” If thesefail to
opacify with contrast material, they areknown as “phantom
calices.”
Ureteric TCC is typically seen as single ormultiple ureteric
filling defects with or withoutsurface stippling and proximal
ureteric dilatation.It is important to remember that
long-standingtumor obstruction of the ureteropelvic junction
orureter may lead to generalized hydronephrosisand poor excretion.
This is a major disadvantageof EU when compared with CT urography,
whichallows assessment of nonfunctioning kidneys. Up-per tract
filling defects may be nonspecific at EU,and obstruction of
pelvicaliceal drainage may ob-scure distal synchronous ureteric
tumors, mean-ing that RP is usually performed to further
assessthese patients.
Retrograde PyelographyRP is usually performed during cystoscopy
or tofurther characterize abnormalities detected atEU, in
inadequately excreting kidneys, or in cases
of contrast material allergy. Although invasive,RP allows
confirmation of the radiologic diagno-sis while also facilitating
ureterorenoscopy withbiopsy or brushing and cytologic examination
oflocalized urine collections. As with EU, renalTCC typically
appears as an intraluminal fillingdefect, which may be smooth,
irregular, orstippled. Opacification of a tumor-involved calix
Figures 1, 2. (1) TCC of the renal pelvis in a 60-year-old man
with painless hematuria. Fifteen-minute EU imageshows a large
irregular filling defect (arrow) involving the right renal pelvis
and extending into the lower pole calicealsystem. (2) TCC of the
renal pelvis in a 65-year-old man. Fifteen-minute EU image shows a
large stippled filling de-fect involving the collecting system of
the right kidney (arrow).
Figure 3. TCC of the upper pole collecting system in
a55-year-old woman. Fifteen-minute EU image showsamputation of the
upper pole calix (arrow) secondary toTCC.
1612 November-December 2005 RG f Volume 25 ● Number 6
Radio
Gra
phic
s
-
may show irregular papillary or nodular mucosa(Fig 4). If TCC
involves an infundibulum, thenan “amputated” calix may be seen with
or with-out focal hydronephrosis and calculi secondary tourinary
stasis (Fig 5). Tumor-filled, distendedcalices are known as
“oncocalices.”
Ureteric TCC classically appears as a solitary,polypoid filling
defect with ureteric dilatation
proximal to the lesion (Fig 6). The ureter itselfmay
occasionally be fixed by diffuse ureteric wallinfiltration from an
intramural lesion. An “applecore” appearance may be observed with
eccentricor encircling ureteric lesions (Fig 7). Malignant
Figure 4. Renal TCC in a 67-year-old woman. RPimage shows a
diffuse infiltrating TCC involving theright lower pole calix with
irregularity of the involvedmucosa (arrow).
Figure 5. Renal TCC in a 67-year-old man with he-maturia. RP
image shows amputation of the upper polecalix due to TCC (straight
arrow). Multiple calculi arealso seen within the lower pole and
interpolar calices(curved arrows).
Figure 6. Ureteric TCC in a 62-year-old man. RPimage shows an
irregular stricture (arrow) with hy-droureter proximal to the site
of the tumor.
Figure 7. Ureteric TCC in a 68-year-old woman. RPimage shows a
long irregular stricture of the left distalureter with proximal
hydronephrosis and “shoulder-ing” (arrow).
RG f Volume 25 ● Number 6 Browne et al 1613
Radio
Gra
phic
s
-
ureteric strictures may be circumferential or ec-centric and can
occasionally be confused withbenign strictures (Fig 8), although
ureteric fixa-tion and nontapering margins are suggestive
ofmalignancy (3). At RP, localized ureteric dilata-tion around and
distal to the filling defect maygive rise to the “goblet” sign (Fig
9), which oc-curs due to slow tumor growth with resultant lu-men
expansion and is not characteristic of moreacute causes of
obstruction.
UltrasonographyCurrently, renal US is frequently requested in
theevaluation of patients with hematuria to assess forrenal
parenchymal masses. However, US is not assensitive as CT in
identifying or characterizingrenal masses (16,20–22); as CT
urographyemerges as an initial imaging investigation for
he-maturia, US will likely play a limited diagnosticrole in the
future. US can be useful in patientswith renal functional
impairment or allergy toiodinated contrast material, although MR
imag-ing is becoming established as the investigation ofchoice in
these patients. US can also allow assess-
ment of the degree of hydronephrosis and guideinterventional
procedures in the setting of acuteobstruction.
At US, renal pelvic TCC typically appears as acentral
soft-tissue mass in the echogenic renalsinus, with or without
hydronephrosis (Figs 10,11) (2). TCC is usually slightly
hyperechoic rela-tive to surrounding renal parenchyma;
occasion-ally, high-grade TCC may show areas of mixedechogenicity
(Fig 12). Infundibular tumors maycause focal hydronephrosis.
Although lesions mayextend into the renal cortex and cause focal
con-tour distortion, typically TCC is infiltrative anddoes not
distort the renal contour (3).
US has a limited role in the evaluation of ure-teric TCC as the
ureters are rarely visualized intheir entirety, even if dilated. If
visualized, thesetumors are typically intraluminal
soft-tissuemasses with proximal distention of the ureter(23). US
also allows limited assessment of peri-ureteric tissues. Recent
developments in high-resolution endoluminal US performed
duringureterorenoscopy have shown promise in theevaluation of upper
tract TCC, offering potentialadvantages over other imaging
techniques, andmay assume a more prominent role in future
di-agnosis (2,24).
Figure 8. TCC of the pelviureteric junction in a 76-year-old man
with painless hematuria. RP image showsconcentric narrowing of the
pelviureteric junction (ar-row) with hydronephrosis. This
appearance simulatesbenign disease but was shown to represent TCC
at his-tologic analysis.
Figure 9. Ureteric TCC in a 58-year-old man. RPimage shows a
duplex right collecting system. TCC inthe midportion of the ureter
draining the upper polemoiety produces the characteristic goblet
sign (straightarrow). Another tumor is identified in the ureter
drain-ing the lower pole moiety (curved arrow).
1614 November-December 2005 RG f Volume 25 ● Number 6
Radio
Gra
phic
s
-
Computed TomographyCT is well established in the preoperative
stagingand assessment of upper tract TCC. CT has alsobeen shown to
be more sensitive than either USor EU in the detection of small
renal mass lesions
and urinary tract calculi (15,25–30). The recentadvent of CT
urography, offering single breath-hold coverage of the entire
urinary tract, im-proved resolution, and the ability to capture
mul-tiple phases of contrast material excretion, offersimproved
diagnostic potential over EU and US inthe assessment of patients
with hematuria due tocalculi or tumor (16,18,29,30). Recent
studieshave also shown higher detection rates for upperand lower
tract urothelial malignancies with CTurography over EU (17,31).
Although the Ameri-can College of Radiology still recommends EU
inthe investigation of hematuria, as CT urographybecomes more
prevalent it is likely to become theinvestigation of choice, as the
urothelium, renalparenchyma, and perirenal tissues can be
assessedat a single examination.
Typically, CT urography consists of a mul-tiphasic helical CT
protocol (Fig 13). A preen-hancement scan is initially performed
from theupper pole of the kidney to the lower edge of thesymphysis
pubis to exclude urinary tract calculi.A late arterial, early
corticomedullary phase scanof the kidney and lower pelvis,
beginning 15–25seconds after contrast material infusion, allows
Figures 10, 11. (10) Renal TCC in a 59-year-old woman. Sagittal
US scan shows a tumor (arrows) in the echogenicrenal sinus. Tumor
tissue is more echogenic than the surrounding renal cortex but less
echogenic than renal sinus fat.(11) Renal TCC in a 55-year-old man.
Sagittal US scan of the left kidney shows a tumor in the upper pole
(arrow).
Figure 12. Renal TCC in a 65-year-old woman. Sag-ittal US scan
shows a large mass of mixed echogenicity(arrows) involving the
upper pole calix and overlyingrenal parenchyma. At histologic
analysis, the mass wasshown to represent high-grade TCC.
RG f Volume 25 ● Number 6 Browne et al 1615
Radio
Gra
phic
s
-
1616 November-December 2005 RG f Volume 25 ● Number 6
Radio
Gra
phic
s
-
evaluation for vascular abnormalities. In the inter-est of
decreasing radiation exposure and time ofexamination, however, this
scan may be omittedunless a vascular abnormality is suspected.
Anephrographic phase scan of the kidney, per-formed 80–140 seconds
after contrast materialinfusion, allows assessment of the renal
paren-chyma. An excretory phase scan from the upperpole of the
kidney to the symphysis pubis, per-formed 4–8 minutes after
contrast material infu-sion, allows assessment of the urothelium.
Someauthors advocate a two-phase hematuria protocolwhere a
nephropyelographic phase is performedonly if the initial
nonenhanced scan does notdemonstrate a satisfactory cause for the
patient’shematuria (18).
Three-dimensional reformations typically in-clude thick and thin
slab coronal and sagittalMIPs for the kidneys, ureters, and
bladder, al-
though other 3D reformation techniques can beused. Coronal
reformation in particular demon-strates the longitudinal extent of
a lesion, allowsassessment for multicentric tumors, and
providesurologists with a familiar imaging format (Fig 13).Viewing
the opacified system at wider windowsettings such as bone windows
can also aid inidentifying and differentiating subtle lesions(Fig
13).
Unlike EU, imaging is not dependent on afunctioning kidney and
the tract distal to a lesioncan be evaluated. CT urography may
revealcauses of hematuria other than tumor or calculi,such as
papillary necrosis, inflammatory lesions,or infarcts. Precontrast
scans are necessary to de-tect calculi and obtain accurate
attenuation valuesfor nonopaque filling defects, whereas
postcon-trast scans aid in confirming lesion location andextent. On
precontrast images, TCC is typicallyhyperattenuating (5–30 HU) to
urine and renalparenchyma (Fig 13) but less attenuating thanother
pelvic filling defects such as clot (40–80HU) or calculus (�100
HU).
Renal TCC is typically seen as a sessile fillingdefect in the
excretory phase, which expands cen-trifugally with compression of
the renal sinus fat(Fig 14). Other appearances include
pelvicalicealirregularity, focal or diffuse mural thickening,
on-cocalix, and focally obstructed calices. Early tu-mors confined
to the muscularis are separatedfrom the renal parenchyma by renal
sinus fat orexcreted contrast material and have normal-ap-pearing
peripelvic fat (Fig 13). Advanced TCCextends into the renal
parenchyma in an infiltrat-ing pattern that distorts normal
architecture (Figs15, 16). However, reniform shape is typically
pre-served (Figs 17, 18), unlike in renal cell carci-noma.
Figure 14. TCC of the renal pelvis in a 66-year-oldman with
hematuria. Axial nephrographic phase CTscan shows a sessile filling
defect (arrow), which is typi-cal of renal pelvic TCC.
Š Figure 13. TCC of the renal pelvis in a 43-year-old man with
flank pain and hematuria. (a) Axial nonenhancedCT scan shows a mass
(arrow) in the right renal pelvis. The mass is slightly
hyperattenuating relative to the urine andrenal parenchyma. (b)
Axial nephrographic phase CT scan shows that the mass (arrow) has
characteristic early en-hancement, which is less than that of the
surrounding renal parenchyma. (c) Axial excretory phase CT scan
showsthe mass within the renal pelvis with surrounding excreted
contrast medium. (d) Axial excretory phase CT scan(bone window)
shows the lesion more clearly (arrow). (e) Coronal maximum
intensity projection (MIP) image showsthe tumor (arrow) in EU
format. (f) Detail of a coronal MIP image shows the lesion more
clearly (arrow).
RG f Volume 25 ● Number 6 Browne et al 1617
Radio
Gra
phic
s
-
Both TCC and renal cell carcinoma can showearly enhancement and
de-enhancement aftercontrast material administration (30). Renal
cellcarcinoma, being hypervascular, tends to enhancemore, although
the two tumors often cannot be
differentiated. Parenchymal invasion may be seenas a focal delay
in all or part of the cortical nephro-gram, although superimposed
pyelonephritis orobstruction alone can also have these
appear-ances. A large infiltrating renal TCC may occa-sionally
manifest with areas of necrosis and mustbe differentiated from
lymphoma, metastases,
Figure 15. TCC of the upper renal pole in a 61-year-old woman.
(a) Axial nonenhanced CT scan shows a mass(arrow) in the upper pole
calix of the left kidney. (b) Axial nephrographic phase CT scan
shows characteristicearly enhancement of the tumor with extension
into the surrounding upper pole parenchyma (arrow). (c) Axial
ex-cretory phase CT scan shows the diffuse tumor with a small
amount of excreted contrast medium centrally (arrow).(d) Coronal
MIP image shows the extent of the tumor (arrow) in EU format.
1618 November-December 2005 RG f Volume 25 ● Number 6Radio
Gra
phic
s
-
and xanthogranulomatous pyelonephritis, whichcan have a similar
appearance (3,5).
Hydronephrosis is the most frequent finding inureteric TCC and
hydroureter can often be seento the point of obstruction, where
Hounsfield unitvalues for attenuation and enhancement usually
allow differentiation of TCC from calculus andclot. Ureteric
wall thickening (eccentric or cir-cumferential), luminal narrowing,
or an infiltrat-ing mass are other features of disease (29).
Athickened enhancing ureteric wall with peri-ureteric fat stranding
is suggestive of extramuralspread (Fig 19).
Figure 16. Renal TCC in a 54-year-old man. Axialexcretory phase
CT scan shows TCC expanding cen-trifugally from the right renal
pelvis. Note the paren-chymal invasion with a delay in the cortical
nephro-gram (arrow).
Figure 17. Renal TCC in a 53-year-old man. Axialnephrographic
phase CT scan shows diffuse tumor in-filtration of the left kidney
with preservation of its reni-form contour.
Figure 18. Renal and ureteric TCC in a 76-year-oldwoman with
gross hematuria and flank pain. Axialnephrographic phase CT scan
shows extensive involve-ment of the left kidney and proximal ureter
with TCCand perinephric extension of the tumor (arrow). Notethe
preservation of the reniform contour.
Figure 19. Ureteric TCC in a 62-year-old man withright flank
pain. Axial nephrographic phase CT scanshows enhancing TCC in the
wall of the right ureter(arrow) with periureteric stranding and
tumor exten-sion.
RG f Volume 25 ● Number 6 Browne et al 1619Radio
Gra
phic
s
-
CT urography allows identification of lesionsat an early stage,
thereby allowing nephron-spar-ing surgery. Axial source images also
allow assess-ment of surrounding structures. Adequate disten-tion
and opacification are fundamental factors inthe thorough evaluation
of the urothelium at CTurography. The increased radiation exposure
isestimated at only 50%–80% over a complete EUseries (30). Although
reformatting and review ofmultiple images with different window
settings istime-consuming, CT urography has the potentialto stand
alone as a comprehensive “one-stop”frontline study for hematuria
and therefore detec-tion of TCC.
MR ImagingMR imaging is infrequently used in the
primaryassessment of upper tract TCC, and the MR im-aging
characteristics of this tumor are not welldescribed. In general, MR
imaging has not played
a leading role in renal tumor imaging due to limi-tations in
image quality, time-consuming se-quences, and susceptibility to
artifacts. Recently,however, the development of newer fast
se-quences has led to increasing use and MR imag-ing has been shown
to equal CT in the detectionand diagnosis of renal masses
(32–34).
MR imaging offers inherently high soft-tissuecontrast, is
independent of excretory function,and allows multiplanar imaging,
which permitsdirect image acquisition in the plane of tumorspread.
The coronal plane is often advantageousbecause it allows evaluation
of the kidneys, renalvessels, inferior vena cava, and spine in a
smallnumber of sections. As with CT, MR imaging can
Figure 20. Bilateral ureteric TCC in a 57-year-oldwoman. (a, b)
Coronal T2-weighted MR images (rep-etition time msec/echo time msec
� 1655/99.6) showlow-signal-intensity tumors in the distal right
(arrow ina) and distal left (arrow in b) ureters. Note the
high-signal-intensity urine surrounding the tumors. (c)
Axialgadolinium-enhanced T1-weighted MR image (616.7/10) obtained
with fat saturation shows enhancement ofthe right ureteric tumor
(large arrow). Note the gado-linium contrast material in the left
ureter (small arrow)above the tumor.
1620 November-December 2005 RG f Volume 25 ● Number 6
Radio
Gra
phic
s
-
demonstrate tumor involvement of the renal pa-renchyma,
perinephric tissues, or periureteric tis-sues and distant
metastases (35,36).
TCC has lower signal intensity than the nor-mally
high-signal-intensity urine on T2-weightedimages, permitting good
demonstration of tumorin a dilated collecting system (Fig 20).
However,TCC is nearly isointense to renal parenchyma onT1- and
T2-weighted images, meaning that gado-linium contrast material is
necessary for accurateassessment of tumor extent (3,33).
AlthoughTCC is a hypovascular tumor, moderate en-hancement is seen
with gadolinium contrast ma-terial, although not to the same degree
as renalparenchyma (37). Postcontrast imaging may beperformed by
using 3D sequences to allow dy-namic evaluation of the kidney (Fig
21). This al-lows assessment of the renal vasculature in arte-rial
and venous phases and of the renal paren-chyma in corticomedullary
and nephrographicphases. Vascular invasion of the renal vein or
infe-rior vena cava, although rare, may be demon-strated without
gadolinium contrast material byusing T2-weighted or gradient-echo
flow-sensi-tive sequences (Fig 22).
Figure 21. Renal TCC in a 68-year-old woman.(a, b) Nephrographic
phase (a) and excretoryphase (b) coronal gadolinium-enhanced 3D
fastlow-angle shot source MR angiograms (3.64/1.37)show a
moderately enhancing TCC (arrow) in theupper pole of the right
kidney. (c) Gadolinium-en-hanced 3D MIP MR angiogram shows the
tumormore clearly. Note the retroaortic segment of the leftrenal
vein.
Figure 22. Bilateral renal TCC in a 77-year-old manwith
hematuria. Axial T2-weighted fast spin-echo MRimage (8000/104)
obtained with fat saturation showsTCC in the upper poles of both
kidneys (arrows) withinvasion of the right renal vein and inferior
vena cava(arrowhead).
RG f Volume 25 ● Number 6 Browne et al 1621
Radio
Gra
phic
s
-
MR imaging evaluation of upper tract TCCshould include MR
urography, which may bestatic or may be dynamic by using
gadoliniumcontrast material (36). Static MR urography per-formed by
using heavily T2-weighted sequencescan permit accurate localization
of ureteric ob-struction (Fig 23), although imaging of
undilatedsystems may be suboptimal (19,38) (Fig 24). Dy-namic
gadolinium-enhanced T1-weighted MRurography performed with or
without a diureticovercomes this problem and allows delayed
ac-quisitions at various time intervals depending onthe degree and
level of obstruction (Fig 25). Thistechnique is helpful in patients
in whom urogra-phy with iodinated contrast material is not
pos-sible. Data postprocessing (eg, MIPs) allows 3Drotation and
evaluation of suspected areas of dis-ease without superimposition
of other structures.This can be performed for both vessels and
thecollecting system. The latter images, which re-semble
conventional EU images, are readily ac-ceptable to clinicians.
Figure 23. TCC of the renal pelvis in a 65-year-old man. Coronal
(a) and sagittal (b) heavilyT2-weighted (half-Fourier rapid
acquisition with relaxation enhancement) source MR
urograms(1500/116) show focal hydronephrosis and irregularity of
the upper pole and interpolar calices ofthe right kidney
(arrow).
Figure 24. Ureteric TCC in a 56-year-old woman.Coronal MIP
half-Fourier rapid acquisition with relax-ation enhancement MR
urogram (1500/116) showshydronephrosis and a filling defect due to
a tumor inthe mid left ureter (arrow). Note the poor demonstra-tion
of the nondilated right collecting system.
1622 November-December 2005 RG f Volume 25 ● Number 6
Radio
Gra
phic
s
-
Ureteric TCC is isointense to muscle on T1-weighted images and
slightly hyperintense on T2-weighted images (39). At MR urography,
uretericTCC typically appears as an irregular mass,whereas calculi
appear as sharply delineated fill-ing defects, although
differentiation betweensmall calculi and tumor may be difficult
(19). Tu-mor enhancement after administration of gadolin-ium
contrast material can also help distinguish itfrom calculi (Fig
20). Soft-tissue stranding in theperiureteric fat is suggestive of
periureteric exten-sion, although prior surgery, radiation, and
in-flammation can also give these appearances. MRimaging may help
differentiate these entities,however, as fibrosis will appear
hypointense onT2-weighted images, particularly in
long-standingcases (39).
These comprehensive MR protocols can imageall the anatomic
components of the urinary tractin a single test and offer
advantages over othertechniques, including lack of iodinated
contrastmedium and radiation exposure. Although MRimaging remains
second line to CT, it offers fur-ther noninvasive imaging of masses
that are notadequately characterized with other imaging mo-
dalities (36). The main disadvantage of MR im-aging is the
inability to reliably detect urinarytract calcifications, calculi,
and air, which limitsits use as a first-line test in the
investigation ofhematuria. Although the sensitivity of renal
pa-renchymal MR imaging with gadolinium for as-sessing renal masses
and abnormalities of thenephrogram is considered similar to that of
CT,spatial resolution is poor compared with that ofintravenous
urography or CT urography, makingdetection of subtle urothelial
malignancies lesslikely (17). Furthermore, complete
characteriza-tion of renal masses may require multiple
time-consuming sequences before and after adminis-tration of
gadolinium contrast material (17,18).
Staging and TreatmentThe tumor stage at diagnosis influences the
devel-opment of local recurrence and metastases andhence overall
survival (1,4). Furthermore, treat-ment and prognosis are largely
determined by thedepth of tumor infiltration, the degree of
lymph
Figure 25. Ureteric TCC in a 68-year-old man with hematuria. (a)
Coronal gadolinium-enhanced 3D fast low-angle shot MR urogram
(3.64/1.37) obtained at 15 minutes shows hydronephrosis and
proximal hydroureter to thelevel of the mid ureter. An enhancing
tumor is seen in the ureteric wall at this level (thick arrow). No
gadolinium con-trast material is seen in the left collecting
system. Deep vein thrombosis of the left femoral vein is
incidentally noted(thin arrow). (b) Coronal gadolinium-enhanced 3D
fast low-angle shot MR urogram (3.64/1.37) obtained at 2 hoursshows
the site of ureteric obstruction due to TCC (arrow).
RG f Volume 25 ● Number 6 Browne et al 1623
Radio
Gra
phic
s
-
node and distant metastases, and the histologictumor type,
making exact staging imperative(Tables 1, 2).
Conventional imaging methods such as EUand RP cannot demonstrate
extension into theperipelvic or periureteric fat or metastases.
Cross-sectional imaging with CT or MR is now rou-tinely employed in
the presurgical work-up ofthese patients. These techniques can
demonstrateintra- and extrarenal local extension of tumor andthe
presence of nodal or distant metastases with ahigh degree of
accuracy. They are used in con-junction with ureterorenoscopy and
biopsy forstaging before surgery.
CT has become routine in the further charac-terization of upper
tract lesions demonstratedwith other modalities and, despite
varying reportson staging accuracy, is currently the
preoperativeimaging modality of choice (9,15,40) (Table 1).As
studies show higher detection rates for urothe-lial malignancies
with CT urography than withEU (29,31), this technique is being
advocated asa one-stop diagnostic and staging assessment
ofsuspected urothelial malignancy. Although CTdoes not allow
distinction between stage 0–II tu-
mors, it does allow differentiation of early-stageTCC confined
to the collecting system wall fromadvanced disease with local
extension or distantmetastases, which is important for defining
surgi-cal management (8,41). Early-stage tumors (stage0–II)
confined to the muscularis are separatedfrom the renal parenchyma
by renal sinus fat orexcreted contrast material and have
normal-ap-pearing peripelvic fat (Fig 13).
More-advanced tumors infiltrating beyond themuscularis into the
peripelvic fat typically showincreased, inhomogeneous peripelvic
attenuation(Fig 18), although this finding may also be seenwith
superimposed infection, hemorrhage, or in-flammation and should be
interpreted with cau-tion to avoid overstaging (39). Metastatic
spread
Table 1TNM Classification of Renal TCC
Stage Histopathologic Findings
Tx Primary tumor cannot be assessedT0 No evidence of a primary
tumorTa Papillary noninvasive carcinomaTis Carcinoma in situT1
Tumor invades subepithelial connective tissueT2 Tumor invades the
muscularisT3 Tumor invades beyond the muscularis into the
periureteric fat or renal parenchymaT4 Tumor invades adjacent
organs, the pelvic or abdominal wall, or through the kidney into
perinephric fatNx Regional lymph nodes cannot be assessedN0 No
regional lymph node metastasisN1 Metastasis in a single lymph node
�2 cm in greatest dimensionN2 Metastasis in a single lymph node �2
cm but �5 cm in greatest dimension or in multiple lymph nodes
�5 cm in greatest dimensionN3 Metastasis in a lymph node �5 cm
in greatest dimensionMx Distant metastasis cannot be assessedM0 No
distant metastasisM1 Distant metastasis
Table 2Histopathologic Grading of Renal TCC
Grade Histopathologic Findings
Gx Grade of differentiation cannot beassessed
G1 Well-differentiated tumorG2 Moderately differentiated
tumorG3, G4 Poorly differentiated or undifferentiated
tumor
1624 November-December 2005 RG f Volume 25 ● Number 6
Radio
Gra
phic
s
-
via urinary or hematogenous routes usually mani-fests as
multifocal mucosal nodules or wall thick-ening, whereas direct
invasion produces a short orlong stricture (22). Extrarenal spread
can occur ator through the renal hilum, and common sites
ofmetastases include the lungs, retroperitoneum,lymph nodes, and
bones (40). Rarely, invasion ofthe renal vein or inferior vena cava
is seen and canbe well demonstrated with comprehensive CTurography
protocols. The overall accuracy of CTin predicting the pathologic
stage ranges from36% to 83% in the literature (2), which meansthat
ureterorenoscopy and biopsy remain essen-tial additional tools.
As with CT, MR imaging can demonstratetumor involvement of the
renal parenchyma, peri-nephric fat, or periureteric fat and distant
metas-tases. It therefore offers an alternative staging mo-dality
and has been shown to allow accurate stag-ing of TCC lesions larger
than 2 cm (Fig 22)(37). As with CT, however, limitations exist
indetecting superficial invasion of the renal paren-chyma and small
lesions may be missed becauseof motion artifacts (2,37). It is the
preferred stag-ing examination in patients who cannot
tolerateiodinated contrast material and in whom multi-planar and
vascular imaging is required for preop-erative assessment.
The traditional treatment of upper tract TCCinvolves total
nephroureterectomy with excisionof the ipsilateral ureteric orifice
and a contiguouscuff of bladder tissue (25). However, the
develop-ment of endoscopic and minimally invasive surgi-cal
techniques allows renal preservation in se-lected patients,
particularly those with a solitarykidney, bilateral tumor, poor
renal function, low-grade tumor, or prohibitive surgical risk, with
re-sults comparable to those of radical surgery(2,25,42–45).
Accurate radiologic detection andstaging of tumor is therefore
essential to deter-mine appropriate surgical therapy, especially
ifconservative surgery is being considered, or theintensity of
chemotherapy for advanced-stage tu-mors (25,40).
Follow-upThere is still no widely accepted protocol for
theradiologic follow-up of patients with primaryTCC of the upper
urinary tract. Current data
suggest that routine follow-up imaging strategiesshould be
individually tailored on the basis of pri-mary tumor
characteristics (2). Annual EU is rec-ommended, especially in the
first 2 years afterinitial diagnosis (10,19,42). RP should be
soughtif EU fails to depict or adequately distend the en-tire upper
tract, especially if cystoscopy is beingperformed to assess the
bladder. This vigilance isjustified in order to detect early
recurrence afterconservative surgery or, in patients who have
onlyone remaining kidney, to detect contralateral le-sions at an
early stage when local excision may befeasible.
Owing to the high rate of metachronous tumorin the bladder,
frequent cystoscopy should also beperformed. At our institution
this is performedevery 3 months for the first year, every 6
monthsfor the second year, and yearly thereafter. In thefuture, CT
urography will likely become the pri-mary radiologic method of TCC
follow-up, al-lowing assessment of the entire urothelium andalso
facilitating virtual cystoscopy, although con-ventional cystoscopy
is still necessary for directvisualization and biopsy.
ConclusionsConventional imaging modalities such as EU,RP, and US
still play a key role in the assessmentof hematuria, in combination
with endourologictechniques. However, multiphasic CT
urographyoffers superior detection of calculi, urothelial tu-mor,
and parenchymal tumor over EU and USand allows accurate staging of
detected lesions atthe same examination. MR imaging, includingthe
newer techniques of MR angiography andMR urography, offers
comparable evaluation inpatients who cannot tolerate iodinated
contrastmaterial and in whom multiplanar, vascular, andcollecting
system imaging is required. Recogni-tion by the radiologist of the
variety of appear-ances of upper tract TCC with all imaging
mo-dalities is necessary to detect and stage tumorsaccurately. In
addition, atypical appearances, par-ticularly in advanced tumors,
should be recog-nized. Synchronous or metachronous tumors of
RG f Volume 25 ● Number 6 Browne et al 1625
Radio
Gra
phic
s
-
the urothelium are common, necessitating vigi-lant urologic and
radiologic follow-up. In the fu-ture, CT urography will likely
become the defini-tive radiologic examination for diagnosis and
fol-low-up in patients with suspected TCC.
References1. Hall MC, Womack S, Sagalowsky AI, Carmody T,
Erickstad MD, Roehrborn CG. Prognostic factors,recurrence, and
survival in transitional cell carci-noma of the upper urinary
tract: a 30-year experi-ence in 252 patients. Urology
1998;52:594–601.
2. Kirkali Z, Tuzel E. Transitional cell carcinoma ofthe ureter
and renal pelvis. Crit Rev Oncol Hema-tol 2003;47:155–169.
3. Wong-You-Cheong JJ, Wagner BJ, Davis CJ Jr.Transitional cell
carcinoma of the urinary tract:radiologic-pathologic correlation.
RadioGraphics1998;18:123–142.
4. Chan V, Pantanowitz L, Vrachliotis TG, RabkinDJ. CT
demonstration of a rapidly growing transi-tional cell carcinoma of
the ureter and renal pelvis.Abdom Imaging 2002;27:222–223.
5. Lee TY, Ko SF, Wan YL, et al. Unusual imagingpresentations in
renal transitional cell carcinoma.Acta Radiol
1997;38:1015–1019.
6. Melamed MR, Reuter VE. Pathology and stagingof urothelial
tumors of the kidney and ureter. UrolClin North Am
1993;20:333–347.
7. Barentsz JO, Jager GJ, Witjes JA, Ruijs JH. Pri-mary staging
of urinary bladder carcinoma: therole of MRI and a comparison with
CT. Eur Ra-diol 1996;6:129–133.
8. Urban BA, Buckley J, Soyer P, Scherrer A, Fish-man EK. CT
appearance of transitional cell carci-noma of the renal pelvis. I.
Early-stage disease.AJR Am J Roentgenol 1997;169:157–161.
9. Buckley JA, Urban BA, Soyer P, Scherrer A, Fish-man EK.
Transitional cell carcinoma of the renalpelvis: a retrospective
look at CT staging withpathologic correlation. Radiology
1996;201:194–198.
10. Keeley FX, Kulp DA, Bibbo M, McCue PA, Bag-ley DH.
Diagnostic accuracy of ureteroscopic bi-opsy in upper tract
transitional cell carcinoma.J Urol 1997;157:33–37.
11. Herranz-Amo F, Diez-Cordero JM, Verdu-Tar-tajo F,
Bueno-Chomon G, Leal-Hernandez F,Bielsa-Carrillo A. Need for
intravenous urographyin patients with primary transitional
carcinoma ofthe bladder? Eur Urol 1999;36:221–224.
12. Grossfeld GD, Litwin MS, Wolf JS, et al. Evalua-tion of
asymptomatic microscopic hematuria inadults: the American
Urological Association bestpractice policy panel. I. Definition,
detection,prevalence, and etiology. Urology 2001;57:599–603.
13. O’Malley ME, Hahn PF, Yoder IC, Gazelle GS,McGovern FJ,
Mueller PR. Comparison of excre-tory phase, helical computed
tomography withintravenous urography in patients with
painlesshaematuria. Clin Radiol 2003;58:294–300.
14. Millan-Rodriguez F, Palou J, de la Torre-Holguera P,
Vayreda-Martija JM, Villavicencio-Mavrich H, Vicente-Rodriguez J.
ConventionalCT signs in staging transitional cell tumors of
theupper urinary tract. Eur Urol 1999;35:318–322.
15. Scolieri MJ, Paik ML, Brown SL, Resnick MI.Limitations of
computed tomography in the pre-operative staging of upper tract
urothelial carci-noma. Urology 2000;56:930–934.
16. Gray Sears CL, Ward JF, Sears ST, Puckett MF,Kane CJ, Amling
CL. Prospective comparison ofcomputerized tomography and excretory
urogra-phy in the initial evaluation of asymptomatic
mi-crohematuria. J Urol 2002;168:2457–2460.
17. Kawashima A, Glockner JF, King BF Jr. CT urog-raphy and MR
urography. Radiol Clin North Am2003;41:945–961.
18. Maher MM, Kalra MK, Rizzo S, Mueller PR,Saini S.
Multidetector CT urography in imagingof the urinary tract in
patients with hematuria. Ko-rean J Radiol 2004;5:1–10.
19. Yousem DM, Gatewood OM, Goldman SM,Marshall FF. Synchronous
and metachronoustransitional cell carcinoma of the urinary
tract:prevalence, incidence, and radiographic detection.Radiology
1988;167:613–618.
20. Perlman ES, Rosenfield AT, Wexler JS, GlickmanMG. CT
urography in the evaluation of urinarytract disease. J Comput
Assist Tomogr 1996;20:620–626.
21. Joffe SA, Servaes S, Okon S, Horowitz M. Multi–detector row
CT urography in the evaluation ofhematuria. RadioGraphics
2003;23:1441–1455;discussion 1455–1456.
1626 November-December 2005 RG f Volume 25 ● Number 6
Radio
Gra
phic
s
-
22. Kim JK, Cho KS. CT urography and virtual en-doscopy:
promising imaging modalities for urinarytract evaluation. Br J
Radiol 2003;76:199–209.
23. Jung P, Brauers A, Nolte-Ernsting CA, Jakse G,Gunther RW.
Magnetic resonance urography en-hanced by gadolinium and diuretics:
a comparisonwith conventional urography in diagnosing thecause of
ureteric obstruction. BJU Int 2000;86:960–965.
24. Hadas Halpern I, Farkas A, Patlas M, Zaghal
I,Sabaq-Gottschalk S, Fisher D. Sonographic diag-nosis of ureteral
tumors. J Ultrasound Med 1999;18:639–645.
25. Tawfiek ER, Bagley DH. Upper-tract transitionalcell
carcinoma. Urology 1997;50:321–329.
26. Silverman SG, Lee BY, Seltzer SE, Bloom DA,Corless CL, Adams
DF. Small (� or � 3 cm) re-nal masses: correlation of spiral CT
features andpathologic findings. AJR Am J Roentgenol
1994;163:597–605.
27. Szolar DH, Kammerhuber F, Altziebler S, et al.Multiphasic
helical CT of the kidney: increasedconspicuity for detection and
characterization ofsmall (�3-cm) renal masses. Radiology
1997;202:211–217.
28. Fielding JR, Silverman SG, Rubin GD. HelicalCT of the
urinary tract. AJR Am J Roentgenol1999;172:1199–1206.
29. Caoili EM, Cohan RH, Korobkin M, et al. Uri-nary tract
abnormalities: initial experience withmulti–detector row CT
urography. Radiology2002;222:353–360.
30. Lang EK, Macchia RJ, Thomas R, et al. Improveddetection of
renal pathologic features on multipha-sic helical CT compared with
IVU in patients pre-senting with microscopic hematuria.
Urology2003;61:528–532.
31. McCarthy CL, Cowan NC. Multidetector CTurography (MD-CTU)
for urothelial imaging (ab-str). Radiology 2002;225(P):237.
32. Cholankeril JV, Freundlich R, Ketyer S, SpiritoAL,
Napolitano J. Computed tomography inurothelial tumors of renal
pelvis and related fillingdefects. J Comput Tomogr
1986;10:263–272.
33. Walter C, Kruessell M, Gindele A, BrochhagenHG, Gossmann A,
Landwehr P. Imaging of renallesions: evaluation of fast MRI and
helical CT.Br J Radiol 2003;76:696–703.
34. Kreft BP, Muller-Miny H, Sommer T, et al. Diag-nostic value
of MR imaging in comparison to CTin the detection and differential
diagnosis of renalmasses: ROC analysis. Eur Radiol
1997;7:542–547.
35. Semelka RC, Shoenut JP, Magro CM, KroekerMA, MacMahon R,
Greenberg HM. Renal cancerstaging: comparison of contrast-enhanced
CT andgadolinium-enhanced fat-suppressed spin-echoand gradient-echo
MR imaging. J Magn ResonImaging 1993;3:597–602.
36. Milestone B, Friedman AC, Seidmon EJ, RadeckiPD, Lev-Toaff
AS, Caroline DF. Staging of ure-teral transitional cell carcinoma
by CT and MRI.Urology 1990;36:346–349.
37. Pretorius ES, Wickstrom ML, Siegelman ES. MRimaging of renal
neoplasms. Magn Reson ImagingClin N Am 2000;8:813–836.
38. Weeks SM, Brown ED, Brown JJ, Adamis MK,Eisenberg LB,
Semelka RC. Transitional cell car-cinoma of the upper tract:
staging by MRI. Ab-dom Imaging 1995;20:365–367.
39. Rothpearl A, Frager D, Subramanian A, et al. MRurography:
technique and application. Radiology1995;194:125–130.
40. Urban BA, Buckley J, Soyer P, Scherrer A, Fish-man EK. CT
appearance of transitional cell carci-noma of the renal pelvis. II.
Advanced-stage dis-ease. AJR Am J Roentgenol 1997;169:163–168.
41. Bagley DH, Liu JB, Goldberg BB. The use of en-doluminal
ultrasound of the ureter. Semin Urol1992;10:194–198.
42. McCoy JG, Honda H, Reznicek M, Williams RD.Computerized
tomography for detection and stag-ing of localized and
pathologically defined uppertract urothelial tumors. J Urol
1991;146:1500–1503.
43. Keeley FX Jr, Bibbo M, Bagley DH. Uretero-scopic treatment
and surveillance of upper urinarytract transitional cell carcinoma.
J Urol 1997;157:1560–1565.
44. Chen GL, Bagley DH. Ureteroscopic manage-ment of upper tract
transitional cell carcinoma inpatients with normal contralateral
kidneys. J Urol2000;164:1173–1176.
45. Elliott DS, Segura JW, Lightner D, Patterson DE,Blute ML. Is
nephroureterectomy necessary in allcases of upper tract
transitional cell carcinoma?Long-term results of conservative
endourologicmanagement of upper tract transitional cell carci-noma
in individuals with a normal contralateralkidney. Urology
2001;58:174–178.
RG f Volume 25 ● Number 6 Browne et al 1627
Radio
Gra
phic
s
This article meets the criteria for 1.0 credit hour in category
1 of the AMA Physician’s Recognition Award. To obtaincredit, see
accompanying test at http://www.rsna.org/education/rg_cme.html.