Genitourinary Steven S. Eilenberg, Scott A. Mirowitz,

MD MD

#{149} Joseph #{149} V.

Marie

Renal Masses: Gd-DTPA-enhanced

K. T. Lee, MD Tartar, MD

Evaluation

Index terms: Cadolinium #{149} Kidney, cysts, 81.3111 #{149} Kidney, MR studies, 81.1214 #{149} Kidney neoplasms, diagnosis, 81.3141, 81.32 #{149} Kidney neoplasms, MR studies, 81.1214 #{149} Magnetic resonance (MR), contrast enhancement

‘From From

ary

the the

masses

14, 1990; RSNA,

MD

(5-7), of renal

RSNA

revision 1990

PATIENTS

of renal the do-

main of computed tomography (CT) and ultrasound. The diagnostic accuracy of these imaging modalities in the characterization of renal lesions approaches 100% (1-4). Magnetic resonance (MR) imaging has met with some success in the detection and characterization of renal cysts and neoplasms staging

#{149}

Gradient-Echo MR Imaging’

as well as in the cell carcinoma (8,9).

Institute

of Radiology,

assembly.

received

March

Received

16; accepted

METHODS

Patients Fifteen

ther

patients

a simple

neoplasm

with

renal

were

CT

cyst

evidence

or a solid

prospectively

of ei-

renal

examined

with the use of MR imaging from April 1989 to September 1989. The 10 men and five women had a mean age of 59 years (range, 25-80 years). One 55-year-old man with no identifiable risk factors had three

renal

cytoma;

cell

carcinomas

an additional

and

an

onco-

six carcinomas

simple renal cysts were identified in six patients (a seventh patient also had a small renal cell carcinoma). The single angiomyolipoma was discovered during chest CT in a 35-year-old woman with ad-

solid

els

neoplasms

and

renal

parenchy-

techniques

during

undertaken

vanced

to evaluate

the

enhancement patterns of simple renal cysts and solid neoplasms and to investigate the usefulness of MR imaging in the differentiation of these lesions.

5105

Kingshighway

December

April

29,

23. Address

Blvd. 1989;

St Louis,

in six other

pulmonary

sis. All patients less

than

Seven

lymphangiomyomatohad

1.4

patients.

serum

mg/dL

creatinine

(120

1ev-

zmol/L).

suspend-

ed respiration (11). By adjusting the flip angle, the repetition time (TR), and the echo time (TE), variable degrees of Ti and T2 weighting can be achieved (12). Further improvement in lesion detection and characterization can be accomplished with the administration of gadolinium diethylenetniaminepentaacetic acid (DTPA) (13-15). This contrast material is handled by the kidney in much the same way that iodinated intravenous contrast agents are (16). This was

present

Eleven contralateral kidneys with normal morphologic features from this same patient population were also studied.

ma (10). A significant reduction in breathing artifacts can be achieved with the use of gradient-echo (GRE) imaging

were

Other

Imaging

All 15 patients dium-enhanced

MO

63110.

revision requested Februreprint requests to J.K.T.L.

Modalities

contrast me3 weeks of MR imaging. CT was performed with either a Somatom DRH or PLUS unit (Siemens Medical Systems, Iselin, NJ) with 8mm beam collimation and 10-mm table increments during the rapid infusion of 100-150 mL of Conray 60 (iothalamate meglumine, Mallinckrodt, St Louis) or a similar dose of Optiray 300 (ioversol; Mallinckrodt). All seven renal cysts met strict

underwent CT within

contrast-enhanced

CT criteria

for simple

cysts, including a sharp interface with nal parenchyma, an attenuation value near that of water, absence of a perceptible wall, and lack of contrast enhancement (2-4). The nine renal cell carcino-

re-

mas and the oncocytoma were prospectively diagnosed at contrast-enhanced CT as solid neoplasms, and the diagnosis was confirmed

scientific

AND

Despite reports to this effect, renal MR imaging remains largely a problem-solving tool. Limitations associated with standard spin-echo (SE) renab imaging include image degradation caused by breathing artifacts and suboptimal lesion conspicuity due to the similarity of Ti and T2 values of

176:333-338

Mallinckrodt

1989

assessment remains largely

ADIOLOGIC

study

1990;

with Dynamic

Dynamic contrast material-enhanced gradient-echo magnetic resonance (MR) imaging was performed on 15 patients with 18 renal masses (seven simple renal cysts, nine renal cell carcinomas, one angiomyolipoma, and one oncocytoma). Fifteen sequential images were obtained while the patients held their breath during a 2.5-3.5minute interval during and immediately after the intravenous administration of gadolinium diethylenetriaminepentaacetic acid (DTPA); delayed images were also obtained for 15 minutes. Time-intensity curves showed that renal cortical enhancement reached maximal intensity 80 seconds after the injection of Gd-DTPA. Medullary enhancement reached maximal intensity at 120 seconds. None of the simple renal cysts showed enhancement; each cyst displayed a signal intensity less than that of the renal cortex on precontrast images. All renal cell carcinomas were isointense with the renal cortex and demonstrated variable enhancement. Three patterns of enhancement were observed: predominantly peripheral, heterogeneous, and homogeneous. Both the angiomyolipoma and the oncocytoma showed brisk, homogeneous enhancement. This MR imaging technique appears to be useful in the detection and characterization of simple renal cysts and solid neoplasms.

Radiology

J. Brown,

#{149} Jeffrey

Radiology

at surgery.

Abbreviations:

DTPA

pentaacetic acid, GRE = gradient

FLASH echo, SE

echo

repetition

time,

TR

The

angiomyoli-

diethylenetriamine-

= = =

fast low-angle spin echo, TE

shot, =

time.

333

3.

2.

1.

11 normal kidneys studied with the dyof renal cortex and medulla were normalized to the signal intensity of retroperitoneal fat (SI tissue/SI fat). (2) Comparison of time-intensity curves for seven simple renal cysts, the enhancing portions of nine renal cell carcinomas, and normal renal cortex of 11 kidneys. All signal intensities were normalized to the signal intensity of retroperitoneal fat. (3) Comparison of time-intensity curves for angiomyolipoma, oncocytoma, simple renal cysts, and the enhancing portions of the Figures

1-3.

namic

(1) Time-intensity

curve

Gd-DTPA-enhanced

renal

FLASH

cell carcinomas.

All

signal

for

renal

technique.

intensities

Imaging

were

Protocol

MR imaging

was performed

commercially

available

with

1 .0-T

a

msec/TE

msec)

transaxial

x

256

sequence

in either

or the coronal

matrix,

section

plane

the

with

thickness

a 192

of 8 mm,

intersection gap of 2 mm, four signals averaged, and an imaging time of 6.5 mmutes.

In

all

15 cases,

the

kidneys

and

the

renal mass were localized during suspended respiration with use of a singlesection GRE (fast low-angle shot [FLASH]) imaging sequence (30/10; flip angle, 40#{176}; 128 X 256 matrix; section thickness, onds). This

8 mm; imaging sequence was

time, selected

from

signal

intensity

of retroperitoneal

Interpretation

from

5 to 80 mm.

sities

of retroperitoneal medulla,

geneous),

renal

The

entire and

mass

mean

inten-

cortex

renal

mass

were

(if homo-

portion

measured.

ty curves were generated ratio of the mean signal

the

signal

fat, renal

enhancing

tissue studied fat. On isolated

of the

Time-intensiwith intensity

use

of the of the

to that of retroperitoneal images acquired during

dynamic

phase

of various

examina-

tions, we observed an apparent increase in the signal intensity of the mass relative to the appearance on the preceding or subsequent image; in each case, this ap-

to be related was

protocol

more

to breathing

prevalent

1 than

FLASH

imaging

was

per-

formed in one of two ways: FLASH protocol i (10 cases) consisted of five cycles of three consecutive 6-second images ob-

with

artifact,

with

FLASH

FLASH

protocol

J.K.T.L.,

J.J.B.) subjectively

Ti-weighted

sequence

2.

postcontrast spicuity.

GRE

images

Statistical

Analysis

compared with

the

the

pre-

for lesion

tamed

during

a 20-second

and

con-

10-second

images,

pause.

each

The

second

followed

by

flushing

with

protocol

re-

40 mL of normal

saline. Delayed FLASH images were obtained at 5, 7, 10, 12, and 15 minutes, followed by an incremental examination of the entire kidney with use of the singlesection FLASH sequence. The total examination time was 35-45 minutes. Radiology

#{149}

analyses

were

performed

independent

RESULTS

a

placed the first to shorten the apneic period and to improve patient cooperation. Imaging was begun during the 50-60-second intravenous infusion of Gd-DTPA (0.1 mmol/kg of body weight) (Schering, Berlin). This was immediately followed by rapid

statistical

use of a two-sample

test.

breath-holding

period, with each cycle followed by a 10second pause. FLASH protocol 2 (five cases) consisted of i5 individual 6-second single-breath

All

ty data

All patients completed the dynamic portion of the examination. Evaluation of the mean signal intensity of the retropenitoneal fat on the preand post-Gd-DTPA FLASH images (n = 15) demonstrated no change throughout the 15-minute study period. Detailed time-intensity curves showing the ratios of the mean signal intensities of the various tissues studied relative to the signal intensity of fat were generated without standard deviations (Figs 1-3). The Table pro-

a summary along

of the with

time-intensi-

the

standard

devi-

ations. Subjective assessment of the Tiweighted and pre- and postcontrast GRE images for lesion conspicuity demonstrated that contrast-enhanced GRE imaging was superior in 100% of cases. Comparison of the noncontrast GRE sequence with the Ti-weighted sequence showed it to be of equivalent value in 50% of cases, superior in 17% of cases, and inferior in 33% of cases. Parenchymal Patterns

Enhancement

Precontrast kidneys

Signal intensities were not measured from these isolated frames; excluded data constituted only 5% of the total measurements. In addition, three of us (S.S.E.,

with

fat.

vides

Objective assessment of the precontrast, dynamic, and delayed FLASH images was performed with use of operator-defined circular regions of interest that ranged

which

6 secbecause

obtained

in four.

Dynamic

334

MR

were

intensities

to the

peared

of its short imaging time and moderate Ti weighting. It produces images free of breathing artifact that optimally demonstrate contrast medium enhancement. The optimal imaging plane was then selected: coronal in six cases, transaxial in five, and sagittal

signal

normalized

and

Magnetom

SP unit (Siemens). The kidneys of 11 patients were first localized and examined with use of a Ti-weighted SE 500/15 (TR

Measurements

The mean

poma was diagnosed at CT on the basis of identification of a small amount of intratumoral fat.

MR

parenchyma.

without

FLASH

images

malignant

from neo-

plasms demonstrated the mean cortical signal intensity to be approximately 12% greaten than that of the medulla (P .057) (Fig 1). Cortical enhancement was brisk in all cases and peaked approximately 80 seconds after contrast material administration; peak cortical signal intensity was 190% greater than that of retropenitoneal fat. Cortical enhancement then gradually decreased throughout the remaining 15 minutes to 70% of its peak

value.

Medullany

enhance-

ment lagged behind that of the contex, becoming maximal and isointense at 120 seconds (P .56). For the remaining 1 1 minutes, the signal intensity of the medulla remained nearly constant, becoming slightly hypenintense relative to the cortex at 4 minutes (P = .18-.026) and demonstrating maximal contrast difference relative to cortex at 12 minutes (19%, P = .026). Renal

Cysts

Precontrast seven simple ed homogeneous

FLASH images of the renal cysts demonstratand relatively low August

1990

a.

Figure precontrast

4.

Selected

sagittal

image

shows

images

relative

from

a dynamic

hypointensity

ity of the cyst, which has no perceptible 190 seconds, the cortex and medulla are

wall nearly

b.

a.

ci-

C.

b. contrast-enhanced

of the cyst relative and does isointense.

not

display (d) At

MR

study

to the cortex. enhancement. 15 minutes,

the

of a patient

with

(b) Image

obtained

Peak cortical nonenhancing

a small

simple

renal

at 90 seconds

signal cyst

intensity remains

cyst

shows

occurred unchanged

(arrow).

increased

(a) The

conspicu-

at this time. (c) At and well defined.

d.

C-

Figure 5. Selected axial images from a dynamic contrast-enhanced MR study of a patient with a renal cell carcinoma with an enhancing rim. (a) The precontrast image shows the carcinoma as a centrally hypointense, largely exophytic cortical mass (arrow). (b) At 90 seconds, portions of the tumor rim (arrow) display enhancement similar to that of the adjacent renal cortex. (c) This pattern is maintained at 220 seconds with only minimal central enhancement. The renal medulla is isointense with the cortex. (d) At 15 minutes, the overall intensity (parenchyma and tumor) has decreased modestly. The central portion of the tumor (*), representing an area of coagulation necrosis, demonstrates only minimal interval enhancement.

Renal

signal intensity (approximately 75% of that of renal cortex [P = .008]). Examination of the time-intensity curves revealed no enhancement during the 15-minute imaging period (Figs 2, 4). Comparison with CT scans demonstrated agreement in size (average, 4.7 cm; range, 1.5-8.0 cm) and appearance of the cysts. In each case, Volume

176

Number

#{149}

2

the cyst was round; had smooth, imperceptible, nonenhancing walls; maintained a low signal intensity; and had a sharp interface with the surrounding renal panenchyma on postcontnast images. In every case, the conspicuity of the renal cyst became much greater after administration of Gd-DTPA.

Cell

Carcinomas

Precontrast FLASH imaging of the nine renal cell carcinomas demonstrated signal intensities similar to the intensity of renal cortex (P 1.0) (Table, Fig 2). Visual inspection of the renal cell carcinomas demonstrated three distinct enhancement patterns: (a) irregular peripheral enhancement with only minimal, heterogeneous central enhancement; (b) heterogeneous enhancement; and (c) mild homogeneous enhancement. Five cases demonstrated the first pattern, with some degree of enhancement of an irregular rim on periphery and with minimal heterogeneous enhancement of the more central portion (Fig 5). The average tumor size in this group was 5 cm (range, 3-8 cm). In each case, tumors that demonstrated regions of little to no contrast enhancement contained foci of pathologically proved coaguRadiology

335

#{149}

lation and/or liquefaction necrosis. Two cases demonstrated the second pattern of enhancement. Both were large tumors, measuring 8.0 and 8.5 cm in diameter, that lacked a clear margin with adjacent parenchyma and demonstrated patchy, hetenogeneous areas of brisk contrast enhancement intermixed with areas of minimal to no enhancement (Fig 6). In both cases enhancing tumor thrombus extended from the renal vein into the inferior vena cava. In the remaining two cases, mild homogeneous enhancement was depicted throughout the mass (Fig 7). The two tumors in this group were the smallest in the series, measuring 1 .5 and 2.5 cm in diameter. Both had irregulan margins with the surrounding panenchyma.

Oncocytoma

to that

throughout study period.

the

of renal

cortex

remainder

of the

The precontrast GRE images of the angiomyolipoma demonstrated fairly homogeneous signal intensity that was nearly isointense with surrounding cortex. The small amount of intratumoral fat cleanly identified at CT and SE MR imaging was almost mapparent. After administration of conmaterial, the enhancement and homogeneous and

was approxi-

mated that of renal cortex; however, the fat was completely obscured. The time to peak enhancement was 80 seconds, at which time the tumor was isointense with renal cortex. A decline in signal intensity followed this peak and closely paralleled that of renab cortex. Delayed images were not 336

Radiology

d.

Figure

6.

Selected

images

large, infiltrating, age demonstrates nephnic struction. less

collateral

(b) At

110

enhancement but

hypointense

nical

seconds,

still

areas

obtained

a dynamic

in this

(arrowheads) heterogeneous

than

in

the

heterogeneous

within

normal

due

have

MR

cell carcinoma.

to a tech-

problem.

involved

left

kidney.

Note

developed

in

response

to

(c)

kidney. of

the

left

Our

data

show

that

The

slower

have

aging

and

guished

can on

the

from

be reliably basis

renal neoGRE im-

distin-

of their

appear-

ance and enhancement pattern. Renal cortex follows a predictable pattern of brisk early enhancement, with a gradual decrease in intensity the

remainder

of the

15-mm-

ute imaging period. Early medullany enhancement lags behind that of the cortex but becomes isointense within the first few minutes. The medulla maintains its enhancement to a greater degree than does the cortex and becomes

slightly

hypenmntense

rela-

tive

to cortex after about 4 minutes. Contrary to other reports, we did not observe a period of early decreasing medullary signal intensity in normal kidneys, which has been thought

medullary

to be

related

to transient

superconcentration

of the

Gd-DTPA (13,17,18). This phenomenon may depend somewhat on technical factors, such as the imaging sequence (18), as well as the rapidity of

the

contrast

material

infusion

(17).

At

220

penvein

there

At

15

a

im-

is noted,

seconds, (d)

rate

material

obwith

has

been

minutes,

of infusion

used

may

document

simple

the renal

left kidney

kidney.

with

precontrast

the

in partially.

10 sec)

cysts behave differently plasms at contrast-enhanced

(a) The

of the

trast

DISCUSSION

of a patient

that right

filled

study

renal of the

enhancement

enhancement

the tumor

patient

contrast-enhanced

enhancing distortion

and

vessels

overall

progressive

from

heterogeneously enlargement

venous

during

Angiomyolipoma

trast brisk

b.

C.

The precontrast FLASH images of the oncocytoma demonstrated homogeneous signal intensity that was isointense with renal parenchyma (Figs 3, 8). Focal cortical expansion from this 5.5-cm mass was apparent. In addition, several small, high-signal-intensity foci that were isointense with renal hilar vessels were noted at the periphery; these foci represented enlarged capsular yessels. Enhancement with Gd-DTPA was brisk and homogeneous. In the 1st minute, the time-intensity curve fell between the curves of renal contex and medulla. The peak enhancement of the tumor occurred at 80 seconds after contrast administration, when the tumor was isointense with renal cortex. The signal intensity was similar

a.

explain this

shown

of con-

in our

our

event.

a loss

study

(60

inability

Several

of this

vs

to groups

medullany

phenomenon in the presence of significant renal disease, including hydronephrosis in humans (17,18) and chemically induced acute tubular nephrosis in the rabbit (19); this loss is thought to be due to a loss of medullary concentrating ability. To the best

of our neys

knowledge, we

the

studied

1 1 normal

were

free

kid-

of medical

and surgical renal disease: They were of normal size, were not obstructed, had distinct tions seen

corticomedullary at MR imaging,

and

juncwere

from patients with normal serum creatinine values. The seven simple renal cysts studied had several common features, including a precontrast signal intensities less than that of the surrounding parenchyma, sharp margins, or oval shape, imperceptible and lack of contrast material hancement. These criteria,

round wall, enwhich

have proved trast-enhanced

to be reliable CT, should

at conbe equally

reliable hanced

applied to contrast-enMR imaging.

The

when GRE

nine

demonstrated

renal

cell

several

carcinomas common

August

fea-

1990

a.

b.

d.

C.

Figure

7. Selected sagittal images from a dynamic contrast-enhanced MR examination of a patient with a small, homogeneously enhancing renal cell carcinoma (arrow) and a cyst (arrowhead). (a) The precontnast image demonstrates the hypointense cyst; however, the carcinoma is not apparent. (b) At 100 seconds, the carcinoma is clearly seen as an irregular, mildly enhancing cortical mass. The cyst increased in conspicuity but did not enhance. (C) At 200 seconds, the carcinoma continues to enhance minimally while the signal intensity of the medulla ap-

proaches

that

of the cortex.

(d) At

15

minutes,

the carcinoma

a.

remains

conspicuous,

with

irregular

margins.

hancement patterns of 16 renal carcinomas, little or no enhancement was found in seven cases (20). While our sample size is small, we believe that measurable enhancement is likely to be present in most renal carcinomas. Our increased sensitivity in the detection of enhancement may be due to higher field strength (1.0 vs 0.5 T), which increases the magnetic susceptibility artifact, as well as to shorter TE (iO vs 17 msec), which increases Ti weighting and, accordingly, the sensitivity to Gd-DTPA. Even in the absence of enhancement, the carcinomas in our series had sufficiently clear and identifiable features (eg, relatively high and often heterogeneous signal intensity and irregular margins) to place them in the indeterminate renal mass category. We did not specifically address the evaluation of indeterminate renal masses. Our anecdotal experience with the contrast-enhanced GRE MR technique in this category of renal masses has met with mixed success. In most cases, indeterminate renal

b.

masses

d.

C.

Figure

8.

patient

with

Selected

image

demonstrates

tense

oncocytoma

rowheads),

upper

ment

with

pole

images

renal two

in

the

pole

and

left lower

no

enhancement

mogeneously,

shows

similar

and right

enlarged

almost

tumor

from

cell carcinomas

left-sided

is seen

representing

right upper

coronal

multiple

to

mild the

a dynamic, (arrows) one right-sided

lower

feeder

pole

pole

vessels,

renal

(*).

contrast-enhanced MR examination of a and an oncocytoma. (a) The precontrast renal cell carcinoma. A nearly isoin-

Several

surround

high-signal-intensity this

cell carcinomas

of their

cortex.

(c)

At

(b)

mild while

The oncocytoma 210

seconds,

foci (an-

At 90 seconds,

centers,

enhancement.

surrounding

mass.

demonstrate

low-signal-intensity

heterogeneous

curvilinear

the

rim enhancethe

small

enhanced

left

ho-

enhance-

ment of the more central portions of the larger of the two renal cell carcinomas is depicted. The oncocytoma has become isointense with the renal parenchyma. (d) At 15 minutes, the renal cell carcinomas remain largely unchanged, while the oncocytoma has become mildly hypointense

relative

to

the

renal

parenchyma.

tunes:

(a) The precontrast signal intensity was similar to that of the surrounding parenchyma and was clearly greater than that of a simple cyst. (b) All lesions had irregular margins. (c) All

Volume

lesions

176

demonstrated

Number

#{149}

some

2

de-

The cyst did not enhance.

gree of contrast material enhancement. (d) All lesions became much more conspicuous after Gd-DTPA administration. In a previously published report, in which a similar technique was used to study the en-

had

enhanced wall, solid tenuation) firm

some

feature

at contrast-

CT (eg, thick on enhancing components, or high atthat we were able to con-

at contrast-enhanced

GRE

MR

imaging. These lesions still necessitate a biopsy or close follow-up. The precise role of the contrast-enhanced GRE MR imaging technique in this group of lesions remains to be determined. The MR imaging appearance of an oncocytoma has been described with use of SE techniques; one report descnibed a barge, encapsulation

necrotic (21), and

mass with another, a

smaller mass with a characteristic central scan (22). Our series includes what we believe to be the first oncocytoma

studied

trast-enhanced

with

GRE

the

use

of con-

MR imaging. Radiology

337

#{149}

The precontrast appearance was that of a homogeneous mass that was isointense with surrounding renal panenchyma. The enhancement pattern paralleled that of renal cortex, with the main difference being more rapid washout of the Gd-DTPA. The oncocytoma in our study was well defined and did not demonstrate central necrosis on scarring. Both CT and SE MR imaging are highly accurate in the diagnosis of angiomyolipoma (23,24). The contrast-enhanced GRE MR evaluation of the single angiomyolipoma included in our study highlighted the potential

shortcomings

of this

imag-

ing technique, including difficulty in the identification of small amounts of intratumoral fat and obscuration of fat after contrast material administration. The enhancement pattern demonstrated was nonspecific and did not provide additional information in this case; beyond that, the mass was solid and vascular. Although our study was not designed to compare the sensitivity of contrast-enhanced

GRE

MR

GRE

and

1.

HM,

Banner

Mulhern

CB,

differentiating nign

2.

Arger

PH,

BC.

renal

ultrasonography

cystic

cysts.

The

1982;

from

be-

1977;

14.

McClennan

BC, Stanley

Levitt

RC,

Sagel

is cyst

aspiration

RJ, Melson

55.

CT

of the

necessary?

17.

CL,

AJR

16.

Ra-

renal

cyst:

Balfe DM, McClennan Weyman PJ, Sagel SS. masses

5,

6.

Arger

PM,

renal

masses:

magnetic

staging

of renal

Fein

and

1987;

12.

HM, AC.

resonance

neoplasms:

staging

Balfe

DM,

Focal

of renal

of MR

cell

imaging

histopathologic

pulse

sequence

1988;

169:359-362.

optimization.

correlation

of the magnetic

Radiology

Hamm B, Wolf KJ, Felix R. Conventional and rapid MR imaging of the liver with Gd-DTPA. Radiology 1987; 164:313-320. Te Strake L, Freling NJM, Kamman RL, Mooyaart EL, Doorenbos H, Schrafford Koops H. Abdominal applications of fast MR imaging: a comparison of fast field echo (FFE) and spin echo (SE) pulse sequences. Magn Reson Imaging 1989;

rabbit kidney resonance

on Tl images.

Acta Radio! 1987; 28:479-482. Carr DH, Brown J, Bydder GM, Gadolinium-DTPA as a contrast MRI: initial clinical experience tients. AJR 1984; 143:215-224. Kikinis

R, von

Normal

and

Schulthess

MR

et al. agent in in 20 pa-

GK,

Jager

hydronephrotic

of renal

JA,

Acute

20.

P.

kidney:

function

with

imaging.

eva!-

contrast-en-

Radiology

and

1987;

fast HK,

1987;

DS,

decki

of the

Gd-DTPA 161(P):1

HS.

J

19.

MR

Uhlenbrock

AC,

Caroline

Assist

Hartman

DF.

kidney:

D,

giomyolipoma

between computed phy

for

imaging

Comput

DS,

Scar

sign

Fischer of the

magnetic

C, Beyer

resonance

tomography, diagnosis.

imaging Urol (hamarto-

and Acta

diagno-

every

kidney:

Radio!

Ra-

of renal

a preoperative

sis is possible in virtually Radiol 1981; 3:135-142. 24.

M,

of renal

oncocytoma: magnetic resonance appearance and lack of specificity. Radio! 1986; 8:46-48. Bosniak MA. Angiomyolipoma ma)

reAssist

11:1085-1087.

Friedman

PD,

et al.

W, Deimling

and

oncocytoma.

Radi-

to evaluate

imaging

Seo

enpre-

J Comput

1987; SY,

Weiss

of gadolinium-

rabbit.

imaging Kim

HA,

imaging

MR

Radiology

Tomogr

23.

MR

Dynamic

of a rena!

Ball

fast

1987; 11:488-495. M, Kornmesser R.

Sohn

Austin

use

in the

Tomogr Laniado

(abstr).

22.

PL,

necrosis:

function

tumors: 21.

Choyke

tubular

Felix

AJR

and

L, Dannels W, CouME. Fast field-echo

19.

DiagCT.

Springer-

MR imaging with Gd-DTPA: physiologic evaluation of the kidneys and liver. Radiology 1986; 160:561-563. Mencey JM, Ackerman N, Kaude JV, et a!. Gadolinium-DTPA enhancement of VX-2

nal

148:749-753.

York:

GH, Inscoe SW, Black JL. Dynamic hanced MR imaging of the kidneys: liminary clinical experience (abstr). ology 1989; l73(P):269. Carvlin MJ, Arger PH, Kundel HL,

carcinoma:

and

RI, Avruch Bernardino

J,

DTPA

a reassess-

et al.

Pettigrew mans

New

111-131.

Frank

imag-

Hnicak H, Demas B, Williams R, et al. Magnetic resonance imaging in the diagnosis and staging of renal and perirenal neoplasms. Radiology 1985; 154:709-715. Quint LE, Glazer GM, Chenevert IL, et al. In vivo and in vitro MR imaging of renal tumors:

1 1.

Pollack

162:679-684.

AB, Lee JKT,

a comparison

10.

HY,

L, Mamourian

imag-

18.

1982;

ment of MR imaging. Radiology 1988; 166:643-649. Marotti M, Hricak H, Fritzsche P. Crooks LE, Hedgcock MW, Tanagho EA. Complex and simple renal cysts: comparative evaluation with MR imaging. Radiology

nosis

9.

Radiology

morphology resonance

165:837-842.

com-

ing. Radiology 1984; 152:471-477. Hnicak H, Thoeni RF, Carroll PR, Demas BE, Marotti M, Tanagho EA. Detection

1987;

8.

Axel

on

Renal

systems.

1989;

hanced

RJ, of renal

indeterminate

puted tomography. 142:421-428. Choyke PL, Kressel

and

7.

considered

BL, Stanley Evaluation

renal

uation

1979;

133:671-675.

4.

GK. in magnetic

carcinoma weighted

C.

124:359-370.

function

Verlag,

Peters

143:741-745.

Schulthess

and

and

in

neoplasms

Radiology

von

ing. In: von Schulthess GK, ed. Morphology and function in MRI: cardiovascular

accura-

Sagel SS, Stanley RJ, Levitt RG, Geisse Computed tomography of the kidney. diology

3.

MP,

Coleman

cy of gray-scale

Ti-

weighted SE sequences for the delineation of renal masses. In addition, contrast-enhanced GRE MR imaging also provides valuable information concerning contrast material enhancement of renal lesions, which, as is the case with contrast-enhanced CT, should improve lesion characterization. Because this breath-holding contrast-enhanced GRE MR technique uses a widely available technology that is easily implemented on most commercially available MR imaging units, we expect that this technique will expand the utility of MR imaging for the evaluation of known or suspected renal masses. It should be particularly useful in patients

Pollack

J,

13.

15.

References

imaging

with that of other imaging sequences, the data in this small series indicate that it is superior to noncontrast-enhanced

with a contraindication for contrastenhanced CT (eg, allergy to iodinated contrast material or renal insufficiency) and in those with suboptimal or equivocal CT or ultrasound studies. Definition of the role of this imaging technique awaits further investigation. U

case.

Urol

HK.

An-

comparison

imaging, ultrasonogra1988;

29:523-526.

7:297-303.

338

Radiology

#{149}

August

1990