625

Contrast-Enhanced Imaging: Animal

GadoliniumDTPA

NMR
Studies Complex Using

Robert
Hanns-Joachim

C. Brasch1
Weinrnann2

George

E. Wesbey1

Gadolinium (Gd)-DTPA complex was assessed as a nuclear magnetic resonance (NMR) contrast-enhancing agent by experimentally imaging normal and diseased animals. After intravenous injection, Gd-DTPA, a strongly paramagnetic complex by virtue of unpaired electrons, was rapidly excreted into the urine of rats, producing an easily observable contrast enhancement on NMR images in kidney parenchyma and urine. Spin-echo intensity of urine within the renal pelvis increased from 2263 to 4414 units; intensity of renal parenchyma increased from 2901 to 3893 after administration of 0.1 mmol/kg Gd-DTPA. Sterile soft-tissue abscesses demonstrated an obvious rim pattern of enhancement. A focus of radiation-induced brain damage in a canine model was only faintly detectable on spin-echo NMR images before contrast administration; after 0.5 mmol/kg Gd-DiVA administration, the lesion intensity increased from 3867 to 5590. In comparison, the normal brain with an intact blood-brain barrier remained unchanged in NMR characterization. Gd-DTPA is a promising new NMR contrast enhancer for the clinical assessment of renal function, of inflammatory lesions, and of focal disruption of the blood-brain bamer.

Ions and molecules containing unpaired electrons demonstrate a paramagnetic behavior when placed in an external magnetic field. Strongly paramagnetic substances are potentially useful as contrast-enhancing agents for nuclear magnetic resonance (NMR) imaging because they hasten relaxation rates of protons in their microchemical environment; Ti and T2 values are shortened [1-5]. Gadolinium ions (Gd3) from the lanthanide series of rare-earth elements contain seven unpaired electrons in the 4-f electron orbitals, a position well protected from chemical interactions with other atoms. These electrons provide gadolinium with strong paramagnetic properties [6]. Gd3 can be chelated with diethylaminetriaminepentaacetic acid (DTPA) to form a stable complex (Gd-DTPA) with a

Received July 7, 1 983; accepted

November 14, 1983.

after

revision

This work was supported in part by a grant from the National Institute of Arthritis, Diabetes, Digestive and Kidney Diseases, AOl AM31 937. 1 Department of Radiology, University of California School of Medicine, San Francisco, CA 94143. Address reprint requests to A. C. Brasch (M-397). 2 Research Laboratories of Schering. D-1 000

formation constant of i 02223 [7]. The meglumine salt of Gd-DTPA after intravenous administration is rapidly excreted into the urine with a small proportion appearing in feces. The pharmacologic properties of Gd-DTPA including stability, pharmacokinetics, and tolerance are described in the accompanying article [4]; all these factors favor the further investigation of Gd-DTPA as an NMR contrast agent. We describe imaging experiments testing Gd-DTPA as an intravenous contrast agent for proton NMR imaging in animals. The ability of Gd-DTPA to contrastenhance normally functioning kidneys was studied in rats. The potential to alter the
signal
in

intensity barrier

of edematous

inflammatory

lesions

of the

soft

tissues

was damage.

studied Our

Berlin 65,

Germany.

rats and a dog.

was

The utility
investigated

of Gd-DTPA
in a dog

to identify
with focal

focal

disruption

of the bloodNMR

brain
AJR 142:625-630, March 1984

radiation-induced

0361 -803X/84/1 423-0625 to American Roentgen Ray Society

purpose
imaging.

was

to estimate

the potential

clinical

utility

of Gd-DTPA-enhanced

626

BRASCH

ET AL.

AJR:142,

March

1984

Materials NMR Imaging NMR resistive

imaging

and

Methods

spin-echo (SE)-intensity images NMA unit previously described

was

were obtained [8]. The useful period.

on a 0.35-T aperture for

6.5 cm in diameter.
a 7.0-mm

Five contiguous

axial sections

were

obtained

during

data acquisition

Each slice was 4.2

mm thick and spatial resolution was 1 .0 x 1 .0 mm. TA (pulse) interval was either 500 or 1 000 msec; TE (echo delay) was either 28 or 56 msec after the 90#{176} radiofrequency (AF) pulse. The same imaging factors were used throughout the in vitro and rat imaging protocols. For each of the five axial sections, four spin-echo images were

obtained

designated

as SE 500/28,

SE 500/56,

SE i 000/28, related

and SE to the

1000/56 (the numbers indicating TRITE). Mean intensity values, in arbitrary units of intensity voltage of the received AF pulse, were determined

for operator-

defined regions of interest. Each region of interest contained at least 50 pixels. Pixel-to-pixel standard deviation in each region of interest

was also calculated.

not altered between

The receiver

gain setting

and RF output

were
Fig. 1 .-SE 500/28 NMR images of aqueous solutions of Gd-DTPA. Highest

pre- and postcontrast

images.

intensity was observed at concentration of 1 .5 mmol/L. Not visible is 20 mmol/ L sample. which gave no detectable signal due to very short T2 relaxation
time, less than 28 msec.

Imaging

of Gadolinium-DTPA

Solutions

in vitro salt was

Series

Gadolinium-DTPA

(Gd-DTPA)
water for

as a meglumine
in vitro imaging.

prepared

as previously
diluted with

described
distilled

[4] and the 0.5 mol/L

stock solution
of 7.0-mI

was
test

tubes filled with distilled water or gadolinium-DTPA

of 0.01, grouped 0.1, 0.3, 0.75, 1.0, 1.5, 2.0, 5.0,

in concentrations
mol/L were

From 500 and 1 500 msec TA spin-echo

images,

we calculated

Ti

and 20.00

together
Imaging

for NMA imaging.

and T2 images of the dog brain [1 2]. The NMA imaging sequences were performed before and up to 40 mm after administration of 0.5
mmol/kg Gd-DTPA. The injection was made slowly over 60 sec. Four TA was begun

data acquisitions
Urographic mm after injection.

at 500 msec TA were begun at 5, 10, 20, and 25

A data acquisition at 1 500 msec

NMR spin-echo-intensity images were obtained in a series of five normal Sprague-Dawley rats weighing 300-400 g. Gd-DTPA was tested as an intravenous contrast-enhancing agent by injection into tail veins of rats anesthetized with pentobarbital (50 mg/kg intraperi-

32 mm after injection. An external reference solution, iron dextran, was placed in the imaging field to serve as a standard for consistency
of pre- and postcontrast intensity and Ti and T2 values. The pathology induced by the 1251 seed in this model has been described [1 0] and was reconfirmed in this animal. Briefly, a noninflammatory area of necrosis was seen within the white matter with

toneal).
before

SE images

through

the level of the kidneys

were obtained

and at 5, 1 2, 30, 60, and 90 mm after Gd-DTPA injection. The doses of Gd-DTPA ranged from 0.01 to 1 .0 mmol/kg. Intensity region-

loss of nuclei and with axonal

protein-rich fluid in the perivascular

swelling.

Adjacent

vessels
spaces.

showed

and interstitial

of-interest
hancement

measurements
within

permitted

the assessment
and within

of contrast
the urine.

enResults

the renal parenchyma

ExperimentalAbscess

Imaging

Imaging

of Gd-DTPA

Solutions

in vitro

Sterile

inflammatory

lesions

(abscesses)

were induced

to this mucopolysaccharide was allowed to develop over 4 hr. Then rats were imaged by NMR before and up to 120 mm after intravenous injection of 0.1 or 1 .0 mmol/kg Gd-DTPA. Carrageenan previously has been shown to produce an edematous pathologic lesion within soft tissues, reaching a peak effect in 3-5 hr [9].

subcutaneous the kidneys modification

tissues of three Sprague-Dawley by injecting 0.4 ml of carrageenan of the method of Winter et al. [9]. Local reaction

in the rats at the level of (0.1 % wt/vol) by a

Radiation

Cerebritis

Imaging

The 1 .5 mmol concentration of Gd-DTPA produced the highest NMR intensity signal (1 2,1 23 units 463) compared with the baseline intensity of distilled water (2i 72 325) (fig. 1 Intensity was also enhanced for the 0.Oi mmol solution (2428 367) and the 0.1 mmol solution (5335 384). The 20.0 mmol solution of Gd-DTPA had no observable signal and appeared black on SE images. This observation is compatible with a very short T2 relaxation time. A similar decrease in spin-echo intensity has been observed with other paramagnetic agents, ferric ions and nitroxide free radicals, in high concentrations [2, i 3].
).

A small focus of radiation-induced cerebritis was induced in a 7 kg by surgical implantation of a 1251 seed directly into the right cerebral cortex for a 48-hr period [1 0]. The calculated radiation dose to brain tissue 1 cm from the center of the seed was 1500 rad (15 Gy). Seven days after extraction of the radioactive seed, the dog, anesthetized again by intravenous pentobarbital, was imaged using a 0.35 T superconducting NMA system previously described [1 1].
beagle

Renal

Imaging

In the series of normal rats, the nonenhanced kidneys appeared as intermediate-intensity structures outlined by high-intensity perinephric fat. The central collecting structures

AJR: 142, March

1984

Gd-DTPA

NMR

IN ANIMALS

627

TABLE 1: NMR Characterization of in-vivo Rat Renal Compartments before and after Gd-DTPA Administration
Dose of Gd-DTPA

SE 500/28 Intensity Before Contrast

SD After Contrast

(mmol/kg):
Renal

Time of Intensity
Measurement

Compartment

(mm)

1.0:

Parenchyma
Urine

3509
2703

396

2355 4089

297

472 384 1 197 41 6

3819886
0.1:

5 85 5 85 10 10 5
5

Parenchyma Urine 0.05: Parenchyma

Urine 0.01:

2901
2264

41

490
896 31 4 938 391

3893 441 4 3783

4670

562 533

3261
2902

527 573 690 484

Parenchyma
Urine
Fig. 2.-SE 500/28 NMR image of normal rat abdomen and retroperitoneum.

3890 2943

3934 4281

60
5

Low intensity of nonenhanced urine in renal pelvis. Vascular structures with rapidly flowing blood, such as inferior vena cava (IVC), have even lower intensity
than urine.

urine (fig. 2). These obserreported [3, 1 4] (table i). At 5 mm after administration of Gd-DTPA in a dose of 0.i mmol/kg (the first image), there was an increase in renal parenchymal SE 500/28 intensity that reached a maximum of 3893 562 after i 0 mm. By 90 mm, the contrast agent was still evident within the kidney and renal intensity was 3517 51 7. Accompanying the increase in renal parenchymal intensity, the urine within the renal collecting structures became much higher in intensity, appearing white (fig. 3). A lowest effective urographic dose of Gd-DTPA was not determined from this series of animals because the lowest tested dose (0.01 mmol/kg) produced marked contrast enhancement (fig. 4) within the urine. The parenchyma did not enhance significantly at this dose (<i SD). Conversely, at the highest tested dose (1 .0 mmol/kg), there was evidence of prompt renal excretion and diuretic effect; the renal calices became markedly distended (fig. 5). However, with this higher dose, the urine intensity decreased to lower than baseline intensity at 5 mm. This observation, corresponding to the invitro observations of concentrations approximating 20 mmol/ L, indicates marked shortening of T2 relaxation times. Such marked T2 shortening is predictable from the spectrometnc results noted for high concentrations of Gd-DTPA [4]. The same rat, by 85 mm after administration, presumably when Gd-DTPA was more dilute within the urine, had increased SE intensity in the renal pelvis and parenchyma (table 1).
vations agree with those previously Experimental Inflammation Imaging

were

seen to contain

low-intensity

Gd-DTPA within the edematous inflammatory lesion; this effect may be due to the gradual diffusion of the paramagnetic complex from the plasma into the expanded interstitial space. Normal soft tissues in the contralateral flank and the ipsilateral muscles did not change significantly in intensity. The higher dose (1 .0 mmol/kg) of Gd-DTPA similarly produced a marked increase in SE intensity within carrageenan-induced lesions for up to i 20 mm after injection.

Radiation-Induced

Cerebritis

The NMR images of flank inflammatory lesions induced by carrageenan were strongly enhanced after intravenous administration of 0.1 mmol/kg Gd-DTPA (fig. 6). The 5-mm image revealed a rim pattern of enhancement that filled in completely by 35 mm. This change suggests a gradual accumulation of

The focal area of radiation damage in the right cerebral hemisphere of a dog induced by a 1251 seed was barely perceptible as a low-intensity (3867 31 6, SE 500/28) zone on the nonenhanced NMR spin-echo images. A large softtissue swelling over the calvaria at the site of craniotomy was lower in intensity than cerebral tissues (fig. 7). On all SE images at 5-32 mm after intravenous injection of 0.5 mmol/kg Gd-DTPA, there was an obvious intensity increase at the site of radiation damage. The accumulation of contrast agent can be assumed to result from focal disruption of the blood-brain barrier. The maximum SE 500/28 intensity increase (5590 628) was observed at 25 mm. The external reference standard, iron dextran, gave internally consistent SE 500/28 intensity values (mean 12,495 35) for the five acquisitions at 500 msec TA. Similar to the carrageenan-induced flank abscesses, the surgical scalp lesion showed a rim pattern of enhancement after Gd-DTPA. Ti images, calculated from SE 500/28 and SE i 500/28 data, demonstrated that the right cerebral lesion before contrast had a relatively long Ti value (747 i 07 msec) as compared with the contralateral normal brain (650 112 msec) (fig. 8). A corresponding Ti image 34 mm after GdDTPA administration (fig. 8) showed shortening of Ti to 633 i 32 msec in the core of the lesion, suggesting accumulation of the paramagnetic contrast agent. Yet, adjacent to the zone of Ti shortening, a rim of persistently long Ti tissue (729

628

BRASCH

ET

AL.

AJR:142,

March

1984

Fig. 3.-SE 500/28 NMR images of one rat arranged in temporal order (left to right, top to
bottom) mmol/kg before intravenous administration of 0.1 Gd-DTPA and 5. 10, 30. 60, and 90 mm after administration. Increased intensity of renal parenchyma and urine after Gd-DTPA administration (see table 1).

Fig. 4.-SE 1 000/28 NMR images of rat before (A) and 5 mm after (B) intravenous administration
of Gd-DTPA (0.01 mmol/kg). High in kidneys and urine after contrast. signal intensity

Fig. 5.-SE

1000/28

NMR images

of rat kidneys

before (A) and 5 mm after (B) administration of 1 mmol/kg Gd-DTPA. Urine in distended renal pelves after injection is very low in intensity, attributable to
marked shortening 1).

agent. Renal parenchyma

(see table

of T2 with high dose of contrast is also lower in intensity

95 msec) was observed. This peripheral zone of tissue with long Ti is suggestive of an edematous zone without significant accumulation of paramagnetic contrast agent. The Ti time of the normal left cerebral hemisphere remained essentially unchanged 34 mm after Gd-DTPA administration (625 i 07 msec). The calculated Ti time of external standard solution was near constant, less than 0.5% difference between the two calculations.

Discussion Results indicate that Gd-DTPA complex, administered intravenously to animals in experiments, accumulates in urine and certain diseased tissues to enhance intensity and increase contrast on spin-echo NMR images. Because of the rapid urinary clearance of Gd-DTPA, more than 85% of the injected dose appearing in the urine within 3 hr [4], the status

AJR:142,

March

1984

Gd-DTPA

NMR

IN ANIMALS

629

Fig. 6.-SE high signal

500/28 intensity

NMR compared

images with

of rat with chemically-induced

right

flank

abscess. A, Before contrast administration.

Soft-tissue
flank.

thickening and relatively

B, 5 mm after intrave-

contralateral

nous injection of 0.1 mmol/kg Gd-DTPA. Marked intensity increase within abscess, predominately in nm pattern. C, At 35 mm. Gd-DTPA has reached core of abscess, producing more enhancement.

The ability to contrast-enhance sterile inflammatory lesions, either by a chemical irritant or by surgery, is another clinically important property of the Gd-DTPA complex. From our observations it seems likely that Gd-DTPA distributes primarily in the intravascular, extracellular space and quickly passes into the extravascular, interstitial space. A similar pattern of distribution is well established for diatrizoate, a commonly used radiographic contrast agent having a molecular weight of about 600. By comparison, the molecular weight of Gd-DTPA complex is 590. The composition of interstitial fluid within tissues is very similar to the plasma within the vascular compartment. Both fluids have rather long Ti relaxation times and yield low signal intensities on NMR images compared with the signal from highly cellular tissues. We postulate that after intravenous injection, Gd-DTPA in the plasma diffuses into the extravascular, interstitial space and thus will conspicuously enhance tissues with a high proportion of interstitial fluid (e.g., edematous tissue). Conversely, tissues with a relatively large cellular part and small interstitial part would be expected to contain a relatively smaller concentration of Gd-DTPA and thus to demonstrate little or no
induced enhancement The ability on NMR of Gd-DTPA images.

dog

Fig. 7.-Precontrast SE 500/28 (A) and SE 1500/28 (B) NMR images of with focal radiation-induced cerebritis demonstrated as faintly lowerintensity region in A. Edematous surgical wound in soft tissues overlying brain

to contrast-enhance selectively a pathologic focus within the brain is perhaps the most clinically significant of the potential applications demonstrated. The results of our study, limited to one animal with focal radiation cerebritis, suggest that Gd-DTPA contrast enhancement may aid in the differentiation of necrotic tissue from surrounding edema. Before clinical trials, a variety of central-nervoussystem-disease animal models should be tested with GdDTPA NMR imaging.

lesion. C, SE 500/28 image 25 mm after 0.5 mmol/kg Gd-DTPA given intravenously. D, SE 1500/28 image 7 mm after C. Marked postcontrast signalintensity enhancement at site of radiation-induced cerebritis and in scalp wound.

of renal

integrity should be assessable by GdNMR imaging. Additional studies in animals with renal functional abnormalities need to be performed to assess more completely the sensitivity of Gd-DTPA-enhanced NMR urography.
DTPA-enhanced

functional

The diagnostically effective intravenous doses of Gd-DTPA used in this study (0.01 -0.5 mmol/kg) are far less than the LD50 dose in rats (1 0 mmol/kg) [4]. In future imaging tests, using smaller Gd-DTPA doses, the ratio between LD50 and lowest diagnostically useful doses may even be greater than the 1 000:1 ratio suggested here. In summary, testing of Gd-DTPA as an NMR contrastenhancing agent indicates that this paramagnetic complex can be used effectively in animals for a variety of diagnostic

630

BRASCH

ET AL.

AJR:142, March

1984

Fig. 8.-Ti images of same dog shown in fig. 7 calculated from SE intensity data. Long Ti values are depicted as white; short Ti values are relatively black on gray-scale assignments. A, Before contrast. Long Ti value of cerebral and soft-tissue lesions. B, 34 mm after administration of 0.5 mmol/ kg Gd-DTPA. Postcontrast Ti shortening in center of cerebral lesion (blood-brain barrier [BBBI defect)

is shown as black area. Surrounding zone of white indicates region of persistently long Ti , presumably zone of edema without significant accumulation of paramagnetic agent.

purposes.

creted,
not

can

Gd-DTPA is strongly be prepared from

in the body,

paramagnetic, readily available

and was well

is rapidly materials,

exis

metabolized

tolerated

in our

small group of animals. Gd-DTPA, used clinically in patients after appropriate trials, may extend the diagnostic utility of NMR imaging examinations.

Mendon#{231}a-Dias MH, Gaggelli E, Lauterbur PC. Paramagnetic contrast agents in NMR medical imaging. Semin NucI Med 1983;i 3:364-376 6. Pople JA, Schneider WG, Bernstein HJ. High-resolution nuclear
5.

ACKNOWLEDGMENTS
We thank
preparing for help

John

R. Fike and Christopher

of cerebral preparation. radiation

E. Cann for assistance

necrosis, and Frank Gorham

in

the model in manuscript

REFERENCES
1 . Eisinger

J, Shulman
ion

AC, Blumberg
binding

WF. Relaxation

enhancement
acid solutions.

by paramagnetic

in deoxyribonucleic

Nature 1961;i 92:963-966 2. Brasch AC. Methods of contrast enhancement for NMA imaging and potential applications. Radiology 1983;i 47:781-788 3. Brasch AC, London DA, Wesbey GE, et al. Nuclear magnetic resonance study of a paramagnetic nitroxide contrast agent for enhancement of renal structures in experimental animals. Radiology 1983;i 47:773-779 4. Weinmann JH, Brasch AC, Press WA, Wesbey GE. Characteristics of gadolinium DTPA: a potential NMR contrast agent. AJR
1984;i 42:619-624

magnetic resonance. New York: McGraw-Hill, 1959:209 7. Moeller T. Gmelin Handbuch der anorganischen Chemie: rare earth elements, Part Dl. Berlin: Springer-Verlag, 1980 8. Crooks L, Hoenninger J, Arakawa M, et al. Tomography of hydrogen with nuclear magnetic resonance. Radiology 1980;i 36:70i -706 9. Winter CA, Risley ED, Nuss GW. Carrageenin-induced edema in the hind paw of rats as an assay for antiinflammatory drugs. Proc Soc Exp Biol Med 1962;i 11:544-547 10. Fike JR, Cann CE, Bernstein M, et al. Radiation damage to the canine brain after 1251 implantation. Radiat Res 1982;9i :370375 1 1 Crooks L, Arakawa, M., Hoenninger J, et al. Nuclear magnetic resonance whole-body imager operating at 3.5 KGauss. Radiology 1982;i43:i69-i74 1 2. Herfkens A, Davis P, Crooks L, et al. Nuclear magnetic resonance imaging of the abnormal live rat and correlations with
.

tissue 1 3.

characteristics.

Radiology

1981;i

41:211-218

Wesbey GE, Brasch AC, Engeklstad BL, Moss AA, Crooks LE, Brito AC. Nuclear magnetic resonance contrast enhancement study of the gastrointestinal tract of rats and a human volunteer
using nontoxic oral iron solutions. Radiology 1983;i 49:175-

180
1 4.

Hansen

G, Crooks

anatomy
1980;i

with
36:695-700

LE, Davis P, et al. In vivo imaging of the rat nuclear magnetic resonance. Radiology