Downloaded from www.ajronline.org by 110.185.241.20 on 10/17/15 from IP address 110.185.241.20. Copyright ARRS. For personal use only; all rights reserved

539

MR Imaging

in Fibromatosis:

Results in 26 Patients Correlation

Stephen

F.

Scott

Quinn1

J. Erickson2 Paul

M.

Dee3

Arthur Walling4 Donald A. Hackbarth5 Gregory J. Knudson2’6 H. Stephens Moseley7

Fibromatoses

are a diverse

hypointense,

hypointense

deposition.

areas

of the

intermediate that

Center,

101 5 N. W. 22nd Ave.,

Sam Jackson

Park

Rd., Portland,

dress reprint requests Samaritan

of Radiology, 31 81 5. W. OR 97201

.

Ad-

to S. F. Quinn at Good

Hospital.

Department of Radiology, Medical College of Wisconsin, 8700 W. Wisconsin Ave. , Milwaukee, WI 53226. 2

3

Department

Hospital, 4

of Radiology,

Chariottesville,

Florida

Orthopedic

Ave., Tampa,

University

of Virginia

VA 22908. Institute,

4175

E. Fowler

Medical College of Ave.

, Milwaukee,

WI 53226. Present

address:

Radiology

pleton,

424 E. Wisconsin

pleton,

WI 54912.

7

Department

Oncology,

inconsistently

been

vary greatly were either

varied

and

soft-tissue

lesions,

invaded

were

demarcated that fibromatoses

of hypocellulanty

skeletal muscle.

and

adjacent

from lesion to hyperintense, dense

structures,

collagen

but the MR

to be well demarcated

judged

(n = 14),

(n = 6).

have a variable

and

this

MR appearance

reflects

variability

no different

the composition

and

of the lesions.

156:539-542,

March

1991

Fibmomatoses are a diverse group of soft-tissue lesions that have not been consistently classified or treated (Table 1 ) [1 J. Most of the superficial fibromatoses, for example palmar and penile fibromatoses, can be diagnosed by visual inspection. The deep fibromatoses, on the other hand, may need the same evaluation that soft-tissue sarcomas require. At our institutions, MR has become the primary method of evaluating soft-tissue masses, including the fibromatoses. The majority of madiologic articles addressing fibromatoses deal primarily with CT [2-6]. Reports of MR appearances of fibromatoses have been limited to single cases [7, 8] and brief descriptions within broader discussions [9-1 5]. In this report, we have attempted to establish a mange of appearances of fibmomatoses on MR and perform a pathologic correlation in an attempt to explain the variable signal intensity pattern.

FL 33617.

5 Department of Orthopedics, Wisconsin, 8700 W. Wisconsin

6

to be zones

all of the lesions

margins

shows

of other

cellularity AJR

appear

(n = 5), or poorly

Our experience from

and Medical

that have

or of mixed signal intensity relative to adjacent

Microscopically

appearances

Portland, OR 97210 and Department Oregon Health Sciences University,

lesions

that of other soft-tissue lesions, and the signal intensities lesion and within lesions themselves. The fibromatoses

isointense,

Hospital

group of soft-tissue

categorized and treated. The purpose of our study was to establish the range of appearances of fibromatoses on MR images and perform a pathologic correlation to explain the variable signal-intensity patterns. During a 3-year period, 26 patients with deep fibromatoses were examined with MR. The MR images were evaluated for signalintensity characteristics, and findings were correlated retrospectively with the pathologic diagnoses. The results showed that the MR appearance of fibromatoses is similar to

The

Received July 9, 1990; accepted after revision September 12, 1990 1 Department of Radiology. Good Samaritan

with Pathologic

Good

Ave.,

Associates

of Ap-

P.O. Box

117, Ap-

Materials

and Methods

During

a 3-year

majority

(n

examinations.

Four

examination

of Surgery, Samaritan

Division of Surgical Hospital

and

Medical

period,

26 patients

with deep fibromatoses

22) had wide-margin

=

and

of the

MR

imaging.

lesions

surgical were

There

recurrences,

were

9-63 years (average, 33 years). The MR examinations were performed

1 7 women

with 0.35-T,

Center, 1015 N. W. 22nd Ave., Portland, OR 97210.

material

0361 -803X/91/1 563-0539 © American Roentgen Ray Society

1 28 to 256 x 256 with one to four excitations. weighted

was

used.

images

Surface (1 00-600/1

coils 6-20

were

resections

used

[TR/TEJ),

in some

were examined

within

with MR. The

weeks

of

the

as determined

by findings

on

physical

and

and

nine

several men,

the

age

range

MR was

1 .0-T, and 1 .5-T MR units.

No contrast

cases.

from

Matrix

Slice thickness proton-density-weighted

sizes

varied

varied

from images

1 28

x

3 to 5 mm. Ti(1000-2000/

540

QUINN

20-40), and T2-weighted images (2000-2500/60-80) in all patients. Sagittal, axial, and coronal images depending on the location of the lesion. The sites of involvement included the chest wall (n pelvic wall (n

1), lower extremity (n cavity (n = 2), pelvic cavity

Downloaded from www.ajronline.org by 110.185.241.20 on 10/17/15 from IP address 110.185.241.20. Copyright ARRS. For personal use only; all rights reserved

abdominal =

=

were obtained were obtained

= 3), abdominalextremity (n = 5), 1), and brachial plexis (n 9), muscle (n = 10), and

1 2), upper

=

(n

2). The sites of origin included fascia

=

(n

=

or indeterminate sites of origin (n 6). The MR images were evaluated for signal-intensity characteristics and correlated retrospectively with the pathologic diagnoses. mixed

=

Results The MR signal intensity of the fibmomatoses varied markedly (Table 2). All of the lesions exhibited some heterogeneity with

TABLE

1: Fibromatoses

I. Superficial

(fascial)

fibromatoses

Palmar fibromatosis Plantar fibromatosis Penile

(Dupuytrens (Ledderhose

fibromatosis

(Peyronie

contracture) disease)

fibromatoses

Extraabdominal fibromatosis (extraabdominal Abdominal fibromatosis (abdominal desmoid)

lntraabdominal

fibromatosis

desmoid)

(intraabdominal

desmoid)

Pelvic fibromatosis Mesenteric

AJR:156,

March

1991

different regions of varied signal intensities, and only one lesion, which had been previously sampled by biopsy, had surrounding edema. One lesion had a central area of cystic necrosis. The majority of cases with increased signal intensity on the Ti-weighted images were only marginally hypemintense relative to the adjacent skeletal musculature (n 1 0). There were three cases, however, in which the signal intensity on Ti -weighted images was markedly increased relative to skeletal muscle. Histologically, two of these cases had abundant myxoid material (Fig. 1 ), and the third contained a large amount of fat, probably related to muscle atrophy (Fig. 2). The fibmomatoses that were hypointense (n = 3) or hetemogeneous with prominent hypointense foci (n = 4) relative to adjacent musculature were found to be relatively hypocellular with dense deposits of collagen. There was no evidence of hemosiderin, calcium, or hypervascularity (Fig. 3). The lesions of fibmomatosis varied from 1 to 1 1 cm in size, with a mean greatest dimension of 5.6 cm. One case was correctly predicted to be multifocal. The margins of the fibmomatoses were characterized as well demarcated (n = 1 4, Fig. 4), intermediate (n 5), or poorly demarcated (n = 7, Fig. 5) on MR images. Micmoscopically all of the lesions had some regions of invasion into adjacent soft tissues. Three of the fibromatoses had a sumrounding hypointense rim. One lesion invaded the posterior femur. =

-

disease)

Knuckle pads II. Deep (musculoaponeurotic

ET AL.

fibromatosis

Gardner syndrome Note.-Reprinted

TABLE

with permission

2: MR Signal

Image Type Ti-Weighted T2-Weighted

Intensities

Increased 14 18

from

Enzinger

and Weiss

[1].

Discussion

of Fibromatoses

Decrrsed 3 3

Eual

Mixed SI

5 1

Note-All signal intensity (SI) measurements are relative to adjacent muscle. Mixed SI indicates mixed signal intensity relative to adjacent muscle with prominent hypointense areas.

4 4 skeletal skeletal

The fibmomatoses are uncommon lesions. Extmaabdominal fibmomatoses have an incidence of three or four cases per million people [1 6]. Abdominal fibmomatoses were found in 0.03% of 50,346 patients admitted to the hospital for neoplastic disease [1 7]. The initial evaluation of deep fibmomatosis is important because the extent of the lesion needs to be delineated. The surgical margins need to be as wide as reasonably possible to help decrease recurrence [1 ]. Half of the fibromatoses in our series were judged to be well demamFig.

1.-Sagittal

Ti-weighted

shows fibromatosis

MR

of upper extremity

image

with hy-

perintense signal that may be related to proteins within myxoid material. Spindle-shaped mass (arrow) is seen within biceps brachialis muscle (B). Fibromatosis has a higher signal intensity than adjacent musculature has. (Image courtesy of Michael Talbot, Oregon Health Sciences University, Portland, OR).

Fig. 2.-Axial

Ti-weighted

shows fibromatosis

MR image

of calf

with high signal intensity

related to fat. Area of high signal intensity is seen in medial portion of gastrocnemius muscle (arrows).

AJR:156,

Fig.

March

3.-A,

MR

1991

Axial

TI-weighted

shows fibromatosis with hypointense posterior calf (arrows).

MR

OF

FIBROMATOSIS

541

image

signal in

B, Photomicrograph shows dense deposition of collagen and little cellularity, accounting for decreased signal intensity on Ti-weighted im-

Downloaded from www.ajronline.org by 110.185.241.20 on 10/17/15 from IP address 110.185.241.20. Copyright ARRS. For personal use only; all rights reserved

age. (H and E, xlO)

Fig. 4.-Axial

T2-weighted

MR image shows

fibromatosis with well-demarcated margins with cystic necrosis. A focal, well-circumscribed Icsion (arrows) is seen in lateral aspect of anterior compartment of lower extremity. (Image cour-

tesy of Clark ED and Bidell J, Portland, OR).

Fig. 5.-Axial proton-density-weighted MR image of distal thigh shows fibromatosis with poorly demarcated margins. A poorly demarcated lesion is infiltrating posterior compartment (arrows) and is closely applied to femur. At surgery, fibromatosis iosteum.

was found to be invading

per-

cated. This appearance is actually misleading, because all of the lesions microscopically invaded adjacent structures. The fibromatoses display a variety of signal intensities. In some cases with high signal intensity on Ti -weighted images this could be attributed to fat within the lesions. This does not explain the high signal intensity on Ti -weighted images in the cases with myxoid material, because myxomas, in our experience, have prolonged Ti relaxation values relative to adjacent musculature. We speculate that an unidentified protein within these lesions causes the Ti relaxation value to shorten. On T2-weighted images, the majority of the fibmomatoses had notably increased signal intensity, as do the majority of other soft-tissue tumors [1 0]. The majority of the fibromatoses have a high signal intensity on T2-weighted images, but approximately one third of the current cases were notably hypointense or had hypointense foci. We believe that these areas of diminished signal intensity are best explained as patches of hypocellularity and dense collagen deposition. Hemosidemin did not appear to be the cause of the areas of decreased signal intensity in our cases [i 8]. Sundamam et al.

[1 5] described areas of decreased signal intensity on T2weighted images and attributed this to acellulamity and abundant collagen. Our experience shows that fibmomatoses have a variable MR appearance, the same as other soft-tissue lesions do, and this variability reflects the composition and cellularity of the lesions.

ACKNOWLEDGMENTS Thanks

to Mary

to Anthony

D’Agostino

Alice

Payne and

for preparation

Gonzalo

Madiedo

of this for

manuscript

pathologic

and correla-

tion.

REFERENCES Enzinger FM, Weiss SW. Fibromatoses. In: Enzinger FM, Weiss SW, eds. Soft tissue tumors. St. Louis: Mosby, 1988:136-163 2. Francis IA, Dorovini-Zis K, Glazer G, et al. The fibromatoses: CT-pathologic correlation. AJR 1986:147:1063-1066 3. Hudson TM, Vandergriend AA, Springfield DS, et al. Aggressive fibroma1

.

542

QUINN

tosis: evaluation

by computed

1984;150:495-501 4. Young JWR, Aisner

tomography

SC, Levine

AM, Resnik

and angiography. CS, Dortman

Downloaded from www.ajronline.org by 110.185.241.20 on 10/17/15 from IP address 110.185.241.20. Copyright ARRS. For personal use only; all rights reserved

mediastinum

report).

primary 1 3. Weekes

H, McGuire

desmoid tumors (aggressive Radio/ 1988:17:16-19 of an abdominal

desmoid

EM,

J

March

et al. MRI

and

CT evaluation

bone and soft tissue tumors. AJR 1986:146:749-756 AG, Berquist TH, McLeod RA, Zimmer WD. Magnetic with

of the

H, Bland KI. Magnetic

extremities.

multicentric

examination with MR imaging Radiology 1987;164:237-241

computed

resonance charac-

with

of

resonance tomography.

conventional

tumors: modalities.

1 5. Sundaram M, McGuire MH, Schajowicz F. Soft tissue masses: histologic basis for decreased signal (short T2) on T2 weighted MR images. AJR 1987;148:

Skeletal

Comput Radiol 1986;1 0:11-13 9. Kransdorf MJ, Jelinek JS, Moser RP, et al. Soft tissue masses: diagnosis using MR imaging. AJR 1989:153:541-547 1 0. Totty WG, Murphy WA, Lee JKT. Soft tissue tumors: MR imaging. Radiology 1986:160:135-141 11. Petasnick JP, Tumer DA, Charters JR. Gitelis 5, Zacharias CE. Soft tissue

compared

1991

of MA imaging with CT.

14. Pettersson H, Gillespy T, Hamlin DJ, et al. Primary musculoskeletal

Vas W. Synchronous

fibromatosis)

8. Hamlin DJ, Page A, Pettersson teristics

MH,

comparison

imaging of soft tissue tumors: comparison Magn Reson Imaging 1985:3:345-352

6. Magid D, Fishman EK, Witaram M, Siegelman 55. Musculoskeletal desmoid tumors: CT assessment during therapy. J Comput Assist Tomogr 1988;12:222-226 7. Sundaram M, Duifrin

system:

Radiology 1986:160: 125-1 33 12. Aisen AM, Martel W, Braunstein

HD. Computed

(case

AJR:i56,

masses of the locomotor

Radiology

tomography of desmoid tumors of bone: desmoplastic fibroma. Skeletal Radiol 1988;17:333-337 5. Black WC, Armstrong P, Daniel TM, Cooper PH. Computed tomography of aggressive fibromatosis in the posterior Comput Assist Tomogr 1987:1 1 : 153-1 55

ET AL.

1247-1250

1 6. Reitamo JJ, Haury P, Nykyri E, et al. The desmoid tumor. 1 . Incidence, sex, age and anatomical distribution in the Finnish population. Am J C/in

tumor.

17.

Pathol 1982:77:665-685 Pack GT, Ehrlich HE. Neoplasms of the anterior abdominal wall with special consideration of desmoid tumors: experience with 391 cases and collective review of the literature. /nt Abstr Surg 1944;79: 177-198

18. Allen PW. The fibromatoses: 140 cases,

a clinicopathologic classification 1977:1:255-269

Part 1 . Am J Surg Pathol

based on