Download Characterization of Bone and Soft Tissue Tumors at MRI

1
Characterization of Bone and Soft Tissue Tumors at MRI
David M. Panicek, MD
Department of Radiology
Memorial Sloan-Kettering Cancer Center, New York, NY
Learning Objectives
• Describe how various features of a lesion may contribute to a more specific
diagnosis at MRI
• Identify characteristic MRI features of certain musculoskeletal tumors
• Recognize strengths and limitations of MRI in characterizing musculoskeletal
tumors
MRI protocol
Optimal characterization of musculoskeletal tumors at MRI involves certain basic
requirements. High-resolution, small-field-of-view axial T1-weighted spin-echo and fatsuppressed T2-weighted fast (turbo) spin-echo images should be obtained through the
entire tumor. T1-weighted images are critical in showing the presence of fat or
hemorrhage within a mass; distinguishing normal marrow and tumor; and demonstrating
bone tumors against a background of fatty marrow. Fat suppression is essential in the
T2-weighted images to allow evaluation of bone marrow for tumor involvement. Preand post-gadolinium fat-suppressed T1-weighted spin-echo images are frequently
helpful, both for distinguishing cysts from solid masses and for demonstrating necrotic
portions of tumor to be avoided at biopsy.
STIR images are not recommended for characterization of tumors, as many lesions
show similar-appearing, very high signal on STIR images. Proton density images are of
limited utility in the characterization of bone lesions, as tumor and surrounding marrow
edema and marrow fat all may show similar signal, decreasing the conspicuity of the
lesion (unless fat suppression is used).
Approach to characterizing soft tissue tumors at MRI
Characterization of soft tissue masses as benign or malignant often will be incorrect if
one relies on traditional imaging features such as sharpness of the margin, presence of
local invasion, large size, and internal homogeneity of the lesion. However, if a mass
demonstrates certain characteristic imaging features, a specific diagnosis frequently can
be made (e.g., cyst, ganglion, lipoma, fat necrosis, giant cell tumor of tendon sheath,
hemangioma, desmoid/fibromatosis, benign nerve sheath tumor, subacute hematoma,
elastofibroma). In the absence of such characteristic features, the possibility of
malignancy cannot be dismissed.
A specific diagnosis can often be suggested based on one or more of the following five
features:
• Chemical composition of the lesion (e.g., water, hemosiderin, methemoglobin)
• Tissue composition of the lesion (e.g., fat, collagen, myxoid matrix)
Proc. Intl. Soc. Mag. Reson. Med. 19 (2011)
2
• Morphology of the lesion (e.g., serpiginous, target sign, infiltrative “tails”)
• Location of the lesion (e.g., deep to scapular tip, in an intermetatarsal space,
along fascia, associated with a peripheral nerve, in a large joint space)
• Clinical circumstances of the patient (e.g., post-amputation, prolonged
immobilization).
Examples of benign lesions that may be misinterpreted as sarcoma at MRI include
muscle strain, myositis, myositis ossificans, rhabdomyolysis, ischemic fasciitis,
parosteal lipoma, fat necrosis, and Morel-Lavallée lesion (a long-standing chronic
hematoma along a subcutaneous fascial plane).
Malignant lesions that are prone to being misinterpreted at MRI as benign — clearly a
much more serious error — include those sarcomas that are predominantly cystic,
hemorrhagic, or myxoid. Extensive cystic change or hemorrhage within a sarcoma may
overshadow the solid tumor elements present. Solid elements must be carefully sought
and viewed as suspicious for tumor until proven otherwise; gadolinium-enhanced
imaging (including subtraction images if enhancement is equivocal visually) is essential
in this assessment.
Myxoid liposarcoma also may appear deceptively cyst-like on standard T1-weighted and
T2-weighted MR images. The presence of subtle fatty streaks within the lesion often is
a clue to the correct diagnosis of myxoid liposarcoma. Even more importantly,
enhancement within the lesion — ranging from fine and lacy to intense and coalescent
— is typically evident after contrast material administration, and is an invaluable finding
to prevent misdiagnosis of myxoid liposarcoma as a ganglion or other cyst.
Another specific type of myxoid sarcoma, myxofibrosarcoma, is increasingly recognized
as one of the most common fibrous sarcomas of the extremities in the elderly. This
tumor has an unusual, infiltrative growth pattern along fascial and vascular planes,
resulting in wispy streaks or “tails” of edema-like signal that frequently extend for
considerable distances from the primary tumor. These tails of tumor enhance, and
need to be brought to the attention of the treating surgeon. Due to this atypical growth
pattern, even low-grade myxofibrosarcoma tends to recur relentlessly, and subsequent
recurrences are of higher grade. Myxofibrosarcoma metastasizes to sites atypical for
most other soft tissue sarcomas (which generally metastasize to lung), including pleura,
peritoneum, adrenal glands, and bone.
The management of a soft tissue lesion will depend on the level of confidence in the
diagnosis provided at MRI. If the lesion has MRI features characteristic of a specific
innocuous lesion (e.g., lipoma, hemangioma, elastofibroma), clinical follow-up alone will
suffice. If the lesion has typical but not pathognomonic MRI findings (e.g., hematoma,
Morel-Lavallée lesion, myositis ossificans, ischemic fasciitis), close-interval imaging
follow-up (such as every 2-4 months for a year) to assess for growth is a reasonable
option. For all other lesions, biopsy or resection usually is the most prudent course of
action.
Proc. Intl. Soc. Mag. Reson. Med. 19 (2011)
3
Approach to characterizing bone tumors at MRI
Radiographs remain essential in the characterization of many bone lesions, and should
not be neglected in the initial evaluation. Most bone tumors have sharply defined
boundaries with the surrounding normal marrow at MRI; such a sharp boundary should
not be misinterpreted as representing the (purely radiographic) criterion of a narrow
zone of transition.
MRI may not depict calcifications unless they are large or very dense, limiting the utility
of MRI in demonstrating calcified tumor matrix (such as in osteoblastoma and
osteosarcoma).
On fat-suppressed T2-weighted images, hyaline cartilage lesions (enchondroma, lowgrade chondrosarcoma) often can be identified due to the presence of multiple, small
lobules with very high signal intensity (due to the high water content of hyaline
cartilage). The distinction between enchondroma and low-grade chondrosarcoma is a
common and difficult problem, however, resulting from lack of consensus on specific
diagnostic criteria — both among pathologists and among radiologists.
The differential diagnosis for benign bone lesions surrounded by extensive marrow
edema includes osteoid osteoma/osteoblastoma, chondroblastoma, Langerhans cell
histiocytosis, and osteomyelitis. As a general (but not absolute) guideline, the greater
the extent of surrounding marrow edema, the more likely the bone lesion is benign.
Marrow edema may be present around malignant bone tumors, as well, in which case it
constitutes the reactive zone of the tumor.
Calcific tendinitis is an important pitfall in the diagnosis of bone tumors, because it can
produce erosions in subjacent bone (most commonly the humeral tuberosities or
femoral trochanters) and an associated marrow edema pattern. Recognition of the
overlying tendinous abnormality (at MRI and/or at radiography) will help prevent
misinterpretation of the subjacent bony changes as representing a metastasis.
Lytic bone metastases generally show moderately increased signal throughout on fatsuppressed T2-weighted images, whereas blastic lesions often show such signal only in
a surrounding rim ("halo"). This "halo," when present, can be particularly helpful in
distinguishing between a blastic metastasis and a bone island — both of which
otherwise usually show low signal on all pulse sequences.
The presence of dot-like or curvilinear flow voids (representing small, high-flow vessels)
within a lytic bone lesion at MRI suggests a renal cancer primary.
Fluid-fluid (blood-fluid) levels have been reported in a wide range of bone and soft
tissue tumors, benign and malignant. A bone lesion composed solely of fluid-fluid levels
is consistent with an aneurysmal bone cyst. If thick septa or soft tissue nodules also are
present within the lesion, other diagnoses must be considered, such as telangiectatic
osteogenic sarcoma, or a secondary aneurysmal bone cyst engrafted on another bone
Proc. Intl. Soc. Mag. Reson. Med. 19 (2011)
4
tumor. Gadolinium-enhanced images are useful in assessing for soft tissue elements
within an otherwise cystic lesion.
References
Bancroft LW, Kransdorf MJ, Peterson JJ, et al. Benign fatty tumors: classification,
clinical course, imaging appearance, and treatment. Skeletal Radiol 2006; 35:719733.
Chan LP, Gee R, Keogh C, Munk PL. Imaging features of fat necrosis. AJR 2003;
181:955-959.
Choi J-A, Lee KH, Jun WS, Yi MG, Lee S, Kang HS. Osseous metastasis from renal
cell carcinoma: “flow-void” sign at MR imaging. Radiology 2003; 228:629-634.
Flemming DJ, Murphey MD, Shekitka KM, Temple HT, Jelinek JJ, Kransdorf MJ.
Osseous involvement in calcific tendinitis: a retrospective review of 50 cases. AJR
2003; 181:965-972.
Huang H-Y, Lal P, Qin J, et al. Low-grade myxofibrosarcoma: a clinicopathologic
analysis of 49 cases treated at a single institution with simultaneous assessment of
the efficacy of 3-tier and 4-tier grading systems. Human Pathol 2004; 35:612-621.
Ilaslan H, Joyce M, Bauer T, et al. Decubital ischemic fasciitis: clinical, pathologic, and
MRI features of pseudosarcoma. AJR 2006; 187:1338-1341.
James SL, Panicek DM, Davies AM. Bone marrow oedema associated with benign and
malignant bone tumours. Eur J Radiol 2008; 67:11-21.
James SLJ, Hughes RJ, Ali KE, Saifuddin A. MRI of bone marrow oedema associated
with focal bone lesions. Clinical Radiology 2006; 61:1003-1009.
Jee WH, Oh SN, McCauley T, et al. Extraaxial neurofibromas versus neurilemmomas:
discrimination with MRI. AJR 2004; 183:629-633.
Kattapuram TM, Suri R, Rosol MS, et al. Idiopathic and diabetic skeletal muscle
necrosis: evaluation by magnetic resonance imaging. Skeletal Radiol 2005; 34:203209.
Kaya M, Wada T, Nagoya S, et al. MRI and histological evaluation of the infiltrative
growth pattern of myxofibrosarcoma. Skeletal Radiol 2008; 37:1085-1090.
Lee JC, Thomas JM, Phillips S, et al. Aggressive fibromatosis: MRI features with
pathologic correlation. AJR 2006; 186:247-254.
Lu C-H, Tsang Y-M, Yu C-W, et al. Rhabdomyolysis: magnetic resonance imaging and
computed tomography findings. J Comput Assist Tomogr 2007; 31:368-374.
Luna A, Martinez S, Bossen E. Magnetic resonance imaging of intramuscular myxoma
with histological comparison and a review of the literature. Skeletal Radiol 2005;
34:19-28.
May DA, Good RB, Smith DK, et al. MR imaging of musculoskeletal tumors and tumor
mimickers with intravenous gadolinium: experience with 242 patients. Skeletal Radiol
1997; 26:2-15.
Mellado JM, Pérez del Palomar L, Diaz L, et al. Long-standing Morel-Lavallée lesions
of the trochanteric region and proximal thigh: MRI features in five patients. AJR
2004; 182:1289-1294.
Proc. Intl. Soc. Mag. Reson. Med. 19 (2011)
5
Moulton JS, Blebea JS, Dunco DM, et al. MR imaging of soft-tissue masses: diagnostic
efficacy and value of distinguishing between benign and malignant lesions. AJR
1995; 164:1191-1199.
Murphey MD, Carroll JF, Flemming DJ, et al. From the archives of the AFIP: benign
musculoskeletal lipomatous lesions. RadioGraphics 2004; 24:1433-1466.
Naylor MF, Nascimento AG, Sherrick AD, et al. Elastofibroma dorsi: radiologic findings
in 12 patients. AJR 1996; 167:683-687.
O’Donnell P, Saifuddin A. The prevalence and diagnostic significance of fluid-fluid
levels in focal lesions of bone. Skeletal Radiol 2004; 33:330-336.
Schweitzer ME, Levine C, Mitchell DG, Gannon FH, Gomella LG. Bull's-eyes and
halos: useful MR discriminators of osseous metastases. Radiology 1993; 188:249252.
Sung M-S, Kang HS, Suh JS, et al. Myxoid liposarcoma: Appearance at MR imaging
with histologic correlation. RadioGraphics 2000; 20:1007-1019.
The Skeletal Lesions Interobserver Correlation among Expert Diagnosticians (SLICED)
Study Group. Reliability of histopathologic and radiologic grading of cartilaginous
neoplasms in long bones. J Bone Joint Surg Am 2007; 89:2113-2123.
Vanel D, Bittoun J, Tardivon A. MRI of bone metastases. Eur Radiol 1998; 8:13451351.
Waters B, Panicek DM, Lefkowitz RA, et al. Low-grade myxofibrosarcoma: CT and MRI
patterns in recurrent disease. AJR 2007; 188:W193-198.
Author has no relevant financial interests or relationships to disclose.
Proc. Intl. Soc. Mag. Reson. Med. 19 (2011)