This is an enhanced PDF from The Journal of Bone and Joint Surgery The PDF of the article you

requested follows this cover page.

Magnetic Resonance Imaging of Soft-Tissue Tumors: Determinate and Indeterminate Lesions

Derek F. Papp, A. Jay Khanna, Edward F. McCarthy, John A. Carrino, Adam J. Farber and Frank J. Frassica J Bone Joint Surg Am. 2007;89:103-115. doi:10.2106/JBJS.G.00711

This information is current as of December 29, 2007 Reprints and Permissions Click here to order reprints or request permission to use material from this article, or locate the article citation on jbjs.org and click on the [Reprints and Permissions] link. The Journal of Bone and Joint Surgery 20 Pickering Street, Needham, MA 02492-3157 www.jbjs.org

Publisher Information

103
COPYRIGHT 2007
BY

THE JOURNAL

OF

BONE

AND JOINT

SURGERY, INCORPORATED

Magnetic Resonance Imaging of Soft-Tissue Tumors: Determinate and Indeterminate Lesions

By Derek F. Papp, MD, A. Jay Khanna, MD, Edward F. McCarthy, MD, John A. Carrino, MD, MPH, Adam J. Farber, MD, and Frank J. Frassica, MD

Introduction he evaluation of patients with soft-tissue masses must be done in a systematic fashion to prevent management errors. Although most soft-tissue masses (approximately 99%) are benign, an error in the management of a soft-tissue sarcoma can lead to limb loss or adversely affect survival1. Before magnetic resonance imaging became easily available, physicians relied on the patients history, physical examination, conventional radiographs, and computed tomography scans for decision-making. These modalities often were insufficient for establishing a definitive diagnosis. The patients history alone cannot provide enough information for a diagnosis and, in fact, may be misleading. For example, lesions identified after a traumatic episode are not necessarily traumatic in origin; only half of soft-tissue sarcomas are painful at presentation2, and the growth rate may not assist in the diagnosis (slow-growing lesions can be malignant or benign). Similarly, although a patient may present with systemic symptoms, the lack of systemic symptoms does not exclude malignancy. Physical examination may provide some clues that may suggest malignancy, but none are pathognomonic. Conventional radiography and computed tomography are not specific enough in differentiating benign and malignant soft-tissue masses. If one relies solely on these modalities, biopsy often is necessary for diagnosis and management. Biopsy is associated with several hazards, including neurovascular injury, hematoma formation, and delayed wound-healing. The use of magnetic resonance imaging to identify softtissue lesions has markedly altered the treatment algorithm for a number of lesions3,4. In contrast to conventional radiography and computed tomography, magnetic resonance imaging provides clear advantages in terms of diagnosis: the spatial resolution of the images is excellent; the images can be reconstructed in multiple planes; lesions can be identified more readily, and

the lesions signal characteristics can help to narrow the differential diagnosis; areas of hemorrhage, cysts, and vascular structures are better seen; and boundaries between the lesion and unaffected bordering tissues can be evaluated readily, facilitating the planning of surgical approaches. In addition, magnetic resonance imaging used for preoperative planning can help to minimize violation and subsequent contamination of crucial neurovascular structures. The purposes of this report were (1) to provide a diagnostic and treatment algorithm for soft-tissue lesions based on the concept of determinate and indeterminate categorization, and (2) to assist the orthopaedic surgeon in developing the skills necessary to categorize a lesion as determinate or indeterminate in nature and to be able to recognize the commonly occurring determinate lesions. With this classification system, surgeons can plan treatment for determinate lesions without a biopsy and can recommend biopsy for indeterminate lesions to establish a diagnosis before planning treatment. Basics of Magnetic Resonance Imaging he patient is positioned appropriately on the magnetic resonance imaging table and, depending on the body part being imaged, an extremity or surface coil also is used. The orientation of the magnetic field aligns the protons within the patient longitudinally with respect to the axis of the scanner. When the imaging begins, an electromagnetic radiofrequency pulse directed toward the patient realigns the protons within the patient in the transverse plane (or perpendicular to the external field for a standard spin-echo sequence). Another pulse then refocuses the protons. As the protons return to their baseline state, they emit energy. The emission of this energy (or signal) is localized to a particular area (or voxel) within the patient, and the information is processed by the system through a Fourier transformation to create the magnetic resonance image5.

Disclosure: The authors did not receive any outside funding or grants in support of their research for or preparation of this work. Neither they nor a member of their immediate families received payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, division, center, clinical practice, or other charitable or nonprofit organization with which the authors, or a member of their immediate families, are affiliated or associated.

J Bone Joint Surg Am. 2007;89(Suppl 3):103-15 doi:10.2106/JBJS.G.00711

104
THE JOURNAL
OF

BONE & JOINT SURGER Y JBJS.ORG VO L U M E 89-A S U P P L E M E N T 3 2007

M A G N E T I C R E S O N A N C E I M A G I N G O F S O F T -T I S S U E TU M O R S : DETERMINATE AND INDETER MINATE LESIONS

TABLE I Tissue Characteristics on Magnetic Resonance Imaging* Signal Intensity Tissue Type Cortical bone Yellow marrow Red marrow Fat Fluid Muscle Ligaments and/or tendons Hemosiderin Hyaline cartilage Physeal scar Labrum T1-Weighted Image Very low High Low/intermediate High Low Intermediate Very low Very low Intermediate Low Very low T2-Weighted Image Very low High Intermediate High High Intermediate Very low Very low Intermediate Low Very low Fat-Suppressed T2-Weighted Image Very low Very low Low/intermediate Very low Very high Low/intermediate Very low Very low Intermediate Low Very low

*Modified from: Khanna AJ, Cosgarea AJ, Mont MA, Andres BM, Domb BG, Evans PJ, Bluemke DA, Frassica FJ. Magnetic resonance imaging of the knee: current techniques and spectrum of disease. J Bone Joint Surg Am. 2001;83:128-41.In addition to the repetition time and echo time, the degree of brightness of the fat also is related to how the sequence is acquired. For example, fat is brighter on a fast-spin-echo sequence than on a conventional spin-echo sequence (these can be made as T1-weighted, T2-weighted, or proton-density-weighted images).

Types of Pulse Sequences y emphasizing the various characteristics of the radiofrequency pulse (strength, frequency, or the time at which the signal is measured), the image can be weighted to create the different pulse sequences, each of which highlights specific tissue characteristics (Table I). T1-weighted imaging allows for excellent anatomic visualization, given its high spatial resolution. T2-weighted imaging shows abnormal changes, free extracellular water, and tissue edema. Fat suppression techniques subtract the signal produced by adipose tissue and highlight abnormal changes and tissue edema even more than do conventional T2-weighted images. A short tau inversion recovery image is a type of fat-suppressed T2-weighted image that highlights areas of increased fluid and/or edema even more than do conventional T2-weighted techniques. Gradient-echo images, which provide excellent visualization of hemorrhage or hemosiderin deposition, are the sequence of choice when imaging suspected vascular lesions, hematomas, or pigmented villonodular synovitis5-7.

to diagnose a soft-tissue lesion that will guide the clinical management of the patient. It is important to note also that discussion with other specialists, such as musculoskeletal radiologists and pathologists, can provide much needed information and often proves to be crucial in making the correct diagnosis. A multidisciplinary approach to soft-tissue lesions is essential8. Determinate lesions include lipoma, hemangioma, ganglion and Baker cysts, hematoma, myositis ossificans, myonerosis, neurofibroma, muscle tear, pigmented villonodular synovitis, bursitis, and aneurysm. Indeterminate lesions include the lesions that cannot be diagnosed without biopsy, such as various sarcomas (e.g., liposarcoma, synovial sarcoma, epithelioid sarcoma, and malignant fibrous histiocytoma). Determinate Lesions Lipoma he lipoma is the most common of all soft-tissue tumors and is composed of mature adipose tissue. It has a characteristic firm, fatty feel and typically confers no symptoms. Nevertheless, some lipomas can cause pain through compression of adjacent neurovascular structures and, accordingly, lesions arising in the deep tissues more commonly cause symptoms than those in more superficial areas. Lipomas have a characteristic appearance on magnetic resonance imaging corresponding to that of adipose tissue. These benign tumors appear bright on T1-weighted imaging (Fig. 1-A) and moderate to bright on T2-weighted images (Fig. 1-B). In all sequences, the lesion appears isointense compared with the subcutaneous fat. The use of fat-suppressed

Determinate and Indeterminate Lesions he algorithm presented relies on the physicians ability to characterize a lesion as determinate or indeterminate. A determinate lesion is one that can be diagnosed definitively by means of the history and physical examination, and with appropriate imaging modalities such as magnetic resonance imaging or conventional radiographsor, in simple terms, a lesion for which a diagnosis can be given without biopsy. An indeterminate lesion is one that must be biopsied to ensure an accurate diagnosis. Each individual physician has a particular degree of confidence in his or her own ability

105
THE JOURNAL
OF

BONE & JOINT SURGER Y JBJS.ORG VO L U M E 89-A S U P P L E M E N T 3 2007

M A G N E T I C R E S O N A N C E I M A G I N G O F S O F T -T I S S U E TU M O R S : DETERMINATE AND INDETER MINATE LESIONS

Fig. 1-A

Fig. 1-B

Lipoma. Axial T1-weighted image (Fig. 1-A) and axial fat-suppressed T2-weighted image (Fig. 1-B) of an intramuscular lipoma (arrows) in the posterior compartment of the thigh. Note that the lipoma has the same signal characteristics as subcutaneous fat on both pulse sequences. There are a few internal septations that may reflect blood vessels.

T2-weighted or short tau inversion recovery images, or both, confirms the diagnosis9. The diagnosis of lipoma can be made with authority when the signal intensity of the lesion matches that of subcutaneous fat on all pulse sequences, including fatsuppression pulse sequences. The image may show fibrous

septations, which may not enhance with gadolinium10,11. These lesions do not penetrate surrounding structures. When a lesion shows areas with fibrous or myxoid degeneration, the physician should consider liposarcoma in the differential diagnosis. In such cases, biopsy is warranted. Similarly, broad sep-

Fig. 2-A

Fig. 2-B

Hemangioma. Axial T1-weighted image (Fig. 2-A) and axial fat-suppressed T2-weighted image (Fig. 2-B) of the right calf showing a mass (arrows) in the medial head of the gastrocnemius muscle. The mass is minimally high signal on the T1-weighted image and heterogeneous and serpentine high signal on the short tau inversion recovery image.

106
THE JOURNAL
OF

BONE & JOINT SURGER Y JBJS.ORG VO L U M E 89-A S U P P L E M E N T 3 2007

M A G N E T I C R E S O N A N C E I M A G I N G O F S O F T -T I S S U E TU M O R S : DETERMINATE AND INDETER MINATE LESIONS

Fig. 3-A

Fig. 3-B

Ganglion cyst. Coronal short tau inversion recovery image (Fig. 3-A) and coronal contrast--enhanced, fat-suppressed T1-weighted image (Fig. 3-B) showing the typical appearance of a ganglion cyst (arrows) along the dorsum of the midfoot. The lesion is round, well-circumscribed, and hyperintense compared with muscle on the short tau inversion recovery imaging with a thin rim of enhancement (arrow in Fig. 3-B), which is compatible with the fluid-filled, cystic nature of a ganglion cyst.

tations or septations with nodularity make the lesion more consistent with liposarcoma. Hemangioma Although a hemangioma is most commonly a cutaneous lesion found in children, the orthopaedic surgeon may be consulted on deep, intramuscular lesions for which physical examination findings are nonspecific. Overall, these lesions have a female predominance as high as 3:112. Fifty percent of patients describe a history of pain, which is theorized to arise as a result of the lesion causing ischemia in the surrounding soft tissues through a vascular steal phenomenon13. Superficial lesions may feel soft and may increase in size when placed in a dependent position, but it may be difficult to detect this posture-related change with deep lesions14. Although conventional radiographs often show phleboliths, the magnetic resonance imaging findings typically prove diagnostic and show lipomatous tissue (most of which is contained in the margins of the lesion), vascular formations, hemosiderin deposits, and fibrosis. Overall, the lesion has a heterogeneous appearance, bright on T1-weighted images (Fig. 2-A) secondary to the adipose tissue of the lesion and bright on T2-weighted (Fig. 2-B) and short tau inversion recovery imaging, corresponding with the vascularity of the lesion. The lesion may have a serpentine appearance secondary to the multiple vessels in the lesion. Lobules and septations also are found within the lesion4,13. Areas of high-velocity blood flow corresponding to feeding vessels appear as signal

voids. Gadolinium administration results in bright enhancement of the lesion15. Ultrasound can also prove useful in the diagnosis of this lesion13,16. Ganglion Cyst Ganglion cysts that involve the dorsal or volar aspect of the wrist are one of the few types of soft-tissue lesions for which a physical examination is characteristic enough to label the lesion as determinate. These lesions need no imaging other than conventional radiography and can be treated as the patient and physician prefer. Some investigators have suggested that these lesions form when repeated stress causes mucoid degeneration of associated periarticular tissues or tendon sheaths, whereas others hypothesize that lining cells cause production of a hyaluronic-acid-rich substance, which causes the formation of a cystic lesion17-19. Ganglion cysts also commonly occur in the foot and ankle, although they can be associated with any tendon sheath, labrum, or capsule. Unlike a Baker cyst, a ganglion cyst does not communicate with the joint. When such a lesion is found in a location other than the wrist, additional imaging should be performed. On magnetic resonance imaging, the lesion appears smooth and round or ovular, and it may have septations. The wall of the cyst is well-circumscribed and forms a distinct border against the adjacent tissues. Although the septations may enhance with gadolinium administration, the lesion itself should not. The lesion appears with a signal intensity similar to that of waterbright on T2-weighted imaging or short tau

107
THE JOURNAL
OF

BONE & JOINT SURGER Y JBJS.ORG VO L U M E 89-A S U P P L E M E N T 3 2007

M A G N E T I C R E S O N A N C E I M A G I N G O F S O F T -T I S S U E TU M O R S : DETERMINATE AND INDETER MINATE LESIONS

inversion recovery4,9,20 (Fig. 3-A) and dark on T1-weighted imaging (Fig. 3-B). Baker Cyst Originally described by Baker21, the synovial cyst (now known as the Baker cyst) forms when synovial fluid insinuates from the knee joint into communicating bursae or by causing herniation of the synovial membrane itself. This lesion occurs most commonly under the medial head of the gastrocnemius muscle. The lesion is often asymptomatic, but it can cause pain. In addition, the cyst may rupture or leak, causing resultant swelling and/or pain in the affected limb; thrombophlebitis, compartment syndrome, and claudication can occur22,23. The cause of the lesion is intra-articular and can include abnormalities ranging from anterior cruciate ligament damage to osteoarthritis, although meniscal tears are the most common cause24,25. Ultrasound may be useful for diagnosis, but magnetic resonance imaging is the study of choice because it can delineate underlying abnormalities such as meniscal or anterior cruciate ligament tears. Magnetic resonance imaging findings of a Baker cyst are similar to those of a ganglion cyst. The Baker cyst is an ovular, well-defined mass found posterior to the knee joint. The fluid appears dark on T1-weighted images and bright on T2-weighted and short tau inversion recovery imaging (Figs. 4-A and 4-B). The signal intensity of the lesion matches that of joint fluid on all pulse sequences. Hematoma Although most patients with a hematoma do not present to an orthopaedist, it is important to keep this lesion in the differen-

tial diagnosis of soft-tissue lesions. The patient often relates a history of trauma, and residual ecchymosis can sometimes be seen during the physical examination. When a history of trauma is not present, it is important to ask the patient about systemic anticoagulation or clotting deficiencies because these conditions can be associated with chronic expanding hematomas26. Most such lesions will resolve on their own, but the natural history of the lesion may also follow two other courses: the hematoma may begin to calcify on the periphery (eventually becoming myositis ossificans) or it may continue to expand, which may be related to continual irritation caused by hemosiderin breakdown products. In such instances, the lesion does not heal, and it chronically expands secondary to continued capillary bleeding27,28. Continued expansion of the lesion may cause compression of neurovascular structures or pressure erosion of nearby osseous structures27. T1-weighted imaging shows a heterogeneous pattern. Areas of high signal intensity represent areas of continuing hemorrhage. Overall, the lesion is well defined and does not invade other muscle compartments (Fig. 5-A). T2-weighted images show heterogeneity (Fig. 5-B). Areas of low signal intensity, i.e., signal dropout, occur in areas of hemosiderin deposition. Areas of high signal intensity represent granulation tissue. A low-signal pseudocapsule often is seen. Fluidfluid levels have also been described29. Gradient-echo imaging can be used to investigate further the areas of hemosiderin deposition26,27 and will show the profound signal dropout (dark areas) that is characteristically seen on gradient-echo imaging in the presence of hemosiderin or other ferromagnetic materials such as metallic implants. Gadolinium-

Fig. 4-A

Fig. 4-B

Baker cyst. Sagittal T1-weighted image (Fig. 4-A) and short tau inversion recovery image (Fig. 4-B) showing a lesion (arrows) in the popliteal fossa situated between the medial head of the gastrocnemius muscle and the semimembranosus muscle. The lesion appears hypointense compared with muscle on T1-weighted images, and it shows high signal intensity on the short tau inversion recovery image because of its fluid content. The location, configuration, and signal intensity are characteristic for a Baker cyst.

108
THE JOURNAL
OF

BONE & JOINT SURGER Y JBJS.ORG VO L U M E 89-A S U P P L E M E N T 3 2007

M A G N E T I C R E S O N A N C E I M A G I N G O F S O F T -T I S S U E TU M O R S : DETERMINATE AND INDETER MINATE LESIONS

Fig. 5-A

Fig. 5-B

Hematoma. Sagittal axial T1-weighted image (Fig. 5-A) and sagittal fat-suppressed T2-weighted image (Fig. 5-B) of the leg showing a mass lesion in the anterior compartment (arrow in Fig. 5-B). The T1-weighted image shows areas of hyperintensity (arrowheads), reflecting methemoglobin. The highsignal-intensity areas on the fat-suppressed T2-weighted image are related to soft-tissue edema and hemorrhage. Clinical correlation is especially helpful for the diagnosis of hematoma, and attention should be given to the presence or absence of coagulopathy, history of surgery, or other trauma.

Fig. 6-A

Fig. 6-B

Myositis ossificans. Axial T1-weighted image (Fig. 6-A) and axial short tau inversion recovery image (Fig. 6-B) showing a lesion (arrows) with an associated rim of bright T1 signal that may reflect subtle expansion of the surrounding fascia or peripheral methemoglobin. The lesion appears isointense to slightly hypointense compared with muscle on T1-weighted images and has a moderate, increased signal on the short tau inversion recovery images with foci of bright signal that likely reflect fluid components. The location adjacent to the anterior femoral cortex is characteristic. If only the lesion rim enhances, then it is unlikely to represent a neoplasm, and early myositis ossificans is more likely.

109
THE JOURNAL
OF

BONE & JOINT SURGER Y JBJS.ORG VO L U M E 89-A S U P P L E M E N T 3 2007

M A G N E T I C R E S O N A N C E I M A G I N G O F S O F T -T I S S U E TU M O R S : DETERMINATE AND INDETER MINATE LESIONS

enhanced images can be used to confirm the identity of the lesion. When there is no enhancement with gadolinium contrast, the lesion can be considered determinate. Although Liu et al. 26 reported cases of gadolinium enhancement of chronic hematomas, the increased chance of malignancy makes diagnosis uncertain, and these lesions cannot be considered determinate26,30. Myositis Ossificans Some injuries progress from a simple hematoma to myositis ossificans. The exact mechanism for this progression is unknown, although it develops most often during the third decade of life31,32. The most commonly affected muscles are the quadriceps and brachialis31,32. It is thought that these lesions occur more commonly after severe injuries. Diagnosis of this lesion can often be made with conventional radiography. The lesion appears in the pattern of mature, lamellar bone and is well defined33. For early lesions, when calcification is absent, magnetic resonance imaging provides an additional imaging option. On T1-weighted imaging, the lesion most often resides inside the same compartment as the traumatized muscle, and the lesion is isointense or slightly hypointense relative to skeletal muscle. In some instances, the lesion may be subtle, so that it may be recognized only by alteration of the fascial planes32 (Fig. 6-A). On T2-weighted and short tau inversion recovery imaging, the

lesion appears hyperintense compared with muscle (Fig. 6B). Surrounding edema may or may not be present. The periphery of the lesion may have decreased signal intensity, showing the zonal pattern of growth; when this pattern is recognized, it is diagnostic. Myonecrosis (Diabetic and Idiopathic) Myonecrosis may present as a soft-tissue lesion. This condition is associated most commonly with diabetes, but other predisposing factors include alcohol consumption. Diabetic myonecrosis can even occur as the initial event in a patients history, although most patients have severe neuropathy or other more severe sequelae in the disease process34,35. This painful lesion develops quickly with or without a history of trauma. It is thought that the lesion develops secondary to endothelial damage from diabetic microangiopathy, combined with activated coagulation factors and the continued presence of fibrin degradation products36. The only laboratory abnormality associated with the disease is an elevated creatinine serum kinase, but even this finding is not always present35,37. Biopsy should be avoided because these patients are at high risk for wound-healing complications38. Magnetic resonance imaging is the diagnostic modality of choice. Swelling of the muscle occurs and can be visualized best on T1weighted imaging, although fascial layers and muscle fiber

Fig. 7-A

Fig. 7-B

Diabetic myonecrosis. Axial T1-weighted image (Fig. 7-A) and axial fat-suppressed T2-weighted image (Fig. 7-B) showing isointense swelling of the involved muscle, with mildly displaced fascial planes and effacement of the fat signal intensity within the muscle (arrowheads in Fig. 7-A). The T2weighted image shows diffuse heterogeneous high signal intensity in the muscles, suggestive of edema (arrowheads in Fig. 7-B). Perifascial fluid (arrow in Fig. 7-B) and subcutanteous edema are also present. If intravenous gadolinium is administered, a focal area of nonenhancing lesion is seen (not shown). This constellation of imaging findings is highly suggestive of muscle necrosis.

110
THE JOURNAL
OF

BONE & JOINT SURGER Y JBJS.ORG VO L U M E 89-A S U P P L E M E N T 3 2007

M A G N E T I C R E S O N A N C E I M A G I N G O F S O F T -T I S S U E TU M O R S : DETERMINATE AND INDETER MINATE LESIONS

Fig. 8-A

Fig. 8-B

Neurofibroma. Sagittal T1-weighted image (Fig. 8-A) and coronal short tau inversion recovery image (Fig. 8-B) of the ankle showing a wellcircumscribed mass (arrows) in the plantar aspect of the foot that is hypointense compared with muscle on the T1-weighted image. The short tau inversion recovery image shows the typical target sign in which the periphery of the lesion is bright and the center is dark.

Fig. 9-A

Fig. 9-B

Rectus femoris muscle tear. Axial T1-weighted image (Fig. 9-A) and axial short tau inversion recovery image (Fig. 9-B) of the distal part of the thigh showing the lesion (arrows) as a focal contour deformity with swelling of the muscle belly. Short tau inversion recovery imaging shows associated muscle edema, reflecting an acute or subacute tear.

111
THE JOURNAL
OF

BONE & JOINT SURGER Y JBJS.ORG VO L U M E 89-A S U P P L E M E N T 3 2007

M A G N E T I C R E S O N A N C E I M A G I N G O F S O F T -T I S S U E TU M O R S : DETERMINATE AND INDETER MINATE LESIONS

Fig. 10-A

Fig. 10-B

Pigmented villonodular synovitis of the elbow. Coronal T1-weighted image (Fig. 10-A) showing a lobular hypointense soft-tissue mass (arrow) with areas of signal dropout corresponding to hemosiderin deposition. Sagittal short tau inversion recovery image (Fig. 10-B) showing several loci of the pigmented villonodular synovitis process (arrow) with high signal intensity, along with areas of signal dropout corresponding to hemosiderin deposition.

patterns are maintained (Fig. 7-A). T2-weighted imaging shows the lesion more clearly, with diffuse hyperintensity. The overall picture is mixed, with areas of necrosis and muscle regeneration (Fig. 7-B). The lack of invasion of surrounding structures differentiates this lesion from sarcomas. Gadolinium contrast may be used, with enhancing areas corresponding to healing or viable tissue, whereas nonenhancing areas correspond to loci of tissue necrosis35,37. Neurofibroma Patients with a history of neurofibromatosis Type 1 present with soft-tissue neurofibromas that can be mistaken for sarcomas. These patients often do not have any associated neurologic deficit39. The masses occur superficially and in the deep tissues, and they may or may not cause pain. Because patients with neurofibromatosis Type 1 are at high risk for malignancy, magnetic resonance imaging can be useful in determining the nature of the lesion and in ruling out dedifferentiation of the lesion into a neurofibrosarcoma. On T1-weighted imaging, the lesion appears darker than skeletal muscle and often has a nodular appearance (Fig. 8-A). T2-weighted imaging shows areas of high signal intensity (corresponding with myxoid areas) and areas of low signal intensity (corresponding with compressed nerve fibers) (Fig. 8-B). Commonly, a target sign is seen, with an area of low signal intensity within a larger, high-signal-intensity area40,41. It should be noted that continued observation with serial magnetic resonance imaging can rule out dedifferentiation to neurofibrosarcoma.

Muscle Tear A patient with a muscle tear often can recall the exact moment when the injury occurred. There is a feeling of sudden weakness, and a snapping sound may or may not be heard. The muscle in question often swells and becomes quite tender. Other signs, including ecchymosis and edema, occur, although none is specific for muscle tear. Over time, the swelling subsides, but the patient may be left with a lump, i.e., soft-tissue fullness, in the area that corresponds to scarring from the previous lesion and may be mistaken for a soft-tissue tumor42,43. Although this presentation is the most common one, it should be noted that muscle tears can also occur insidiously. One series of muscle tears presenting as soft-tissue masses showed that six of seven occurred without a traumatic event and that only 50% caused pain44. Magnetic resonance imaging is the best imaging modality for making the diagnosis of a muscle tear. T1-weighted imaging shows the ruptured muscle or the formation of scar, or both. Care should be taken to evaluate the course of the involved muscle in the axial, coronal, and sagittal planes and to confirm that the morphology of the injury suggests the presence of a muscle tear. Most frequently, the deformity is clear, and there is no evidence of invasion of the surrounding structures (Fig. 9-A). The time relative to the injury dictates the muscles appearance on T2-weighted imaging. Acute injuries show hyperintense lesions with surrounding edema (Fig. 9-B). Chronic injuries show low to intermediate signal intensity.

112
THE JOURNAL
OF

BONE & JOINT SURGER Y JBJS.ORG VO L U M E 89-A S U P P L E M E N T 3 2007

M A G N E T I C R E S O N A N C E I M A G I N G O F S O F T -T I S S U E TU M O R S : DETERMINATE AND INDETER MINATE LESIONS

Fig. 11-A

Fig. 11-B

Olecranon bursitis. Axial T1-weighted image (Fig. 11-A) and axial T2-weighted image (Fig. 11-B) of the elbow showing a smooth, well-circumscribed ovoid lesion (arrows) exhibiting fluid-like signal intensity, hypointense signal on T1-weighted images (lower than that of skeletal muscle), and hyperintense signal on non-fat-suppressed T2-weighted images (brighter than that of fat) compatible with a cystic structure. The location is characteristic.

Pigmented Villonodular Synovitis Another lesion that might be mistaken for a soft-tissue mass, pigmented villonodular synovitis, occurs infrequently, but it can be diagnosed with magnetic resonance imaging. Patients typically complain of pain in or around the involved joint, with associated swelling. Warmth and effusion of the joint also may occur43,45. The most common location is the knee, followed by the hip, ankle, and elbow. The disease has two different forms: one that affects the entire joint in a diffuse pattern and another that presents in a nodular form. The nodular form is more easily resected, but treatment for either form should be attempted because secondary arthritis will develop if the abnormal tissue is not excised46. Conventional radiographs show a joint effusion with joint erosion after continued disease, although most of the joint space is maintained. Narrowing of the joint occurs concentrically. The lesion has a characteristic appearance on magnetic resonance imaging: low signal intensity on both T1weighted and T2-weighted imaging secondary to the high hemosiderin content of the lesion (Figs. 10-A and 10-B). Gradient-echo images, which are particularly sensitive to ferromagnetic materials such as hemosiderin, show profound signal dropout in patients with pigmented villonodular synovitis and may help to differentiate this lesion from synovial chondromatosis. Thus, the lesion can be described as blooming from the joint capsule45. Although gradient-echo and fast-field-echo imaging may show hemosiderin content better

than T1-weighted or T2-weighted imaging47 does, hemosiderin deposition may not occur if the lesion is imaged too early, i.e., before maturation46,48. If a mass is noted, it is usually indicative of nodular disease. Bursitis When a mass develops near a joint, the physician should consider bursitis among the potential diagnoses. The inflamed bursa can present acutely, subacutely, or chronically; the last presentation may mimic a tumor. The cause of the bursitis may vary, but most commonly it is associated with trauma, infection, gout, or rheumatoid arthritis49-51. Similarly, the affected area varies, and almost any joint can be involved. The most common locations include the prepatellar area and the greater trochanter. In these locations, the physician can make the diagnosis without magnetic resonance imaging. Magnetic resonance imaging is most useful in assisting the physician to make the diagnosis of bursitis in other locations (e.g., about the olecranon, ankle, and pes anserinus insertion)49,51,52. On T1-weighted imaging, the inflamed bursa appears as an ovular or cystic structure that is hypointense compared with muscle (Fig. 11-A). T2-weighted imaging shows the high signal intensity common to all fluid-filled lesions (Fig. 11-B), and the periphery of the lesion may also appear bright. Gadolinium enhancement occurs in infectious or inflammatory cases (e.g., rheumatoid arthritis or gout)49,51,53.

113
THE JOURNAL
OF

BONE & JOINT SURGER Y JBJS.ORG VO L U M E 89-A S U P P L E M E N T 3 2007

M A G N E T I C R E S O N A N C E I M A G I N G O F S O F T -T I S S U E TU M O R S : DETERMINATE AND INDETER MINATE LESIONS

Aneurysm and/or Pseudoaneurysm Although the formation of an aneurysm or pseudoaneurysm in an extremity is uncommon, it may mimic a malignant soft-tissue lesion. The most common cause of such a lesion is trauma, although there are other causes (e.g., congenital disease, atherosclerosis, and mycotic aneurysms). The incidence approaches its peak in individuals who are sixty-five years of age or older, and men are affected more often than women54. The formation of a pseudoaneurysm occurs more commonly than does a true aneurysm in the extremities, and aneurysms rarely form below the knee55. Popliteal artery aneurysms, the most common aneurysm in the lower extremity, can affect nearby neurovascular structures56-58. Symptoms such as paresthesias or painful claudication secondary to compression of adjacent neurovascular structures can occur, although the mass itself may be nonpainful. Palpation of the lesion often reveals a pulse. On both T1-weighted and T2-weighted imaging, the size of the associated vessel is at least 50% larger than that of neighboring sections (Fig. 12). With gadolinium enhancement, the aneurysm appears bright (as does the entire vessel). Mural thrombus may or may not exist and has imaging characteristics similar to those of a hematoma. Indeterminate Lesions and Biopsy Techniques hen the physician cannot reach a diagnosis on the basis of the patient history, physical examination, and advanced imaging, the lesion is considered indeterminate and possibly malignant. Lesions that are hypointense compared with muscle on T1-weighted imaging and hyperintense on T2-weighted imaging are suspicious for soft-tissue sarcomas. Similarly, certain elements of the physical examination, such as a size of >5 cm, a firm mass, subfascial location, and a matted lesion adherent to neighboring tissues, are characteristic of soft-tissue sarcomas. Although suggestive of malignancy, these characteristics are not diagnostic59. If the lesion is indeterminate, a tissue sample must be obtained for diagnosis. Given the consequences of misdiagnosis or failure to diagnose a soft-tissue lesion, the importance of obtaining a biopsy cannot be overemphasized. The three primary techniques include open biopsy, core needle biopsy, and excisional biopsy. The standard in obtaining diagnostic samples of tissue is the open biopsy technique. The technique allows direct visualization of the lesion. Frozen sections obtained at the time of the procedure ensure that the surgeon has obtained a diagnostic section of tissue. When performing an open biopsy, the surgeon must meticulously plan the approach because subsequent procedures will require excision of the tract of the biopsy. Transverse incisions must be avoided, and care should be taken to avoid exposing (and thus contaminating) neurovascular structures during the procedure. At all times, meticulous hemostasis must be maintained to prevent hematoma formation. Hematomas that track under subcutaneous tissue or through intermuscular spaces carry

Fig. 12

Pseudoaneurysm of the popliteal artery as seen on coronal T2weighted imaging. The lesion (arrow) was caused by osteochondroma. The vessel is >50% larger than the previous section of vessel.

tumor cells and may contaminate these areas9. The core needle biopsy, especially when performed at centers that specialize in musculoskeletal tumors, is an alternative to open biopsy. Some investigators have questioned the sensitivity of core biopsy alone60, but numerous studies have shown that it provides an accurate diagnosis in >80% of cases61-64. The numerous advantages of core needle biopsy include decreased patient morbidity, less subcutaneous hematoma formation, and avoidance of the need for open biopsy. The procedure often is performed on an outpatient basis and without the use of general anesthesia. Advanced techniques such as cytogenetic studies and electron microscopy also can be used with this technique9. When available, core needle biopsy often is preferred to open biopsy. In general, excisional biopsy should be avoided, but when used, it should be reserved for lesions measuring <2 cm or lesions for which the surgeon already has a definitive diagnosis. It should not be used for large lesions or lesions that do not have a diagnosis because the consequences of inadequate resection are severe. If excisional biopsy must be repeated, the revision procedure becomes markedly more difficult than the index procedure because of the disruption of soft-tissue planes and because hematoma formation contaminates the wound bed. In addition, repeat excision results in lower rates of successful tumor removal65,66.

114
THE JOURNAL
OF

BONE & JOINT SURGER Y JBJS.ORG VO L U M E 89-A S U P P L E M E N T 3 2007

M A G N E T I C R E S O N A N C E I M A G I N G O F S O F T -T I S S U E TU M O R S : DETERMINATE AND INDETER MINATE LESIONS

Corresponding author: A. Jay Khanna, MD c/o Elaine P. Henze, BJ, ELS, Medical Editor, Department of Ortho-

paedic Surgery, Johns Hopkins Bayview Medical Center, 4940 Eastern Avenue, #A672, Baltimore, MD 21224-2780. E-mail address: ehenze1@jhmi.edu

References
1. Weiss SW, Goldblum JR. General considerations. In: Weiss SW, Goldblum JR, editors. Enzinger and Weisss soft tissue tumors. 4th ed. St. Louis: Mosby; 2001. p 1-19. 2. Hussein R, Smith MA. Soft tissue sarcomas: are current referral guidelines sufficient? Ann R Coll Surg Engl. 2005;87:171-3. 3. Hanna SL, Fletcher BD. MR imaging of malignant soft-tissue tumors. Magn Reson Imaging Clin N Am. 1995;3:629-50. 4. Woertler K. Soft tissue masses in the foot and ankle: characteristics on MR imaging. Semin Musculoskelet Radiol. 2005;9:227-42. 5. Farber A, Fayad L, Johnson T, Cascio B, Shindle M, Neubauer P Khanna AJ. , Magnetic resonance imaging of the shoulder. Current techniques and spectrum of disease. J Bone Joint Surg Am. 2006;88 Suppl 4:64-79. 6. Bitar R, Leung G, Perng R, Tadros S, Moody AR, Sarrazin J, McGregor C, Christakis M, Symons S, Nelson A, Roberts TP MR pulse sequences: what every . radiologist wants to know but is afraid to ask. Radiographics. 2006;26:513-37. 7. Hughes TH, Sartoris DJ, Schweitzer ME, Resnick DL. Pigmented villonodular synovitis: MRI characteristics. Skeletal Radiol. 1995;24:7-12. 8. Frassica FJ, Khanna JA, McCarthy EF. The role of MR imaging in soft tissue tumor evaluation: perspective of the orthopedic oncologist and musculoskeletal pathologist. Magn Reson Imaging Clin N Am. 2000;8:915-27. 9. Frassica FJ, Thompson RC. Evaluation, diagnosis, and classification of benign soft tissue tumors. Instr Course Lect. 1996;45:447-60. 10. Drevelegas A, Pilavaki M, Chourmouzi D. Lipomatous tumors of soft tissue: MR appearance with histological correlation. Eur J Radiol. 2004;50:257-67. 11. Kirby EJ, Shereff MJ, Lewis MM. Soft-tissue tumors and tumor-like lesions of the foot. An analysis of eighty-three cases. J Bone Joint Surg Am. 1989;71:621-6. 12. Bauland CG, van Steensel MAM, Steijlen PM, Rieu PNMA, Spauwen PHM. The pathogenesis of hemangiomas: a review. Plast Reconstr Surg. 2006;117:29e-35e. 13. Olsen KI, Stacy GS, Montag A. Soft-tissue cavernous hemangioma. Radiographics. 2004;24:849-54. 14. Murphey MD, Fairbairn KJ, Parman LM, Baxter KG, Parsa MB, Smith WS. From the archives of the AFIP Musculoskeletal angiomatous lesions: radiologic. pathologic correlation. Radiographics. 1995;15:893-917. 15. Gampper TJ, Morgan RF. Vascular anomalies: hemangiomas. Plast Reconstr Surg. 2002;110:572-88. 16. Ostlere S, Graham R. Imaging of soft tissue masses. Imaging. 2005; 17:268-84. 17. Bui-Mansfield LT, Youngberg RA. Intraarticular ganglia of the knee: prevalence, presentation, etiology, and management. AJR Am J Roentgenol. 1997; 168:123-7. 18. Feldman F, Johnston A. Intraosseous ganglion. Am J Roentgenol Radium Ther Nucl Med. 1973;118:328-43. 19. McCarthy CL, McNally EG. The MRI appearance of cystic lesions around the knee. Skeletal Radiol. 2004;33:187-209. 20. Yilmaz E, Karakurt L, Ozercan I, Ozdemir H. A ganglion cyst that developed from the infrapatellar fat pad of the knee. Arthroscopy. 2004;20:e65-8. 21. Baker WM. On the formation of synovial cysts in the leg in connection with disease of the knee joint. St Bartholomew Hosp Rep. 1877;13:245-61. 22. Fritschy D, Fasel J, Imbert JC, Bianchi S, Verdonk R, Wirth CJ. The popliteal cyst. Knee Surg Sports Traumatol Arthrosc. 2006;14:623-8. 23. Zhang WW, Lukan JK, Dryjski ML. Nonoperative management of lower extremity claudication caused by a Baker's cyst: case report and review of the literature. Vascular. 2005;13:244-7. 24. Miller TT, Staron RB, Koenigsberg T, Levin TL, Feldman F. MR imaging of Baker cysts: association with internal derangement, effusion, and degenerative arthropathy. Radiology. 1996;201:247-50. 25. Sansone V, de Ponti A. Arthroscopic treatment of popliteal cyst and associated intra-articular knee disorders in adults. Arthroscopy. 1999;15:368-72. 26. Liu PT, Leslie KO, Beauchamp CP Cherian SF. Chronic expanding hematoma of , the thigh simulating neoplasm on gadolinium-enhanced MRI. Skeletal Radiol. 2006;35:254-7. 27. Aoki T, Nakata H, Watanabe H, Maeda H, Toyonaga T, Hashimoto H, Nakamura T. The radiological findings in chronic expanding hematoma. Skeletal Radiol. 1999;28:396-401. 28. Reid JD, Kommareddi S, Lankerani M, Park MC. Chronic expanding hematomas. A clinicopathologic entity. JAMA. 1980;244:2441-2. 29. Keenan S, Bui-Mansfield LT. Musculoskeletal lesions with fluid-fluid level: a pictorial essay. J Comput Assist Tomogr. 2006;30:517-24. 30. Imaizumi S, Morita T, Ogose A, Hotta T, Kobayashi H, Ito T, Hirata Y. Soft tissue sarcoma mimicking chronic hematoma: value of magnetic resonance imaging in differential diagnosis. J Orthop Sci. 2002;7:33-7. 31. Hait G, Boswick JA Jr, Stone NH. Heterotopic bone formation secondary to trauma (myositis ossificans traumatica). J Trauma. 1970;10:405-11. 32. Parikh J, Hyare H, Saifuddin A. The imaging features of post-traumatic myositis ossificans, with emphasis on MRI. Clin Radiol. 2002;57:1058-66. 33. McCarthy EF, Frassica FJ. Tumor-like lesions. In: Pathology of bone and joint disorders: with clinical and radiographic correlation. Philadelphia: WB Saunders; 1998. p 291-306. 34. Bunch TJ, Birskovich LM, Eiken PW. Diabetic myonecrosis in a previously healthy woman and review of a 25-year Mayo Clinic experience. Endocr Pract. 2002;8:343-6. 35. Kattapuram TM, Suri R, Rosol MS, Rosenberg AE, Kattapuram SV. Idiopathic and diabetic skeletal muscle necrosis: evaluation by magnetic resonance imaging. Skeletal Radiol. 2005;34:203-9. 36. Bjornskov EK, Carry MR, Katz FH, Lefkowitz J, Ringel SP Diabetic muscle . infarction: a new perspective on pathogenesis and management. Neuromuscul Disord. 1995;5:39-45. 37. Khoury NJ, el-Khoury GY, Kathol MH. MRI diagnosis of diabetic muscle infarction: report of two cases. Skeletal Radiol. 1997;26:122-7. 38. Keller DR, Erpelding M, Grist T. Diabetic muscular infarction. Preventing morbidity by avoiding excisional biopsy. Arch Intern Med. 1997;157:1611-2. 39. Theos A, Korf BR. Pathophysiology of neurofibromatosis type 1. Ann Intern Med. 2006;144:842-9. 40. Kostelic JK, Haughton VM, Sether LA. Lumbar spinal nerves in the neural foramen: MR appearance. Radiology. 1991;178:837-9. 41. Lim R, Jaramillo D, Poussaint TY, Chang Y, Korf B. Superficial neurofibroma: a lesion with unique MRI characteristics in patients with neurofibromatosis type 1. AJR Am J Roentgenol. 2005;184:962-8. 42. Peterson L, Stener B. Old total rupture of the adductor longus muscle. A report of seven cases. Acta Orthop Scand. 1976;47:653-7. 43. Rubin SJ, Feldman F, Staron RB, Zwass A, Totterman S, Meyers SP Magnetic . resonance imaging of muscle injury. Clin Imaging. 1995;19:263-9. 44. Temple HT, Kuklo TR, Sweet DE, Gibbons CLMH, Murphey MD. Rectus femoris muscle tear appearing as a pseudotumor. Am J Sports Med. 1998;26:544-8. 45. Al-Nakshabandi NA, Ryan AG, Choudur H, Torreggiani W, Nicoloau S, Munk PL, Al-Ismail K. Pigmented villonodular synovitis. Clin Radiol. 2004;59:414-20. 46. Tyler WK, Vidal AF, Williams RJ, Healey JH. Pigmented villonodular synovitis. J Am Acad Orthop Surg. 2006;14:376-85. 47. Cheng XG, You YH, Liu W, Zhao T, Qu H. MRI features of pigmented villonodular synovitis (PVNS). Clin Rheumatol. 2004;23:31-4. 48. Bouali H, Deppert EJ, Leventhal LJ, Reeves B, Pope T. Pigmented villonodular synovitis: a disease in evolution. J Rheumatol. 2004;31:1659-62. . 49. Floemer F, Morrison WB, Bongartz G, Ledermann HP MRI characteristics of olecranon bursitis. AJR Am J Roentgenol. 2004;183:29-34. 50. Garcia-Porrua C, Gonzalez-Gay MA, Vazquez-Caruncho M. Tophaceous gout mimicking tumoral growth. J Rheumatol. 1999;26:508-9. 51. Mutlu H, Sildiroglu H, Pekkafali Z, Kizilkaya E, Cermik H. MRI appearance of retrocalcaneal bursitis and rheumatoid nodule in a patient with rheumatoid

115
THE JOURNAL
OF

BONE & JOINT SURGER Y JBJS.ORG VO L U M E 89-A S U P P L E M E N T 3 2007

M A G N E T I C R E S O N A N C E I M A G I N G O F S O F T -T I S S U E TU M O R S : DETERMINATE AND INDETER MINATE LESIONS

arthritis. Clin Rheumatol. 2006;25:734-6. 52. Rennie WJ, Saifuddin A. Pes anserine bursitis: incidence in symptomatic knees and clinical presentation. Skeletal Radiol. 2005;34:395-8. 53. Dawn B, Williams JK, Walker SE. Prepatellar bursitis: a unique presentation of tophaceous gout in an normouricemic patient. J Rheumatol. 1997; 24:976-8. 54. Wain RA, Hines G. A contemporary review of popliteal artery aneurysms. Cardiol Rev. 2007;15:102-7. 55. Kato T, Takagi H, Sekino S, Manabe H, Matsuno Y, Furuhashi K, Sekido Y, Umemoto T. Dorsalis pedis artery true aneurysm due to atherosclerosis: case report and literature review. J Vasc Surg. 2004;40:1044-8. 56. Kars HZ, Topaktas S, Dogan K. Aneurysmal peroneal nerve compression. Neurosurgery. 1992;30:930-1. 57. Katz SG, Kohl RD, Razack N. Bilateral infrapopliteal artery aneurysms. Ann Vasc Surg. 1992;6:168-70. 58. Tatli S, Lipton MJ, Davison BD, Skorstad RB, Yucel EK. From the RSNA refresher courses: MR imaging of aortic and peripheral vascular disease. Radiographics. 2003;23:S59-S78. 59. Sim FH, Frassica FJ, Frassica DA. Soft-tissue tumors: diagnosis, evaluation, and management. J Am Acad Orthop Surg. 1994;2:202-11.

60. Ogilvie CM, Torbert JT, Finstein JL, Fox EJ, Lackman RD. Clinical utility of percutaneous biopsies of musculoskeletal tumors. Clin Orthop Relat Res. 2006;450:95-100. 61. Duda SH, Johst U, Krahmer K, Pereira P Konig C, Schafer J, Huppert P Schott , , U, Bohm P Claussen CD. [Technique and results of CT-guided percutaneous bone , biopsy]. Orthopade. 2001;30:545-50. German. 62. Issakov J, Flusser G, Kollender Y, Merimsky O, Lifschitz-Mercer B, Meller I. Computed tomography-guided core needle biopsy for bone and soft tissue tumors. Isr Med Assoc J. 2003;5:28-30. 63. Jelinek JS, Murphey MD, Welker JA, Henshaw RM, Kransdorf MJ, Shmookler BM, Malawer MM. Diagnosis of primary bone tumors with image-guided percutaneous biopsy: experience with 110 tumors. Radiology. 2002;223:731-7. 64. Logan PM, Connell DG, OConnell JX, Munk PL, Janzen DL. Image-guided percutaneous biopsy of musculoskeletal tumors: an algorithm for selection of specific biopsy techniques. AJR Am J Roentgenol. 1996;166:137-41. 65. Goodlad JR, Fletcher CDM, Smith MA. Surgical resection of primary softtissue sarcoma. Incidence of residual tumour in 95 patients needing re-excision after local resection. J Bone Joint Surg Br. 1996;78:658-61. 66. Noria S, Davis A, Kandel R, Levesque J, OSullivan B, Wunder J, Bell R. Residual disease following unplanned excision of soft-tissue sarcoma of an extremity. J Bone Joint Surg Am. 1996;78:650-5.