DOI: 10.1148/rg.266055184
RadioGraphics 2006;26:1873-1876
© RSNA, 2006
Best Cases from the AFIP
Elastofibroma Dorsi1
Jennifer E. Ochsner, MD,
Sarah A. Sewall, MD,
Gregory N. Brooks, MD and
Rashmi Agni, MD
1 From the Departments of Radiology (J.E.O., G.N.B.) and Pathology and Laboratory Medicine (S.A.S., R.A.), University of Wisconsin Hospitals and Clinics, Madison, WI. Received September 29, 2005; revision requested October 24 and received December 6; accepted December 7. All authors have no financial relationships to disclose.
Address correspondence to J.E.O., Seattle Cancer Care Alliance, 825 Eastlake Ave East, G3-200, P.O. Box 19023, Seattle, WA 98109-1023 (e-mail: ochsner{at}u.washington.edu).
 |
History
|
|---|
A 78-year-old woman with a distant past medical history of breast cancer and right mastectomy presented with a left subscapular mass. The patient believed that the mass had been growing over the past few months. Her remaining past medical history was noncontributory. At clinical examination, the approximately 4-cm subcutaneous mass was mobile, firm, and located inferior to the tip of the scapula. The mass protruded from the subscapular region upon abduction of the left arm and caused some minor physical discomfort. Chest computed tomography (CT) with intravenous administration of contrast material was performed. Although the CT results indicated a benign elastofibroma dorsi, the patient requested that the mass be surgically excised for her own "peace of mind," since she had a remote history of breast cancer and wanted further assurance that this mass was not a malignancy. She recovered from the excision without incident or recurrence.
|
Imaging Findings
|
|---|
Contrast material–enhanced CT demonstrated a 4.3 x 3.5 x 7.5-cm soft-tissue mass centered inferior and medial to the left scapular tip. The borders of the mass were indistinct, but adjacent fat planes were preserved (Fig 1). The majority of the mass had an attenuation similar to that of the adjacent skeletal muscle; however, scattered linear fat attenuation was seen centrally throughout the mass (Figs 1, 2). There were no nodular areas of contrast enhancement, no calcifications, and no adjacent osseous destruction. No mass was seen adjacent to the contralateral scapular tip. No tissue sampling or further imaging was deemed necessary for diagnosis, since these are the classic imaging findings of an elastofibroma dorsi.
View larger version (104K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1a. (a) Contrast-enhanced CT scan demonstrates a 3.5 x 4.3-cm irregular soft-tissue mass (arrowheads) inferior to the tip of the left scapula. The mass has attenuation similar to that of the adjacent skeletal muscle. (b) CT scan (magnified view) more clearly depicts the irregular borders of the mass (arrows) and demonstrates some internal fat attenuation (arrowhead).
|
|
View larger version (97K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1b. (a) Contrast-enhanced CT scan demonstrates a 3.5 x 4.3-cm irregular soft-tissue mass (arrowheads) inferior to the tip of the left scapula. The mass has attenuation similar to that of the adjacent skeletal muscle. (b) CT scan (magnified view) more clearly depicts the irregular borders of the mass (arrows) and demonstrates some internal fat attenuation (arrowhead).
|
|
View larger version (90K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2. Contrast-enhanced CT scan obtained at the level of the inferior left scapula shows the superior aspect of the mass (short arrow), which is poorly defined. Scattered fat attenuation (long arrow) is seen centrally within the mass.
|
|
|
Pathologic Evaluation
|
|---|
The gross specimen was removed in three portions at surgery, with the three portions of excised soft tissue measuring 3.5, 4, and 7 cm in greatest dimension, respectively. All three portions were firm, irregularly shaped, and nonencapsulated. The cut surfaces were variegated and consisted of thick, tan-white fibrous bands interposed between layers of golden yellow adipose tissue (Fig 3). Necrosis and hemorrhage were not present.
View larger version (134K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3. Photograph of the largest portion of excised soft tissue reveals that the mass is irregularly shaped and nonencapsulated and has a variegated cut surface, with thick fibrous bands (*) interposed between layers of adipose tissue (arrows).
|
|
At microscopic examination, the mass consisted of pink collagen bundles intertwined with eosinophilic, dense elastic fibers. The collagen bundles and elastic fibers were intermixed with variable-sized mature adipocytes (Fig 4). Verhoeff stain highlighted the elastic fibers as dark bands, often with serrated edges and occasionally with a beaded appearance (Fig 5).
View larger version (117K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a. (a) Low-power digital photomicrograph (original magnification, x40; hematoxylin-eosin stain) demonstrates thick bands of fibrous tissue (long arrows) intermixed with multiple regions of mature adipose tissue (short arrows). (b) Intermediate-power digital photomicrograph (original magnification, x100; hematoxylin-eosin stain) reveals the fibrous bands to be composed of a mixture of collagen bundles (long arrows) and dense, eosinophilic elastic fibers (arrowheads). Mature adipose tissue (short arrows) is also seen.
|
|
View larger version (124K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b. (a) Low-power digital photomicrograph (original magnification, x40; hematoxylin-eosin stain) demonstrates thick bands of fibrous tissue (long arrows) intermixed with multiple regions of mature adipose tissue (short arrows). (b) Intermediate-power digital photomicrograph (original magnification, x100; hematoxylin-eosin stain) reveals the fibrous bands to be composed of a mixture of collagen bundles (long arrows) and dense, eosinophilic elastic fibers (arrowheads). Mature adipose tissue (short arrows) is also seen.
|
|
|
Discussion
|
|---|
Elastofibroma dorsi is a benign soft-tissue tumor with a characteristic location and appearance at CT. It was first described in 1961 by Jarvi and Saxen (1) and was initially considered rare. However, these tumors are now known to be much more commonplace in the elderly population than was previously thought. The tumor had a reported prevalence of 2% in a study of patients over 60 years old who were undergoing chest CT for evaluation of the pulmonary parenchyma (2). However, in an autopsy series, elastofibroma dorsi was found in 11.2% of men and 24.4% of women (3). The difference in prevalence between the CT study and the autopsy series may be secondary to the small size of the multiple elastofibromas found in the latter. CT lacks sufficient contrast resolution to depict the smallest of these lesions, some of which were described as "only... a streak in the fascia" when seen macroscopically at autopsy (3). Therefore, the reported prevalence of 2% at CT may be more applicable to the practice of radiology.
Elastofibroma dorsi is classically found in the infrascapular regions, deep to the serratus anterior and latissimus dorsi musculature. In less than 1% of cases, elastofibromas may be found in other locations, including adjacent to the ischial tuberosity, adjacent to the olecranon, or within the thoracic wall (4). Unilateral masses have a slight right-sided predilection, but up to 60% of elastofibromas are bilateral (4). Bilateralism further supports the presumptive diagnosis of elastofibroma (2,5). Elastofibroma dorsi is most frequently seen in older women, with a reported female predilection of 5–13:1 and a mean age at diagnosis of 65–70 years (4,5).
At clinical examination, elastofibroma dorsi is often asymptomatic. In up to 50% of patients, however, elastofibroma dorsi may cause moderate pain and clicking, snapping, or clunking of the scapula with movement. In one series, a "snapping scapula" was demonstrated at clinical examination in 60% of patients with elastofibroma dorsi (6). Occasionally, elastofibroma dorsi manifests with shoulder pain and is misdiagnosed clinically as a rotator cuff tendon tear or subacromial bursitis (6). Most patients in whom symptomatic masses are surgically excised remain free of symptoms. Recurrence has been reported to occur on rare occasions but is thought to be secondary to incomplete excision (7).
Elastofibroma dorsi is composed of fibrous tissue with internal fatty streaks. At histologic analysis, it is composed of eosinophilic collagen bundles; groupings of mature adipocyte cells; and linear, globular, and stellate elastic fibers (8–10), findings that correspond to the thick fibrous bands and golden yellow fatty tissue seen at gross pathologic examination. With multiple imaging modalities, the dense fibrous tissue seen at gross pathologic examination correlates with the regions of the mass that have imaging characteristics similar to those of skeletal muscle. The internal linear fat seen at gross pathologic examination correlates with scattered or linear areas of fatty tissue seen at imaging. In addition, the poorly defined margins of the mass demonstrated with multiple imaging modalities are consistent with the lack of a capsule seen at gross pathologic examination.
The CT appearance of elastofibroma dorsi is diagnostic. A poorly defined soft-tissue mass in the infrascapular or subscapular region with attenuation similar to that of the adjacent skeletal muscle is seen. These masses have been reported to have soft-tissue attenuation at CT with internal striations or scattered areas of fat attenuation (5,7). However, elastofibromas have also been reported to have homogeneous attenuation at CT without visible internal foci of decreased attenuation (2,5,7). It has been theorized that smaller elastofibromas may appear more homogeneous at CT compared with larger elastofibromas due to their earlier stage of development (2). In another study, two elastofibromas with homogeneous CT attenuation were noted to have less internal fat at histopathologic examination than did the elastofibromas with internal low-attenuation striations at CT (5). In our case, scattered fat attenuation was seen within the mass at CT, a finding that correlates with the histopathologic finding of plentiful mature adipocytes scattered throughout the elastofibroma. Atypical features, including adjacent bone destruction or intense contrast enhancement, should raise suspicion that the tumor has a different cause (eg, liposarcoma, metastatic disease).
Other imaging modalities have been used to evaluate symptomatic elastofibromas. As at CT, these lesions manifest at magnetic resonance (MR) imaging as a soft-tissue mass in the classic subscapular location. Elastofibromas demonstrate patterns of alternating fibrous and fatty tissue at MR imaging. The fibrous tissue is isointense relative to skeletal muscle with both T1- and T2-weighted sequences, whereas the fatty tissue has high signal intensity with T1-weighted sequences. Therefore, on T1-weighted MR images, the majority of the mass is isointense relative to skeletal muscle with interspersed areas of linear and curvilinear increased signal intensity. The borders of these masses are relatively well defined, but the mass is nonencapsulated. Contrast-enhanced MR images have been reported to show both areas with and areas without enhancement (7).
At ultrasonography (US), elastofibromas have been reported to manifest as a well-defined multilayered pattern of hypoechoic linear areas of fat deposition intermixed with echogenic fibroelastic tissue. This characteristic pattern was described in three pathologically proved elastofibromas, two of which were also evaluated with MR imaging to allow the comparison of US and MR imaging findings (11). It has been found useful to perform US with the patient prone and with the arm abducted for localization of the mass, since the mass often protrudes from the subscapular region upon arm abduction (11,12). Alternatively, it has been suggested that patients stand with the arms slightly elevated forward and adducted for optimal mass protrusion (13).
There have been three reported cases of hyper-metabolism of elastofibroma dorsi at positron emission tomography–CT in two patients (14). The standardized uptake values in these cases ranged between 1.52 and 1.98. Although this finding has been reported in only a small number of patients, hypermetabolism in a subscapular mass with the classic CT or MR imaging findings of elastofibroma dorsi should not change this presumptive diagnosis.
In conclusion, elastofibroma dorsi may be seen in up to 2% of the population, particularly in older women. These indistinct soft-tissue masses have a classic infrascapular location, attenuation similar to that of the adjacent skeletal muscle, and (often) scattered or striated internal fatty attenuation at CT. Therefore, radiologists should be comfortable diagnosing this benign and often asymptomatic and incidentally found soft-tissue tumor at CT. Further imaging studies or tissue sampling is not required for diagnosis when the classic CT findings of elastofibroma dorsi are present.
|
References
|
|---|
- Jarvi O, Saxen AE. Elastofibroma dorsi. Acta Pathol Microbiol Scand 1961;51(suppl 144):83–84.
- Brandser EA, Goree JC, El-Khoury GY. Elastofibroma dorsi: prevalence in an elderly population as revealed by CT. AJR Am J Roentgenol 1998; 171:977–980.[Abstract/Free Full Text]
- Jarvi OH, Lansimies PH. Subclinical elastofibromas in the scapular region at autopsy series. Acta Pathol Microbiol Scand 1975;83:87–108.
- Nagamine N, Nohara Y, Ito E. Elastofibroma in Okinawa: a clinicopathologic study of 170 cases. Cancer 1982;50:1794–1805.[CrossRef][Medline]
- Naylor MF, Nascimento AG, Sherrick AD, McLeod RA. Elastofibroma dorsi: radiologic findings in 12 patients. AJR Am J Roentgenol 1996; 167:683–687.[Abstract/Free Full Text]
- Majo J, Gracia I, Doncel A, Valera M, Nunez A, Guix M. Elastofibroma dorsi as a cause of shoulder pain or snapping scapula. Clin Orthop Relat Res 2001;388:200–204.[Medline]
- Kransdorf MJ, Meis JM, Montgomery E. Elastofibroma: MR and CT appearance with radiologic-pathologic correlation. AJR Am J Roentgenol 1992;159:575–579.[Abstract/Free Full Text]
- Abe S, Miyata N, Yamamoto Y, Yamaguchi T, Tamakawa M. Elastofibroma dorsi: CT, MRI, and pathologic findings. Plast Reconstr Surg 1999; 104:2121–2126.[CrossRef][Medline]
- Nakamura Y, Ohta Y, Itoh S, et al. Elastofibroma dorsi: cytologic, histologic, immunohistochemical and ultrastructural studies. Acta Cytol 1992;36: 559–562.[Medline]
- Domanski HA, Carlen B, Sloth M, Rydholm A. Elastofibroma dorsi has distinct cytomorphologic features, making diagnostic surgical biopsy unnecessary. Diagn Cytopathol 2003;29:327–333.[CrossRef][Medline]
- Bianchi S, Martinoli C, Abdelwahab IF, Gandolfo N, Derchi LE, Damiani S. Elastofibroma dorsi: sonographic findings. AJR Am J Roentgenol 1997; 169:1113–1115.[Abstract/Free Full Text]
- Dalal A, Miller TT, Kenan S. Sonographic detection of elastofibroma dorsi. J Clin Ultrasound 2003;31:375–378.[CrossRef][Medline]
- Kara M, Dikman E, Kara SA, Antasoy P. Bilateral elastofibroma dorsi: proper positioning for an accurate diagnosis. Eur J Cardiothorac Surg 2002; 22:839–841.[Abstract/Free Full Text]
- Pierce JC, Henderson R. Hypermetabolism of elastofibroma dorsi on PET-CT. AJR Am J Roentgenol 2004;183:35–37.[Free Full Text]
Top
History
Imaging Findings
