Hibernoma: a case series with multimodality imaging and pathologic correlation.
Abstract: Hibernoma is an uncommon, benign tumor of brown fat origin. The distribution of this tumor originally was described as following the location of persistent brown fat within the subcutaneous tissue of the thorax (especially the periscapular and interscapular regions), neck, axilla, shoulder, and retroperitoneum. Recently, hibernoma was described as being most common in the thigh.
Subject: Diagnostic imaging
Authors: Yim, Annabella C.K.
Sheikh, Adnan
Dharmadhikari, Rahul
Gravel, Denis
Rakhra, Kawan
Di Primio, Gina
Schweitzer, Mark E.
Pub Date: 10/01/2012
Publication: Name: Bulletin of the NYU Hospital for Joint Diseases Publisher: J. Michael Ryan Publishing Co. Audience: Academic Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2012 J. Michael Ryan Publishing Co. ISSN: 1936-9719
Issue: Date: Oct, 2012 Source Volume: 70 Source Issue: 4
Accession Number: 311498788
Full Text: Lipomatous lesions are the second most common soft tissue neoplasm. These tumors range from the simple lipoma to the liposarcoma. Other non-lipomatous tumors have regions with the imaging appearance of fat including hemangiomas and myositis ossificans. Hibernoma is interesting as it has imaging characteristics quite different from those we associate with fat. Positron emission tomography (PET) is a nuclear medicine modality that takes advantage of the fact that neoplastic cells, unlike normal cells, preferentially use glucose as a substrate for anaerobic metabolism when patients are imaged during fasting. (1) By radioactively labeling a glucose derivative, such as 2-fluoro-2-deoxy-D-glucose (FDG), that accumulates once metabolized inside the neoplastic cells, these hypermetabolic cells may be localized and imaged. The hibernoma is a tumor wherein PET imaging has clinical relevance, since the brown fat contained within it is quite FDG avid.

Case 1

A 48-year-old female presented with a palpable, non-tender, slow growing mass in the right thigh. The patient had noticed the mass 18 months earlier, and it gradually increased in size. On physical examination, a mass was palpable on the lateral aspect of her distal thigh. This was firm, and nontender to palpation. No overlying warmth or erythema was noted, and no lymphadenopathy was found. The patient had no neurological deficit but reported experiencing a pressure sensation in her lower right leg.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

Radiographs demonstrated no radiographic evidence of periosteal reaction, osseous scalloping, or cortical breach. A faintly radiolucent ovoid lesion could be perceived. No internal mineralization could be identified (Fig. 1).

Ultrasound revealed a 10.5 x 3.8 x 6.3 cm (craniocaudal by anterior-posterior by transverse, respectively) encapsulated, smoothly-contoured mass within the vastus intermedius muscle. This mass closely abutted the femur. The mass was homogenously hyperechoic relative to the surrounding musculature and subcutaneous fat with echogenic through-transmission. There was intralesional internal vascularity but no calcification and limited proximal muscular edema (Figs. 2 and 3).

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

The MRI confirmed the presence of an intramuscular, lipomatous mass with comparable dimensions (11.8 cm CC x 3.4 cm AP x 5.4 cm TR). This was well-circumscribed. The mass was homogenously hyperintense to musculature but slightly hypointense relative to the subcutaneous fat on T1-weighted sequences (Fig. 4A). There was incomplete suppression of the intralesional signal in comparison to subcutaneous fat on PD FS (Fig. 4B), and this distinction from subcutaneous fat was more strikingly notable on STIR sequences. While the background signal intensity of the lesion became only slightly hyperintense, there were several internal hyperintense, curvilinear structures, which were contiguous with vessels (Fig. 4C). On postcontrast scans, there was diffuse, slightly heterogeneous, low-level enhancement and enhancement of internal vessels (Fig. 4D).

On gross pathology, the rubbery yellow homogeneous mass while being well circumscribed did not possess a true capsule. There were a few thin fibrous septae. There was no concern for atypia or invasion. On microscopy multi-vacuolated brown tumor cells were identified between mature univacuolated adipocytes (Fig. 4E).

Case 2

A 32-year-old female with a history of neurofibromatosis presented with right knee pain of uncertain etiology. Initial radiographs and MRI (not shown) of the area did not identify a specific cause; however, a bone scan was suggested to further evaluate nonspecific marrow heterogeneity. A dedicated MRI of the buttock was recommended based on the nuclear medicine findings.

[FIGURE 6 OMITTED]

On the blood pool study, there was soft tissue hyperemia localized to the left buttock; an MRI was recommended to further evaluate this (Fig. 5).

On MRI, there was a large, well-circumscribed lipomatous mass deep to the gluteus maximus muscle. This was extramuscular with no intrapelvic extension. The mass measured 9.6 cm CC x 9.7 cm TR x 3.4 cm AP. On T1 there was slight hypointensity relative to subcutaneous fat (Fig. 6A). On T2 fat saturated sequence, there was near complete suppression of signal and persistent flow voids (Fig. 6B). Postcontrast, there was subtle diffuse enhancement with avid enhancement of vessels (Fig. 6C). No grossly concerning septae, invasion, or nodular component was appreciated.

Case 3

A 66-year-old male presented with a 2 year history of large, painless left thigh mass that is firm but non-tender to palpation.

On ultrasound, there was a large circumscribed hyperechoic mass within the subcutaneous tissue (Fig. 7). Doppler confirms vascularity within the lesion (not shown). The mass measures 14.8 cm x 10.8 cm x 6.7 cm.

On MRI, there was an encapsulated mass in the subcutaneous tissue of the left medial thigh. Several presumed vascular structures were noted within it. There was no invasion of the adjacent musculature. On T1-weighted images, the mass was minimally hypointense to adjacent subcutaneous fat (Fig. 8A). On PD fat saturated images, there was a thin uniform capsule with incomplete fat saturation (Fig. 8B). Postcontrast, there was mild diffuse background enhancement, thin septal enhancement, and enhancement of the central vessels (Fig. 8C).

The lesion was resected under the presumed diagnosis of low-grade liposarcoma or atypical lipoma. On gross dissection the mass was homogeneous yellow adipose tissue with small fibrous septations. It had a thin capsule with no extracapsular extension. Microscopy revealed eosinophilic, multi-vacuolated brown fat cells interspersed with mature adipocytes. This was devoid of atypia, necrosis, or invasion (Fig. 9).

[FIGURE 7 OMITTED]

Discussion

Velsch was the first investigator to recognize the distinction between brown and white adipose tissue. (2) In 1906, Merkel used the term "pseudolipoma" to describe benign brown fat tumors. (3) In 1914, the actual term hibernoma was coined when Gery recognized a histologic resemblance between these tumors and the brown fat identified in hibernating animals. (4) Due to the abundance of mitochondria and its high vascularity, brown fat is hypothesized as contributing to the non-shivering thermoregulation of hibernating animals and newborn humans. (5) Brown fat gradually diminishes in volume from infancy and compromises less than 0.1% of total body weight by 70 years of age. (6) Recent 18F-FDG PET studies indicate the persistence of brown fat in the neck, in the thorax, between the subscapularis and pectoralis muscles, around the brachial plexus, and in a paraspinal distribution. (7) PET imaging is useful in the scenario of the commonly encountered lipomatous lesion because it may distinguish the simple lipomas that have only minimal metabolism from those lesions that may have a higher metabolic rate such as hibernomas and malignant liposarcomas.

[FIGURE 8 OMITTED]

[FIGURE 9 OMITTED]

Hibernoma is a rare, benign tumor, with less than 200 cases reported in the literature. (8) They are usually seen in the third to fifth decades of life but can occur at any age. (9) Although previously reported to be more common in females, more recent studies suggests a slight male predominance. (6) Most cases present as an asymptomatic, slowly growing, painless mass, although the mass may occasionally cause symptoms due to compression of adjacent structures. (10,11)

On gross examination, hibernomas are soft to rubbery, encapsulated, yellow-brown macrolobulated masses. (11) These typically range in size from 5 to 10 cm in diameter, although lesions reaching 20 cm have been documented. (12,13) Their brownish hue may be attributed to the mitochondrial and vascular predominance. (14)

Hibernomas manifest radiographically as radiolucent masses with an absence of calcification, osseous, or chondroid matrix and lacking associated aggressive osseous pathology, (14) as illustrated by our first case. On ultrasound these are hyperechoic, circumscribed masses with hypervascularity on Doppler interrogation (14); both of the cases that were evaluated with ultrasound demonstrated these typical characteristics. Due to this hypervascularity or arteriovenous shunting, core needle biopsy is a relative contraindication particularly in deep-seated lesions, (15) where there may be difficulty tamponading a post-intervention hemorrhage. As in case 2, this hypervascularity may even be appreciated on bone scintigraphy as mild to moderate radiotracer uptake on blood pool and static images. (14)

On CT, a fat attenuating mass is characteristic, although it is slightly hyperattenuating to subcutaneous fat. With intravenous contrast, there may be diffuse, mild background enhancement with more prominent enhancement of septae and vessels. (14)

MRI features of hibernoma may vary according to their histological composition. These lesions are minimally heterogeneous and slightly hypointense to subcutaneous fat on T1-weighted images and fail to fully suppress on STIR or fat saturated T2-weighted images. (17) The enhancement pattern is variable, as on CT. (8) Not infrequently, there are associated prominent vascular structures. (16,17) This type of vascularity is not seen in well-differentiated liposarcomas, and this is an important feature for differentiation. (14) All of our lesions clearly demonstrate these typical MRI features.

Differential diagnosis based on the clinical and radiographic features would include other fairly homogeneous, noninvasive lipomatous lesions. Lipomas are more likely to parallel fat on all modalities. However, not all lipomas are homogeneous especially if traumatized or necrosed. Simple lipomas should suppress fully on fat saturation sequences and should not have more than thin capsular or mild septal enhancement. (18) The septae in superficial lipomas are less than 2 mm in thickness, uniform, and may be nonenhancing; however, deeper lesions may have thicker, even nodular septae. (14,19) Low-grade liposarcomas are more difficult to differentiate from hibernomas. The septae in low-grade liposarcomas may be thicker, nodular, and more avidly enhancing than lipomas. (14,19) The prominence of vascular structures are not seen in liposarcomas, and this is distinct to hibernomas. (14) The avid metabolism on 18-FDG PET is unique to hibernomas and more metabolically-active liposarcomas. (19)

Although hibernomas are always benign, their large size may cause compression of neighboring structures, such that surgical excision is usually recommended. (20) There is no recurrence with complete surgical resection and metastases have not been reported. (21)

Conclusion

These three cases of hibernoma clearly demonstrate the features as expected on each modality. Although it is not always possible to definitively diagnose a hibernoma, based on imaging alone, this histologic lipomatous variant should be considered in the differential diagnosis when radiologists encounter a lesion that does not exhibit the typical features of either a simple lipoma or a liposarcoma.

Disclosure Statement

None of the authors have a financial or proprietary interest in the subject matter or materials discussed, including, but not limited to, employment, consultancies, stock ownership, honoraria, and paid expert testimony.

References

(1.) Jadvar H, Parker JA. Clinical PET and PET/CT. London: Springer-Verlag, 2005.

(2.) Kristensen S. Cervical hibernoma: Review of literature and a new case. J Laryngol Otol. 1985 Oct; 99(10):1055-8.

(3.) Merkel H. On a pseudolipoma of the breast. Beitr Pathol Anat. 1906;39:152-7.

(4.) Gery L. Discussions. Bull Mem Soc Anat (Paris). 1914;89:111.

(5.) Dadrick I. Hibernoma: A possible model of brown fat his togenesis. Hum Pathol. 1978 May;9(3):321-9.

(6.) Heaton JM. The distribution of brown adipose tissue in the human. J Anat. 1972 May;112(Pt 1):35-9.

(7.) Cypess AM, Lehman S, Williams G, et al. Identification and Importance of Brown Adipose Tissue in Adult Humans. N Engl J Med. 2009 Apr 9;360(15):1509-17.

(8.) Della Vople C, Salazard B, Casanova D, et al. Hibernoma of the anterolateral thigh. Br J Plast Surg. 2005 Sep;58(6): 859-61.

(9.) Fentiman IS, Davies EE, Ramsay GS. Hibernoma of the thigh. Clin Oncol. 1975 Mar;1(1):71-6.

(10.) Alvine G, Rosenthal H, Murphey M, Huntrakoon M. Hibernoma. Skeletal Radiol. 1996 Jul;25(5):493-6.

(11.) Lele SM, Chundru S, Chaljub G, et al. Hibernoma: a report of 2 unusual cases with a review of the literature. Arch Pathol Lab Med. 2002 Aug;126(8):975-8.

(12.) Christopher D, Unni K, Mertens F. Adipocytic Tumors. WHO Classification of Tumors. Pathology and Genetics: Tumors of soft Tissue and Bone. Lyon, France: IARC, 2002, pp. 19-46.

(13.) Weiss S, Goldblum J. Benign lipomatous tumors. In: Enzinger and Weiss's Soft Tissue Tumors (4th ed). St Louis: Mosby, 2001, pp. 571-639.

(14.) Murphey MD, Carroll JF, Flemming DJ, et al. Benign musculoskeletal lipomatous lesions. Radiographics. 2004 Sep-Oct;24(5):1433-66.

(15.) Angervall L, Nilsson L, Stener B. Microangiographic and histological studies in 2 cases of hibernoma. Cancer. 1964 Jun;17:685-92.

(16.) Anderson SE, Schwab C, Stauffer E, et al. Hibernoma: imaging characteristics of a rare benign soft tissue tumor. Skeletal Radiol. 2001 Oct;30(10):590-5.

(17.) Lee JC, Gupta A, Saifuddin A, et al. Hibernoma: MRI features in eight consecutive cases: Clin Radiol. 2006 Dec;61(12):1029-34.

(18.) Gaskin CM, Helms CA. Lipomas, lipoma variants, and well-differentiated liposarcomas (atypical lipomas): Results of MRI evaluations of 126 consecutive fatty masses. AJR Am j Roentgenol. 2004 Mar;182(3);733-9.

(19.) Chatterton BE, Mensforth D, Coventry BJ, Cohen P. Hibernoma: intense uptake seen on Tc-99m tetrofosmin and FDG positron emission tomographic scanning. Clin Nucl Med. 2002 May;27(5):369-70.

(20.) Baldi A, Santini M, Mellone P, et al. Mediastinal hibernoma: a case report. J Clin Pathol. 2004 Sep;57(9):993-4.

(21.) Carinci F, Carls FP, Pelucchi S, et al: Hibernoma of the Neck. J Craniofac Surg. 2001 May;12(3):284-6.

Annabella C.K. Yim, M.D., Adnan Sheikh, M.D., Rahul Dharmadhikari, M.D., Kawan Rakhra, M.D., Gina Di Primio, M.D., and Mark E. Schweitzer, M.D., are from the Department of Diagnostic Imaging, and Denis Gravel, M.D., is from the Department of Pathology, The Ottawa Hospital, General Campus, University of Ottawa, Ottawa, Canada.

Correspondence: Adnan Sheikh, M.D., Department of Diagnostic Imaging, The Ottawa Hospital, General Campus, University of Ottawa, 501 Smyth Road, Ottawa, Canada KIH 8L6; asheikh@ Ottawahospital.on.ca.
Gale Copyright: Copyright 2012 Gale, Cengage Learning. All rights reserved.