DOI:10.2214/AJR.04.1717
AJR 2006; 186:805-811
© American Roentgen Ray Society
Lipofibromatous Hamartoma of the Upper Extremity: A Review of the Radiologic Findings for 15 Patients
Andoni P. Toms1,
Dimitri Anastakis2,
Robert R. Bleakney3 and
Thomas J. Marshall
1 Department of Radiology, Norfolk and Norwich University Hospital, Colney Ln.,
Norwich, Norfolk NR4 7UY, England.
2 Toronto Western Hospital, University Health Network, Toronto, ON,
Canada.
3 Mount Sinai Hospital, University Health Network, Toronto, ON, Canada.
Received November 7, 2004;
accepted after revision January 31, 2005.
Address correspondence to A. P. Toms
(andoni.toms{at}nnuh.nhs.uk).
Abstract
OBJECTIVE. The purpose of this study was to analyze the radiologic
characteristics of lipofibromatous hamartomas affecting upper limb peripheral
nerves.
CONCLUSION. Although there are pathognomonic features that
characterize lipofibromatous hamartoma on MRI, the range of appearances is
broad. Sonography appears to show equally characteristic features and may be a
useful tool for assessing this condition.
Keywords: lipofibromatous hamartoma • MRI • soft-tissue neoplasms • sonography
Introduction
The term "lipofibromatous hamartoma" is most commonly used to
describe the proliferation of mature adipocytes within peripheral nerves
resulting in a palpable yellow neurogenic mass
(Fig. 1). Mature fat
infiltrates the nerve, separating axonal bundles sheathed in perineurium,
which normally lie adjacent to each other. Perineural and endoneural fibrosis
thickens the axonal bundles, which are interspersed through the proliferative
fat, to produce the characteristic histologic appearances.
View larger version (103K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1 —52-year-old man with lipofibromatous hamartoma. Intraoperative
photograph illustrates typical gross appearance of lipofibromatous hamartoma.
Multiple soft, gray-yellow lobulated masses (arrows) are present
within epineural sheath of radial digital nerve of this right index finger.
Epineurium of digital nerve had extensive perineural fibrosis.
|
|
The World Health Organization tumor classification describes this entity as
lipomatosis of the nerve [1].
Although this description simplifies the numerous descriptions in the
literature, such as neural lipoma and fibrolipoma, it does not do justice to
the fibrous element within the lesion.
Lipofibromatous hamartomas are rare. Patients typically present in their
third or fourth decade of life with a long history of painless swelling in the
distal forearm that often was first noted in childhood
[2-8].
The median nerve is the most commonly involved peripheral nerve
[2,
3,
5,
6,
8-16],
with involvement of other peripheral nerves of the upper extremity reported
less frequently [2,
7,
17-21].
Involvement of peripheral nerves outside the upper extremity has been
reported only as isolated cases
[22-26].
Patients with median nerve involvement commonly present with carpal tunnel
syndrome [6,
8,
11]. Before the widespread
availability of MRI, the diagnosis of lipofibromatous hamartoma was confirmed
after exploratory surgery and biopsy of the lesion
[2]. In the largest reported
series on lipofibromatous hamartoma, approximately one quarter of patients had
macrodystrophia lipomatosa—gigantism produced by mesenchymal overgrowth
confined to the affected nerve territory
[2]. Similar findings have been
reported for a smaller radiologic series
[20] and a number of case
reports
[27-29].
The appearance of lipofibromatous hamartoma on MRI has been well described
in numerous isolated case reports
[5-12].
The MRI features common to all cases of lipofibromatous hamartoma have been
analyzed and are considered to be pathognomonic
[20,
30]. However, the range of
morphologic features of lipofibromatous hamartoma on MRI has not been analyzed
systematically. The sonographic findings for lipofibromatous hamartoma have
been described in a single case report
[6] and appear to correlate
with the MRI appearances. But, to date, the sonographic findings for
lipofibromatous hamartoma have not been analyzed systematically. The purpose
of this study was to analyze the radiologic characteristics of lipofibromatous
hamartoma affecting upper limb peripheral nerves, paying particular attention
to the range of morphologic features on MRI, and to correlate these
appearances with sonographic findings.
Materials and Methods
This study included 15 patients who underwent sonography, CT, or MRI. The
patients came from two centers: 12 from Toronto Western Hospital in Toronto,
Canada, and three from the Norfolk and Norwich University Hospital in Norwich,
England. The patients were identified from hospital and personal
databases.
The radiologic examinations took place between 1998 and 2003. Ten
sonographic examinations, 15 MRI examinations, and one CT examination were
performed. Sonography was performed on an ATL 5000 HDI machine (Wave Imaging)
with a 5- to 12-MHz linear-array probe. CT was performed on a LightSpeed
machine (GE Healthcare) with 5-mm axial slices acquired with a pitch ratio of
1.5 and reconstructed every 5 mm. MRI studies were performed on a 1.5-T Signa
machine (GE Healthcare). Axial T1-weighted and fat-saturated T2-weighted
sequences were performed on all patients, with at least two additional
sequences performed in at least one other plane. All radiologic studies were
archived on a PACS (Centricity, GE Healthcare). Data for all studies were
recalled and reviewed retrospectively by two musculoskeletal radiologists
using a DICOM viewer (eFilm Workstation, Merge Healthcare) on a
high-resolution 2K monitor.
The shape, length, cross-sectional area, and number of neural bundles of
each lesion were recorded along with sonographic, CT, and MRI characteristics.
Any associated hypertrophy of local osseous and soft tissues was also noted.
Any reporting discrepancies were resolved by consensus.
Results
The results are summarized in Table
1. Of the 15 patients reviewed, seven were men and eight were
women (age range, 22-55 years; average, 37.4 years). Six patients had isolated
median nerve lesions, six had isolated ulnar nerve lesions, and three had
lipofibromatous hamartoma involving multiple upper limb nerves (of which two
had brachial plexus involvement). Four patients had macrodystrophia
lipomatosa; two of these had isolated lipofibromatous hamartoma of the median
nerve, whereas one had extensive involvement of the median, ulnar, and radial
nerves along with the proximal brachial plexus. Seven of the 15 patients had
histologic confirmation of the radiologic diagnosis. Tissue samples were
obtained as part of a carpal tunnel release in two patients, after debulking
in four, and after amputation in one.
View larger version (48K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A —52-year-old man with lipofibromatous hamartoma of median nerve and
macrodystrophia lipomatosa. Photograph of volar aspect of right hand and
forearm shows local gigantism of thumb and radial half of index finger
(arrowheads), with associated peripheral cyanosis. Scar
(arrow) extending from thenar eminence proximally into forearm bears
witness to previous surgical exploration and decompression of "fatty
mass."
|
|
View larger version (94K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —52-year-old man with lipofibromatous hamartoma of median nerve and
macrodystrophia lipomatosa. Conventional radiograph of same patient shows
regional osseous gigantism in median nerve distribution. Premature
osteoarthrosis of interphalangeal joints (arrowheads) has been
treated by arthrodesis and internally fixed with cerclage wires.
|
|
Isolated Median Nerve Lipofibromatous Hamartoma
Six patients had isolated median nerve lesions that measured between 2.4
and 20 cm in length (mean, 11.3 cm) and had maximal cross-sectional areas
ranging from 0.6 to 2.4 cm2 (mean, 1.35 cm2). In five of
the six patients, the proximal margins were found at the level of the pronator
quadratus whereas the distal position was more variable. In one patient, the
proximal end of the lesion, which extended into the forearm, was not included
in the field of view of the MRI study. Within these isolated lesions, between
15 and 20 neural bundles measuring 1-3 mm in diameter were identified, with
evenly interspersed fat.
Isolated Ulnar Nerve Lipofibromatous Hamartoma
Six patients had isolated ulnar nerve lesions. Five patients had diffuse
ulnar nerve involvement whereby three or four thickened neural bundles were
evenly interspersed with fat. Lesions measured between 2 and at least 8 cm in
length (mean, 6.25 cm). Cross-sectional areas ranged from 0.3 to 0.7
cm2 (mean, 0.37 cm2). In four patients, the proximal
margins of the lesions were included in the MRI study and were identified
within Guyon's canal. In the other two patients, the proximal edge of the
lesion was not included in the field of view. One of the three ulnar nerve
lesions had a homogeneous fatty component (with a cross-sectional area of 17
cm2) that extended eccentrically from Guyon's canal through the
palm and into the web spaces.
Multiple Nerve Lipofibromatous Hamartoma
Two patients presented with involvement of the brachial plexus and its
branches, including the axillary, musculocutaneous, radial, median, and ulnar
nerves. The total length of these lesions was 80 cm, and the maximal
cross-sectional areas were 2.2 and 4.5 cm2. The neural bundles
numbered 15-16 and were approximately 2 mm in diameter. Unlike the isolated
median nerve lesions, the fatty component was typically unevenly distributed
through the nerve, with multiple eccentric, lobulated areas.
Four patients with lipofibromatous hamartoma of the median nerve underwent
a sonographic examination that revealed 15-16 nearly anechoic bundles encased
in echogenic soft tissue with no detectable Doppler signal. CT of the wrist in
one patient showed neural bundles of soft-tissue attenuation encased in fat
within the carpal tunnel. MRI studies in all but one patient showed a
characteristic signal: Neural bundles were of low signal intensity on all
sequences and, on T1- and T2-weighted images, were encased in soft tissue of
fatty signal intensity that was completely saturated on spectral fat
suppression or inversion recovery sequences. In one patient with isolated
median nerve involvement and macrodystrophia lipomatosa, the neural bundles
were of intermediate signal intensity and surrounded by fat of relatively low
signal intensity on T1-weighted images. Soft-tissue edema was not noted on any
sequences.
Discussion
The average age and range of this series of patients were similar to the
two previously reported series of patients with lipofibromatous hamartoma
[2,
20]; the incidence of
macrodystrophia lipomatosa (four of 15 patients) was also nearly the same as
the 25% reported by the same authors (Figs.
2A and
2B). The near-equal numbers of
men and women were typical.
Consistent with previous reports, the median nerve was the most common
anatomic location for lipofibromatous hamartoma in this series (nine of 15
patients). However, the incidence of ulnar nerve involvement was much higher,
at nearly 50% (eight of 15 patients), than the approximately 10% previously
reported [2,
20]. Two of the patients with
ulnar nerve involvement had this as part of a generalized involvement of all
upper extremity nerves, resulting in a 40% incidence (six of 15 patients) of
isolated ulnar nerve involvement. On cross-sectional imaging, lipofibromatous
hamartoma of the ulnar nerve can be more difficult to appreciate than is
lipofibromatous hamartoma involving the median nerve. Ulnar nerve lesions
typically have a cross-sectional area of 0.3 cm2 in the cubital
tunnel, range from 2 to 8 cm in length, and contain three or four neural
bundles of low signal intensity measuring 1 mm in diameter
(Fig. 3). Continued growth of
the lesion must proceed in less restricted surroundings proximal or distal to
the cubital tunnel (Fig.
4).
View larger version (127K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3 —31-year-old woman who presented with paresthesia in ulnar nerve
distribution. Axial T1-weighted MR image through wrist at level of distal
carpal row shows Guyon's canal distended with fat (arrowhead) and
thickened fascicles of ulnar nerve. Conductive delay was confirmed with
electromyelography.
|
|
View larger version (141K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4 —41-year-old man with lipofibromatous hamartoma of median nerve.
Coronal T1-weighted MR image of hand shows that tumor (arrow) is
predominantly lipomatous and extends from Guyon's canal. Palmar mass had first
been noticed at age of 6 years and grew slowly for 35 years until
presentation.
|
|
View larger version (85K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5 —40-year-old woman with lipofibromatous hamartoma of median nerve.
Axial T1-weighted MR image of wrist at level of proximal carpal row shows that
median nerve is enlarged, as are individual neural bundles, but that signal
characteristics are those of normal nerve. Neural bundles have higher T1
signal (arrow) than does normal muscle, and no discrete lipomatous
component is present at this level.
|
|
The radiologic characteristics of lipofibromatous hamartoma of the median
nerve are consistent with previously reported descriptions (Figs.
5,
6A, and
6B). The pathognomonic pattern
of an enlarged nerve containing 15 or 16 coaxial "cables" or
bundles of axons encased in epineural fibrous tissue was the norm. The bundles
were characteristically larger than in the ulnar nerve, measuring 2 mm and
occasionally 3 mm in diameter. In the region of the carpal tunnel, the neural
bundles were evenly interspersed with fat in all patients, such that the
cross-sectional area of the median nerve was increased to 0.6-1.6
cm2. Even in the two patients with extensive lipofibromatous
hamartoma of the upper extremity nerves, where the distribution of fat in the
lesion was, in most places, wildly eccentric (Figs.
7A and
7B), the components within the
carpal and cubital tunnels conformed to this description of evenly
interspersed fat and neural bundles. The fibroosseous tunnels appear to
restrict the distribution of fat in the nerve because of their inability to
expand; above and below these restrictions the fatty component has room to
proliferate more freely and erratically.
View larger version (109K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6A —31-year-old woman with extensive lipofibromatous hamartoma of upper
limb nerves. Sonograms of enlarged median nerve acquired in sagittal
(A) and axial (B) planes show hypoechoic cablelike neural
bundles (arrows) separated by hyperechoic fat.
|
|
View larger version (109K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6B —31-year-old woman with extensive lipofibromatous hamartoma of upper
limb nerves. Sonograms of enlarged median nerve acquired in sagittal
(A) and axial (B) planes show hypoechoic cablelike neural
bundles (arrows) separated by hyperechoic fat.
|
|
View larger version (167K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 7A —55-year-old man with lipofibromatous hamartoma of median nerve.
Coronal T1-weighted MR image shows lipofibromatous hamartoma of brachial
plexus (arrow) at level of formation of terminal branches.
|
|
Sonography showed the characteristic hypoechoic coaxial cabling encased by
an echogenic substratum (Figs.
6A and
6B). In all but one patient,
the MRI findings were characteristic, with low-signal-intensity axonal bundles
surrounded by fat (Fig. 5). One
patient showed less typical signal characteristics but a morphologic
arrangement typical of lipofibromatous hamartoma of the median nerve.
Throughout the lesion, the neural bundles were hyperintense to muscle and the
surrounding substratum was isointense to muscle on T1-weighted images
(Fig. 8). These signal
characteristics may represent an unusual developmental distribution of the
fibrous and fatty elements within this lesion. Although not the pathognomonic
appearances of lipofibromatous hamartoma on MRI, these appearances previously
have been attributed to lipofibromatous hamartoma
[31]. The lack of high signal
intensity on the fat-saturated T2-weighted sequences excludes hereditary
hypertrophic interstitial neuritis
[20]. The patient did not have
a biopsy of this lesion, and therefore histologic correlation was not
possible. This case therefore must be considered a potential atypical variant
based solely on imaging characteristics that are similar to the classic
description of lipofibromatous hamartoma, illustrating a limitation of this
study. Only seven of the 15 patients had histologic confirmation of the
diagnosis, as is consistent with the view that the MRI appearances of
lipofibromatous hamartoma preclude the need for histologic confirmation
[20]. This view therefore is
used to justify the diagnosis of lipofibromatous hamartoma in this series.
View larger version (99K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8 —31-year-old man. Axial T1-weighted MR image through wrist shows
enlarged median nerve and divided flexor retinaculum. Signal intensities of
nerve are reversed from typical pattern seen in lipofibromatous hamartoma,
with neural fascicles of high fat intensity (arrow) and intervening
substratum of intermediate soft-tissue intensity.
|
|
Of the six patients with isolated lipofibromatous hamartoma of the median
nerve, the proximal end of the lesion was identified in the distal forearm at
the level of the pronator quadratus for five patients, in all of whom a
normal-caliber median nerve was identified proximal to the pronator quadratus.
The proximal end of the lesion in the sixth patient was not included in the MR
images.
Two patients had an extensive lipofibromatous hamartoma affecting the
brachial plexus and all major upper extremity nerves. Maximal axial diameters
were 2.2 and 4.5 cm2 and occurred in the proximal median nerve in
both patients. The distribution of fat within the nerve was typically
eccentric, with lipomatous masses bulging from the nerve at multiple sites and
extending into adjacent muscles (Figs.
7A and
7B). The lesions extended
proximally into the brachial plexus to the level of the clavicle. The proximal
margin was not defined because imaging of the cervical spine was not
performed.
The sonographic data from this study illustrate, for what is to our
knowledge the first time, consistent echo patterns in a series of patients
with lipofibromatous hamartoma, and we suggest that these patterns are
characteristic. The sonographic features of a thickened hyperechoic nerve
containing discrete hypoechoic cablelike fascicles corresponded, in all
patients, to the MRI signal patterns. In theory, sonography should be quicker
and easier to perform than MRI for both diagnosis and delimitation of
lipofibromatous hamartoma, particularly in patients with extensive disease.
These patients require MRI studies with multiple, time-consuming fields of
view, whereas sonography can cover the lesion in a matter of minutes. Our
anecdotal experience suggests that sonography is the more convenient
technique. However, in all but one patient, the sonography was performed with
full knowledge of the diagnosis made on MRI, precluding any direct comparison
of the utility of the two techniques in this retrospective study.
In conclusion, lipofibromatous hamartoma is a rare disorder that most
commonly manifests in the nerves of the upper extremity and increasingly is
being recognized and reported. The MRI features are considered pathognomonic,
but this series shows the morphologic variability; sonographic and CT features
have been described less frequently but are also characteristic of the lesion.
The sonographic findings consistently corresponded to the MRI findings and may
be characteristic of this condition. Sonography may be useful for extensive
lesions, on which complete assessment of the affected nerve can be obtained
more easily than with MRI. The histologic features are the same for all
lipofibromatous hamartomas, but the gross morphologic appearances may vary
widely from small, simple lesions of the ulnar nerve to extensive, complex
lipomatosis involving all upper extremity nerves. For lipofibromatous
hamartoma, knowledge of characteristic radiologic findings—both
sonographic and MRI—may obviate diagnostic biopsy in centers with
extensive peripheral-nerve experience.
References
- Nielsen GP. Lipomatosis of nerve. In: Fletcher CDM, Unni KK,
Mertens F, eds. Pathology and genetics of tumours of soft tissue
and bone. Lyon, France: IARC Press, 2002:24
-25
- Silverman TA, Enzinger FM. Fibrolipomatous hamartoma of nerve: a
clinicopathologic analysis of 26 cases. Am J Surg
Pathol 1985; 9:7
-14[CrossRef][Medline]
- Sondergaard G, Mikkelsen S. Fibrolipomatous hamartoma of the median
nerve. J Hand Surg Br 1987;12
: 224-226[CrossRef][Medline]
- Drut R. Ossifying fibrolipomatous hamartoma of the ulnar nerve.
Pediatr Pathol 1988;8
: 179-184[Medline]
- Walker CW, Adams BD, Barnes CL, Roloson GJ, FitzRandolph RL. Case
report 667: fibrolipomatous hamartoma of the median nerve. Skeletal
Radiol 1991; 20:237
-239[Medline]
- Declercq H, De Man R, Van Herck G, Tanghe W, Lateur L. Case report
814: fibrolipoma of the median nerve. Skeletal Radiol1993; 22:610
-613[Medline]
- Oleaga L, Florencio MR, Ereno C, et al. Fibrolipomatous hamartoma
of the radial nerve: MR imaging findings. Skeletal
Radiol 1995; 24:559
-561[Medline]
- Sagtas E, Orguc S, Elcin F, Memis A, Menzilcioglu S. Quiz case of
the month: fibrolipomatous hamartoma of the median nerve. Eur
Radiol 2000; 10:547
-548[CrossRef][Medline]
- Feyerabend T, Schmitt R, Lanz U, Warmuth-Metz M. CT morphology of
benign median nerve tumors: report of three cases and a review.
Acta Radiol 1990;31
: 23-25[Medline]
- Kransdorf MJ, Moser RP Jr, Meis JM, Meyer CA. Fat-containing
soft-tissue masses of the extremities. RadioGraphics1991; 11:81
-106[Abstract]
- Cavallaro MC, Taylor JA, Gorman JD, Haghighi P, Resnick D. Imaging
findings in a patient with fibrolipomatous hamartoma of the median nerve.
AJR 1993; 161:837
-838[Free Full Text]
- Evans HA, Donnelly LF, Johnson ND, Blebea JS, Stern PJ. Fibrolipoma
of the median nerve: MRI. Clin Radiol1997; 52:304
-307[CrossRef][Medline]
- Canga A, Abascal F, Cerezal L, Bustamante M, Perez-Carro L,
Vazquez-Barquero A. Fibrolipomatous hamartoma of the median nerve: case
illustration. J Neurosurg 1998;89
: 683[Medline]
- Berry MG, Mallucci P, Banwell PE, Heywood AJ. Fibrolipoma of the
median nerve. J R Soc Med 1999;92
: 408-409[Medline]
- Lowenstein J, Chandnani V, Tomaino MM. Fibrolipoma of the median
nerve: a case report and review of the literature. Am J
Orthop 2000; 29:797
-798[Medline]
- Khanna G, Sundaram M, Rotman M, Janney CG. Fibrolipomatous
hamartoma of the nerve. Orthopedics 2001;24
: 836, 919-920[Medline]
- Price AJ, Compson JP, Calonje E. Fibrolipomatous hamartoma of nerve
arising in the brachial plexus. J Hand Surg Br1995; 20:16
-18[CrossRef][Medline]
- Meyer BU, Roricht S. Fibrolipomatous hamartoma of the proximal
ulnar nerve associated with macrodactyly and macrodystrophia lipomatosa as an
unusual cause of cubital tunnel syndrome. J Neurol Neurosurg
Psychiatry 1997; 63:808
-810[Free Full Text]
- Kronberger P, Rainer C, Hittmair A, Anderl H. Lipofibromatous
hamartoma (neural fibrolipoma) of a flexor nerve of the index finger.
Scand J Plast Reconstr Surg Hand Surg1998; 32:237
-239[CrossRef][Medline]
- Marom EM, Helms CA. Fibrolipomatous hamartoma: pathognomonic on MR
imaging. Skeletal Radiol 1999;28
: 260-264[CrossRef][Medline]
- Wong CH, Chow L, Yen CH, Ho PC, Yip R, Hung LK. Uncommon hand
tumours. Hand Surg 2001;6
: 67-80[CrossRef][Medline]
- Bibbo C, Warren AM. Fibrolipomatous hamartoma of nerve.
J Foot Ankle Surg 1994;33
: 64-71[Medline]
- Berti E, Roncaroli F. Fibrolipomatous hamartoma of a cranial nerve.
Histopathology 1994;24
: 391-392[Medline]
- Donley BG, Neel M, Mitias HM. Neural fibrolipoma of the foot: a
case report. Foot Ankle Int 1996;17
: 712-713[Medline]
- Sobel E, Giorgini RJ, Potter GK, Schwartz RD, Chieco TM.
Progressive pedal macrodactyly surgical history with 15 year follow-up.
Foot Ankle Int 2000;21
: 45-50[Medline]
- Ogose A, Hotta T, Higuchi T, Katsumi N, Koda H, Umezu H.
Fibrolipomatous hamartoma in the foot: magnetic resonance imaging and surgical
treatment—a report of two cases. J Bone Joint Surg
Am 2002; 84:432
-436[Free Full Text]
- Boren WL, Henry RE Jr, Wintch K. MR diagnosis of fibrolipomatous
hamartoma of nerve: association with nerve territory-oriented macrodactyly
(macrodystrophia lipomatosa). Skeletal Radiol1995; 24:296
-297[Medline]
- Meyer BU, Roricht S, Schmitt R. Bilateral fibrolipomatous hamartoma
of the median nerve with macrocheiria and late-onset nerve entrapment
syndrome. Muscle Nerve 1998;21
: 656-658[CrossRef][Medline]
- Brodwater BK, Major NM, Goldner RD, Layfield LJ. Macrodystrophia
lipomatosa with associated fibrolipomatous hamartoma of the median nerve.
Pediatr Surg Int 2000;16
: 216-218[CrossRef][Medline]
- De Maeseneer M, Jaovisidha S, Lenchik L, et al. Fibrolipomatous
hamartoma: MR imaging findings. Skeletal Radiol1997; 26:155
-160[CrossRef][Medline]
- Bordalo-Rodrigues M, Amin P, Rosenberg ZS. MR imaging of common
entrapment neuropathies at the wrist. Magn Reson Imaging Clin N
Am 2004; 12:265
-279[CrossRef][Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
