DOI:10.2214/AJR.04.1879
AJR 2006; 186:812-818
© American Roentgen Ray Society
MRI for Differentiating Ganglion and Synovitis in the Chronic Painful Wrist
Suzanne E. Anderson1,
Lynne S. Steinbach2,
Edouard Stauffer3 and
Esther Voegelin4
1 Department of Diagnostic, Interventional, and Pediatric Radiology, University
Hospital of Bern, Inselspital, Freiburgstrasse, CH-3010 Bern,
Switzerland.
2 Department of Radiology, University of California, San Francisco, San
Francisco, CA.
3 Department of Pathology, University Hospital of Bern, Inselspital, Bern,
Switzerland.
4 Department of Orthopedic, Plastic, and Reconstructive Surgery, Division of
Hand Surgery, University Hospital of Bern, Intesspital, Bern,
Switzerland.
Received December 10, 2004;
accepted after revision February 2, 2005.
Presented at the 2003 annual meeting of the Radiological Society of North
America, Chicago, IL.
Address correspondence to S. E. Anderson
(suzanne.anderson{at}insel.ch).
Abstract
OBJECTIVE. The purpose of our study was to determine if preoperative
MRI can differentiate between occult ganglion and synovitis in the chronic
painful wrist.
CONCLUSION. MRI is accurate in preoperatively distinguishing between
ganglion and synovitis in the setting of chronic dorsal wrist pain. Four main
criteria were useful: margin, shape, internal structure, and enhancement after
administration of contrast material, with shape and internal structure being
most helpful.
Keywords: ganglion • hand • MRI • musculoskeletal imaging • synovitis • wrist
Introduction
Chronic wrist pain is a therapeutic challenge for hand surgeons. Dorsally
located wrist pain over the scapholunate ligament is a common problem;
however, there is little information available in the radiologic literature
regarding diagnosis. Literature on wrist pain concentrates largely on
rheumatoid arthritis.
Among the known etiologies of chronic dorsal wrist pain are dorsal occult
ganglion cysts, posterior interosseous neuroma or neuropathy, dorsal impaction
syndrome, avascular necrosis of the lunate or scaphoid, ligamentous tears,
fractures, tumors, inflammatory arthropathies, and extensor tenosynovitis
[1,
2]. Occult wrist ganglions are
small, benign cystic lesions usually located at the dorsal scapholunate
interval but they may occur in the radiopalmar or radioulnar aspects in the
wrist. They may be the result of a chronic irritation to the scapholunate or
other intercarpal ligaments. When there is no swelling in combination with
pain and there is no underlying carpal instability, these lesions are not
detectable by physical and routine radiographic examinations.
Therapy for recalcitrant chronic wrist pain that has not responded to
conservative treatment is surgical exploration. Unfortunately, some of these
surgical explorations yield no specific diagnosis because cystic lesions may
vary in size and have a more or less central mucinous content. While chronic
tenosynovitis of the extensor tendons can be easily distinguished from a
dorsal (not occult) ganglion, swelling and puckering seen with digital
extension are clinically not present in occult synovitis.
Tenosynovitis refers to inflammation of the synovial lining of a tendon
sheath or a joint and is associated with rheumatoid arthritis. Other
inflammatory-type etiologies include deposition diseases such as amyloidosis,
crystalline tendinopathy such as calcific tenosynovitis or gout, and septic
tenosynovitis [3]. More common
is the reactive tenosynovitis occurring about the narrow fibroosseous canals
that provide fulcrums for acute angulation of wrist tendons. Reactive occult
tenosynovitis of the wrist extensors and flexors is not rare. The surgical
approach for this condition varies from that for occult ganglion cysts.
The use of MRI in the diagnosis of dorsal occult ganglion cysts has been
previously reported [4,
5]. However, the question
remains whether MRI can differentiate between ganglion and other perisynovial
fibrous tissue that is a noninflammatory synovitis.
This retrospective study assesses the efficacy of high-resolution MRI in
patients with recalcitrant dorsal wrist pain who have an otherwise negative
workup. Because no MRI criteria were available in the literature allowing
clear distinction between occult ganglion and synovitis, review of a
consecutive patient cohort was performed and these findings are presented. MRI
criteria were used to define an MRI diagnosis that was subsequently correlated
with intraoperative findings and histology. In addition, MRI features were
analyzed to differentiate between ganglion and synovitis.
View larger version (93K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A —23-year-old woman with MRI showing characteristic features of
ganglion of dorsal aspect of wrist. Focal, relatively large mass
(arrows) is evident on T1-weighted image. Placement of vitamin
capsule to demarcate patient's pain site has slightly displaced soft tissues.
Clinically, however, no soft-tissue swelling was evident.
|
|
View larger version (51K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B —23-year-old woman with MRI showing characteristic features of
ganglion of dorsal aspect of wrist. Corresponding fast spin-echo T2-weighted
axial image shows focal, fluid cystic mass (arrows) with internal
septa (asterisk).
|
|
View larger version (91K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1C —23-year-old woman with MRI showing characteristic features of
ganglion of dorsal aspect of wrist. After administration of IV contrast agent,
wall enhancement is evident (arrows). Surgery confirmed ganglion as
diagnosis. There was no joint involvement.
|
|
Materials and Methods
Patients
Eighty consecutive patients underwent MRI for chronic wrist pain between
1998 and 2002. Patients included in the study were those with chronic dorsal
wrist pain of longer than 3 months' duration who subsequently had surgery and
either surgical or histopathologic correlation. The pain time duration was
arbitrarily chosen because there is no current definition in the hand surgery
literature and no data are available. Patients included in the study had no
clinical evidence of swelling. Patients with triangular fibrocartilage or
full-thickness scapholunate ligament tears, traumatic or posttraumatic
injuries, previous surgery to the wrist, vascular abnormalities of carpal
bones, or a rheumatologic inflammatory joint disease such as rheumatoid
arthritis were excluded from the study. The patient list was generated through
the database of the hand surgery department at a tertiary referral center.
Thirty-four patients fulfilled the inclusion criteria. Forty-six patients were
treated conservatively and are not included in the study because no gold
standard is available. Radiographs were reviewed for the presence of
soft-tissue swelling or calcification and for any underlying bone or joint
disease and had to be normal for the patient to be included in the study.
Imaging Technique
Frontal and lateral radiographs were available in all patients (n
=34). Preoperative MRIs (n = 34) were performed on an
outpatient ambulatory care 1.5-T superconducting magnet (Signa, GE
Healthcare). MRI standard sequences covering the wrist joint region were used
with a vitamin capsule demarcating the pain region. A dedicated, 4-channel
phased-array wrist coil was used in all cases with the patient in the supine
position with the hand by the patient's side. A standard MRI protocol was used
for all patients. T1-weighted images were acquired in the coronal and axial
planes with TR range/TE, 400-600/17; matrix size, 512 x 256; number of
excitations, 4; and field of view, 10 with 3-mm slice thickness and a slice
gap of 1 mm.
Fat-saturated T2-weighted MR images were acquired in the axial plane by
using a fast spin-echo technique with 3,400-6,000/85-102; matrix size, 256
x 256; number of excitations, 4; field of view, 10; and slice thickness,
3 mm with a slice gap of 1 mm. STIR images were obtained with 3,540/17;
inversion time, 150; matrix size, 256 x 192; number of excitations, 4;
3-mm slice thickness; and 0-mm gap. Coronal plane gradient-echo images with a
256 x 192 matrix were performed with a slice thickness of 1 mm and no
gap.
All patients received IV gadopentetate dimeglumine (Magnevist, Schering),
at a dose of 0.1 mmol/kg of body weight. Contrast-enhanced fat-suppressed
T1-weighted images (n = 26) were acquired in both the coronal and
axial planes. Fat suppression was accomplished by using selective
presaturation of lipid-resonant frequency. All axial images were performed at
the exact same location to allow correlation of findings and anatomy.
Imaging Interpretation
The two senior radiologists were blinded to the surgical and histopathology
results at the time of readout. MRI criteria were analyzed for their
usefulness to differentiate ganglion from synovitis. The criteria were lesion
location, size, shape, margins, presence of internal septa, signal
characteristics, and contrast-enhancement pattern. Size measurements were
acquired from the axial plane T2-weighted images from the maximal side-to-side
and superior-to-inferior margins of the lesion with the size being calculated
as size (mm2)= a x b, where a =
width of the lesion and b = depth of the lesion. Shape was divided
into two forms, either diffuse and crescentic or uni- or multilocular and
rounded. Margins were characterized as either diffuse or defined. Internal
structure within the lesion was defined as with or without the presence of
septa. Contrast-enhancement pattern was defined as diffuse enhancement within
and throughout the lesion, wall enhancement at the periphery of the lesion, or
a combination of both.
View larger version (113K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A —19-year-old woman with MRI showing characteristic features of dorsal
occult ganglion. Axial T1-weighted image shows small, mass-like region
(arrow). Vitamin capsule was placed at level of pain and to side so
as not to obscure small lesions.
|
|
View larger version (95K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —19-year-old woman with MRI showing characteristic features of dorsal
occult ganglion. Corresponding axial fast spin-echo T2-weighted image shows
focal high-signal-intensity mass (arrow).
|
|
View larger version (117K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2C —19-year-old woman with MRI showing characteristic features of dorsal
occult ganglion. After contrast administration, T1-weighted fat-suppressed
image shows wall enhancement (arrow). This was evident in 50% of
dorsal occult ganglion cysts. Small dorsal occult ganglion characteristically
is located at dorsal aspect of scapholunate ligament, superficially overlying
midcarpal dorsal ligament.
|
|
View larger version (133K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2D —19-year-old woman with MRI showing characteristic features of dorsal
occult ganglion. Histopathology image shows myxoid central portion devoid of
gelatin myxoid material (asterisks) and chronic inflammatory wall of
ganglion. Note absence of synovial lining. At surgery, dorsal occult ganglion
was seen arising from scapholunate ligament, which was subsequently
débrided. (H and E, x10)
|
|
The site was divided into areas associated with the dorsal aspect of the
wrist joint, associated closely with the scapholunate ligament and dorsal
midcarpal ligament, or anywhere within the wrist. Tenosynovitis was defined as
an enlarged fluid-filled tendon sheath, and tendonosis as an enlargement of
the tendon with inhomogeneous signal intensity within the tendon. Lesion
signal characteristics were analyzed on both T1- and T2-weighted images and
compared with the signal intensity of muscle. Review for signs of inflammation
of the posterior dorsal nerve with enlargement and contrast enhancement was
performed on axial images before and after contrast administration.
All images were interpreted by consensus by two senior radiologists. The
radiographs and MR images were interpreted separately and at different
sessions. Data were recorded on a specifically created data sheet. Diagnoses
included ganglia or synovitis. Ganglia were divided into ganglion and dorsal
occult ganglion cysts. Ganglia could be located anywhere in the dorsal aspect
of the wrist, and the dorsal occult ganglia were defined by their origin at
the scapholunate ligament.
Surgery
Surgery was performed by the senior hand surgeons of the hand surgery unit.
They recorded their surgical findings as part of usual clinical practice,
often with diagramatic sketches. These notes were later correlated with the
MRI diagnoses. At surgery, the main cyst may be single or multiloculated with
a smooth, white, and translucent appearance and is strictly localized compared
with the diffuse extension of synovitis, which surrounds tendons and may
involve the underlying joint. A ganglion cyst may occur anywhere between
tendons, but its pedicle connects to the involved joint usually over the
scapholunate ligament (as in dorsal occult ganglion cysts) in contrast to the
synovitis that shows diffuse extension without a pedicle. The main cyst and
its pedicle are mobilized down to the underlying joint capsule. The ganglion
and its capsular attachments including the pedicle are tangentially excised
off the involved ligament to prevent recurrence. Posterior dorsal nerve
inflammation was defined as a thickened and obviously visible nerve. In
tenosynovitis, there is marked thickening of the tendon sheath, which is
completely excised.
View larger version (108K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A —21-year-old woman with MRI showing characteristic features of dorsal
occult ganglion. Axial T1-weighted image shows small focal mass
(arrow) adjacent to dorsal aspect of scapholunate ligament.
|
|
View larger version (100K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3C —21-year-old woman with MRI showing characteristic features of dorsal
occult ganglion. After contrast administration and fat suppression, diffuse
contrast enhancement (arrow) is seen within the lesion.
|
|
View larger version (110K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3D —21-year-old woman with MRI showing characteristic features of dorsal
occult ganglion. Histopathologic image shows chronically thickened myxoid wall
with very small central pseudolumen (asterisk) (gelatinous material
is removed during pathology processing) and with small lesion size, these
features presumably account for presence of 50% of diffuse contrast
enhancement of dorsal occult ganglion cysts. (H and E, x10)
|
|
Histopathology
MRI findings were correlated with intraoperative findings and
histopathology.
Statistics
The differences between dorsal occult ganglion cysts, ganglion, and
synovitis using six MRI parameters were assessed by exact Pearson chi-square
tests (categoric parameters) and Kruskal-Wallis tests (continuous parameters).
A p value of less than 0.05 was considered to indicate a
statistically significant difference. Discriminant analysis was used to find
combinations of parameters that differentiate between diagnoses. The agreement
between MRI findings and intraoperative findings was analyzed using the
proportion of overall agreement with a Blyth-Still-Casella confidence interval
and the kappa coefficient (proportion of agreement after chance agreement is
removed) with a 95% confidence interval. Sensitivity and specificity for each
diagnosis (i.e., synovitis, dorsal occult ganglion, and ganglion) were
calculated as well. Calculations were done in SAS software, version 8 (SAS
Institute) and StatXact, version 5 (Cytel Software).
Results
Patients
There were 34 operated patients (23 women, 11 men) ranging in age from 18
to 41 years, with an average age of 29.5 years. There was no statistical
difference for patient age or sex.
Radiographs
All radiographs (n = 68) were normal in appearance. There was no
evidence of soft-tissue swelling, calcification, or bone or joint
abnormality.
MRI
Thirty-five MRI abnormalities were found in 34 operated patients, including
25 ganglia (Figs. 1A,
1B, and
1C), 16 of which were dorsal
occult ganglia (Figs. 2A,
2B,
2C,
2D,
3A,
3B,
3C, and
3D), and six cases of
synovitis (Figs. 4A,
4B,
4C,
4D, and
4E)
(Table 1). MR diagnosis was
confirmed by surgery with an overall agreement of 71% (95% confidence interval
[CI], 55-85%) and an overall kappa of 0.57 (95% CI, 0.38-0.76). In four cases,
the MRI diagnosis (two ganglion cysts and two synovitis) could not be
confirmed at surgery. In two cases, MRI diagnosed ganglia that
intraoperatively were found to be synovitis. Two cases of synovitis were
diagnosed as ganglia on MRI and intraoperatively found to be dorsal occult
ganglion cysts. In one case, the MRI diagnosis was consistent with a fibrosed
tendon but was found to be tenosynovitis at surgery and pathology. There were
no cases of dorsal occult ganglia associated with posterior dorsal nerve
inflammation.
View larger version (82K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4B —40-year-old-man with MRI showing characteristic features of
synovitis. Corresponding fast spin-echo fat-suppressed T2-weighted image shows
diffuse region of altered signal intensity (arrows).
|
|
View larger version (128K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4E —40-year-old-man with MRI showing characteristic features of
synovitis. At higher power, chronic granulation tissue is also evident within
synovial thickening. Note synovial epithelial lining (arrows). At
surgery, synovitis was extensive, requiring major stripping. (H and E,
x10)
|
|
Correlation with Surgery or Surgery and Histopathology
The MRI diagnosis was correlated with surgery in all 34 patients and, in
addition, with histopathology in 24 (70%) of 34 patients
(Table 2). There was
sensitivity for ganglia of 89% (95% CI, 56-99%) and for dorsal occult ganglion
cysts of 94% (95% CI, 70-100%). Regarding the small synovitis group, there was
very low correlation mirrored by a low sensitivity of 33% but with an
acceptable specificity of 86%.
Analysis of MRI Features to Differentiate Ganglion and Synovitis
Six MRI criteria were analyzed for their usefulness to differentiate
ganglion from synovitis. MRI signal characteristics were not useful in
distinguishing between ganglion and synovitis. The margins, either diffuse or
defined, proved to be statistically significant in differentiating between
synovitis and ganglion (p < 0.001). The diffuse crescentic shape
was typical for synovitis in comparison with the uni- or multilocular rounded
shape of a ganglion (p < 0.001). The site was statistically
significant in separating dorsal radiocarpal joint-associated synovitis from
ganglia located anywhere in the wrist, and from the dorsal occult ganglion
cyst with its origin at the scapholunate ligament (p < 0.001).
Ganglia were larger in size (> 0.5 mm2) than dorsal occult
ganglion cysts and synovitis (< 0.5 mm2). In addition, synovitis
was differentiated in size from ganglia, with synovitis covering a broad and
thin area and ganglia having a narrower width (p < 0.001). Ganglia
and dorsal occult ganglion cysts commonly had a width less than 8 mm, whereas
synovitis was commonly greater than 8 mm in width. Ganglia commonly had a
depth greater than 3 mm, whereas dorsal occult ganglion cysts and synovitis
commonly had a depth less than 3 mm. In synovitis, there is usually no
septation in the internal structure; however, septa are commonly present in
two thirds of ganglia and in one half of dorsal occult ganglion cysts
(p = 0.003). Internal septa were present in a single case of
synovitis. Contrast enhancement was diffuse in synovitis compared with a wall
enhancement in ganglia (p < 0.001), and in dorsal occult ganglion
cysts there was either diffuse (50%) or wall (50%) enhancement. The main
criteria and results are summarized in
Table 3.
With multivariate analysis, the criteria margin (p < 0.001) and
site (p < 0.001) are sufficient for differentiating between
ganglion and synovitis (100%). Margin separates synovitis from dorsal occult
ganglion cysts and ganglion; site separates dorsal occult ganglion cysts and
ganglion.
Discussion
Dorsal chronic wrist pain is clinically difficult. Patients have often
suffered long-term pain, sought advice from many medical sources with no
effect, and may even have been labeled as malingerers. Often the diagnosis is
not known. Though a large number of etiologies
[1,
2] may be associated with the
chronic painful dorsal wrist, very little information is available regarding
causation and diagnosis with the literature dominated by the diagnosis of
pannus and synovitis associated with rheumatoid arthritis. Some etiologies
such as the dorsal occult ganglion have been assessed with MRI
[4,
5]. The natural history of
undiagnosed wrist pain, particularly in the young, may be disabling, with some
40% significantly troubled after long-term follow-up of more than 10 years
[6]. A subset of ganglia are
thought to cause chronic pain secondary to or related to internal derangements
of the wrist and ligament tears
[7,
8].
The significance of being able to distinguish between the two entities is
important, and preoperative MRI may allow a specific diagnosis. Patients with
chronic synovitis may undergo a longer trial period of conservative treatment,
and if surgery is required it may be more extensive. Patients with chronic
dorsal wrist pain due to a ganglion are more likely to go to surgery earlier
and the operation is more targeted with resection of the ganglion and
inspection and débridement of any associated ligament tears. Our series
has shown that MRI of dorsal chronic wrist pain is accurate in distinguishing
between synovitis and ganglion. After review of the results, many MRI features
are intuitive, but they have not been previously formally tested.
Chronic synovitis and ganglion cysts are histopathologically different.
Synovitis consists of a synovial proliferation with an inflammatory cellular
response with lymphocytes, plasma cells, and occasional macrophages. A
ganglion cyst consists of a myxoid wall with benign fibroblastic collagen and
has no epithelial synovial lining. There may be evidence of a light chronic
inflammatory cellular reaction in the myxoid wall, and there is myxoid
degeneration with a gelatinous fluid center
[2].
Because the wrist region may be associated with a large number of
incidental ganglion cysts [2],
we recommend the usual practice of demarcating the patient's pain with a
marker, such as a vitamin capsule, before the MRI. This was done in our series
and we had no confusion with additional incidental findings. We found the use
of contrast administration very helpful in this clinical setting. It was
statistically significant in distinguishing between ganglion and synovitis.
Some overlap of a diffuse contrast-enhancement pattern usually associated with
synovitis was seen in some of the small dorsal occult ganglion cysts, which
related to their smaller size and presumably thicker wall with a smaller
central luminal surface area and thus shorter time for intracystic dispersal
of contrast material (Figs. 3A,
3B,
3C, and
3D).
A pitfall of our study was its retrospective nature. Another was that not
all MRI diagnoses correlated with the surgical or histologic diagnosis. This
presumably relates to both entities having some form of dynamic inflammation
or irritation. We obtained 2D measurements of the lesions because 3D volume
sequences had not been performed. The 3D measurements may add further
information. We did not compare MRI with sonography because the aim of our
study was to determine MRI criteria useful in distinguishing ganglion from
synovitis in dorsal chronic wrist pain.
Sonography has been found to be generally useful in imaging the hand in
patients with persistent pain and suspected occult ganglion
[8]. The efficacy of MRI and
sonography in diagnosing the presence of dorsal occult ganglion cysts is
similar
[9-11];
however, MRI offers an objective, reproducible display of anatomic
relationships. MRI is not as user-dependent as sonography, though with
improved technique, for example with advanced use of panorama sonography, in
the future sonography may be the imaging technique of first choice and MRI may
be used for more difficult cases. If so, we hope the presented criteria may be
helpful. Another positive aspect to this study is that all pain is not
necessarily related to ganglia, and having a definitive diagnosis, as well
excluding other sinister diagnoses, is helpful in this clinical setting.
In conclusion, MRI is a helpful preoperative tool in distinguishing between
ganglion and synovitis in the chronic painful dorsal wrist. MRI is
diagnostically helpful with chronic wrist pain without obvious swelling.
Characteristically, ganglia have defined margins, are spherical in shape, are
multilocular with septa, and have wall enhancement. Characteristically,
synovitis will have diffuse margins, be crescentic in shape over a broad thin
area, lack septa, and have evidence of diffuse enhancement.
Acknowledgments
Many thanks to Daniel Dietrich, Department of Statistics, University of
Bern, for statistical analysis and support. Thanks also to Karin Kohler and
Elisabeth Haefeli for image preparation and to Susanne Furrer for manuscript
support.
References
- Chidgey LK. Chronic wrist pain. Orthop Clin North
Am 1992; 23:49
-64[Medline]
- Angelides AC. Ganglions of the hand and the wrist. In: Green DP,
ed. Operative hand surgery, 4th ed. New York, NY:
Churchill Livingstone, 1999:2171
-2183
- Wolfe SW. Tenosynovitis. In: Green DP, ed. Operative
hand surgery, 4th ed. New York, NY: Churchill Livingstone,1999
: 2022-2044
- Hollister AM, Sanders RA, McCann S. The use of MRI in the diagnosis
of an occult ganglion cyst. Orthop Rev1989; 28:1210
-1212
- Vo P, Wright T, Hayden F, Dell P, Chidgey L. Evaluating dorsal
wrist pain: MRI diagnosis of occult dorsal wrist ganglion. J Hand
Surg Am 1995; 20:667
-670[Medline]
- Ryley JP, Langstaff RJ, Barton NJ. The natural history of
undiagnosed wrist pain in young women: a long-term follow-up. J
Hand Surg Br 1992; 17:51
-54[Medline]
- Povlsen B, Pekett WR. Arthroscopic findings in patients with
painful wrist ganglia. Scand J Plast Reconstr Surg Hand
Surg 2001; 35:323
-328[Medline]
- el-Noueam KI, Schweitzer ME, Blasbalg R, et al. Is a subset of
wrist ganglia the sequela of internal derangements of the wrist joint? MR
imaging findings. Radiology 1999;212
: 537-540[Abstract/Free Full Text]
- Osterwalder JJ, Widrig R, Stober R, Gachter A. Diagnostic validity
of ultrasound in patients with persistent wrist pain and suspected occult
ganglion. J Hand Surg Am 1997;22
: 1034-1040[Medline]
- Cardinal E, Buckwalter K, Braunstein EM, Mih A. Occult dorsal
carpal ganglion: comparison of US and MR imaging.
Radiology 1994;193
: 259-262[Abstract/Free Full Text]
- Blam O, Bindra R, Middleston W, Gelbermann R. The occult dorsal
ganglion: usefulness of magnetic resonance imaging and ultrasound in
diagnosis. Am J Orthop 1998;27
: 107-110[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
