DOI:10.2214/AJR.05.0764
AJR 2007; 188:198-202
© American Roentgen Ray Society
Extensor Retinaculum of the Wrist: Sonographic Characterization and Pseudotenosynovitis Appearance
Brian L. Robertson1,
David A. Jamadar1,
Jon A. Jacobson1,
Monica Kalume-Brigido1,
Elaine M. Caoili1,
Zvi Margaliot2 and
Michel O. De Maeseneer1,3
1 Department of Radiology, University of Michigan Hospitals, 1500 E Medical
Center Dr., TC2910, Ann Arbor, MI 48109.
2 Division of Plastic Surgery, Department of Surgery, Trillium Health Centre,
Mississauga, ON, Canada.
3 Present address: Division of Radiologic Sciences, Wake Forest University,
Winston-Salem, NC 27157-1088.
Received May 4, 2005;
accepted after revision July 25, 2005.
Address correspondence to D. A. Jamadar.
Abstract
OBJECTIVE. We have found in our practice that the normal extensor
retinaculum of the wrist may appear hypoechoic on sonography and, because it
is closely applied to the extensor tendons, may simulate tenosynovitis. This
study prospectively evaluates the extensor retinaculum in 50 healthy adult
volunteers, characterizing its sonographic appearance.
CONCLUSION. The extensor retinaculum has a characteristic appearance
on sonography. A hypoechoic appearance from anisotropy should not be confused
with tenosynovitis.
Keywords: anatomy • hand • sonography • tenosynovitis • wrist
Introduction
Sonography of the hand and wrist has been used successfully to evaluate a
number of musculoskeletal disorders. Several examples include the diagnosis of
carpal tunnel syndrome, ganglionic cyst, and tendon abnormalities, such as a
tendon tear and tenosynovitis
[1]. Particularly in the distal
extremities, where there is less soft-tissue attenuation, high-frequency
probes may be used to produce exquisite detail of anatomy and any
abnormality.
At the dorsum of the wrist, the extensor retinaculum is a fibrous band that
holds the extensor tendons in place and prevents bow-stringing (Figs.
1 and
2). By extending fascial
attachments to the underlying bones and periosteum, the retinaculum forms six
compartments over the dorsal wrist, each of which contains various tendons
[2,
3].
In our clinical practice, we have observed a variable appearance of the
normal extensor retinaculum on sonography that, because of its close proximity
to the extensor tendons, may be misinterpreted as complex fluid in the tendon
sheath or in the adjacent soft tissues and thus prompt the diagnosis of
tenosynovitis. This study characterizes the sonographic appearance of the
extensor retinaculum at the wrist in a population of healthy adult
volunteers.
Subjects and Methods
Institutional review board approval was obtained before commencing this
study. The initial in vivo study population consisted of 51 healthy
volunteers. Exclusion criteria consisted of a history of wrist trauma or of a
systemic disease, such as rheumatoid arthritis or psoriasis, or a history of
tenosynovitis at the wrist. The final study population consisted of 50
subjects, with one subject with psoriasis excluded.
One hundred wrists of 50 healthy subjects (15 men, 35 women) were examined.
The subjects had a mean age of 33.4 years (age range, 20-58 years). Six of the
50 subjects were left hand-dominant, 43 were right hand-dominant, and one was
ambidextrous.
After written informed consent was obtained from the study subjects,
sonographic examinations were performed using a 7- to 15-MHz compact linear
array transducer (model HDI 5000, Philips-Advanced Technology Laboratories).
All scanning was performed by a board-certified sonographer (8 years of
experience) who had been specifically trained in musculoskeletal sonography
(2.5 years of musculoskeletal sonography experience). For the sonography
examinations, each subject sat with the arm and hand extended and placed in a
prone position on the examination table. The table was positioned to level the
arm comfortably. Liberal transmission gel was used in place of a standoff pad.
Sonographic scanning was performed on the prone dorsal wrist where the
retinaculum was identified by its expected anatomic location and compact
fibrillar sonographic texture.
Images were obtained along the long axis (longitudinal) and along the short
axis (transverse or cross section) of the retinaculum of each wrist. A minimum
of five images of each wrist were obtained. Along the short axis of the
retinaculum, sonographic images were obtained with the sound beam
perpendicular to the retinaculum and in obliquity to determine the effect of
transducer positioning on echogenicity. The sonographer also evaluated the
extensor retinaculum with power Doppler sonography (color power angiography,
78%; Wall filter, medium; pulse repetition frequency, 700). Hand dominance was
recorded; both right and left wrists were examined in all subjects.
View larger version (81K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A —37-year-old woman with normal extensor retinaculum. Sonograms
along long axis of extensor retinaculum (A) and along short axis of
extensor retinaculum (B) show normal subtle compact fibrillar extensor
retinaculum (arrows) to be relatively hypoechoic compared with
adjacent extensor digitorum tendons (ED). Note radius (R) in B, lunate
(L) in A and B, capitate (C) in B, and scaphoid (S) in
A; distal is to right in B.
|
|
View larger version (80K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B —37-year-old woman with normal extensor retinaculum. Sonograms
along long axis of extensor retinaculum (A) and along short axis of
extensor retinaculum (B) show normal subtle compact fibrillar extensor
retinaculum (arrows) to be relatively hypoechoic compared with
adjacent extensor digitorum tendons (ED). Note radius (R) in B, lunate
(L) in A and B, capitate (C) in B, and scaphoid (S) in
A; distal is to right in B.
|
|
The sonographic images were retrospectively analyzed by a
fellowship-trained musculoskeletal radiologist (8 years of experience in
musculoskeletal sonography) and by the same sonographer who had obtained the
images. The time interval between image acquisition and retrospective image
analysis was 7-10 days. Sonographic images obtained along the short axis of
the extensor retinaculum were used for measurements. On these static images,
the extensor retinaculum was identified by its characteristic location and
compact fibrillar echotexture. Retinaculum thickness and the short-axis width
were measured by the radiologist using measurement calipers at an image
workstation (Kinet Dx Workstation 3000, Siemens Medical Solutions).
View larger version (92K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4 —50-year-old woman with normal extensor retinaculum. Sonogram
along short axis of extensor retinaculum shows its characteristic fusiform
thickening and location (arrows). Note hypoechoic appearance relative
to extensor digitorum tendons (ED), although internal echoes are still
visible. Note radius (R), lunate (L), and capitate (C) bones; distal is to
right.
|
|
View larger version (88K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5A —37-year-old man with normal extensor retinaculum. Sonograms
show effect of angling transducer between relatively perpendicular (A)
to retinaculum (arrows, A) and at angle (B) to
retinaculum (arrows, B). Retinaculum appears more hypoechoic
when beam is at angle. Note extensor digitorum tendons (ED) and radius (R),
lunate (L), and capitate (C) bones; distal is to right.
|
|
View larger version (99K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5B —37-year-old man with normal extensor retinaculum. Sonograms
show effect of angling transducer between relatively perpendicular (A)
to retinaculum (arrows, A) and at angle (B) to
retinaculum (arrows, B). Retinaculum appears more hypoechoic
when beam is at angle. Note extensor digitorum tendons (ED) and radius (R),
lunate (L), and capitate (C) bones; distal is to right.
|
|
View larger version (95K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6 —48-year-old man with 19-year history of psoriasis and
asymptomatic dorsal wrist. Sonogram shows retinaculum retains fusiform shape
(arrows), but flow adjacent to retinaculum is increased and flow
around extensor digitorum tendons (ED) is increased. Note radius (R). Distal
is to right.
|
|
For each wrist, the echogenicity of the extensor retinaculum was
characterized by consensus agreement as hypoechoic, isoechoic, or hyperechoic
relative to the adjacent tendon tissue. If a retinaculum had several
echogenicities on different images from the same wrist, each was recorded. The
shape of the retinaculum was also characterized on short-axis images of the
retinaculum. The static power Doppler images were also reviewed by consensus
and were characterized as showing positive or negative flow within the
retinaculum. The power Doppler settings were configured for the first subject
at a threshold just short of aliasing to optimize for visualization of
low-velocity flow and were then reproduced for the remaining subjects.
Approval for use of a cadaveric specimen was obtained from the anatomic
donations department. The skin and subcutaneous tissues were dissected, by one
of the authors, off the dorsum of the right wrist of a single formalin-fixed
adult male cadaver to expose the superficial aspect of the wrist
(Fig. 2).
Results
Sonography revealed the extensor retinaculum in all wrists located
superficial to the extensor tendons at the level of the radiocarpal joint
(Figs. 3A and
3B). Along the long axis of the
extensor retinaculum (Fig. 3A),
the continuous homogeneous compact fibrillar echotexture was identified
superficial to the extensor tendons and deep to the subcutaneous fat. Along
the short axis of the extensor retinaculum, a fusiform appearance was noted
with thickening of the central portion at the level of the radiocarpal joint,
tapering both distally and proximally (Fig.
3B). This configuration was observed in all 100 retinacula, with
variation only in the long-axis dimension and thickness.
Sonography along the short axis of the extensor retinaculum showed a
maximum thickness of 0.6-1.7 mm in men (mean ± SD, 1.16 ± 0.24
mm) and 0.5-1.6 mm in women (mean, 1.06 ± 0.22 mm). The short axis of
the retinaculum varied from 8.2 to 23 mm in men (mean, 15.6 ± 2.73 mm)
and from 8.0 to 19.8 mm in women (mean, 12.9 ± 2.40 mm).
Using the two-tailed Student's t test, we found a statistically
significant difference in the short-axis dimension of the retinaculum between
men and women (p < 0.001). Significant differences between men and
women were also noted in the retinacular short-axis dimensions in both right
(p = 0.003) and left (p = 0.001) hands. A trend was noted
when assessing the differences in thickness of the extensor retinaculum
between men and women (p = 0.6): No significant differences in
thickness (men: p = 0.718; women: p = 0.873) or short-axis
dimension (men: p = 0.718; women: p = 0.603) were noted when
comparing the effects of handedness, or hand dominance.
In all 100 extensor retinacula, when imaged perpendicular to the
retinaculum, a subtle compact fibrillar echotexture was seen slightly
hypoechoic relative to tendon (Figs.
3A and
3B). When imaged obliquely, all
retinacula became more hypoechoic, although internal echoes were still
identified (Fig. 4). None of
the retinacula were found to be hyperechoic relative to tendon. In all wrists,
there was no flow with power Doppler imaging.
The cadaveric part of the study was performed on an elderly male whose age
at the time of death and cause of death were unknown. The dissection revealed
a fibrous band in the expected location of the extensor retinaculum extending
obliquely across the dorsal wrist. A photograph was obtained to illustrate
gross anatomy (Fig. 2).
Discussion
The extensor retinaculum is a fibrous thickening of the investing fascia,
extending around the wrist to become continuous with the volar carpal
ligament. Proximally the extensor retinaculum arises from the antebrachial
fascia of the forearm and distally it tapers, becoming indistinguishable from
the loose fascia over the metacarpals, which is part of the superficial fascia
of the dorsum of the hand [2,
3]. At the level of the
radiocarpal joint, the extensor retinaculum courses obliquely across the
dorsal carpus and is thickened centrally along its long axis, tapering
proximally and distally. The function of the extensor retinaculum is to
prevent bowstringing by keeping the various extensor tendons in close
proximity to the underlying radius and ulna but separate from each other at
the wrist [2,
3]. Portions of the extensor
retinaculum may be used as a graft in the reconstruction of interosseous
ligaments of the wrist [4] or
in the reconstruction of pulleys in the fingers
[5].
In our clinical practice, we have noted a variable sonographic appearance
of the extensor retinaculum. We found that the fusiform shape and possible
hypoechogenicity may cause the normal extensor retinaculum to simulate
tenosynovitis of the extensor tendons. This study characterizes the extensor
retinaculum in healthy volunteers with regard to its shape, size, and
echogenicity.
Our results show that the normal extensor retinaculum has a characteristic
appearance superficial to the extensor tendons, tapering proximally and
distally with a subtle compact fibrillar echotexture, at the level of the
radiocarpal joint. The maximal thickness of the retinaculum ranged from 0.5 to
1.7 mm, with no effect related to handedness. The difference in the short-axis
dimension of the retinaculum between men and women may be in part explained by
the overall size differences between these two groups, although this was not
directly assessed in this study.
With regard to echogenicity, the normal extensor retinaculum showed a
subtle fibrillar echotexture when imaged perpendicular to the sound beam and
was relatively hypoechoic compared with adjacent tendon. The extensor
retinaculum appeared even more hypoechoic when imaged obliquely. This can be
explained by anisotropy, which may be seen when scanning other structures in
the musculoskeletal system, such as tendons and ligaments.
The sonographic appearance of compact linear fibrillar structures as
hyperechoic, hypoechoic, or even anechoic is due to the sonographic phenomenon
of anisotropy
[6-8].
Anisotropy is a sonographic property of linearly organized tissues, such as
tendons, ligaments, and nerves, where sonographic appearance is determined in
part by the angle of insonation of the ultrasound beam. When the beam is
perpendicular to a structure such as the extensor retinaculum, that structure
will appear reflective and the internal stacked parallel linear structure will
be visible. When the beam is angled away from perpendicular, these linear
reflective structures appear less hyperechoic. Even an angle of as little as
2° off the perpendicular can produce this phenomenon
[6]. Angling the transducer
back and forth during real-time scanning to optimize reflectivity (Figs.
5A and
5B) will help to distinguish
anisotropy from hypoechoic abnormalities such as tenosynovitis
[8].
Tenosynovitis is the result of a disorder of the synovium surrounding the
tendon that causes synovial proliferation and the production of fluid that
accumulates around the involved tendon
[9]. The etiology of
tenosynovitis includes infectious, inflammatory, and posttraumatic causes.
Sonographically, although simple fluid surrounding a tendon typically appears
anechoic, both the synovium and complex fluid may appear hypoechoic
[10]. The presence of flow on
color or power Doppler imaging may help differentiate complex fluid from
synovitis, with the latter possibly showing increased flow, helping to
differentiate inflamed synovium from a normal extensor retinaculum. Knowledge
of the anatomy of the wrist, the location of the synovial tendon sheaths in
relation to the tendons and retinaculum, and patient symptoms also help to
differentiate between the two. In the subject with psoriasis, one of our
exclusion criteria, sonography showed flow adjacent to the extensor
retinaculum (Fig. 6).
We found that the normal extensor retinaculum may appear as abnormal
hypoechoic tissue surrounding the extensor tendons because of anisotropy;
however, the characteristic location of the extensor retinaculum at the level
of the radiocarpal joint and its typical shape, width, and thickness should
indicate that the tissue is the normal extensor retinaculum. By angling the
transducer during real-time imaging to obtain a perpendicular orientation
between the sound beam and the extensor retinaculum, the subtle compact
fibrillar echotexture of the normal extensor retinaculum can be detected.
Comparison with the contralateral side, another advantage of sonography, may
also be helpful because symmetry is often found in normal structures.
One potential limitation of this study is the absence of a gold standard.
However, the known anatomic location of the extensor retinaculum and typical
compact fibrillar appearance of ligaments and ligamentlike structures made
identification and evaluation possible. In addition, the consistent
identification of this structure in all asymptomatic wrists is also evidence
that a normal anatomic structure was identified. Last, interobserver
variability for extensor retinaculum identification and measurements was not
tested; however, characterization and measurements were made with agreement by
consensus.
In summary, the normal extensor retinaculum of the wrist has a
characteristic location and sonographic appearance that will allow
identification and prevent misinterpretation as tenosynovitis.
References
- Middleton WD, Teefey SA, Boyer MI. Hand and wrist sonography.
Ultrasound Q2001; 17:21
-26[Medline]
- Palmer AK, Skahen JR, Werner FW, Glisson RR. The extensor
retinaculum of the wrist: an anatomic and biomechanical study. J
Hand Surg Br 1985;10:11
-16[Medline]
- Taleisnik J, Gelberman RH, Miller BW, Szabo RM. The extensor
retinaculum of the wrist. J Hand Surg Am1984; 9:495
-501[Medline]
- Weiss AP. Scapholunate ligament reconstruction using a
bone-retinaculum-bone autograft. J Hand Surg Am1998; 23:205
-215[Medline]
- Guelmi K, Barbato B, Kolb F. Reconstruction of a
metacarpophalangeal pulley in the thumb with a free extensor retinaculum
graft. J Hand Surg Br1997; 22:274
-276[CrossRef][Medline]
- Connolly DJA, Berman L, McNally EG. The use of beam angulation to
overcome anisotropy when viewing human tendon with high frequency linear array
ultrasound. Br J Radiol2001; 74:183
-185[Abstract/Free Full Text]
- Crass JR, van de Vegte GL, Harkavy LA. Tendon echogenicity: ex vivo
study. Radiology1988; 167:499
-501[Abstract/Free Full Text]
- Martinoli C, Bianchi S, Derchi LE. Tendon and nerve sonography.
Radiol Clin North Am1999; 37:691
-711[CrossRef][Medline]
- Rasmussen OS. Sonography of tendons. Scand J Med Sci
Sports 2000;10:360
-364[CrossRef][Medline]
- Breidahl WH, Stafford Johnson DB, Newman JS, Adler RS. Power
Doppler sonography in tenosynovitis: significance of the peritendinous
hypoechoic rim. J Ultrasound Med1998; 17:103
-107[Abstract]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
