Ultrasonographically Assisted

Carpal Tunnel Release

Ken-ichi Nakamichi, MD, Shintaro Tachibana, MD, Tokyo, Japan

An operative technique of carpal tunnel release using intraoperative ultrasonography is

described. In this technique, "safe line" is defined in the transverse carpal ligament and the
adjacent deep forearm fascia midway between the ulnar margin of the median nerve and the
radial margin of the ulnar artery. After ultrasonographic design of a 1.0 to 1.5-cm skin inci-
sion along the safe line at the distal carpal tunnel, the distal ligament is released under direct
vision. Proximal release is performed along this line under ultrasonographic monitoring
using a device that consists of a basket punch and an outer metal tube. In a prospective ran-
domized study, the outcomes were compared for carpal tunnel release using either this tech-
nique in 50 hands of 50 patients or conventional open release in 53 hands of 53 patients.
Follow-up assessment at 3, 6, 13, 26, 52, and 104 weeks showed no significant difference
with respect to numbness and paresthesias, static two-point discrimination, findings on
Semmes-Weinstein monofilament testing, findings on manual muscle testing of the abductor
pollicis brevis, and electrophysiologic findings. The ultrasonographic-release group had bet-
ter outcomes regarding pain, tenderness of the scar, and key-pinch strength at 3, 6, and 13
weeks, and grip strength at 3 and 6 weeks after surgery. The scar was more aesthetic in this
group. There were no complications with either technique. (J Hand Surg 1997;22A:853-862.)

In the surgical treatment of carpal tunnel syn-
drome (CTS), open carpal tunnel release (OCTR) is
considered a reliable and simple technique with a
minimal complication rate. Recently, with the advent
of endoscopic carpal tunnel release (ECTR) tech-
niques, limited-dissection release has been credited
with less postoperative discomfort, a shorter period
of disability, and a smaller incision, 1-5 but complica-
tions and costs are the major concern. 6-9
The neurovascular structures that should be pro-
tected when dividing the transverse carpal ligament
(TCL) and the adjacent forearm fascia include the
median nerve and the ulnar neurovascular bundle.
On the basis of the carpal tunnel anatomy, this can
From the Departmentof OrthopaedicSurgery,ToranomonHospital,
Tokyo,Japan.
Received for publication Sept. 6, 1996, accepted in revised form
April 15, 1997.
No benefitsin any formhavebeen receivedor will be receivedfroma
commercialpartyrelateddirectlyor indirectlyto the subjectof this article.
Reprint requests: Ken-ichi Nakamichi, MD, Department of
OrthopaedicSurgery,ToranomonHospital,2-2-2, Toranomon,Minato-
ku, Tokyo105,Japan.
be accomplished by releasing within the area be-
tween the median nerve and the ulnar artery, and
safety seems to be maximized by release along the
line bisecting this area (safe line, Fig. 1). For this
purpose, both the median nerve and the ulnar artery
should be identified. Our concern about E C T R is the
difficulty in knowing the proximity of the median
nerve to an operating instrument and the inability to
identify the ulnar neurovascular bundle. It has been
pointed out that the former risks the median nerve 10
and that the latter is associated with the inadvertent
release of Guyon's canal or a risk of injuries to its
contents. 1~ These considerations have led us to
develop a limited-dissection technique in which
release is performed along the safe line with ultra-
sonographic assistance; an important advantage of
this technique is that expensive disposable instru-
ments are not used.
To define the role of this technique--ultrasono-
graphic carpal tunnel release ( U C T R ) - - w e obtained
institutional approval and started a prospective ran-
domized study on the outcomes of both U C T R and
OCTR in February 1993. The purpose of this report

The Journal of Hand Surgery 853

854 Nakamichi and Tachibana/Ultrasound and Carpal Tunnel Release
Safe I i ne ulnar
artery
/ Ulnar
nerve
Medi an nerve
Figure 1. Distal-to-proximal view of the carpal tunnel at
the level of the hook of hamate. A safe line is defined in
the transverse carpal ligament and the adjacent deep fore-
arm fascia midway between the ulnar margin of the
median nerve and the radial margin of the ulnar artery.
Note this line is approximately palmar to the middle (m)
or ring (r) finger flexor digitorum superficialis tendon.
is to describe the technique and present the results of
the study. To provide a uniform model, we included
only homemakers with idiopathic CTS in the study.
Materials and Methods
Selection of Patients
One hundred three hands of 103 female homemak-
ers (age range, 45-81 years; mean, 58 years) with
idiopathic CTS were included.
The diagnosis of CTS was made clinically and
electrophysiologically. The clinical evaluation in-
cluded questioning about sensory symptoms, admin-
istration of Phalen's test, and tests for sensibility and
muscle strength. In the electrophysiologic evalua-
tion, we measured the median distal motor nerve
latency and the median sensory nerve conduction
velocity across the carpal tunnel. To test for the for-
mer, we stimulated the nerve 2 cm proximal to the
wrist crease, and for the latter, the orthodromic con-
duction velocity was calculated based on the dis-
tance between stimulation at the middle crease of the
middle finger and the recording needle electrode 2
cm proximal to the wrist crease. The upper limit of
normal for the former was 4.2 ms, and the lower
limit of normal for the latter was 45 m/s. We used a
sensory-conduction velocity instead of a latency to
adjust for differences in hand size that would result
in varied distances among individuals between the
stimulation and the recording electrode. These crite-
ria have a 4.8% false negative rate. 12
Every patient was screened for local pathology by
wrist imaging consisting of plain radiographs and
ultrasonograms. In addition, we assessed rheumatoid
factor, thyroid and growth hormone levels, and
serum and urine electrophoresis to rule out rheuma-
toid arthritis, hypothyroidism, gigantism, and multi-
ple myeloma or M-proteinemia/-proteinuria.
Our diagnostic criteria for idiopathic CTS were
based on exclusion of the following: rheumatoid
arthritis, multiple myeloma, M-protein in serum or
urine, chronic renal failure treated by hemodialysis,
diabetes, gout, hypothyroidism, gigantism, Colles
fracture, space-occupying lesions, chondrocalcino-
sis, osteoarthritis, and tuberculosis.
Surgery was performed on hands with thenar atro-
phy or intractable sensory symptoms with poor
response to conservative treatment for 3 months,
including avoidance of overuse, splinting, and local
steroid injection.
On the basis of random drawing, UCTR was per-
formed in 50 hands of 50 patients and OCTR in 53
hands of 53 patients. All patients gave informed con-
sent. There were patients who later underwent
surgery on the opposite hands. These hands, how-
ever, were not included in this study because we
wished to avoid the possibility that some patients
would refuse when the random drawing assigned dif-
ferent procedures for the opposite hands.
Preoperative Evaluation
The hands were examined preoperatively by a
hand therapist who was blinded to the procedures to
be done. Numbness and paresthesias were graded as
0 (absent), 1 (mild), 2 (moderate), or 3 (severe). Sen-
sibility was quantified in the middle finger by static
two-point discrimination and Semmes-Weinstein
monofilament testing. Motor tests included the man-
ual muscle testing of the abductor pollicis brevis
(APB) (results graded 0-5) and grip and key-pinch
measurements. Electrophysiologic data were reob-
tained just before surgery.
Surgical Technique
Ultrasonographic Carpal Tunnel Release. The
UCTR technique involved release along the safe line
(Fig. 1) distally under direct vision and proximally
under ultrasonographic monitoring using a cutting
device (Fig. 2).
 The Journal of Hand Surgery/Vol. 22A No. 5 September 1997 855

Figure 2. Cutting device for proximal release. (A) It consists of a basket punch (2.7-mm outer diameter) and an outer metal
tube (3.5-mm outer diameter) that has been taken from an arthroscopic shaver. (B, C) The tube is applied on the basket
punch. The outer tube has a side window at its rounded end through which the upper jaw of the basket punch is opened. The
tube has four roles: (1) to hold the device tip against the undersurface of the transverse carpal ligament, (2) to mildly retract
the flexor tendons to prevent them from being caught by the punch, (3) to collect the resected debris, and (4) to create clear
acoustic shadows in ultrasonographic images for recognition of the device position. (B) Side view of the device tip. The fully
opened upper jaw has a height of 2.8 mm. (C) Top view of the device tip. The upper jaw has a width of 1.2 mm.

In addition to tools in an ordinary hand tray, we
used an ultrasound system with a 10-MHz scanner
(SSD-650CL, Aloka Inc., Tokyo, Japan), a sterile
plastic bag to wrap the scanner, sterile jelly, small
self-retaining retractors, and the aforementioned cut-
ting device. This procedure was performed without
an assistant.
The hand was elevated by being rested on the bot-
tom surface of a small bowl that was placed upside
down because the handle of the basket punch, when
this was used, was below the hand level. A pneu-
matic tourniquet was not used for pulsation of the
ulnar artery to be recognized.
The procedure began with ultrasonographic loca-
tion of the key structures. The scanner was angled to
obtain the best contrast of the target structure, which
was considered the most important tip on the scan
technique. In longitudinal images, the distal TCL
edge was identified (Fig. 3A). The T C L was best
visualized at the level of the middle finger flexor dig-
itorum superficialis (FDS) tendon as a thick structure
immediately palmar to this tendon. Its distal edge
was usually recognized in static images, but real-
time observation during passive motion of the mid-
dle finger was also helpgul. The edge was appreci-
ated at the junction between the distal thin fascia,
which was vertically mobile during the tendon
motion, and the TCL proper, which was solid. At this
time, the pulsation of the superficial palmar arch
(SPA) was seen usually 1-2 cm distal to the distal
T C L edge. In axial images, the median nerve and the
ulnar artery were located to define the safe line (Fig.
3B). The flexor tendons, which were readily identi-
fied by the passive finger motion, were used as land-
marks for recognition of the nerve. It was visualized
as a low-echogenic oval structure surrounded by the
856 Nakamichiand Tachibana/Ultrasound and Carpal Tunnel Release

: ~age

Figure 3. Ultrasonographic marking of a skin incision. (A) A longitudinal image at the level of the middle finger flexor
digitorum superficialis tendon (Middle f. FDS). The distal edge of the transverse carpal ligament (TCL) is recognized at
the junction between the TCL and the distal thin fascia over the superficial palmar arch (SPA). It is marked on the palm
(dotted line in Fig. C). (B) An axial image of the distal carpal tunnel. The safe line runs in the TCL midway between the
ulnar margin of the median nerve (Median N.) and the radial margin of the ulnar artery (Ulnar A.). It is projected on the
palm. (C) A 1.0- to 1.5-cm skin incision (solid line) is marked on the projected safe line, with its center crossed by the dis-
tal edge of the TCL (dotted line).

TCL palmarly, the flexor pollicis longus tendon
radiodorsally, the index FDS tendon dorsally, and the
middle finger FDS tendon ulnodorsally. The ulnar
artery was confirmed ulnarly by its pulsation.
On the basis of this observation, a skin incision
was marked on the palm (Fig. 3C). After local anes-
thesia, the incision was made and deepened perpen-
dicular to the TCL. Mild bleeding was controlled by
placement of a small self-retaining retractor and with
bipolar cautery. After longitudinal division of the
palmar aponeurosis, the distal TCL was divided
under direct vision (Fig. 4). The adipose tissue
around the SPA was approached. Release was done
in this operative field as proximally as possible.
Next, the middle or ring finger FDS tendon was seen
through the synovium as noted in Figure 1. At the
same time, we investigated, under direct vision,
whether there was high division of the median nerve
or whether the motor branch passed around the distal
TCL edge. (If the former was present, we intended to
 The Journal of Hand Surgery / Vol. 22A No. 5 September 1997 857

Figure 4. Division of the distal transverse carpal ligament

(arrows) under direct vision. Arrowhead indicates the
flexor tendon synovium. Right, radial. Upper, proximal.

Figure 6. Proximal release is done with the cutting device

convert the procedure to an OCTR. Regarding the
under ultrasonographic monitoring. The scanner is ster-
latter, we thought the procedure could be continued
ilely wrapped. The self-retaining retractor has been
by keeping the motor branch out of harm's way.) changed to a smaller one to make room for the scanner.
During this investigation, the synovium was pro-
tected as much as possible to minimize bleeding.
After distal release, the cutting device was inserted
extrabursally with the upper jaw opened (Fig. 5). the tunnel after the jaw was closed. Third, the device
The device was localized along the safe line under was slightly withdrawn (2-3 mm), the upper jaw was
ultrasonographic monitoring (Fig. 6). Release was reopened through the gap of the divided TCL, and
done in a bite-by-bite fashion in the proximal direc- then we repeated the first step. During this cycle, it
tion (Fig. 7). Each bite consisted of 3 steps. First, the was not necessary to apply upward pressure to the
opened upper jaw was pushed snugly against the device tip, because the tension produced by the
TCL. Second, it was closed to cut the TCL. During
this step, forward pressure was not applied to the
device; otherwise, the device tip would migrate into

Figure 7. Ultrasonogram showing the position of the cut-

ting device at the midcarpal tunnel. An acoustic shadow is
created under the device. It is recognized between the
Figure 5. Extrabursal insertion of the cutting device with ulnar margin of the median nerve (Median N.) and the
the upper jaw opened. Right, radial. Upper, proximal. radial margin of the ulnar artery (Ulnar A.).
858 Nakamichi and Tachibana / Ultrasound and Carpal Tunnel Release
Figure 8. Ultrasonogram showing the position of the cut-
ting device at the proximal carpal tunnel. An acoustic
shadow is created under the device. It is recogized between
the ulnar margin of the swollen median nerve (Median N.)
and the radial margin of the ulnar artery (Ulnar A.).
flexor tendons held the tip against the undersurface
of the TCL. At the proximal carpal tunnel, the cross-
section of the median nerve was larger because of
its neuroma-like swelling proximal to compression
(Fig. 8). Release was done to exceed the level proxi-
mally where the swollen nerve had the largest cross-
sectional area, which was usually 5-10 mm proximal
to the wrist crease.
Release was confirmed at the distal two thirds of
the tunnel under direct vision. At the proximal one
third, communication was confirmed using a hemo-
stat between the inside and the outside of the tunnel
under palpation or ultrasonography. After irrigation,
the wound was closed.
Open Carpal Tunnel Release. This technique
also was performed without an assistant. Using local
anesthesia and a pneumatic tourniquet, we made a
curved incision 5 mm ulnar and parallel to the thenar
crease beginning just distal to Kaplan's cardinal line.
It extended proximally to the wrist crease. The pal-
mar aponeurosis and the TCL were divided longi-
tudinally. The adjacent deep forearm fascia was
divided 1 cm proximal to the wrist crease. Neither
synovectomy nor neurolysis was performed. After
the tourniquet was deflated, hemostasis was ob-
tained. The wound was irrigated and closed.
Time and Cost
We recorded the procedure time required for each
group: from ultrasonographic marking of a skin inci-
sion to wound closure in the UCTR group and from
administration of local anesthesia to wound closure
in the OCTR group.
Surgical costs, including tools, equipment, and the
operating room, were calculated based on the gov-
ernment-controlled medical billing system.
Postoperative Evaluation
No splinting was used in either group. All the
patients were encouraged to use the hand as toler-
ated. The dressing and sutures were removed 7-10
days after surgery. All patients were asked to visit us
every month to minimize dropout.
At 3, 6, 13, 26, 52, and 104 weeks after surgery,
the hands were examined by a hand therapist who
was blinded to the procedures. Before each examina-
tion, an adhesive soft tape was attached to the proxi-
mal palm to obscure the incision. In addition to the
parameters evaluated preoperatively, pain and ten-
derness of the scar were scored as 0 (absent), 1
(mild), 2 (moderate), or 3 (severe). Electrophysio-
logic studies were repeated at 13, 26, 52, and 104
weeks. Scars were classified at 104 weeks into
almost invisible, visible but fiat, contracted, or
hypertrophic. Return-to-job status was not assessed
because the patients were all homemakers.
Statistical Analysis
We used analysis of covariance for comparison at
each time interval, and repeated-measures analysis
of covariance to compare the results of the two
groups. No-shows were excluded from the analysis.
Variables were presented as mean and standard devi-
ation. The reported p values were two-tailed. The
level of significance was p < .05.
Results
Number of Patients
The dropout rate was 8% or less in the UCTR
group and 11% or less in the OCTR group at any
time interval.
Sensory Testing
Preoperatively, numbness and paresthesias were
severe in 41 hands in the UCTR group and in 46
hands in the OCTR group (Table 1). Moderate symp-
toms were noted in 9 hands in the former group and
in 7 hands in the latter group. At 104 weeks, 7 hands
in the UCTR group and 6 hands in the OCTR group
 The Journal of Hand Surgery / Vol. 22A No. 5 September 1997 859

still had mild symptoms, while the remaining hands Strength

were symptom-free.
Preoperatively, static two-point discrimination was
more than 15 mm in 10 hands in the UCTR group and
in 14 hands in the OCTR group. It ranged from 11 to
15 mm in 16 hands in the UCTR group and in 15
hands in the OCTR group and from 6 to 10 mm in 20
hands in the former group and in 24 hands in the latter
group. It was 5 mm in 4 hands in the UCTR group. At
104 weeks, it was more than 10 mm in 3 hands in both
groups. It ranged from 6 to 10 mm in 14 hands in the
UCTR group and in 13 hands in the OCTR group. In
the remaining hands, it recovered to a level of 5 mm
or less. The preoperative Semmes-Weinstein mono-
filament testing revealed diminished light touch
(3.22-3.61) in 8 hands in the UCTR group and in 3
hands in the OCTR group, and diminished protective
sensation (3.84-4.31) in 27 hands in the former group
and in 29 hands in the latter group. Eleven hands in
the UCTR group and 21 hands in the OCTR group
had loss of protective sensation (4.56-6.65), and 4
hands in the former group had scores of more than
6.65. The values at 104 weeks revealed diminished
light touch (3.22-3.61) in 24 hands in the UCTR
group and in 29 hands in the OCTR group, dimin-
ished protective sensation (3.84 4.31) in 7 hands in
the former group and in 4 hands in the latter group,
and loss of protective sensation (4.56-6.65) in 3 hands
in the former group. The scores for the remaining
hands returned to the normal range, 1.65-2.83. We
found no significant differences in improvement of the
above sensory data between the two groups at any
time interval.
Weakness of the APB was noted preoperatively in
36 hands in the UCTR group and in 44 hands in the
OCTR group (Table 2). It was graded as 0 preopera-
tively and strength was not recovered even at 104
weeks in 7 hands in the former group and in 4 hands
in the latter group. The remaining hands showed
improvement. We found no significant differences
between the 2 groups at any time interval.
The UCTR group had significantly greater grip
strength at 3 and 6 weeks and greater key-pinch val-
ues at 3, 6, and 13 weeks.
Pain and Tenderness
The UCTR group had significantly less pain and
tenderness, present in the anteromedial aspect of the
wrist, of the scar at 3, 6, and 13 weeks (Table 3). In
this group, 12 hands had pain on grip in the antero-
medial aspect of the distal forearm that was relieved
by 13 weeks. The OCTR group generally had pain
and tenderness over the proximal half of the scar. At
26 weeks, 3 hands in the UCTR group and 5 hands
in the OCTR group had mild pain, and 4 hands in the
former group and 6 hands in the latter group had
mild tenderness.
Electrophysiologic Data
Motor potentials were undetectable preoperatively
in 21 hands in the UCTR group and in 23 hands in
the OCTR group (Table 4). It was still undetectable
at 104 weeks in 8 hands in the former group and in 6

Table 1. Sensory Testing

Scores
Parameters Preoperative A t 3 Wk At 6 Wk A t 13 Wk A t 26 Wk A t 52 Wk A t 104 Wk
Numbness and paresthesias
UCTR 2.8+0.4 2.6+0.6 2.1+0.8 1.3+1.0 0.7+0.9 0.4+0.6 0.1+0.4
OCTR 2.9+0.3 2.8+0.5 2.2+0.8 1.4+0.9 0.8+0.8 0.4+0.6 0.1+0.3
p value 951 .41 .82 .88 .92 .36 .25
Two-point discrimination (mm)
UCTR 12.1+4.9 11.4+5.4 9.8+5.7 8.1+4.5 6.6+3.2 6.1+3.0 5.9+3.0
OCTR 13.2+4.8 12.2+4.7 10.0+4.8 8.1+4.5 7.0+3.8 6.1+3.0 5.7+2.2
p value .28 .78 .84 .93 .68 .92 .63
Semmes-Weinstein
monofilament testing
UCTR 4.44+0.72 4.40+0.75 4.30+0.12 4.02+0.76 3.77+0.55 3.63+0.56 3.52+0.56
OCTR 4.37+0.22 4.35+0.24 4.21 +0.05 4.00+0.42 3.77+0.51 3.46+0.55 3.38+0.46
p value 950 .87 .86 .69 .50 .34 .54

OCTR, open carpal tunnel release; UCTR, ultrasonographic carpal tunnel release.
860 Nakamichi and Tachibana / Ultrasound and Carpal Tunnel Release

Table 2. Strength
Scores
Parameters Preoperative A t 3 Wk A t 6 Wk At 13 Wk A t 26 Wk At 52 Wk A t 104 Wk

Abductor pollicis
brevis power
UCTR 2.5• 2.5• 2.8• 3.1• 3.5• 3.8• 3.9•
OCTR 2.3• 2.3+2.0 2.6• 3.3• 3.8• 4.1• 4.3•
p value .60 .94 .94 .38 .16 .15 .10
Grip strength (kg)
UCTR 24.2• 16.7• 19.1 • 21.6• 23.8• 24.3• 24.2 • 6.0
OCTR 23.6• 13.6• 16.5• 19.4+4.8 21.9• 23.0• 23.4•
p value .57 < .01 .03 .13 .20 .54 .86
Key-pinch
strength (kg)
UCTR 3.84• 2.73• 3.10• 1.16 3.48• 3.73• 3.89• 3.89+1.13
OCTR 3.74• 1 . 1 4 2.20• 2.58• 2.91 • 3.57• 1 . 0 7 3.88• 1.08 3.96• 1.12
p value .64 .02 .04 .03 .79 .68 .44

OCTR, open carpal tunnel release; UCTR, ultrasonographic carpal tunnel release.

hands in the latter group. The remaining hands Anomalies

showed improvement.
Sensory potentials were undetectable preopera-
tively in 29 hands in the UCTR group and in 27
hands in the OCTR group. At 104 weeks, it was still
undetectable in 1 hand in the UCTR group. In this
hand, motor potentials were also undetectable
throughout the course, but there was relief of sensory
symptoms. The remaining hands showed improve-
ment. There were no significant differences between
the 2 groups at any time interval.
Scar
At 104 weeks, the scar was almost invisible in 25
hands and visible but flat in 21 hands in the UCTR
group. In the O C T R group, it was visible but fiat in
40 hands, contracted in 6 hands, and hypertrophic in
1 hand.
No hands had neurovascular anomalies.
Time and Cost
The mean operating time was 54 minutes in the
U C T R group and 48 minutes in the OCTR group.
The mean cost was $513 for the UCTR and $487
for the OCTR. The difference was due to the costs
for intraoperative use of the ultrasound system ($20);
for sterilization of the cutting device, a plastic bag,
and small self-retaining retractors ($1); and for ster-
ile jelly ($5).
Complications
There were no nerve, vascular, or tendon injuries
using either technique. In the UCTR group, no pro-
cedures were converted to the OCTR. There were no
hands with persistent or recurrent symptoms.

Table 3. Pain and Scar Tenderness

Scores
Parameters Preoperative At 3 Wk A t 6 Wk At 13 Wk A t 26 Wk A t 52 Wk At 104 Wk

Pain
UCTR 0.0• 2.1• 1.4• 0.5• 0.1• 0.0• 0.0•
OCTR 0.0• 2.7• 2.2• 1.2• 0.1• 0.0• 0.0•
p value <.01 <.01 <.01 .52
Scar tenderness
UCTR 0.0• 2.4• 1.7• 0.7• 0.1• 0.0• 0.0•
OCTR 0.0• 2.9• 2.6• 1.4• 0.1• 0.0• 0.0•
p value < .01 < .01 < .01 .57

OCTR, open carpal tunnel release; UCTR, ultrasonographic carpal tunnel release.
 The Journal of Hand Surgery / Vol. 22A No. 5 September 1997
Table 4. Electrophysiologic Data*
Parameters Preoperative
Median distal motor latency (ms)
UCTR 6.7 + 1.7
OCTR 6.5 + 1.6
p value .66
Median sensory conduction velocity (m/s)
UCTR 27.5+7.2
OCTR 26.0 + 6.1
p value .47
*The hands in which preoperative potentials were undetectable were not included in the statistical analysis.
OCTR, open carpal tunnel release; UCTR, ultrasonographic carpal tunnel release.
Discussion
The UCTR was primarily designed to protect
all vital structures at risk during carpal tunnel
release. 13a4 Release along the safe line protects the
median nerve as well as the ulnar neurovascular bun-
dle. It is crucial to avoid the ulnar neurovascular
bundle because it often exists radial to the hook of
the hamate. 15 The SPA and the digital nerves are pro-
tected by an imaging-based skin incision and distal
release under direct vision. (The SPA is usually dis-
tal to the skin incision and thus out of harm's way.)
The flexor tendons are mildly retracted and protected
by the cutting device. Complete release is confirmed
as described in the surgical technique. In addition, a
thick cannula is not passed through the tunnel, which
eliminates pain and transient nerve compression
associated with cannula insertion.
We believe the cutting device is safer than a knife.
Even if the device tip migrates into the carpal tunnel,
it can be safely withdrawn as long as the jaw is not
closed. The thickness of the TCL and the forearm fas-
cia 16,17is important to consider in determining the size
of a basket punch (Fig. 2B). Regarding the shape, the
upper jaw should have the same width from the tip
down to its base (Fig. 2C). A jaw with a wide base
will be caught, when fully opened, in the gap of the
divided TCL and will not release smoothly. The outer
tube should fit the punch (Fig. 2B, C).
The palmar incision has 2 aims. The first is to rule
out the median nerve anomalies. On the basis of the
Lanz classification, ~8 Groups II and IV are not at risk
because of release along the safe line. If, however, the
motor branch is found to be bent around the distal
edge of the T C L (a subgroup of group I), the branch
should be protected. Another anomaly that should be
ruled out is high division of the median nerve (group
III). The second aim is to confirm the complete divi-
861
Scores
At 13 Wk A t 26 Wk At 52 Wk A t 104 Wk
4.9 + 1.0 4.3 + 0.8 4.0 + 0.6 4.0 + 0.6
4.8 + 0.8 4.3 + 0.7 4.1 + 0.6 4.0 + 0.6
.77 .80 .46 .96
38.1 +4.0 41.9+3.9 43.7+4.3 43.9+4.3
35.1 + 6.6 38.8 + 4.9 40.8 + 4.7 41.7 + 4.5
.35 .22 .30 .56
sion of the distal TCL. It is important to identify
these structures under direct vision because the cur-
rent ultrasonographic resolution is not good enough
to visualize them. (We have had no experience with
high division of the nerve and thus are not sure
whether this anomaly is detectable by ultrasonogra-
phy. Therefore, we rule this out under direct vision as
well as by preoperative imaging.)
In this comparison study, it was shown that the
U C T R provided neurologic improvement equal to
that from the OCTR regarding numbness and pares-
thesias, sensibility, muscle power of the APB, and
electrophysiologic data. Better outcomes were noted
in the early postoperative period regarding grip and
key-pinch strength, pain, and scar tenderness. At the
final evaluation, the palm incision was small and
often had become almost invisible, which resulted in
a cosmetically appealing scar.
Different degrees of pain were noted between the
2 groups. The UCTR group generally had antero-
medial wrist pain, while the OCTR group had pain
around the proximal scar. In the former group, 12
hands had pain on grip in the anteromedial aspect of
the distal forearm. This difference may be explained
by the release line. In the OCTR, it was approxi-
mately in the midpalmar aspect of the wrist, while in
the UCTR, the safe line ran'ulnar to the midline.
Although there were no complications, this series
was not large enough to determine the possible types
and rate of complications in the U C T R compared
with the OCTR. This early experience with no com-
plications, however, has made us believe in the
safety of the UCTR, based on the above considera-
tions. To our knowledge, there are no other tech-
niques in which an operating instrument is localized
and directed based on imaging of the key structures.
We also took care to reduce costs. The additional
costs in the UCTR included intraoperative use of an
862 Nakamichi and Tachibana / Ultrasound and Carpal Tunnel Release
ultrasound system, sterile jelly, and sterilization of
the cutting device, a bag for the scanner, and small
retractors. We used no expensive disposable instru-
ments, however. We now use saline instead of sterile
jelly, which saves $5.
Since exploration of the carpal tunnel is limited in
the UCTR, we always rule out local pathology and
anomalies by wrist imaging preoperatively. We use
plain radiographs of the wrist and carpal tunnel, look-
ing for bony and calcific lesions, and ultrasonograms,
looking for soft lesions. Patients should be screened
for malunited fractures, osteophytes secondary to
osteoarthritis, calcific deposition, space-occupying
lesions, anomalous muscles, and synovial swelling.
Space-occupying lesions should be ruled out espe-
cially when a patient has unilateral involvement. 19,20
Synovial swelling often indicates inflammation, as in
rheumatoid arthritis or infection51 Generous expo-
sure should be considered for these conditions.
The detection of a pulsating median artery by
ultrasonography may indicate high division of the
median nerve. Under these circumstances, the
OCTR is indicated. Use of UCTR with these
anomalies will result in the close proximity of the
cutting device to these and other neurovascular
structures. By contrast, the motor branch anomalies
can be protected in the UCTR because of the distal
release under direct vision.
During preoperative imaging, we also evaluate the
distance between the median nerve and the ulnar
artery. We have noted considerable variation of this
distance. The significance of this is that the proxim-
ity of the cutting device to the median nerve and
ulnar artery depends on this distance. A distance of 3
mm or less makes us less confident of the safety of
UCTR and favor OCTR instead.
In summary, the UCTR is a imaging-based lim-
ited-dissection technique that can be performed
without an assistant and expensive disposable instru-
ments. In this comparison study, it provided the same
neurologic recovery as did OCTR. Better early out-
comes were noted regarding grip and key-pinch
strength, pain, and scar tenderness. At the final eval-
uation, the palm scar was found to be more aesthetic.
There were no complications.
References
1. Okutsu I, Ninomiya S, Takatori Y, Ugawa Y. Endoscopic
management of carpal tunnel syndrome. Arthroscopy
1989;5:11-18.
2. Chow JCY. Endoscopic release of the carpal ligament: a
new technique for carpal tunnel syndrome. Arthroscopy
1989;5:19-24.
3, Agee JM, McCarroll HR Jr, Tortosa RD, Berry DA, Szabo
RM, Peimer CA. Endoscopic release of the carpal tunnel: a
randomized prospective multicenter study. J Hand Surg
1992; 17A:987-995.
4. Mirza MA, King ET Jr, Tanveer S. Palmar uniportal extra-
bursal endoscopic carpal tunnel release. Arthroscopy 1995;
11:82-90.
5. Tsai TM, Tsuruta T, Syed SA, Kimura H. A new technique
for endoscopic carpal tunnel decompression. J Hand Surg
1995;20B:465-469.
6. Newmeyer WL. Thoughts on the technique of carpal tun-
nel release [editorial]. J Hand Surg 1992; 17A:985-986.
7. Brown RA, Gelberman RH, Seiler JG III et al. Carpal tun-
nel release. A prospective, randomized assessment of open
and endoscopic methods. J Bone Joint Snrg 1993;75A:
1265-1275.
8. Evans DM. Endoscopic carpal tunnel release--the hand
doctor's dilemma [editorial]. J Hand Surg 1994; 19B :3-4.
9. Louis DS. Commentary: progress?--at what price? J Hand
Surg 1995;20A: 172.
10. Seiler JG Ili, Barnes K, Gelberman RH, Chalidapong P.
Endoscopic carpal tunnel release: an anatomic study of the
two-incision method in human cadavers. J Hand Surg
1992;17A:996-1002.
11. Cobb TK, Carmichael SW, Cooney WP. The ulnar neu-
rovascular bundle at the wrist: a technical note on endo-
scopic carpal tunnel release. J Hand Snrg 1994; 19B:24-26.
12. Tachibana S, Nagano A, Okinaga S. The role of electro-
physiological study in carpal tunnel syndrome [in
Japanese; abstract in English]. J Jpn Soc Surg Hand 1992;
8:873-880.
13. Bozentka D J, Osterman AL. Complications of endoscopic
carpal tunnel release. Hand Clin 1995;11:91-95.
14. Urhaniak JR, Desai SS. Complications of nonoperative
and operative treatment of carpal tunnel syndrome. Hand
Clin 1996;12:325-335.
15. Cobb TK, Carmichael SW, Cooney WP. Guyon's canal
revisited: an anatomic study of the carpal ulnar neurovas-
cular space. J Hand Surg 1996;21A:861-869.
16. Schmidt H-M, Lanz U. Anatomy of the median nerve in
the carpal tunnel. In: Gelberman RH, ed. Operative nerve
repair and reconstruction. Philadelphia: JB Lippincott,
1991: 889-898.
17. Cobb TK, Dalley BK, Posteraro RH, Lewis RC. Anatomy
of the flexor retinaculum. J Hand Surg 1993;18A:91-99.
18. Lanz U. Anatomical variations of the median nerve in the
carpal tunnel. J Hand Surg 1977;2:44-53.
19. Nakamichi K, Tachibana S. Unilateral carpal tunnel syn-
drome and space-occupying lesions. J Hand Surg 1993;
18B :748-749.
20. Erdmann MWH. Endoscopic carpal tunnel decompression.
J Hand Surg 1994;19B:5-13.
21. Nakamichi K, Tachibana S. The use of ultrasonography in
detection of synovitis in carpal tunnel syndrome. J Hand
Surg 1993;18B:176-179.