The triangular fibrocartilage complex (TFCC) is a well-known structure that acts as
a stabilizer of the distal radioulnar joint (DRUJ) and a shock absorber of the ulnocarpal
joint. Recent anatomical studies have proven that the distal radioulnar ligament consists
of superficial and deep bundles, which attach on the fovea and provides DRUJ stability.[1]
[2]
[3] Haugstved et al demonstrated that the deep ligaments provide greater stability of
the DRUJ than the superficial ligaments in a biomechanical study.[4] This indicates that disruption of the TFCC at the fovea insertion could lead to
DRUJ instability, resulting in disability in daily living. In this situation, a foveal
tear might be repaired. Several procedures have been recommended to repair a foveal
tear, and they can be divided into open and arthroscopic repairs. However, there have
been few reports that have compared the effectiveness of the open repair and the arthroscopic
repair.[5]
[6]
[7] The purpose of this study was to compare open with arthroscopic repair of foveal
tears of the TFCC.
Materials and Methods
This was a retrospective study of a group of patients who had complained disability
of the wrist and were found to have a foveal TFCC tear at the time of wrist arthroscopy.
Informed consent was obtained from all patients, and the study protocol adhered to
the ethical guidelines of the 1975 Declaration of Helsinki. This clinical investigation
was conducted with the approval of our institutional review board. The preoperative
findings were recorded following a retrospective record review.
Our indication for repair of a TFCC foveal tear was symptomatic DRUJ instability that
had not responded to nonsurgical treatment for over 3 months. Therefore, in this study,
the patients with a foveal tear associated with a fresh distal radius fracture (DRF)
were excluded. The patients who had concomitant scapholunate ligament (SL) injury
were also excluded. Furthermore, the patients who had foveal tear with an ulnar positive
variance of more than +1 mm and were initially treated with ulnar shortening osteotomy
were also excluded.
Since December 2004 to January 2014, 42 wrists of 42 patients with a TFCC foveal tear
were treated surgically. These included 5 patients with acute DRF, 1 patient associated
with SL injury, and 7 patients treated with ulnar shortening osteotomy. Thus, 29 wrists
of 29 patients with a TFCC foveal tear treated surgically were investigated. There
were 13 men and 16 women, 14 right and 15 left wrists, and 16 dominant and 13 nondominant
hands. The mean age of the patients was 30 (range, 14–72) years. Sixteen patients
suffered the injury during sports activities, and 12 patients suffered the injury
during working, by a fall, or twisting the wrist. One patient could not remember the
clear history of wrist trauma. Five patients had a previous history of DRF that had
healed uneventfully with normal alignment by cast immobilization.
The first eight patients between December 2004 and October 2008 underwent open repair
(group O). Twenty-one patients between November 2008 and January 2014 were repaired
arthroscopically (group A). The mean duration of symptoms before surgery was 7.1 months,
ranging from 3 to 20 months. The follow-up period ranged from 24 to 70 months, with
an average of 34.4 months. The age and duration of symptoms before surgery in each
group are represented in [Table 1].
Table 1
Patients' demographics and preoperative data in each group
|
Group O
(n = 8)
|
Maximum–Minimum
|
Group A
(n = 21)
|
Maximum–Minimum
|
|
Mean
|
Mean
|
|
Age
|
22
|
14–40
|
34
|
14–72
|
|
Period from onset to surgery (mo)
|
11
|
3–20
|
8.5
|
3–20
|
|
Ulnar variance (mm)
|
–0.6
|
–2.5 to +0.3
|
–0.7
|
–3.0 to +0.3
|
|
Atzei's classification
|
Class 2: 4
Class 3: 4
|
|
Class 2: 6
Class 3: 15
|
|
|
NRS
|
10
|
Only 10
|
10
|
Only 10
|
|
Extension (°)
|
71.7
|
64–80
|
72.6
|
54–86
|
|
Flexion (°)
|
61.0
|
48–84
|
59.6
|
45–81
|
|
Pronation (°)
|
83.3
|
70–90
|
81.3
|
60–90
|
|
Supination (°)
|
89.1
|
75–90
|
86.9
|
45–90
|
|
Grip strength (%)
|
81.6
|
38–91
|
80.2
|
38–100
|
Abbreviation: NRS, numerical rating scale.
Clinical and Radiological Evaluation
All patients complained of ulnar-sided wrist pain with wrist extension and forearm
rotation. All patients also felt ulnar head instability during forceful forearm rotation.
Some specific physical examinations for the foveal tear were examined. A positive
fovea sign represented ulnar-sided wrist tenderness in the ulnar fovea.[8] The ulnar head ballottement test was examined by the piano key sign with neutral
forearm rotation and 90° flexion of the elbow, holding the radius and the carpal bones.
Obvious palmar and dorsal ulnar head instability compared with the contralateral wrist
was diagnosed as positive in this test. All patients underwent a radiographic evaluation
including neutral rotation posteroanterior and lateral X-rays, and 1.5T coronal plane
magnetic resonance imaging (MRI). On the X-rays, none represented ulnar styloid nonunion
or DRUJ arthrosis. The mean ulnar variance was –0.7 mm (–3.0 to + 0.3 mm). Three patients
showed a distended DRUJ joint over 1 mm compared with the contralateral X-ray. One
patient showed a small fragment just distal to the fovea. On MRI, 18 patients showed
lack of continuity of TFCC at the fovea, whereas 11 patients seemed to demonstrate
continuity at the fovea on MRI. These were graded by the agreement of two hand surgeons
including the first author (Y.A.). All patients underwent an initial trial of conservative
treatment, such as cast immobilization, splinting, and administration of nonsteroidal
anti-inflammatory drugs, all of which failed.
All patients were assessed with wrist arthroscopy by single surgeon (Y.A.) including
radiocarpal (RC) and DRUJ arthroscopy, and they were confirmed to have a foveal tear.
According to Atzei's classification,[9] there were 4 Class 2 (repairable complete tear) and 4 Class 3 (repairable proximal
tear) in group O, and there were 6 Class 2 and 15 Class 3 in group A. Eleven patients
who demonstrated continuity on MRI had fragile scar tissue at the fovea. To ensure
interobserver reliability of the arthroscopic examinations, the foveal tear was confirmed
by the first author and a scrub doctor intraoperatively, and reconfirmed by another
surgeon who did not participate in the surgery through photographs postoperatively.
These surgeons were qualified by our National Orthopaedic Association, and all surgeons
agreed with the diagnosis of the foveal tear.
Preoperative Data
A fovea sign and ulnar head ballottement test were positive in all patients. The mean
ulnar variance was –0.6 mm (–2.5 mm to +0.3 mm) in group O, –0.7 mm (–3.0 mm to +0.3
mm) in group A. Preoperative pain was scored as 10 in all patients with numerical
rating scale (NRS). The mean extension of the wrist was 71.7° (range: 64°–80°), and
mean flexion was 61.0° (range: 48°–84°) in group O; the mean extension was 72.6° (range:
54°–86°), and the mean flexion was 59.6° (range: 45°–81°) in group A. The mean pronation
of the forearm was 83.3° (range: 70°–90°), and the mean supination was 89.1° (range:
75°–90°) in group O; the mean pronation was 81.3° (range: 60°–90°), the mean supination
was 86.9° (range: 45°–90°) in group A. The mean grip strength was 81.6% (range: 38–91%)
in group O and 80.2% (range: 38–100%) in group A ([Table 1]).
Postoperative Evaluation
The final evaluation included pain, measurements of wrist and forearm motion, grip
strength, ulnar head instability, the Disabilities of the Arm, Shoulder, and Hand
(DASH) questionnaire, and the Mayo modified wrist score (MMWS). Postoperative pain
was evaluated with a NRS, and preoperative pain was scored as 10. Wrist flexion–extension
was assessed with a goniometer (TTM-KO, Sakai Medical Co., Tokyo, Japan). Forearm
supination and pronation were assessed with the elbow flexed 90° at the patient's
side. Grip strength was measured with a calibrated dynamometer (TKK 5401, Takei Machinery
Co., Niigata, Japan) and reported as the ratio to the contralateral side. Ulnar head
instability was examined with the ulnar head ballottement test and assessed with Nakamura
et al's DRUJ instability score (0: no end point in any direction; 1: at least one
endpoint either in dorsal or palmar; 2: looser than the intact contralateral side;
4: stable DRUJ).[7]
Statistical Analysis
The operating time, NRS score, range of motion (ROM), grasping power, the ulnar head
instability score, and the DASH score for both procedures were compared using the
t-test. The MMWS were compared with the chi-squared test. A p-value of < 0.05 was regarded as significant.
Surgical Technique
The patient is placed in a supine position under general anesthesia with the affected
arm on a hand table and a tourniquet placed on the proximal arm and inflated. The
wrist is suspended in vertical traction and examined by arthroscopy. Generally, two
dorsal arthroscopic portals are used: a 3–4 portal and a 4–5 portal to examine the
RC joint. A 1.9- or 2.3-mm arthroscope with a 30° angle is introduced through the
3–4 portal, and a probe, a shaver, and a radiofrequency device are interchangeably
inserted through the 4–5 portal. The 6U portal is used as an outflow portal with the
wet technique. If a foveal tear is present, TFCC tension becomes loose; therefore,
loss of the trampoline effect is recognized. A peripheral tear (ulnar styloid tear[10]) of the TFCC should be also investigated through a hook test. Then, the TFCC foveal
insertion is evaluated through the DRUJ portal. DRUJ arthroscopy can directly visualize
a foveal tear ([Fig. 1A, B]). The TFCC is thoroughly inspected through these portals.
Fig. 1 Distal radioulnar joint (DRUJ) arthroscopy for the left wrist showed complete tear
of the fovea (A). DRUJ arthroscopy for the right wrist represented the elongated ligament without
tension (B).
Both open and arthroscopic repair are subsequently performed after diagnostic arthroscopy
with keeping the forearm suspended by finger traps. In the open procedure, we recognized
mild swelling at the surgical field after diagnostic wet arthroscopy; however, it
did not affect the surgical approach so much. The open repair is started with about
a 3-cm straight skin incision on the ulnar side of the ulnar neck. The tendon sheath
of the extensor carpi ulnaris (ECU) tendon is incised, and the ECU tendon is retracted
palmarly or dorsally and freely mobilized. The ulnar wrist capsule is cut longitudinally,
exposing the TFCC disc and the fovea under loupe magnification. A foveal lesion of
the TFCC is recognized as a torn ligament or continuous scar tissue which is loose.
The scar tissue should be debrided the minimum to confirm precise point of the attachment
of the TFCC.
Two osseous tunnels are made by inserting two parallel 1.5-mm Kirschner wires (K-wire)
from the ulnar neck to the foveal region. The point of K-wire insertion is the ulnar
surface of the ulnar neck, 1.5 to 2.0 cm proximal from the distal end of the ulnar
styloid. The place of bone tunnels is confirmed to be appropriate with image intensifier.
The two sutures, 3–0 PDS (Ethicon, Somerville, NJ) and 3–0 Vicryl (Ethicon), are threaded
horizontally at the ulnar peripheral lesion of the TFCC and pass through the two bone
tunnels to the ulnar surface of the ulnar neck ([Fig. 2]). After loosening the traction of the forearm, and applying compression between
the radius and ulna by the assistant in neutral forearm rotation, the threads were
tied up with manual maximum tension to directly attach the TFCC to the fovea.
Fig. 2 The open approach. The suture passes through the two bone tunnels with open approach.
The arthroscopic repair is performed through a similar but shorter skin incision,
and the ECU tendon is freely mobilized and expose the ulnar capsule. Two bone tunnels
are created in a similar fashion with the direct repair. Two K-wires should be protruded
within half to one-third ulnar side of TFCC, confirmed with RC arthroscopy. The location
of the bone tunnels is also confirmed with an image intensifier. The two 21-gauge
needles are inserted with a lasso loop of a 3–0 nylon suture through the two bone
tunnels and the torn ulnar edge of the TFCC in the RC joint. The two looped sutures
are retrieved through the 4–5 portal using blunt mosquito forceps ([Fig. 3A]), and then the two sutures, 3–0 PDS and 3–0 Vicryl, are threaded through the loop
suture and introduced into the RC joint. Traction on the looped sutures then pulls
the PDS and the Vicryl sutures through the TFCC and out through the two bone tunnels.
The TFCC is tightly attached to the fovea with tying up the threads with manual maximum
tension ([Fig. 3B]).
Fig. 3 The arthroscopic approach. The two loop sutures passed through the bone tunnels are
retrieved through the 4–5 portal (A), the triangular fibrocartilage complex (TFCC) is tightly attached to the fovea with
tying up the thread (B).
Postoperative Management
After repair, the postoperative protocol was consistent with both procedures. The
wrist was fixed with a long-arm cast for 2 weeks with 90° of elbow flexion and neutral
forearm rotation. A short arm cast was applied for an additional 2 weeks. Gentle ROM
exercise including rotation of the forearm was started at 4 weeks after surgery, and
grip strengthening was started at 2 months. The patients were instructed that they
could return to preoperative sports or work 3 to 6 months after surgery.
Results
All outcomes for both procedures are shown in [Table 2]. The average operation time was 89.2 minutes (75–110 minutes) in group O and 55.3
minutes (30–80 minutes) in group A, significantly shorter than in group O (p = 0.002). There was no patient who complained of wrist pain at the final follow-up
in group O; the average NRS was 0. In group A, three patients felt mild ulnar-sided
wrist pain during heavy activities. The average NRS was 0.2 (0–2). The mean extension
of the wrist was 66.6° (range: 60°–73°), and the mean flexion was 63.0° (range: 50°–70°)
in group O; the mean extension was 72.9° (range: 60°–85°), and the mean flexion was
66.3° (range: 50°–80°) in group A. The mean pronation of the forearm was 83.4° (range:
80°–90°), and the mean supination was 90.0° (range: 85°–95°) in group O; the mean
pronation was 83.3° (range: 75°–90°), and the mean supination was 89.4° (range: 80°–90°)
in group A. The mean grip strength was 96.9% (range: 92–100%) in group O and 97.6%
(range: 74–115%) in group A. DRUJ instability of all patients was evaluated as 4 in
group O; 18 patients were assessed as 4, and 3 patients were evaluated as 2 in group
A, the average was 3.7. The mean DASH at final follow-up was 7.8 (0–15.3) in group
O and 5.7 (0–14.7) in group A. The final results according to the MMWS were all excellent
in group O, with 18 excellent and 3 good in group A. There were no significant differences
between the groups in the t-test (p > 0.05) and the chi-square test (p > 0.05) except for the operating time. There were no complications, and no patients
needed reoperation in both groups.
Table 2
Postoperative data in each group
|
Group O
(n = 8)
|
Maximum–Minimum
|
Group A
(n = 21)
|
Maximum–Minimum
|
Statistical
results
|
|
Mean
|
Mean
|
|
Operating time (min)
|
89.2
|
75–110
|
55.3
|
30–80
|
p = 0.002
|
|
NRS
|
0
|
Only 0
|
0.2
|
0–2
|
p > 0.05
|
|
Extension (°)
|
66.6
|
60–73
|
72.9
|
60–85
|
p > 0.05
|
|
Flexion (°)
|
63.0
|
50–70
|
66.3
|
50–80
|
p > 0.05
|
|
Pronation (°)
|
83.4
|
80–90
|
83.3
|
75–90
|
p > 0.05
|
|
Supination (°)
|
90.0
|
85–95
|
89.4
|
80–90
|
p > 0.05
|
|
Grip strength (%)
|
96.9
|
92–100
|
97.6
|
74–115
|
p > 0.05
|
|
DRUJ instability (score)
|
4
|
Only 4
|
3.7
|
2–4
|
p > 0.05
|
|
DASH score
|
7.8
|
0–15.3
|
5.7
|
0–14.7
|
p > 0.05
|
|
MMWS
|
E: 8
|
|
E: 18, G: 3
|
|
p > 0.5
|
Abbreviations: DASH score, Disabilities of the Arm, Shoulder and Hand score; DRUJ,
distal radioulnar joint; E, excellent; G, good; MMWS, Mayo modified wrist score; NRS,
numerical rating scale.
Note: There were no significant differences between the two groups except for the
operating time.
Discussion
In this study, we clarify that transosseous repair for TFCC foveal tear through both
open and arthroscopic approach could provide feasible results. These results suggested
that reattachment of the TFCC to the precise location, which was confirmed with macroscopy,
arthroscopy, and image intensifier, is the critical point to reconstruct the TFCC
foveal tear using the open or arthroscopic procedures. This investigation revealed
that the ulnar attachment of the TFCC is divided into two sections: attachment to
the ulnar styloid and the fovea. The tear on the ulnar side of TFCC inevitably occurs
at these attachments. Nakamura and Makita described the detailed three-dimensional
structure of the TFCC. During forearm rotation, the dorsal and volar portions of the
distal radioulnar ligament show a nearly isometric length pattern at the fovea.[3] Atzei has analyzed Palmer's 1B tear[11] in detail and classified it into a distal component, a proximal component, or both,
suggesting that fundamentally these tears can be repaired.[9] Abe et al reported various patterns of TFCC tears and distinguished a foveal tear
from an ulnar styloid tear.[10]
The foveal tear should be diagnosed precisely. The patient complains of a slack sensation
during forearm rotation and loses strong grasping. A positive fovea sign is suggestive
but not specific, because this sign may suggest not only a foveal tear, but a tear
from the ulnar styloid and the inflammation at the surrounding structure. The ulnar
head ballottement test is a reliable physical test. It must be evaluated bilaterally;
the instability is more evident if it is examined under general anesthesia. MRI can
delineate a foveal detachment clearly. A gradient echo sequence T2-weighted image
provides a high-delineation image of the TFCC structure. However, evaluation of MRI
findings is sometimes confusing when the ligament is continuous with scar tissue like
in this study. DRUJ arthroscopy is a definitive procedure to diagnose a foveal tear.
DRUJ arthroscopy is still a technically demanding procedure because the joint space
is very narrow. However, when a foveal tear exists, the foveal region can easily be
visualized through DRUJ arthroscopy, because the DRUJ is loose. The quality of the
remnant fibers should be evaluated. If the remnant fibers are severely disrupted,
primary repair is not indicated.
Several procedures for open and arthroscopic repair have been described. Moritomo
described the open repair through a volar approach with the concept that foveal detachment
would initially occur from the volar element.[12] Atzei[9] and Kim et al[13] described a hybrid approach in which they explored the foveal lesion arthroscopically
and used an open technique to reattach the foveal insertion using a bone anchor. Iwasaki
et al described arthroscopic reattachment by creating an osseous tunnel, 2.9 mm in
diameter, from the ulnar neck to the foveal surface, and their 2- to 4-year follow-up
results were good.[14] Nakamura et al reported a three-dimensional mattress suture technique that can create
an anatomical reconstruction using an open ulnar approach.[15] Nakamura et al also described arthroscopic transosseous repair using their original
targeting device. Their comparative study between the open and arthroscopic approaches
showed that both procedures could obtain excellent clinical results.[7] Shinohara et al performed arthroscopically assisted foveal repair primarily in accordance
with the method of Nakamura, and they showed satisfactory outcomes with a mean follow-up
of 30 months.[16]
There have been few reports of comparative studies between open and arthroscopic approaches.
Anderson et al stated that there was no significant difference in clinical outcome
after open versus arthroscopic repair.[5] However, in this report, TFCC tear was classified as a 1B tear with Palmer's classification,
and it was unclear whether the tear was a foveal tear or avulsion from the ulnar styloid.
Arthroscopic repair was not described precisely. What procedure did the outside-in
repair they mentioned mean; the capsular repair for avulsion from the ulnar styloid
or transosseous repair for foveal tear? Luchetti et al reported successful outcomes
with open and arthroscopically-assisted repairs.[6] They confirmed the foveal detachment through DRUJ arthroscopy and repaired it using
a suture anchor. They showed no significant postoperative differences between the
two groups except for the DASH, which was significantly better in the arthroscopic
group.
Our procedure that TFCC was reattached to the fovea by two threads pull-out technique
through two bone tunnels in both approaches was totally different from previous comparative
reports except for Nakamura et al's report. They obtained excellent clinical results
in both procedures.[7] However, they pointed out that the clinical results were unsatisfactory when arthroscopic
repair was performed at an average of 19 months after injury. They also stated that
the patients with +2 mm positive ulnar variance resulted in unsatisfactory results
when arthroscopic repair was performed. In this study, we obtained satisfactory results
in all cases. This is because the patients with an ulnar positive variance of more
than +1 mm were excluded, and the mean duration of symptoms before surgery was 7.1
months. Shinohara et al stated that a patient with a traumatic foveal tear without
ulnar abutment may be a good candidate for arthroscopic foveal repair even 7 months
after the injury, and the present clinical results may have proven that. In addition,
foveal repair was not indicated if DRUJ arthroscopy showed the unrepairable remnant
of the ligament. DRUJ arthroscopy is essential to determine the indication for foveal
repair.
The operating time was significantly shorter in the arthroscopic approach than in
the open approach. This is because, in the open approach, we need to find the precise
location of the fovea under loupe magnification. Reverse L-incision to the capsule
may be suitable to expose the TFCC disc and the foveal lesion widely. However, we
set the longitudinal incision to capsule not to open the RC and DRUJ joint widely.
This procedure may be one of the reasons that open procedure took longer time than
the arthroscopic approach because visual field was very narrow. In any case, threading
the suture horizontally at the critical point of the TFCC through the open approach
was a tough procedure.
Limitations
This study has several limitations. The number of cases was small, especially the
number treated with the open approach. The indication for each procedure was determined
by the time period of surgery; first, 8 patients were repaired with open approach
and the following 21 patients were repaired arthroscopically. All surgeries were consecutive,
and the open repairs were all conducted sequentially before the author converted to
an arthroscopic approach. This is a potential significant source of bias of surgical
skill. All surgery was performed by a single surgeon, and these cases were subject
to the learning curve. The surgeon's ability to perform both the open and arthroscopic
approach, especially the DRUJ arthroscopy, and suture naturally is fundamentally linked
to the learning curve. That may affect the difference of the operating time.
