Keywords
arthroscopy - perilunate - dislocation - fracture-dislocation - percutaneous
Perilunate injuries are highly unstable carpal dissociations characterized by a complete
loss of contact between the lunate and surrounding carpal bones. They can be pure
perilunate dislocations (PLDs) or perilunate fracture-dislocations (PLFDs) associated
with carpal fractures around the lunate.[1]
[2]
[3]
[4] The key to successful treatment of perilunate injuries is early surgical intervention
to restore normal alignment of the carpal bones and stability.[3]
[5] Generally accepted treatment has consisted of open primary repair or reconstruction
of the ligaments with open reduction and internal fixation of the fractures.[1]
[3]
[6]
[7]
[8]
[9]
[10]
[11]
[12] However, it is clear that open surgery introduces additional surgical trauma to
the important capsular and ligamentous structures, which may be associated with a
high rate of complications, such as the development of joint stiffness due to capsular
fibrosis or failure of proper bone healing because of damage to the blood supply.[4]
[7] Furthermore, posttraumatic arthritis, which may modify the functional outcome, is
major concern following the open surgery, with an incidence of 38–86%.[3]
[4]
[13]
Arthroscopic technique has the theoretical advantage of facilitating the healing of
fractures and torn ligaments because it can minimize capsular and adjacent soft tissue
injury and provide preservation of an already tenuous blood supply.[14]
[15]
[16]
[17] In fact, several pioneers suggested that an arthroscopic reduction and percutaneous
fixation, as an alternative to an open approach, can effectively treat acute perilunate
injuries.[12]
[18] We propose that arthroscopic treatment results in successful restoration of carpal
alignment and good functional outcomes, and this technique can be a reliable option
for the treatment of these challenging injuries.
Patients and Methods
A retrospective review of 20 patients with a PLD or PLFD who were treated with arthroscopic
reduction and percutaneous fixation has been performed and reported.[19] According to the sagittal displacement classification system described by Herzberg,[3] there were seven stage I (the lunate remained in place under the radius), ten stage
IIa (the lunate was palmarly dislocated but rotated by less than 90°), and three stage
IIb (the lunate was palmarly dislocated and rotated more than 90°). Our series included
five PLDs, 12 transscaphoid PLFDs, and three transscaphoid, transtriquetral PLFDs.
There were 19 male patients and one female patient with a mean age of 37.3 years (range
19–57 years). The dominant wrist was involved in eight patients. Fifteen patients
sustained an injury during falls from a height onto the outstretched hand. Three injuries
were caused by motorcycle accidents, and the remaining two injuries occurred during
snowboarding. The median time from the injury to operation was 3.9 days (range 1–20
days). Sixteen patients underwent the surgery within the first week following the
accident. Six patients had concomitant extremity injuries: one olecranon fracture
in the same side, two distal radius fractures with or without elbow fracture-dislocation
in the contralateral side, and three lower extremity fractures including an intertrochanteric
fracture, a patellar fracture, and a calcaneal fracture.
All patients were assessed postoperatively with Disabilities of the Arm, Shoulder,
and Hand (DASH) questionnaire[20] and Patient Related Wrist Evaluation (PRWE) scores[21] as well as clinical examination, such as range of motion and grip strength. Overall
outcomes were assessed according to the Modified Mayo Wrist Score (MMWS) system, of
which rating of this system was divided by four categories as follows: excellent,
90–100 points; good, 80–89 points; fair, 65–79, and poor, less than 65 points.[1]
Radiographs were evaluated by measuring the scapholunate gap, the lunotriquetral gap,
the scapholunate angle, and the carpal height ratio at the time of surgery and at
the last follow-up examination.[22]
[23] Midcarpal and radiocarpal arthritis were rated based on plain radiographs according
to the system of Knirk and Jupiter.[24]
Surgical Technique
Before arthroscopy was performed, a closed reduction had been attempted. When the
gross reduction was not amenable to initial closed reduction, this maneuver was not
repeated in order to prevent further cartilage or soft tissue damage. Fifteen of 20
patients (75%) achieved a gross reduction of the capitolunate joint by manipulation.
The other five patients failed the gross reduction.
Wrist arthroscopy was performed under brachial plexus block or general anesthesia.
The hand was suspended using a Traction Tower (Linvatec, Largo, FL, USA) with 10 to
15 lb of traction. The forearm was wrapped with a compressive elastic bandage, and
continuous saline irrigation was instilled by gravity infusion from an elevated bag
to minimize fluid extravasation. After clot and debris were removed through the 3–4
and 4–5 portals, the palmar capsular ligaments, scapholunate and lunotriquetral ligaments,
and triangular fibrocartilage complex (TFCC) were carefully evaluated. Arthroscopic
findings of five patients who had failed to achieve a gross reduction by a closed
manipulation revealed that palmar capsular ligaments were torn and interposed between
the lunate and capitate ([Figs. 1a–g]). In those cases, the lunate, which had volarly dislocated with or without proximal
fragment of the scaphoid, was effectively reduced by pulling it or them dorsally with
use of a probe (Video 1). Extensive injury to the radioscaphocapitate ligament with
relatively intact long and short radiolunate ligaments was the consistent finding,
which can be explained anatomically based on the pathology of perilunate injury.
Fig. 1 (a) Three-dimensional CT images of a 20-year-old woman who had failed to achieve a gross
reduction by a closed manipulation showed volar dislocation of the lunate. (b,c) Arthroscopic findings showing that palmar capsular ligaments were torn and interposed
between the lunate and capitate (arrowhead), which was reduced by pulling the lunate
with use of a probe. Asterisk indicates the dorsal capsular tissue attached to the
dorsal limb of the lunate. (d–f) The scapholunate and lunotriquetral intervals were reduced by manipulating K-wires,
which were inserted percutaneously into the scaphoid and triquetrum, as viewed directly
in the midcarpal portals. The wires were driven across the intercarpal intervals into
the lunate once the anatomic reduction was achieved. (g) Eighteen-month follow-up radiographs showing a normal carpal alignment with no evidence
of arthritis. (Cp, capitates; Hm, hamate; Lu, lunate; Sc, scaphoid; SLIL, scapholunate
interosseous ligament;Tq, triquetrum).
After evaluation of the radiocarpal joint, our attention was directed toward the midcarpal
joint, in which the main disruption had occurred. The midcarpal space was entered
at the midcarpal radial (MCR) and midcarpal ulnar (MCU) portals. Bone or cartilage
fragments and frayed edges of torn palmar capsular ligaments were thoroughly débrided
or removed to facilitate reduction of the proximal intercarpal joint ([Fig. 2a–f]). Arthroscopic evaluation of the midcarpal joint showed focal cartilage damage of
the capitate in nine wrists and volar rim fracture of the lunate in seven wrists.
Probing the intercarpal joint of the proximal carpal row showed gross instability
of both the scapholunate and lunotriquetral articulations in PLDs and the lunotriquetral
joint in transscaphoid PLFDs from the complete tear. Six of 15 patients with transscaphoid
injuries showed combined scapholunate ligamentous injuries graded as II or III according
to the classification suggested by Geissler et al.[25]
Fig. 2 (a) Initial radiographs of a 38-year-old man showing a transscaphoid transtriquetral
PLFD. (b,c) Midcarpal view showing torn palmar capsular ligaments (arrowheads) that were interposed
between the carpal rows and between the fracture fragments of the scaphoid (arrows).
(d,e) After accurate reduction was obtained by manipulating K-wires and advancing these
wires across the fracture site, a scaphoid screw was percutaneously inserted as well
as K-wires into the triquetrum and capitate. (f) Twenty-six-month follow-up radiographs showing union of fractures with no evidence
of arthritis (Cp, capitate; Lu, lunate; Sc, scaphoid).
Once the injury pattern was identified, Kirschner wires (K-wires) were inserted percutaneously
into the scaphoid and triquetrum and an additional K-wire placed dorsally into the
lunate under a fluoroscope ([Fig. 1d]). After longitudinal traction was released, the scapholunate and lunotriquetral
intervals were reduced by manipulating the K-wires as joysticks, as viewed directly
in the midcarpal portals ([Fig. 1e]). When the anatomic reduction was restored, the wires were then driven across the
intercarpal intervals into the lunate ([Fig. 1f]).
For transscaphoid-type injuries, the reduction of the scaphoid fragments was attempted
by manipulating provisional K-wires driven into the distal fragment and not across
the fracture site, and an additional K-wire placed on the dorsum of the proximal pole
of the scaphoid. These wires were used as joysticks to reduce the fracture further
anatomically, while viewing the articular surface from the midcarpal portal. K-wires
in the distal pole of the scaphoid were advanced across the fracture site into the
proximal fragment after a congruous articular surface was obtained. Percutaneous scaphoid
fixation using a cannulated headless autocompression screw (Acutrak; Acumed, Beaverton,
OR, USA) was attempted in 11 patients who did not have comminution in the scaphoid.
The Acutrak screw was introduced after a guide wire was properly advanced from dorsal
to volar along the central axis of the scaphoid under fluoroscopic control, as described
by Slade et al ([Fig. 2d]).[26] The patients who had combined intercarpal ligamentous injuries had additional 1.2-mm
K-wire fixation into the intercarpal joint under arthroscopic assistance. Finally,
the wires were bent, cut, and buried underneath the skin ([Fig. 2e]).
After the operation, the wrist was immobilized in a short-arm thumb spica cast. In
pure ligamentous injuries, the K-wires were removed at 10 weeks. In patients with
a scaphoid fracture that had been fixed with K-wires, the K-wires were removed when
there was radiographic evidence of a union. Intensive physiotherapy was started.
Results
Acute carpal tunnel syndrome developed in 8 patients (40%) following the injury. The
symptoms resolved within 48 hours after arthroscopic reduction, and no patient required
release of the transverse carpal ligament. Besides the failed union of scaphoid fractures,
there were no conspicuous complications, except skin irritation by the buried pins
in two patients, which resolved after the pin was removed.
The average time to scaphoid union in transscaphoid injuries was 14 weeks (range 9–21
weeks) except in two patients with nonunion. Of the two nonuions, one patient, who
had a transscaphoid, transtriquetral PLFD and underwent surgery 20 days after injury,
showed loss of the carpal alignment with an avascular change of the proximal fragment
of the scaphoid at 6-month follow-up evaluation; a scaphoid excision and four-corner
fusion was subsequently performed. The other patient, who had had a transscaphoid
PLFD, demonstrated nonunion from the computed tomography (CT) perspective at 12 months
after the surgery. No further treatment was performed for this patient because he
did not complain of wrist pain or functional disability.
The mean follow-up period was 31.2 months (range 18–61 months). At the final evaluation,
the average extension and flexion were 53° (range 30–70°) and 51° (range 25–70°),
respectively. The average radial and ulnar deviation were 17° (range 10–26°) and 30°
(range 18–42°), respectively. The average pronation and supination were 71° (range
50–90°) and 79° (range 55–90°), respectively. The range of motion of the injured wrist
ranged from 79% to 89% in comparison with the contralateral side ([Fig. 3]). The average grip strength was 78% (range 62–94%) of the contralateral wrist. The
mean DASH and PRWE outcome scores were 18 (range 1–36) and 30 (range 5–52), respectively.
The overall functional outcomes according to the MMWS were excellent or good in 11
patients and fair or poor in 9 patients ([Fig. 4]). The one patient who underwent a four-corner fusion with excision of the scaphoid
was considered as having a poor result. The other patient who had a poor result developed
a diffuse stiffness of the wrist and hand.
Fig. 3 Physical examination at final follow-up.
Fig. 4 Overall functional outcomes according to the MMWS.
Radiologic measurements are summarized in [Table 1]. There were significant radiologic changes in scapholunate gap, scapholunate angle,
and carpal height at the final follow-up compared with those immediate postoperative
measurement (p = 0.038; p < 0.001; and p = 0.038, respectively). The scapholunate gap and scapholunate angle increased at
final follow-up by mean 0.3 mm (range 0–0.9 mm) and 3° (range 0–8°). The average carpal
height ratio had decreased by 0.02 (range 0–0.04). However, 15 patients had normal
scapholunate angle (normal, 30–60°),[22] and 16 patients had normal carpal height ratio (normal 0.51–0.57)[23] at final follow-up. No patients had radiographic evidence of arthritis of either
radiocarpal or midcarpal joints.
Table 1
Results of radiographic measurements after arthroscopic reduction and percutaneous
fixation
|
Parameter
|
Immediately after surgery
|
Follow-up
|
p value
|
|
Scapholunate gap (mm)
|
1.6 (1.3 to 2)
|
1.9 (1.4 to 2.6)
|
0.038
|
|
Lunotriquetral gap (mm)
|
1.7 (1.3 to 2.2)
|
1.8 (1.3 to 2.2)
|
0.295
|
|
Scapholunate angle (°)
|
55 (48 to 61)
|
58 (48 to 64)
|
<0.001
|
|
Carpal height
|
0.54 (0.52 to 0.54)
|
0.52 (0.49 to 0.54)
|
0.038
|
Discussion
Most surgeons have suggested that the key to a successful long-term result in the
treatment of a PLD or PLFD is open repair of interosseous and capsular ligaments and
internal fixation of the fracture.[6]
[7]
[27]
[28]
[29] Recently, arthroscopic techniques as an alternative minimally invasive method for
the treatment of perilunate injuries have been suggested by several authors.[12]
[18] They asserted that arthroscopic treatment allows anatomic reduction of intercarpal
articulations and proper reestablishment of carpal stability; however, the results
were based on only a few patients.
Our previous study has several limitations. It is a retrospective case series without
a control group, and the number of subjects was relatively small with a heterogenous
combination of pure dislocations and fracture-dislocations. In addition, this study
includes an insufficient follow-up period to prove the efficacy of arthroscopic techniques
in preventing the development of posttraumatic arthritis.
Souer et al[30] evaluated 18 patients treated with open reduction followed by temporary intercarpal
K-wire fixation or screw fixation at an average of 44 months. The average flexion-extension
ranges of motion (ROMs) of the wrist and grip strength were ∼60% and 70% of the contralateral
wrists, respectively. Four patients (22%) had wrist arthrodesis due to a deep infection,
scaphoid nonunion, crushing injury, or early development of arthritis. Forli et al[13] reported on 11 patients with PLDs and 7 patients with transscaphoid PLFDs treated
with open reduction and internal fixation. The mean flexion-extension ROM and grip
strength of the injured wrist were 76% and 87%, respectively, compared with the contralateral
side. They also observed signs of arthritis in 12 of 18 patients (67%) with a minimum
10-year follow-up. Otherwise, several authors have demonstrated better results of
an open approach where the average ROM of wrist flexion-extension and grip strength
were as high as 80% of the uninjured opposite wrist.[3]
[9]
[31] Therefore, the clinical results of our series seem to compare favorably with those
of open techniques.
Our initial concern of the previous study was whether the proximal interosseous ligaments,
which serve as the key stabilizers of the carpal joint, can achieve reliable healing
that is sufficient to maintain carpal stability without a direct repair. The results
of our study suggest that the capsular structures can heal adequately with a good
vascularity when they are properly approximated and protected for some period and
that open repair of interosseous ligaments is not necessary, given the additional
soft tissue and vascular injuries.[14]
[15]
[16]
[17] Therefore, we believe that arthroscopic management has the great advantage of facilitating
the healing of fractures and torn ligaments since it can minimize capsular and adjacent
soft tissue injury and provide preservation of an already tenuous blood supply.[32]
[33]
[34]
