Keywords wrist arthritis - carpal instability - scaphoid nonunion - distal scaphoid resection
- scaphoid nonunion advanced collapse (SNAC) arthritis
Introduction
The perfect carpal kinematics depends on bone and ligament integrity, that together
form the carpal ring, a concept instituted by Linchman et al.[1 ] Carpal collapse may ensue due to a solution continuity of that ring, as caused by
scaphoid nonunion following neglected or inadequately treated fractures. Such lesion
determines a degeneration pattern with well-defined stages known as scaphoid nonunion
advanced collapse (SNAC), which initiates at the radioscaphoid joint, evolves to the
scaphocapitate and lunocapitate joints, and ends at the radiolunate joint.[2 ]
In type II SNAC, the radioscaphoid joint is compromised at the most distal part of
the styloid, sparing the segment in the vicinity of the scaphoid proximal pole. Several
publications have demonstrated that scaphoid bone distal pole resection is a simple
and practical solution for the treatment of type II SNAC lesion, thus promoting improvement
of pain, range of motion, and grip strength. Nonetheless, despite of all advantages,
the technique carries the risk of dorsal intercalar scaphoid instability (DISI) pattern
as an undue result.[3 ]
[4 ]
[5 ]
We propose a new treatment technique for type II SNAC lesions by resection of the
distal scaphoid associated to tenodesis of the remaining proximal pole with a portion
of the extensor brevis carpi radialis (EBCR) tendon ([Figs. 1A ] and [1B ]). The technique yields the maintenance of lunocapitate angles within physiological
range (< 15°), thus avoiding DISI type instability.
Fig. 1 Proposed technique for the treatment of type II scaphoid nonunion advanced collapse
lesions by the resection of the distal scaphoid associated to a tenodesis of a portion
from the extensor brevis carpi radialis to the remaining proximal pole, together with
bone anchor fixation onto the dorsum of the capitate and the proximal scaphoid bones.
The figure shows the wrist on dorsal (A ) and lateral (B ) views.
Methods
This is a retrospective, observational study. After obtaining approval of the institutional
ethics committee, type II SNAC patients were submitted to distal scaphoid resection
(DSR), associated to a tenodesis to the proximal EBCR tendon. The study length was
from February 2016 to March 2018. Six patients were enrolled. Only patients with a
comprehensive assessment, and a minimum of 6 months postoperative follow-up were included.
Surgical indication criteria and chosen technique included pain and functional impairment
lasting longer than 6 months, along with no improvement with conservative measures
(such as physical therapy and orthoses). Our retrospective review assessed clinical
and radiographic results, along with possible complications due to the surgery.
Patients with previous surgeries, infections, or those with less than 6 months follow-up
were not included in the study. We have not indicated this technique for SNAC patients
caused by a proximal scaphoid fracture.
Patient data such as age, gender, occupation, and dominance, along with injury characteristics,
previous therapies, complaints, and time span between the lesion appearance and surgery
were all analyzed. Patient degree of satisfaction was also taken into consideration.
Pain was assessed both preoperatively and at the final assessment through visual analogue
scale (VAS) (from 0 to 10, in which 0 means absence of pain, and 10 means the worst
pain one can feel). Regarding the procedure, surgery duration and postoperative immobilization
time were recorded. A subjective assessment and wrist range of motion and grip, as
compared with the contralateral side, besides possible complications, such as infection,
hypertrophic scarring, rigidity, instability, or degeneration, were all taken into
consideration at the final analysis of the results. The time to return to work (or
to normal activities, in the case of a retired patient) was recorded (Mayo Clinic
Score).[6 ]
After the procedure, patients were seen at weekly intervals during the 1st month, and at every 30 days until 6 months of follow-up; by then, the final assessment
was performed. All patients were personally examined by one of the authors ([Table 1 ]).
Table 1
Patient demographics and data before and After the scaphoid distal pole resection
associated to extensor brevis carpi radialis tenodesis
Patient
Age (years)
Gender
Side
Nonunion
Pain scale (VAS)
Flexion-extension mobility
Grip strength
Postop follow-up (months)
Complications
Evolution time (months)
Type
Preop
Postop
Preop
Postop
Preop (kgf)
Postop (kgf)
1
28
M
R
36
Stable
Mid-third
Transverse
8
0
90°
140°
22
36
17
None
2
30
F
R
48
Stable
Mid-third
transverse
8
0
80°
135°
22
30
22
None
3
46
M
R
72
Stable
mid-third
transverse
10
2
85°
124°
17
28
5
None
4
45
F
R
44
Unstable
mid-third
oblique
9
1
70°
127°
22
29
16
None
5
36
M
R
34
Unstable
mid-third
oblique
9
5
55°
100°
18
28
12
Proximal scaphoid necrosis
6
44
M
R
54
Stable
mid-third
transverse
9
1
80°
137°
20
31
17
None
Abbreviations: M, male; F, female; R, right; VAS, visual analogue scale; Preop, preoperative; Postop,
postoperative.
Surgical Technique
Patient lies in supine position, with brachial plexus block, and a tourniquet is applied.
A longitudinal, italic “S” incision is placed between the central and radial thirds
at the dorsum of the wrist. Radial nerve sensitive branches are radially retracted.
An opening of the extensor retinaculum is performed with the elevation of a 1-cm wide
flap that is detached from the radial portion of the first extensor compartment, keeping
the insertion at the ulnar border of the 4th compartment. This facilitates retracting of the extensor tendons, thus exposing the
joint capsule that is longitudinally incised at the central portion for the elevation
of two flaps: one radially, and the other ulnarly.
Next, the distal radial styloid resection and neurectomy of posterior interosseous
nerve capsular branches are routinely performed. Then, the scaphoid distal resection
is performed through the nonunion plane ([Figs. 2A ] and [2B ]).
Fig. 2 The radius styloid process (#) and the distal pole of the scaphoid bone (¥) are seen
through a longitudinal, italic “S” incision on the back of the wrist (A ). After distal scaphoid and distal radius styloid resection, the capitate (§) and
the proximal scaphoid (£) are visualized (B ). Fixation of the proximal scaphoid to the capitate with a 1.5-mm K-wire, keeping
the lunate bone at 15° of flexion (C ).
The bleeding surface of the proximal pole is smoothed, and a 1.5-mm K-wire is introduced
on the back of the lunate bone to be used as a joystick to assist the alignment correction
of the first carpal row. Next, a stripe of the EBCR tendon with a third of its thickness
is elevated, keeping the distal insertion at the base of the third metacarpal bone.
There should be enough length so that the tendon can be inserted on the proximal scaphoid
bone fragment, along with a 2-cm long excess for a loop, and to be re-sutured back.
At this time, a temporary fixation of the proximal scaphoid to the capitate is performed
with a second 1.5-mm K-wire, keeping the lunate bone at 15° of flexion ([Fig. 2C ]). The exposure of the proximal scaphoid is facilitated by a previous radius styloidectomy.
The K-wire may also be passed from the triquetrum to the capitate when the proximal
scaphoid fragment is small. After carpal stabilization, the lunate dorsal K-wire may
be removed.
The EBCR tendon stripe is initially fixated to the capitate body and then to the scaphoid
with 2.5 mm anchors. The double-anchorage mount aims to avoid a possible bowstring
formation on the back of the wrist ([Figs. 3A ] and [3B ]). The tenodesis should be firmly tensioned, which can be achieved by suturing the
tendon to itself, after the fixation on the scaphoid. The K-wire introduced in the
scaphoid is also employed to assist on tenodesis stabilization and tension ([Fig. 3C ]). A proximal, dorsal scaphoid scarification is performed to improve the bone fixation
of the transferred tendon by forming a groove near the anchor site. Fluoroscopy is
employed to assess the correct placement of bone anchors and K-wires, along with the
lunate and capitate bones alignment.
Fig. 3 Elevation of extensor brevis carpi radialis tendon trip with one third of its thickness
(A ). Extensor brevis carpi radialis tendon fixation onto the scaphoid with 2.5 mm bone
anchor (B ). Tenodesis fixation with sufficient tension and suture onto itself after scaphoid
fixation (C ). Suture of the wrist joint capsule (D ).
The capsule and extensor retinaculum are sutured by planes ([Fig. 3D ]). The wrist is immobilized with a short forearm cast for 8 weeks; by that time,
the K-wire is removed. Patients are followed-up on a weekly basis for a month and
are usually submitted to a rehabilitation protocol with a hand therapist. They are
also oriented to perform lengthening and range-of-motion exercises at home.
Results
Of the 6 patients assessed, 4 were male and 2 were female, with ages ranging from
28 to 46 years (mean, 38.1 years). All compromised limbs were the dominant ones. The
elapsed time between nonunion diagnosis or scaphoid bone fracture and definitive treatment
ranged from 34 to 72 months, with an average of 48.0 months. Mean follow-up time between
the surgery and final results assessment was 15.3 months, ranging from 8 to 22 months.
All patients proved to have stage II SNAC by computerized tomographic assessment,
showing compromise of the joint between the radius and the scaphoid, but with no proximal
capitate lesion. All six patients of our series had mid-third scaphoid bone nonunion;
in four patients, the nonunion line was transverse, and, in two patients, the nonunion
line was oblique.
Pain assessment was performed by means of a VAS on a ruler. Preoperatively, mean pain
was 8.8, ranging from 8 to 10. At the 6-months postoperative follow-up, mean pain
was 1.5, ranging from 0 to 5.
We added wrist flexion and extension to measure joint range of motion. Preoperatively,
the mean measure was 76.6°, ranging from 55° to 90°. Postoperatively, the mean was
127.1°, ranging from 110° to 140°.
Grip strength was assessed with a Jamar dynamometer on position #3, with the elbow
flexed at 90°, and the non-supported upper limb over the table. Preoperatively, patients
had mean strength of 20.1 kgf, ranging from 17 to 22. Postoperative mean strength
was 30.3 kgf, ranging from 28 to 36.
Preoperative lateral wrist radiographs showed an increased angle between the lunate
and capitate bones, a DISI-type dissociation pattern with an average of 25°, ranging
from 19 to 32°. Regarding the radiologic pattern, the measurement of lunocapitate
angle remained < 15° in all patients, and a DISI-type instability evolution was not
observed. ([Fig. 4 ]).
Fig. 4 Preoperative anteroposterior views showing scaphoid nonunion and type II scaphoid
nonunion advanced collapse lesion (A ). Radiographs showing the technique comprising resection of the distal scaphoid associated
to the tenodesis of a portion from the extensor brevis carpi radialis onto the remaining
proximal pole, along with bone anchor fixation onto the dorsum of the capitate and
the proximal scaphoid. Correction of dorsal intercalar scaphoid instability (DISI)
pattern as an undue result deformity, as revealed in the lateral view (C ).
One patient presented proximal scaphoid necrosis as a complication at 8 months postoperatively.
The patient was treated with proximal row carpectomy with a good clinical outcome;
however, he did not return to his original job activities. There were no other serious
complications such as infection, hypertrophic scarring, or complex regional pain syndrome.
The other five patients did return to their previous labor activities.
There was a significant improvement in 5 patients, who have risen from 42 ± 14 points
to 91 ± 5 points on the clinical evolution as measured by the Mayo Clinic Score. One
patient had little improvement, from 53 to 64 points; he evolved later to a necrosis
of the proximal pole.
We asked patients to rank their improvement as very satisfied; satisfied; little satisfied;
and unsatisfied regarding treatment outcomes. Of a total of six patients, four reported
to be very satisfied; one patient was satisfied; and the other little satisfied.
Discussion
The treatment of carpal collapse due to scaphoid nonunion has been well-stablished
in the literature, with definite treatment options for each stage. Regarding specifically
to stage II, it has been cited: simple DSR;[3 ]
[4 ]
[5 ]
[6 ] total scaphoid resection associated to the tenodesis of the flexor carpi radialis;[7 ] total scaphoid resection associated to carpal mini-arthrodesis; total scaphoid resection
associated to capsulodesis;[8 ] and the carpectomy of the first-row carpal bones.[9 ]
All those procedures aim to relieve pain, improve the range of motion, and increase
strength. Simple DSR is a practical, easily reproducible technique, and its results
testify that the procedure does improve range of motion, relieves the pain, and increases
grip strength. Nonetheless, it has been observed that there is an evolution toward
a DISI-type instability pattern, leading to an increased lunocapitate angle, a fact
that worsen patients' clinical evolution.[5 ] Besides, Karmal et al cited that the DISI pattern associated to midcarpal dissociation
that may occur after scaphoid distal fragment excision is aggravated by the dorsal
carpal ligament injury, which would work as a container between the triscaphoid and
the triquetral-hamate joints, thus leading to a painful midcarpal instability.[10 ]
Luchetti et al[7 ] created a technique that associates the transfer of the flexor carpi radialis segment
that is sutured over itself after looping the radio-triquetral ligament to avoid carpal
ulnar translation, and a DISI pattern after total resection of the scaphoid. They
have had employed the technique in 18 patients, and described procedure failure in
4 cases; another 4 patients were lost to follow-up; and the 10 remaining patients
presented satisfactory clinical results on the Mayo Wrist Score, which had a gain
of 23 points—from 52 to 75 points. However, they had emphasized that the radiologic
pattern was rather precarious, as characterized by increased DISI deformity, lunate
and triquetrum ulnar translation, and, in some cases, a contact between the capitate
and the lunate facet, bearing an evident chondral damage. Comparing that technique
to the one proposed by the authors of this article, we have noted that there was neither
a disharmonic pattern nor a proximal migration of the capitate bone in any of our
patients, thus maintaining carpal height along with equitable radial distribution
of axial forces from the capitate head and lunate and the remaining pole of the scaphoid
bone.
The dorsal and volar wrist capsule suture procedure at the immediately radial portion
to the lunocapitate joint, after complete scaphoid excision, has been described to
keep the capitate bone aligned to the lunate bone, so that the need of an arthrodesis
is avoided. Good functional results have been described with that technique;[8 ] nonetheless, besides being more technically demanding to be performed, the procedure
creates an overload onto the lunate bone due to the absence of the scaphoid by keeping
the radial facet of the capitate head uncovered. Our technique has the advantage of
preserving carpal architecture by avoiding capitate head radial translation by keeping
this structure in a bony framework formed by the lunate and the proximal portion of
the scaphoid bones.
The proximal carpectomy has the advantage of keeping a good range of motion, despite
grip strength limitation. That procedure has a restrict indication for cases where
there is no chondral lesion, either of the proximal capitate, or of the radius lunate
fossa. In a retrospective study with a follow-up of more than 15 years, there was
evidence that the procedure alleviates the pain, but it does not eliminate the pain.
As time goes by, there is also a gradual reduction of the range of motion achieved
initially, a fact that had direct correlation to the degeneration degree of the radiocapitate
joint.[9 ] Biomechanical studies have suggested that the progression of the degenerative process
of that joint is due to the morphologic incongruity between the surfaces of the capitate
and radius lunate fossa, which generates increased pressure onto a small contact area.[9 ] Ali et al suggested that in young, high-demand patients other treatment alternatives
should be considered, but not carpectomy.[9 ]
The total resection of the scaphoid bone associated to some kind of partial carpal
arthrodesis (four-corner arthrodesis or radiolunate arthrodesis) is in general indicated
for the treatment of type III SNAC, which is normally associated to a capitate head
chondral lesion.
Comparative data from the literature show discrepant results among main wrist-preserving
techniques.[3 ] Regarding increased grip strength that is postoperatively achieved, when compared
with the contralateral side, there is an equivalence: 75% for DSR; 70% for total scaphoid
resection associated to partial carpal arthrodesis; and 78% in cases of proximal carpectomy.
Nonetheless, regarding mean increased range of motion for flexion-extension, we found
43° for DSR; 21° for total scaphoid resection associated to partial carpal arthrodesis;
24° in cases of proximal carpectomy;[3 ] and 10° in cases of total scaphoid resection associated to capsule suturing.[8 ] The following results were reported regarding radioulnar deviation: 41° for DSR;
15° for total scaphoid resection associated to partial carpal arthrodesis; 24° for
proximal carpectomy;[3 ] and the loss of 2° from total scaphoid resection associated to capsule suturing.[8 ] These data demonstrate the superiority of our results when compared with techniques
that advocate partial carpal fusions associated to total scaphoid excision, or to
scaphoid resection and capsule suturing.
There was no evidence of arthrosis from the lunocapitate or scaphocapitate joints
in our series, which overlaps with the literature findings.
A severe complication ensued in patient 5, scaphoid necrosis. The lesion was diagnosed
on the late postoperative period (8 months) and was treated by proximal carpectomy,
with good clinical results. There could have been signs of vascular compromise at
the nonunion when exams prior to the first surgical procedure were assessed, although
with there no evidence has been found. Maybe that patient did not present conditions
to undergo the proposed technique. We had no cases of either pin-tract or soft-tissue
infection, or any case of reflex sympathetic dystrophy.
Criticisms to the procedure are bound to technical restrictions, such as the conjoint
insertion of a bone anchor and K-wire onto the scaphoid proximal fragment. This fact
may have influenced on the proximal scaphoid necrosis of patient 5. Hence, we stress
that K-wire insertion to fix the scaphoid to the capitate shall be performed in one
attempt, if possible.
The DSR associated to tenodesis with a segment of the EBCR renders the alignment between
both carpal rows, thus reconstructing the carpal kinematics. That avoids DISI instability,
which is the main complication of other techniques. Future comparative studies need
to be conducted in order to clarify whether there is a real advantage in increasing
the complexity of the technique toward the neutralization of the first carpal row
positioning, along with its influence on the final results, as compared with the classical
techniques.
Conclusion
Despite the small series of cases, we understand that distal scaphoid resection associated
to EBCR segment tenodesis has demonstrated to be a useful method for the treatment
of type II SNAC. The technique preserves carpal height and grip strength besides keeping
the stability between the proximal portion of the capitate and the lunate joint surface,
together with the proximal scaphoid. The method is restricted for cases in which there
are no chondral changes of the capitate head, having the additional benefit of a possible
reversion to any other technique if the need arises in the future (such as carpectomy,
partial carpal arthrodesis, or a total wrist arthrodesis). Having in mind that cost
and learning curve must be taken into account, our conclusion is that the technique
is an effective alternative to those classic procedures employed for the treatment
of type II SNAC lesions.
