Keywords trapeziometacarpal prosthesis - ligamentoplasty - joint replacement - teview - failure
Introduction
Osteoarthritis of the base of the thumb is a disabling pathology, much more common
in middle-aged women, with a ratio of 10:1 compared to men. The prevalence is high,
affecting up to 10% of women in middle age[1 ]. The main goal of treatment for painful osteoarthritis of the thumb is the restoration
of thumb function with a mobile, stable, pain-free, and stable joint. with a preserved
force. When conservative measures fail, different surgical treatments have been used
to achieve these objectives, but the vast majority are associated with a certain loss
of length and strength in the thumb. For this reason, prosthetic replacement of the
trapeziometacarpal (TMC) joint has increased in popularity in recent years due to
an improvement in implant design, reproducible clinical results, and a 10-year prosthetic
survival curve of up to 93%.[2 ]
[3 ] The recommendation for the use of prostheses in TMC osteoarthritis is made on the
theoretical basis of faster recovery and greater strength compared to other treatments,
but the long-term complications and cost of using implantation continue to be a matter
of debate.[2 ] Herren et al. published that patients with TMC prostheses recover significantly
faster in the first 3 postoperative months compared to patients in whom suspension
arthroplasty with ligamentous reconstruction is performed[4 ] and that the postoperative results at 1 year are similar in almost all patient parameters
evaluated for both groups, but patients with prostheses have greater clamp strength
than patients treated with TMC suspension arthroplasty.
The most common prosthetic complications are related to dislocation and loosening
of components (almost exclusively the trapezius component), requiring revision surgery
to restore a pain-free hand with proper function. Implant failure rate and revision
rates vary between different series and different implants: cemented Avanta® implants,
from 7% to 20%, de la Caffinière® implants, from 12% to 23%, Elektra® implants, from
18% % to 35%, ceramic devices, 33%, and the ARPE® implant, 6% to 7%.[2 ]
[3 ]
[5 ]
[6 ] According to Kaszap et al., the results of secondary trapeziectomy after bone replacement
arthroplasty of the failed trapeziometacarpal joint differ little from the primary
cases but conclude that more studies are needed.[6 ] Regarding TMC prosthetic replacement, if the surgical technique is precise, the
metacarpal stem is rarely a problem in these implants, since the integration is practically
constant. The problem may arise during revision surgery. In cases where the metacarpal
stem is not prominent, it can be left in place.[7 ] However, in some cases, it must be removed due to trapeziometacarpal or metacarpal-scaphoid
conflict due to axial instability of the thumb, and it is a technically difficult
step. difficult during revision surgery. The objective of this work is to present
a new surgical technique to remove the metacarpal stem of the prosthesis and to review
the clinical results of 17 patients in whom a secondary trapeziectomy with ligamentous
reconstruction was performed because of a TMC prosthesis in the form of a patella
(type “ball-and-socket”) failed.
Material and Methods
We conducted a retrospective case series study that included 12 patients who underwent
revision surgery due to a failed trapeziometacarpal (TMC) prosthesis between the years
2007 and 2019. Epidemiological and clinical data were recorded and analyzed ([Table 1 ]). The inclusion criteria were a failed patella-shaped TMC prosthesis at least 12
months after the initial surgery since removal of the implants in the recent postoperative
environment may be easier than in later stages due to osseointegration of the implant.
Seven patients came from other hospitals to our department to treat implant complications,
while the rest were initially treated at our institution. All patients were treated
using the same surgical technique. A descriptive statistical analysis was performed
using SPSS® 20.0 software. The Wilcoxon test was used to compare groups.
Table 1
GENDER
Frequency
Percentage
WOMAN
11
91,6
MEN
1
8,4
Total
12
100,0
AFFECTED SIDE
Frequency
Percentage
RIGHT
9
75
LEFT
3
25
Total
12
100,0
DOMINANT HAND
Frequency
Percentage
RIGHT
12
100
LEFT
0
0
Total
12
100,0
PROSTHESIS TYPE
Frequency
Percentage
ARPE
3
25,0
ELEKTRA
4
33,3
IVORY
2
16,7
MAIA
3
25,0
Total
12
100,0
Surgical Technique
A “V”-shaped dorsal approach is performed at the level of the TMC joint, with the
apex at the junction between the palmar and dorsal skin. The sensory branches of the
radial nerve and the radial artery are identified and protected in the anatomical
snuffbox. The TMC joint is approached between the tendons of the extensor pollicis
longus and extensor pollicis brevis muscles, along the axis of the first metacarpal.
Next, a longitudinal capsulotomy and subperiosteal dissection of the base of the first
metacarpal and trapezius is performed, exposing the implant ([Figure 1 ]). It is very common to find a loss of capsular tissue replaced by a thick scar.
It is recommended to preserve this tissue as a capsule, as it provides some degree
of axial stability of the metacarpal at the end of the procedure. It is also common
to observe bone resorption at the base of the metacarpal due to stress shielding,
as occurs in other implants such as the stem of the radius head prosthesis or the
distal ulna prosthesis,8 completely exposing the proximal part of the metacarpal stem.
Fig. 1 Exposure of the prosthesis through a dorsal approach.
Once the prosthesis is exposed, the first step consists of dorsal dislocation and
removal of the neck of the prosthesis. When treating a chronic dislocation, we generally
find the head of the prosthesis dislocated dorsally, on the dorso-radial aspect of
the trapezius, often associated with a bone defect in the trapezius.
The second step is to remove the trapezius while preserving the flexor carpi radialis
(FCR) tendon. Trapeziectomy allows complete visualization of the stem and removal
of fibrous tissue if necessary. Minimal exposure of the metacarpal base is done, just
to allow the metacarpal base to move dorsally for good access to the stem. Preservation
of the volar-ulnar scar tissue provides some degree of stability against the axial
collapse of the metacarpal. A Hohmann retractor is placed in the volar cortex of the
base of the metacarpal ([Figure 1 ]).
At this point, a broken screw extraction set (Depuy Synthes® or similar) is used ([Figure 2 ]). This set has different tools designed to remove screws with a damaged head. The
tapered tip of the extraction screw (2.4 or 2.5 mm) engages the stem ([Figure 3 ]).
Fig. 2 Screw removal set.
Fig. 3 The frustoconical tip of the extraction screw (2.4 or 2.5 mm) engages the internal
thread of the stem and engages by rotating the handle counterclockwise.
With gentle rotary movements and with the other hand holding the thumb at the level
of the first metacarpal, the rotary force breaks the bone bridges that fix the stem
and can be removed by pulling and turning counterclockwise ([Figure 4 ] and [5 ]). If this step is difficult due to stem integration, holding your thumb in your
hand may not be enough to remove the stem. In this case, a reduction bone clamp (Setter
type or similar) can be used at the base of the metacarpal to counteract the rotational
force applied to the extraction tool. Care must be taken not to cause a fracture.
If necessary, additional distal exposure of the metacarpal can be performed to apply
the clamp to the cortical bone. In cases in which the metacarpal stem fills the metacarpal
shaft and rotational movement is not allowed, it could cause a fracture of the metacarpal,
although to date this has not happened to us on any occasion.
Fig. 4 Schematic representation of the stem extraction maneuver.
Fig. 5 Removal of the metacarpal stem by repeated rotations of the T-handle.
Once the stem is removed, tendon interposition and suspension arthroplasty are performed
as in primary cases ([Figure 6 ]). We use a suspension trapeziectomy technique and ligamentous reconstruction with
abductor pollicis longus around the flexor carpi radialis tendon9 and reinforce with
the capsule or scar tissue preserved during exposure. But, since the stem has been
removed, any technique can be used. Finally, capsular closure is performed with a
3/0 absorbable suture, skin with 5/0 monofilament, and a plaster splint is placed.
The stitches are removed after 8 days, and the plaster splint is maintained until
3 weeks. At this point, hand therapy is initiated, and an orthosis is worn part-time
for 3 weeks, which can be removed during the day to begin progressive mobilization
exercises.
Fig. 6 Preoperative x-ray of a patient with a painful prosthesis due to loosening of the
trapezius cup and x-ray at 23-month follow-up after prosthesis removal.
Results
Twelve patients were reviewed, whose data are summarized in [Tables 1 ] and [2 ]. The average age of the patients at the time of surgery was 61 years (range 56-65).
Eleven patients were women, and one was a man. Four different types of prostheses
were reviewed: 3 Arpe®, 4 Elektra®, 2 Ivory®, and 3 Maia®. None of the Elektra® implant
patients were initially treated at our institution. The average survival time of the
prosthesis was 25.6 months (range 12-38), and the average follow-up time after revision
surgery was 35.5 months (range 20-50). Two cases were due to late post-traumatic dislocation
and the remaining cases were due to mobilization of the trapezius dome. The mean visual
analog scale (VAS) score was 3.3 (95% CI: 1.8-4.7) with a range of 1 to 8. The mean
Quick DASH questionnaire was 48.6 (95% CI: 35.0-62.2). The metacarpal stem was removed
in all patients. Differences between pinch strength and fist strength with the contralateral
side were compared using the Wilcoxon test. The average clamp force was 1.9 kg, representing
50% of the contralateral side, with a p-value <0.002. The mean fist force was 12.7 kg,
which is 71.3% of the contralateral side, with a p-value <0.0001 ([Table 2 ]).
Table 2
Value
Error Std.
AGE (YEARS)
Mean
61,3
2,1
95% CI
Lower limit
56,8
Upper limit
65,8
SD
8,1
IMPLANT SURVIVAL (MONTHS)
Mean
25,6
6,2
95% CI
Lower limit
12,3
Upper limit
38,8
SD
23,9
FOLLOW-UP TIME (MONTHS)
Mean
35,5
6,9
95% CI
Lower limit
20,7
Upper limit
50,2
SD
26,6
VISUAL ANALOG SCALE
Mean
3,3
0,7
95% CI
Lower limit
1,8
Upper limit
4,7
SD
2,7
OPPOSITION (KAPANDJI SCALE)
Mean
7,9
0,4
95% CI
Lower limit
7,0
Upper limit
8,8
SD
1,6
QUICK DASH
Mean
48,6
6,3
95% CI
Lower limit
35,0
Upper limit
62,2
SD
24,6
CLAMP FORCE (Kg)
Mean
1,9
0,4
95% CI
Lower limit
1,1
Upper limit
2,7
SD
1,4
FIST STRENGTH (Kg)
Mean
12,7
1,8
95% CI
Lower limit
8,8
Upper limit
16,6
SD
7,1
CONTRALATERAL CLAMP FORCE (Kg)
Mean
3,8
0,4
95% CI
Lower limit
2,9
Upper limit
4,7
SD
1,6
CONTRALATERAL FIST STRENGTH (Kg)
Mean
17,8
1,6
95% CI
Lower limit
14,3
Upper limit
21,2
SD
6,2
Discussion
The most common complications of the prosthesis are related to dislocation and loosening
of the components, which implies the need for revision surgery to restore a pain-free
hand with adequate functionality.
The long-term results of trapeziometacarpal prostheses have not yet been established
for large groups of patients. In Martin Ferrero's series of 69 TMC prostheses, the
survival rate was 93% at 10 years with good mobility, pincer, and grip strength.[2 ]
[3 ] In the Apard and Saint-Cast series, the survival rate Survival of the ARPE prosthesis
was less encouraging: 85% at 5 years and 79% at 11 years, but only 32 prostheses were
available for evaluation.[10 ]
Few articles have been published on the complications of patella-type TMC prostheses.
According to Bricout et al., the complication rate of the Maia® prosthesis in their
series was 35.9%, ranging from minor complications that did not require treatment
to major complications that required surgical revisions. In total, 18 surgical revisions
were performed on the 156 implanted prostheses.[11 ] Martin Ferrero, in his series of 64 patients with a follow-up of 10 years, 60 implants
(92.3%) were functional and five were not (7.7%). The survival estimate for functional
implants at 10 years was 93.9% (95% confidence interval: 82.3-97.9). Sinking of the
dome of the trapezius was observed in 15.8%.[2 ]
Toffoli et al., in 80 patients treated with the Maia® implant, observed 5 failures
(5.2%); of which, 4 loosening of the trapezius dome in the first 3 years requiring
revision surgery. Two cases had a secondary trapeziectomy with ligamentous reconstruction,
but the metacarpal stem was not removed. Although they reported that the clinical
results of these 5 rescue procedures were satisfactory, the average postoperative
value of the DASH scale was 39.[12 ] This DASH score is relatively high, considering that 0 represents no disability
and 100 the most severe disability. Furthermore, preoperative DASH values were not
evaluated, so clinical significance cannot be accurately assessed, although it does
give an idea of the patient's clinical status.
Cootjans et al., in a series of 166 prostheses (in 156 patients, 10 bilateral) with
an average follow-up of 80 months (median, 75 months), had 8 indications for revision.
Two patients were asymptomatic and did not require treatment. The remaining 6 indications
were made for review. The prosthesis was removed in only 1 patient (0.6%). A trapeziectomy
with tendon interposition (flexor carpi radialis muscle, Weilby procedure) was performed,
leaving the metacarpal stem in place.[13 ] There is no information on the results.
Although some characteristic complications such as loosening and/or sinking of the
implant, periprosthetic fracture, or dislocation are not frequent, if they cause symptomatology
they may require revision surgery. The revision strategy after a prosthetic TMC joint
replacement includes implant revision, implant removal, and partial or total trapeziectomy
with or without ligamentous reconstruction. During revision surgery, if implant removal
is performed, the dome of the trapezius is not a problem. Trapeziectomy can be performed
as in primary trapeziectomies. However, the metacarpal stem is usually completely
integrated, and its removal can be difficult as well as challenging. For this reason,
and because it is generally believed that removal of the stem is not necessary, the
vast majority of authors leave the metacarpal stem in place.[2 ]
[12 ]
[13 ]
[14 ]
However, rigid fixation of the stem can cause proximal resorption of the bone at the
base of the first metacarpal known as “stress shielding.” Stress shielding is known
to occur around rigidly fixed implants, as it can occur in other implants such as
radial head prostheses, regardless of stem design. However, it is usually mild and
non-progressive.[8 ] This stress shielding leaves the metal base of the stem exposed ([Figure 7 ]). With thumb clamping, proximal displacement of the first metacarpal can occur and
result in a painful thumb due to conflict between the first metacarpal and the trapezius.
In this situation, stabilization of the first metacarpal is crucial and, with the
metacarpal stem in place, two problems arise: protrusion of the metal base of the
stem and conflict with the scaphoid or sometimes the trapezoid.
Fig. 7 Stress shielding leaves the metal base of the stem exposed, which can cause conflict
with the scaphoid or trapezoid.
We have been using a useful technique to remove the metacarpal stem with a broken
screw removal set (De Puy Synthes® or similar) ([Figure 2 ]). This set has different tools designed to remove screws with damaged heads. The
frustoconical tip of the extraction screw (2.4 or 2.5 mm) engages perfectly in the
internal thread of the stem ([Figure 3 ]). Since using this technique, we have been able to remove all but one case of the
metacarpal stems. The case in which we were unable to remove the stem was in a patient
who had had a previous revision surgery in which the metacarpal stem was replaced
with the larger (size 10) Maïa® prosthesis. In this case, due to the risk of fracture,
it was not removed.
The extraction of the stem allows any suspension arthroplasty technique to be performed
with ligamentous reconstruction, and we believe that it also has a beneficial psychological
impact for the patient, since, if pain persists in the event of non-extraction of
the stem, it is easy to attribute the problem to the existence of the stem, although
this is not always the case.
To date, there is little information available on whether the results of primary and
secondary trapeziectomies might be similar. We present a series of 12 patients in
whom salvage surgery was performed after a failed TMC prosthesis. The results indicate
that although patients improved after revision surgery, the Quick DASH questionnaire
mean was 48.6 (95% CI: 34.9-62.2). The average pinch force was 1.8 kg, which is 47%
of the contralateral side with a value of p <0.002, and the average fist force was
12.7 kg, which is 71.3% of the contralateral side with a value of p <0.0001 ([Table 2 ]). The results may not be as good as those reported by Kaszap et al.6 with an average
DASH value of 17.2.
Limitations of the present study include its retrospective nature and the lack of
information prior to the rescue surgery (VAS, Quick DASH, clamp force), which limits
the interpretation of the results, and the small number of cases, although the follow-up
is long enough. However, when compared with the study by Kaszap et al.,6 in which
the mean Quick DASH score for secondary trapeziectomy was 16 (SD 4.3), in our series
the results differ by more than 30 points, with a mean of 48.6 (95% CI 34.9-62.2),
a very high value considering that the range is from 0 to 100, with 0 being the best
possible value.
As a result of the present study, we can conclude that in cases of failure of ball-and-socket
TMC prosthesis, secondary trapeziectomy with ligamentous reconstruction is a reproducible
treatment option. Removal of the metacarpal stem, if necessary, can be performed in
most cases with this relatively simple technique. However, the results of secondary
trapeziectomy may be worse than those of trapeziectomy with primary ligamentous reconstruction.
