ARTHROPLASTY OF THE HAND

The primary indication for arthroplasty of the
metacarpophalangeal (MCP) joints is advanced painful destruction of the
joint by rheumatoid arthritis. The systemic effects of rheumatoid
arthritis on the metacarpophalangeal joints are mediated by the
formation of a pannus that is destructive to both the soft tissue and
bone. In patients without advanced bone destruction at the MCP joints,
intermediate procedures such as synovectomy and crossed intrinsic
transfers can be utilized to improve function and decrease pain,
although eventual progression and bony destruction will most commonly
occur. Patients who suffer from isolated posttraumatic arthritis of the
MCP joint, most commonly secondary to fracture, may also benefit from
arthroplasty; the increased demands of these generally younger patients
must be taken into account in assessing implant use, however. In
osteoarthritis, implant use is most satisfactory in nonborder digits
and in patients who will not accept joint fusion as a primary treatment
modality.

The ideal candidate for arthroplasty is the rheumatoid
patient who presents with significant pain or severe destruction of the
MCP joints involving palmar subluxation of the proximal phalanx and
early fixed hyperextension deformity of the proximal interphalangeal
joint (Fig. 71.1) (20).
The concept of MCP joint implant arthroplasty was first reported in
1959, with goals to both relieve pain and correct deformity (5).
Since that time, a variety of implants have been developed and used for
MCP replacement, with improved durability in materials coming with
time. The primary prostheses used today are fashioned of silicone. The
major implants in use today are the original Swanson implant (Wright
Medical, Arlington, TN) (20), now used with
metal grommets in an attempt to prevent implant failure; the neutral
Avanta (Avanta, San Diego, CA) block design implant; and the newer
anatomically neutral NeuFlex (DePuy, Warsaw, IN) implant, designed to
decrease implant failure and improve functional arc of motion.

Carefully assess the rheumatoid patient prior to upper
extremity surgery. Functional status, lower extremity problems, and
level of pain are important issues prior to deciding whether or not
implant surgery should go forward. Many patients with rheumatoid
arthritis can function well with painless advanced deformities of the
joints of the hand, although deformity, despite being relatively
painless, should not be allowed to progress to a point where
arthroplasty becomes technically very difficult. In addition, the
status of the wrist also affects MCP arthroplasty outcome in that
uncorrected fixed radial deviation of the wrist in the face of MCP
joint replacement often leads to rapid recurrent ulnar drift and
implant failure at the MCP joints postoperatively. Most authors
advocate staged or simultaneous wrist fusion or arthroplasty to provide
a stable fulcrum for MCP joint arthroplasty in these patients (17).
In the majority of patients with rheumatoid arthritis, all four MCP
joints are generally replaced at a single setting. The thumb MCP joint
is generally not treated by implant arthroplasty due to the high forces
that occur during function, and the relative lack of disability
associated with fusion of the thumb MCP joint. In patients who have
significant disease at the thumb MCP joint, fusion of this joint can
frequently be undertaken simultaneously with implant arthroplasty of
the index through small-finger MCP joints.

The majority of patients gain significant pain relief
and increased level of function after MCP arthroplasty. The correction
of ulnar deviation (Fig. 71.2), improved
ability to perform activities of daily living, and improvement of
extensor lag have been documented in several outcome studies. Change in
range of motion has been variable, with some series reporting losses or
gains of up to 10° from preoperatively (4,11,14,16).
Until recently, the main objection to MCP joint arthroplasty has been
the inability to obtain full flexion postoperatively. Anatomically
neutral implants appear to allow improved functional flexion of the
digits without a significant increase in extensor lag, thereby
improving the overall arc of motion (Fig. 71.3).
In addition, in anatomically neutral implants the compressive and
tensile forces on the silicone is more evenly distributed through the
functional arc of motion. Nevertheless, no long-term data are available
on failure rates for anatomically neutral implants, and whether
theoretical improvements lead to clinical improvements remains to be
seen. All implants available in the United States and abroad have
initially been found to be satisfactory, but with long-term follow-up
they tend to fail.

Two different surgical approaches are available. Use either a double longitudinal incision (B₁, B₂ Fig. 71-4),
with one incision between the index and middle finger metacarpal heads
and the second between the ring and small finger metacarpal heads, or a
transverse incision at the level of

the MCP joints (A, Fig. 71.4A).
We favor the double longitudinal incision technique since postoperative
range of motion exercises do not cause tension on the wound; if wound
complications do occur, the implants are not directly exposed even in
cases of wound dehiscence.

For appropriate implant sizing, utilize sequential
broaching in both the metacarpal and phalangeal shafts. Use the
presized broaches to rasp the intramedullary canals to as large a size
as possible.

Patients are generally admitted overnight for observation and 24 hours of intravenous antibiotics and pain medication.

The splint and dressing remain in place for 5 to 10 days
postoperatively (longer in patients who have had significant MCP joint
palmar subluxation and shortening preoperatively). Remove the drains
prior to hospital discharge. Remove sutures at the first postoperative
visit. Have the patient evaluated by a hand therapist and have a molded
full-extension pan splint placed as well as a dynamic extension splint
to allow early range of motion of the digits while maintaining the
extensor attitude of the MCP joints themselves (Fig. 71.12).
We recommend splints for the first 4 to 6 weeks, with gradual reduction
of their use by 2 to 3 months. In general, maximal functional range of
motion is obtained at 3 to 4 months postoperatively (Fig. 71.13). These basic principles and techniques

are germane to the use of all silicone arthroplasty implants available.

The proximal interphalangeal (PIP) joint is susceptible
to destruction by synovial proliferation as seen in the MCP joints in
rheumatoid arthritis. Painful PIP joints with loss of function are the
primary indication for replacement (21).
Advanced osteoarthritic destruction of the PIP joint can also produce
similar symptoms. PIP joint arthrosisin osteoarthritis is far more
commonly seen than at the MCP joint level (18).
Frequently, the distal interphalangeal joint (DIP) is also involved in
this process. Although arthroplasty for the DIP joint has been reported
(2,3), results in
general have been less than satisfactory. Most surgeons recommend
fusion of the DIP joint by one of several methods as a definitive form
of treatment for painful deformities (3). PIP
joint arthroplasty for osteoarthritis, although not commonly performed,
is a treatment option after failed conservative therapy. Arthroplasty
of the PIP joints is best reserved for the nonborder digits, as is the
case with the MCP joints due to the high forces generated by pinch and
use of the hand (18). The same principles apply
to the rheumatoid patient, although in general deformity is far more
advanced and bone quality poor in the rheumatoid patient.

Options for implants at the PIP joints include silicone
replacements similar to those used for the MCP joints as well as
various metal/high-density-polymer designs (3).
Recent reports have demonstrated excellent decrease in pain and
relatively good long-term durability with silicone implant arthroplasty
(1). Overall range of motion postoperatively,
however, has demonstrated mixed results, generally equaling the
preoperative abilities. Combining various metal/high-density-polymer
designs with the use of bone cement simulates a more anatomic
articulation; however, prosthetic loosening, fracture, and protrusion
through the cortex have been noted (15).
Nevertheless, the attractive anatomic contouring of these devices has
led several investigators to continue their development and refinement.

Dorsal, volar, and ulnar lateral approaches are used to
approach the PIP joint. We favor a dorsal approach although we have
used the volar approach quite successfully. Regardless of the approach
used, several basic principles, if followed, allow appropriate implant
placement and acceptable postoperative function.

Broaching for PIP joint arthroplasty is challenging, as
adequate bone stock must be maintained to prevent osseous fracture, and
yet as large a size prosthesis as possible should be implanted. In
addition, the gap between the proximal and middle phalanx shafts needs
to be wide enough to allow placement of the implant. This may require
some additional bone resection at the base of the middle phalanx or
distal portion of the proximal phalanx. After appropriate broaching,
several steps need to be taken to ensure appropriate arthroplasty.

Keep the patient in a postoperative extension splint for
7 to 10 days, at which time the sutures are removed. Then fit a
removable Orthoplast splint to maintain the PIP joint in full extension
until 2 weeks after surgery, at which time gentle range-of-motion
exercises can begin. Use the extension splint between exercises to
prevent any extensor lag from occurring. If a volar approach is
utilized, less attention needs to be directed toward extension
splinting. With the volar approach, start early flexion as soon as
possible to prevent adhesions to the flexor tendons. Recurrence of
ulnar deviation (which is commonly seen preoperatively) can occur
following PIP joint arthroplasty. Splint training of the digit may be
required.

Painful degeneration of the thumb carpometacarpal (CMC)
joint is most often secondary to osteoarthritis. This clinical entity
is most commonly seen in postmenopausal women. Rheumatoid arthritis,
repetitive stress injury, and hormonally induced changes have also been
implicated as causes for basilar joint arthritis, but this occurs far
less frequently than degenerative osteoarthritis (17).
The extent of degenerative change in the CMC joint can be staged
radiographically according to the classification of Eaton and Littler (Table 71.1) (9).

Initial treatment of thumb CMC arthritis is
conservative. A trial of nonsteroidal antiinflammatory medication and
hand-based splinting of the thumb CMC joint is often

successful
in early and moderate cases. In addition, an injection of
corticosteroid into the CMC joint can occasionally provide dramatic
pain relief and increased function in patients, although long-term
relief is not common. Some patients with extensive radiographic disease
have minimal pain, and others with minimal radiographic disease have
extensive pain; therefore, base treatment on the patient’s symptoms
rather than on the radiographs alone.

Historically, multiple surgical procedures have been
performed for painful thumb CMC arthritis. Silicone implantation to
replace the resected trapezium was developed by Swanson in 1968 (20).
Smaller silicone wafers utilized in a hemitrapezial excision have
provided excellent early symptomatic improvement, although silicone
synovitis resulting in pain and bony destruction has occurred in
long-term studies. As a result, silicone implantation of the thumb CMC
joint has been all but abandoned. Metal implants have recently been
advocated, although experience is limited and outcome uncertain (7).
Arthrodesis of the thumb CMC joint is a viable option and is useful in
younger, manual laborers who perform heavy work. It gives a predictable
painless result despite sacrificing some range of motion (17,19).

Currently, some combination of excision of the trapezium
and tendon interposition or ligament reconstruction is most commonly
performed for arthroplasty of the thumb CMC joint (8).
In 1970, Froimson introduced the concept of using a ball of tendon as
an interpositional spacer between the carpus and thumb metacarpal after
excision of the trapezium (10). Due to the
potential for proximal migration of the thumb metacarpal, others have
advocated ligamentous reconstruction utilizing the flexor carpi
radialis tendon woven through the base of the thumb metacarpal (6). Outcome studies have not demonstrated that one technique provides substantially improved results over another (13). In fact, some authors have advocated excision of the trapezium alone without any interposition or reconstruction at all (12).
Having utilized both techniques, we currently favor excision of the
trapezium and tendon interposition without ligament reconstruction.

Maintain the splint and dressing until 7 to 10 days
postoperatively, then remove the sutures. Apply a below-elbow thumb
spica cast allowing interphalangeal thumb motion and full MCP motion of
the other digits. Maintain the cast for 4 weeks. Then remove the
Kirschner wire in the office and place the patient in a removable
Orthoplast splint. Begin hand therapy at 4 weeks with gradually
increasing range-of-motion and strengthening exercises. Four weeks of
therapy is usually required.