Abstract
Background: Interposition arthroplasty is an option for the
treatment of arthritis of the elbow. Conversion to a total elbow arthroplasty
can be considered later, when the patient reaches a suitable age. We
investigated the results of conversion of an interpositional elbow
arthroplasty to a semiconstrained total elbow arthroplasty in a series of
patients.
Methods: The results of twelve consecutive linked semiconstrained
total elbow arthroplasties in twelve patients who had had a prior
interposition arthroplasty for the treatment of degenerative arthritis of the
elbow were evaluated at an average of ten years postoperatively. The average
age at the time of the total elbow arthroplasty was fifty years, and the
average interval from the interposition arthroplasty to the total elbow
arthroplasty was 9.9 years. Pain and elbow performance as measured with the
Mayo Elbow Performance Score were assessed in a retrospective chart review and
an evaluation of questionnaires, and postoperative radiographs were
reviewed.
Results: At the time of the latest follow-up, postoperative pain was
rated as mild or none in ten of the twelve patients, and the result was rated
as subjectively satisfactory in ten patients. The average Mayo Elbow
Performance Score improved from 32.1 points (range, 10 to 70 points)
preoperatively to 80.4 points (range, 40 to 100 points) postoperatively (p
< 0.001). According to these objective criteria, there were three
excellent, six good, one fair, and two poor results. All of the elbows were
stable following the arthroplasty. Radiographs demonstrated a well-fixed
prosthesis in all but one patient who had extensive osteolysis at the site of
the humeral component. One other patient had radiographic evidence of bushing
wear. Both of these patients required revision procedures.
Conclusions: Semiconstrained total elbow arthroplasty can be
performed successfully in patients with a prior interposition arthroplasty.
Reliable pain relief and a satisfactory result can be achieved in most
patients.
Level of Evidence: Therapeutic Level IV. See Instructions
to Authors for a complete description of levels of evidence.
Surgical treatment of elbow arthritis in young patients is a
difficult problem. A total elbow prosthesis may not be able to withstand the
functional demands placed on prostheses by younger, more active individuals.
Schneeberger and two of us (R.A. and B.F.M.) reported a 27% complication rate
(eleven of forty-one) following semiconstrained total elbow arthroplasty
performed in younger patients with posttraumatic
arthritis1. King and
two of us (R.A. and B.F.M.) reported an equally high complication rate after
revision elbow
arthroplasty2.
Interposition arthroplasty has been used with success as a nonprosthetic
alternative in younger
patients3-15.
Resection arthroplasty was initially described for the treatment of
tuberculous arthrosis by Park and Moreau in
180610, and a
variety of techniques utilizing various interposition materials (silicone,
metal, tendon, muscle, fat, skin, and fascia) have since been described. The
most popular technique has been skin or fascial arthroplasty, and success
rates ranging from 26% to 94% have been
reported3-15.
Recent modifications of this technique have included the use of a hinged
external fixator to prevent the postoperative instability that occurs with
aggressive softtissue
release3,16.
One of the theoretical advantages of interposition arthroplasty is that it
can be converted to a total elbow arthroplasty after the patient has reached a
more advanced age, ideally more than sixty years. This strategy is designed to
avoid the potential complications associated with the failures and revisions
of the total elbow prosthesis that occur more commonly in younger patients.
Interposition arthroplasty preserves bone stock and motion so that total elbow
arthroplasty can be performed successfully when the patient reaches an age
when less stress is anticipated on the implant. However, the results of total
elbow arthroplasty in patients with a prior interposition arthroplasty have
not been previously investigated, to our knowledge. This study was designed to
determine the validity of the assumptions underlying this approach.
Patient Selection
Sixteen patients who had had conversion of a prior interposition
arthroplasty to a total elbow arthroplasty at our institution over an
eighteen-year period (1976 to 1994) were identified. Three of these patients
had an unconstrained prosthesis and were excluded, leaving thirteen patients
with a linked arthroplasty. One patient died thirteen months postoperatively
and was dropped from the study as well. The medical records of the remaining
twelve patients were retrospectively reviewed. Nine of the twelve patients had
a radiographic examination at the time of or after the clinical follow-up. The
study design and protocol for human subjects were approved by the
institutional review board at our institution. There were nine women and three
men with an average age of fifty years (range, twenty-eight to eighty years)
at the time of the total elbow arthroplasty. The dominant arm was involved in
six patients. Eleven patients had posttraumatic arthritis, and one had Reiter
syndrome. Ten patients had had prior interposition of a fascia lata autograft.
Bone wax was used in one patient, and there was fibrous scarring at the site
of a resection arthroplasty in one. Interposition of fibrous tissue was
observed at the time of the conversion to the total elbow arthroplasty in all
patients. The average interval from the interposition arthroplasty to the
total elbow arthroplasty was 9.9 years (range, five to 576 months).
Indications
All patients had had the interposition arthroplasty performed to treat
elbow arthritis, which was posttraumatic in eleven and inflammatory in one.
The primary indication for the semiconstrained total elbow arthroplasty was
pain, which was rated as severe by all but one patient. Eight of the elbows
were considered unstable at the time of the operation. One patient had a stiff
elbow with a fixed flexion deformity of 100°. An average of two prior
procedures (range, one to four procedures) had been performed prior to the
semiconstrained total elbow arthroplasty. Radial head excision had been
performed as part of the interposition arthroplasty procedure in nine
patients.
Implants
There was an evolution of semiconstrained total elbow prostheses during the
study period (1976 to 1994). Three types of semiconstrained prostheses with
similar design features were used in this study. The Pritchard-Walker
prosthesis was an early semiconstrained design with metal humeral and ulnar
stems and a polyethylene articular surface and
bushing17-19.
The semiconstrained Coonrad-II implant was used from 1978 to 1981, and the
same implant with an anterior flange over the distal part of the humerus (the
Coonrad-Morrey total elbow prosthesis) has been used since 1981. Subsequent
minor modifications of the Coonrad-Morrey implant have included changes to the
metal composition, surface coating, and locking pin. In the present
investigation, the Pritchard-Walker prosthesis was used in three patients; the
Coonrad-II prosthesis, in two; and the Coonrad-Morrey prosthesis, in seven.
All were implanted through a posterior Bryan-Morrey approach. An autogenous
anterior bone graft was applied beneath the flange of all of the
Coonrad-Morrey prostheses. The radial head was retained in the three patients
in whom it had not been previously resected.
Surgical Technique
The technique of semiconstrained total elbow arthroplasty has been
described
previously20.
Briefly, the patient was positioned supine with the arm draped across the
chest. The Bryan-Morrey triceps-elevating approach was used with a midline
posterior skin incision. The ulnar nerve was identified and was transferred
anteriorly in all eight patients in whom it had not been transposed
previously. A medial incision was made over the ulnar periosteum, and the
extensor mechanism was reflected laterally as a sleeve by transecting the
Sharpey fibers at their insertion into the ulna. The anconeus was elevated
from its ulnar and humeral attachments. The medial and ulnar collateral
ligaments were released in the presence of stiffness to avoid iatrogenic
fracture of the supracondylar columns.
Fibrous tissue overlying the distal part of the humerus from the prior
interposition was carefully stripped from the bone, and the humerus and ulna
were prepared in a routine fashion for implantation of the components. For the
Coonrad-Morrey prostheses, an autogenous bone graft was prepared from the
resected trochlea and was placed beneath the anterior flange of the humeral
component. All components were cemented. The triceps was then repaired with
number-5 braided nylon nonabsorbable sutures through drill-holes in a cruciate
and transverse pattern. The tourniquet was released, and hemostasis was
obtained. Drains were placed if bleeding was excessive, and the wound was
closed in layers. A compressive dressing and an anterior plaster splint were
applied with the elbow held in extension. The average duration of the
operation was 156 minutes (range, 100 to 295 minutes).
Postoperative Care
The arm was placed in an anterior molded splint in extension, and the
splint was worn for the first three days postoperatively. The arm was elevated
for twenty-four hours, after which the dressings were changed and any drains
were removed. The patient then began range-of-motion exercises but avoided
strengthening exercises. No formal physical therapy was used. The patient was
allowed to perform routine activities of daily living at home but was
instructed not to lift more than 1 lb (0.45 kg) for the first three months and
to permanently avoid using the treated arm to lift =10 lb (=4.5 kg) as a
single event and =5 lb (=2.3 kg) as a repetitive activity.
Analysis of Outcome
The Mayo Elbow Performance Score (MEPS) was used to determine outcome, on a
100-point scale, on the basis of subjective, objective, and functional
criteria20. Pain is
graded in increments of 15 points on a scale of 0 to 45 points. Five points
are awarded for each of five functional activities that the patient can carry
out, with a maximum total of 25 points. A maximum of 20 points is awarded for
the arc of motion, and up to 10 points are awarded for stability. A score of
=90 points is considered an excellent result; 75 to 89 points, a good
result; 60 to 74 points, a fair result; and <60 points, a poor result.
Patients who required revision of the prosthesis were considered to have a
poor result. MEPS scores were determined through a combination of chart review
and evaluation of mailed questionnaires completed by the patients. Follow-up
questionnaires were completed at the time of the study by seven of the twelve
patients.
Radiographic Analysis
Standard anteroposterior and lateral radiographs were routinely made
preoperatively as well as postoperatively at three months, one year, and every
two to three years thereafter (Figs. 1-A,
1-B, 1-C, 1-D,
2-A, 2-B,
2-C). With use of previously
described
methods21, all
postoperative radiographs were assessed for implant fixation, bushing wear,
radiolucent lines, and incorporation of the bone graft. The implant was
considered to be well fixed when there was a solid bone-cement interface and
the position of the implant had not changed on consecutive postoperative
radiographs. Radiolucencies were graded, according to previously published
criteria21, as
none, type I (a progressive radiolucent line of <1 mm in thickness
involving <50% of the bone-cement interface), type II (a progressive
radiolucent line of =1 mm in thickness involving <50% of the interface),
type III (a progressive radiolucent line of =1 mm in thickness involving
=50% of the interface), type IV (a progressive circumferential radiolucent
line of >2 mm in thickness), or type V (a progressive circumferential
radiolucent line of >2 mm in thickness with evidence of prosthesis shifting
or subsidence).
Bushing wear was assessed by determining the angle formed between the
articular portion of the ulnar component and the humeral yoke on a true
anteroposterior radiograph. An angle of >7° indicated bushing wear in
the Coonrad-Morrey prosthesis. Bushing wear was deemed to be present in the
Pritchard-Walker prosthesis when this angle exceeded the 6° varus-valgus
laxity for the prosthesis. Bone-graft incorporation was assessed on the most
recent, best postoperative lateral radiograph and was graded as complete,
incomplete, or resorbed.
Statistical Methods
The paired Student t test was used to compare preoperative and
postoperative numerical data. Results were considered to be significant when
the likelihood of an event occurring by chance was <0.05.
Mayo Elbow Performance Score
The average Mayo Elbow Performance Score (MEPS) improved from 32.1 points
(range, 10 to 70 points) preoperatively to 80.4 points (range, 40 to 100
points) postoperatively (p < 0.001) (see Appendix). There were three
excellent, six good, one fair, and two poor results. One patient had a good
result according to the MEPS but was considered to have a poor result because
of the need for a revision of the prosthesis. A second patient had moderate
postoperative pain and an MEPS of 40 points. All patients had a better score
postoperatively than preoperatively, and all, including the two with a poor
result, were satisfied with the outcome and would be willing to have the
procedure again under similar circumstances.
Pain Relief
All but one patient rated the preoperative pain as severe. On a scale of 0
points (severe pain) to 45 points (no pain), the average preoperative pain
score was 3.8 points (range, 0 to 45 points). Following the semiconstrained
arthroplasty, the average pain score increased by an average of 27.5 points,
to 31.3 points (range, 15 to 45 points) (p < 0.0001). Postoperatively, the
pain was rated as mild or none by ten patients and moderate by two. No patient
had severe pain postoperatively.
Motion
Preoperatively, the patients had an average of 115° (range, 70° to
130°) of flexion, 37° (range, 0° to 100°) of extension,
67° (range, 40° to 85°) of pronation, and 67.5° (range, 0°
to 90°) of supination. Postoperatively, the average ranges of motion
improved to 129° (range, 95° to 145°) of flexion, 29° (range,
0° to 65°) of extension, 60° (range, 30° to 90°) of
pronation, and 65° (range, 30° to 90°) of supination. This
represented an average improvement in the flexion-extension arc from 78.3°
(range, 0° to 130°) preoperatively to 99.8° (range, 55° to
130°) postoperatively (p < 0.002).
Radiographic Analysis
Radiographic follow-up was performed for all but one patient at an average
of ninety-five months (range, twelve to 225 months) after the arthroplasty.
Nine of these radiographic evaluations were carried out at the time of or
after the latest clinical examination. Postoperative radiographs demonstrated
a well-fixed prosthesis in all but one patient. Nine patients had no evidence
of radiolucency, one patient had type-II radiolucency, and one patient had
type-IV radiolucency. Two patients had an articular component angle that was
greater than the normal varus-valgus laxity of the prosthesis, indicating the
presence of bushing wear; one of these patients was the patient with type-IV
radiolucency, who eventually required revision five years postoperatively.
This patient had a Pritchard-Walker prosthesis, which has no flange and has
been shown to have an increased rate of
loosening17. The
other patient had the evidence of bushing wear seven years after the
arthroplasty and required three bushing exchange procedures, at seven, ten,
and eighteen years after the primary total elbow arthroplasty (Figs.
2-A, 2-B, and
2-C). Both patients were
satisfied with the result following the revision procedures. With the numbers
available, no differences in loosening or wear rates could be demonstrated
among the types of prostheses. The bone graft beneath the anterior flange of
the prosthesis was incorporated in six of the seven patients in whom the
Coonrad-Morrey prosthesis had been used (Figs.
1-A, 1-B,
1-C, 1-D).
Complications
There were no surgical or perioperative complications and no infections.
Triceps weakness was noted in three patients at the time of follow-up, and one
patient with preoperative ankylosis had recurrent heterotopic ossification
postoperatively. Bushing wear developed in two patients, one of whom had
prosthetic loosening, as described above (Figs.
2-A, 2-B, and
2-C). In one patient, ulnar
nerve irritation developed postoperatively as a result of synovitis from
bushing wear.
This study demonstrated that reliable pain relief and restoration of
function can be attained with semiconstrained total elbow arthroplasty in
patients with a prior interposition arthroplasty performed for the treatment
of elbow arthritis. Ten of the twelve patients had a subjective satisfactory
result at an average of ten years postoperatively, with an average improvement
in the Mayo Elbow Performance Score from 32.1 points preoperatively to 80.4
points postoperatively. Our results compare favorably with those in several
reported series in which semiconstrained total elbow arthroplasty had been
performed for other
indications21-23.
Schneeberger and two of us (R.A. and B.F.M.) previously reported a good or
excellent result in 83% (thirty-four) of forty-one patients with posttraumatic
elbow arthritis1,
whereas two of us (B.F.M. and R.A.) reported a good or excellent result in 91%
(fifty-three) of fifty-eight elbows affected by rheumatoid arthritis and pain
relief in 91% (fifty-three) of the
elbows20. Ramsey
and two of us (R.A. and B.F.M.) recently reported a satisfactory result in
sixteen of nineteen patients in whom total elbow arthroplasty had been
performed to treat
instability21.
The results in the present series also compare favorably with the results
of revision semiconstrained total elbow arthroplasty. King and two of us (R.A.
and B.F.M.) reported the results of revision semiconstrained total elbow
arthroplasty in forty-one patients who had been followed for an average of six
years2. Twenty
patients had rheumatoid disease, and twenty-one had posttraumatic arthritis.
The average postoperative MEPS was 87 points, with 85% (thirty-five) of the
patients having a good or excellent result. Complications were reported in
eleven patients (27%), and eight (20%) had additional elbow surgery.
Although the rate of complications in this series was high, it is
comparable with those in other series. Fibrosis and scarring related to
multiple prior procedures (an average of two) in these patients may explain
the one case of ulnar nerve irritation as well as the triceps weakness noted
in three patients. While the average operative time was not excessive (156
minutes), the procedures took longer than primary total elbow arthroplasties,
which may also partly explain the increased rate of complications. The
recurrence of heterotopic ossification in one patient who had had preoperative
ankylosis suggests that total elbow arthroplasty should be performed with
caution in patients with severe stiffness following interposition
arthroplasty. Prophylaxis with indomethacin (75 mg of slow-release
indomethacin administered daily for three weeks) is currently recommended when
there is concern about heterotopic bone formation in these patients.
The prosthetic wear rates in these patients were comparable with the wear
rates in other
series2,20-23.
Schneeberger and two of us (R.A. and B.F.M.) found bushing wear in 5% (two) of
forty-one patients in whom semiconstrained arthroplasty had been performed for
the treatment of posttraumatic
arthritis1. In the
series of nineteen patients in whom semiconstrained arthroplasty had been
performed for the treatment of instability, Ramsey and two of us (R.A. and
B.F.M.) reported that no patient had evidence of bushing wear but one patient
had progressive radiolucency around the humeral
component21. In the
present series, postoperative radiographs demonstrated a well-fixed prosthesis
in all but one patient. Two patients had radiographic and clinical evidence of
bushing wear at five and seven years postoperatively. Extensive periprosthetic
osteolysis requiring revision with cortical strut grafts occurred in one of
those patients. That patient had a Pritchard-Walker implant, which has been
shown to have an increased rate of
loosening17. The
other patient, who had clinical failure and radiographic evidence of bushing
wear in a Coonrad-Morrey prosthesis, required three consecutive bushing
exchange procedures.
The limitations of the present study include the small number of patients
and the fact that different types of prostheses were used. These limitations
are a consequence of the rarity of both interposition arthroplasty and
revision total elbow arthroplasty. At a tertiary referral center for elbow
reconstruction, only sixteen patients presented with this indication over an
eighteen-year period. Three were excluded because they had an unconstrained
prosthesis, and an additional patient was excluded because of inadequate
follow-up. Because of the duration of the study period and the evolution of
elbow prostheses, changes in design were unavoidable. Because the Coonrad-II,
Coonrad-Morrey, and Pritchard-Walker semiconstrained prostheses are similar in
design, all three were included in an effort to provide a clinical sample of
adequate size to be studied.
The present study supports the role of semiconstrained total elbow
arthroplasty as a salvage procedure for patients in whom a prior interposition
procedure for the treatment of elbow arthritis has failed. On the basis of the
present and prior results of interposition arthroplasty, we currently consider
performing interposition arthroplasty in active, younger patients (less than
sixty years of age) who are reluctant to avoid strenuous physical activity.
Semiconstrained total elbow arthroplasty is performed in less active, older
patients (more than sixty years of age) or in those with a prior interposition
arthroplasty who no longer have satisfactory pain relief or elbow stability.
This review has shown that the intraoperative and postoperative complication
rates are comparable with those following a primary semiconstrained total
elbow arthroplasty performed for the treatment of noninflammatory elbow
arthritis. Revision rates are not higher than those following primary
arthroplasty, but exchange of the bushings may be required more
frequently.
Tables describing the preoperative and postoperative patient data are
available with the electronic versions of this article, on our web site at
(go to
the article citation and click on "Supplementary Material") and on
our quarterly CD-ROM (call our subscription department, at 781-449-9780, to
order the CD-ROM). ?
Schneeberger AG, Adams R, Morrey BF.
Semiconstrained total elbow replacement for the treatment of posttraumatic
osteoarthrosis. J Bone Joint Surg Am.1997;79:
1211-22.791211
1997
[PubMed]
King GJ, Adams RA, Morrey BF. Total
elbow arthroplasty: revision with use of a non-custom semiconstrained
prosthesis. J Bone Joint Surg Am.1997;79:
394-400.79394
1997
[PubMed][CrossRef]
Cheng SL, Morrey BF. Treatment of the
mobile, painful arthritic elbow by distraction interposition arthroplasty.
J Bone Joint Surg Br.2000;82:
233-8.82233
2000
[PubMed][CrossRef]
Cobb TK, Morrey BF. Use of distraction
arthroplasty in unstable fracture dislocations of the elbow. Clin
Orthop.1995;312:
201-10.312201
1995
Froimson AI, Silva JE, Richey D. Cutis
arthroplasty of the elbow joint. J Bone Joint Surg Am.1976;58:
863-5.58863
1976
[PubMed]
Kita M. Arthroplasty of the elbow using
J-K membrane. An analysis of 31 cases. Acta Orthop Scand.1977;48:
450-5.48450
1977
[PubMed][CrossRef]
Ljung P, Jonsson K, Larsson K, Rydholm
U. Interposition arthroplasty of the elbow with rheumatoid arthritis. J
Shoulder Elbow Surg.1996;5:
81-5.581
1996
[CrossRef]
Morrey BF. Distraction arthroplasty.
Clinical applications. Clin Orthop.1993;293:
46-54.29346
1993
[PubMed]
Oyemade GA. Fascial arthroplasty for
elbow ankylosis. Int Surg.1983;68:
81-4.6881
1983
[PubMed]
Park H, Moreau PF. Cases of the
excision of carious joints with observations by James Jeffrey.
Glasgow: J Scrymgeour; 1806.
1806
Regan WD, Reilly CD. Distraction
arthroplasty of the elbow. Hand Clin.1993;9:
719-28.9719
1993
[PubMed]
Shahriaree H, Sajadi K, Silver CM,
Sheikholeslamzadeh S. Excisional arthroplasty of the elbow. J Bone
Joint Surg Am.1979;61:
922-7.61922
1979
Smith MA, Savidge GF, Fountain EJ.
Interposition arthroplasty in the management of advanced haemophilic
arthropathy of the elbow. J Bone Joint Surg Br.1983;65:
436-40.65436
1983
[PubMed]
Wright PE 2nd, Burns EB. Resection and
anatomic elbow arthroplasty with and without interposition: indications and
results. Instr Course Lect.1991;40:
57-60.4057
1991
Modabber MR, Jupiter JB. Reconstruction
for posttraumatic conditions of the elbow joint. J Bone Joint Surg
Am.1995;77:
1431-46.771431
1995
Morrey BF. Posttraumatic contracture of
the elbow. Operative treatment, including distraction arthroplasty. J
Bone Joint Surg Am.1990;72:
601-18.72601
1990
Madsen F, Gudmundson GH, Sojbjerg JO,
Sneppen O. The Pritchard Mark II elbow prosthesis in rheumatoid arthritis.
Acta Orthop Scand.1989;60:
249-53.60249
1989
[PubMed][CrossRef]
Pritchard RW. Long-term follow-up study:
semi-constrained elbow prosthesis. Orthopedics.1981;4:
151.4151
1981
Pritchard RW. Total elbow arthroplasty
in patients with rheumatoid arthritis. Semin Arthritis Rheum.1991;21:
24-9.2124
1991
[PubMed][CrossRef]
Morrey BF, Adams RA. Semiconstrained
arthroplasty for the treatment of rheumatoid arthritis of the elbow. J
Bone Joint Surg Am.1992;74:
479-90.74479
1992
Ramsey ML, Adams RA, Morrey BF.
Instability of the elbow treated with semi-constrained total elbow
arthroplasty. J Bone Joint Surg Am.1999;81:
38-47.8138
1999
[PubMed]
Kraay MJ, Figgie MP, Inglis AE, Wolfe
SW, Ranawat CS. Primary semiconstrained total elbow arthroplasty. Survival
analysis of 113 consecutive cases. J Bone Joint Surg Br.1994;76:
636-40.76636
1994
[PubMed]
Schlein AP. Semiconstrained total elbow
arthroplasty. Clin Orthop.1976;121:
222-9.121222
1976
[PubMed]