Extract
While total shoulder arthroplasty has a successful clinical track
record for pain relief and improvement of function, there are substantial
concerns regarding the prevalence of glenoid lucent lines and therefore the
possibility of loosening of the glenoid component over the long
term1,2.
Options for treatment of symptomatic glenoid loosening include revision of the
component or conversion to a hemiarthroplasty by removal of the loose glenoid
component with or without bone-grafting of the
glenoid3. The
philosophy behind glenoid component removal is based on reports of humeral
hemiarthroplasty that have shown that it is indeed possible to have
satisfactory pain relief without a glenoid
component4-12.
The purpose of this report is to present an arthroscopic method for removing a
symptomatic loose glenoid component and the underlying cement mantle.
While total shoulder arthroplasty has a successful clinical track
record for pain relief and improvement of function, there are substantial
concerns regarding the prevalence of glenoid lucent lines and therefore the
possibility of loosening of the glenoid component over the long
term1,2.
Options for treatment of symptomatic glenoid loosening include revision of the
component or conversion to a hemiarthroplasty by removal of the loose glenoid
component with or without bone-grafting of the
glenoid3. The
philosophy behind glenoid component removal is based on reports of humeral
hemiarthroplasty that have shown that it is indeed possible to have
satisfactory pain relief without a glenoid
component4-12.
The purpose of this report is to present an arthroscopic method for removing a
symptomatic loose glenoid component and the underlying cement mantle.
This study was approved by our institutional review board, and all patients
had, on file, a signed consent form indicating their willingness to have their
data used in a research publication.
Between June 1995 and May 2000, five patients, sixty-seven to
seventy-eight years of age, required surgical intervention to treat a painful
failed total shoulder prosthesis with isolated loosening of the glenoid
component. Each patient had a relative contraindication against either
reimplantation of a glenoid component or the performance of a major operation
(Table I). All five patients
had received the primary total shoulder arthroplasty at our institution. The
right (dominant) shoulder was affected in three patients, and the left
(nondominant) shoulder was affected in two. The shoulder arthroplasty had
initially been performed to treat primary osteoarthritis in four patients and
osteoarthritis secondary to synovial osteochondromatosis in one.
The glenoid components were all standard (large) ultra-high molecular
weight polyethylene Cofield components (Richards, Memphis, Tennessee) that had
been inserted ten to 144 months prior to the removal. The components had all
been cemented after reaming of the glenoid face. The glenoid cavity had been
prepared with guides for reaming and drilling and then cleaned with pulsed
lavage. The glenoid cavity had been pressure-packed with epinephrine-soaked
sponges, and the cement had been inserted with finger-pressurization. In one
patient (Case 2), substantial medial erosion of the glenoid had been noted at
the time of the original implantation, and this had required trimming of the
keel to permit insertion. Osseous bleeding was difficult to control in that
patient, and a thin complete lucent line was apparent on the postoperative
radiographs. The other four patients demonstrated good initial bone-cement
interfaces.
The indication for revision surgery in each patient was severe pain and a
progressive, complete 1 to 2-mm-thick radiolucent line around the glenoid
component. One patient demonstrated a shift in the position of the glenoid
component. We chose to remove the glenoid component and convert the total
shoulder arthroplasty to a hemiarthroplasty either (1) to obviate the need for
a difficult open revision of the glenoid component or (2) because of a large
rotator cuff tear since such tears predispose the glenoid component to
loosening by the rocking-horse
phenomenon13.
Two of the procedures were done with the patient in the lateral
decubitus position, and three were performed with the patient in the
beach-chair position. We do not have a specific preference with regard to the
position of the patient. The operations were performed (with only slight
variation) according to the following steps. Standard posterior and anterior
portals are used initially, and accessory anterior and/or posterior portals
are made as required. As recommended by Hersch and Dines, blunt trocars are
used to avoid damage to the humeral
head14. A 7-mm
cannula is placed in the high anterior portal. We used a standard 30°,
4-mm arthroscope for all procedures. We found, as others
have14, that
orientation can be difficult because of mirror reflections off the polished
metal humeral head. The first step is to perform a local synovectomy and
débridement around the anterior, superior, and posterior margins of the
glenoid component and to confirm the looseness of the glenoid component with a
probe. Next, a 4 to 6-mm osteotome inserted through the anterior portal is
used to make three sets of cuts
(Fig. 1) so that the
prosthesis can be removed in four pieces. The first cut (cut 1a in Figs.
1 and
2) transects the glenoid
prosthesis diagonally from anterosuperior to posteroinferior
(Fig. 2). The second part (cut
1b) of that set divides the remaining intact polyethylene between the cut
portion of the faceplate and the underlying keel. The superior third of the
glenoid faceplate can then be withdrawn through the anterior portal by
grasping it at the tip and taking advantage of the wedge shape created by the
diagonal cut (Figs. 2 and
3). The second set of cuts
(cuts 2a and 2b in Figs. 1 and
3) detaches the inferior third
of the glenoid faceplate from the middle portion and from the keel
(Fig. 3). The osteotome is then
passed beneath the remaining middle third of the faceplate (cut 3 in Figs.
1,
4-A, and 4-B), separating it
from the keel (Fig. 4).
Finally, the keel is removed by backing it out with a twisting motion, apex
first. Strong graspers are recommended, as the soft tissues around the portal
need to stretch to permit extraction of the component. Prior to removing the
pieces, the surgeon should be certain that the portals are of adequate size (1
to 2 cm). A Ferris-Smith grasper, normally used to extract intervertebral disc
or intramedullary cement, is useful. The cement is then removed in pieces,
with the surgeon breaking it with a curved osteotome as necessary. A curet is
used to clean the glenoid cavity. A total synovectomy is performed if
extensive synovitis is present. Finally, a capsulotomy can be performed to
improve motion if necessary.
Several other techniques were tried without success in the first patient.
These included grinding the glenoid component with a burr and cutting it into
small pieces with bone-biting instruments such as a Ferris-Smith rongeur. We
found the plastic to be too difficult to cut with anything other than the
osteotomes. In one patient, the unwise decision was made to start by cutting
off the keel, but that rendered the remainder of the glenoid component
unstable and difficult to cut.
Postoperatively, the patients were given a sling for comfort for the first
few days, but they were encouraged to start moving the limb on the day after
the surgery and to pursue activities as tolerated. Two procedures were
performed on an outpatient basis. The other patients stayed in the hospital
overnight, principally because of medical concerns related to their ages and
general medical status.
From a technical perspective, the operations were successful in all
five patients, establishing that it is practical to remove a loose glenoid
component arthroscopically. In each patient, glenoid loosening was confirmed
with use of a probe to lift the component away from the underlying bone. The
cement mantle appeared to be fractured in all patients, although it is unclear
whether the fracture had occurred while the component was being cut or whether
it had existed previously. In three patients, the surgery took less than an
hour. The most difficulty was encountered in Case 2, who was obese. The
shoulder of this patient was so large that the arthroscope had to be inserted
all the way into it just to achieve visualization, and the switching stick
would not reach the skin on the other side of the shoulder. Blood loss was
negligible. The largest portal, which was 1.5 cm, was made in the first
patient, from whom the glenoid component was removed in three pieces. In the
subsequently treated patients, the portal size was kept standard and the
component was cut into four pieces (or five if necessary). Of the five
components, one was removed in three pieces; two, in four; and two, in
five.
The average duration of follow-up since the arthroscopic conversion to the
hemiarthroplasty was forty-nine months (range, thirty-four to eighty-three
months). From a clinical perspective, the procedures were beneficial. No
additional surgical procedures were performed in any of the five shoulders.
Subjective pain relief was partial (40% and 50%) in two patients and complete
in three. One patient had partial relief for almost two years but almost
complete relief by three years. Her situation was complicated by severe
lymphedema from prior radiation, and the lymphedema was worse for many months
after each shoulder operation. Functional improvement, not surprisingly,
paralleled pain relief; the three patients with complete pain relief gained
55° to 70° of elevation and 30° to 35° of external rotation.
The other two patients, with partial pain relief, did not gain motion.
Improvements in strength were noted but undoubtedly were related to pain
relief. No complications were encountered. Each patient believed that the
operation had been worth-while and stated that he or she would undergo it in
the future if faced with a similar decision.
There were no operative or perioperative complications except for the
temporary exacerbation of the pre-existing lymphedema in one patient (Case 5).
Final follow-up radiographs revealed mild-to-moderate medialization of the
humeral head in the glenoid, depending on the extent of the glenoid deficiency
(Figs. 5-A,
5-B, and
5-C); however, progressive
medial erosion was not evident. Remodeling of the bone at the base of the
glenoid cavity restored a concentric articulation for the humeral head in all
patients.
The options available for treating a loose glenoid component that is
symptomatic are either to revise it with another glenoid component or to
remove
it2,3,15.
Revision shoulder surgery is challenging, and removal of a glenoid component
with all of the underlying cement requires a large exposure, transection of
the subscapularis tendon, and opening of the rotator interval, all of which
create substantial morbidity. Performing the procedure percutaneously by
arthroscopic means can greatly reduce the morbidity. Also, because it can be
performed on an outpatient basis, a cost-savings can be expected. The concept
is practicable, but the questions that remain to be answered are: Is it
predictable and successful, and what are the indications, contraindications,
and complications?
We observed that the articular surface of the humeral component was usually
marked following this procedure. This occurred even though only plastic
cannulae and blunt trocars were used, as recommended by Hersch and
Dines14. The issue
of damage to total arthroplasty components from arthroscopic instruments has
received scant attention in the
literature14,16,
and its effect on the longevity of the implant is unknown. This may be an
important consideration if, in the future, a revision glenoid component needs
to be implanted. Theoretically, a scuffed or roughened humeral component
articulating with a polyethylene glenoid component could increase polyethylene
wear and the risk of early failure of the prosthesis. Therefore, later
reimplantation of a glenoid component should probably be combined with
revision of the humeral head, especially if the humeral component is modular.
At the time of writing, there were no plans to reimplant a glenoid component
in any of our five patients. A larger series with longer follow-up needs to be
evaluated to answer this question.
Little has been written about the potential role of arthroscopy in total
shoulder arthroplasty, and we found no evidence that it had been used to
remove a loose component. Bonutti et al. used arthroscopy to evaluate
loosening of the glenoid component in nine patients and found it to be a
valuable technique for assessing glenoid
stability17. They
confirmed their arthroscopic findings with a subsequent open procedure, with
which they addressed the glenoid component loosening. They recommended that
=2 mm of motion of the glenoid component, as seen arthroscopically,
indicates a loose component. We found that a better indicator of loosening is
the ability to insert a probe under the glenoid component and lift it away
from the bone, as described by Wirth and
Rockwood2.
Hersch and Dines reported the results of diagnostic arthroscopy in ten
patients with a total of twelve failed total shoulder
arthroplasties14.
Arthroscopic treatment included acromioplasty, débridement of the
biceps tendon, capsular release, and removal of loose bodies, including cement
fragments. They found arthroscopy to be useful not just diagnostically, but
also therapeutically. In our series, no additional procedures, such as
acromioplasty, other than synovectomy (a potentially important part of the
procedure) were performed. Thus, the benefits noted in our series could be
attributed to the removal of the loose glenoid component and cement.
The indications for arthroscopic removal of a loose glenoid component
together with its cement are uncommon. In most patients, a comprehensive
approach to revision of the components and/or soft-tissue reconstruction and
balancing is most appropriate. Our current indications for arthroscopic
glenoid removal are painful loosening of a polyethylene glenoid component in
the presence of a large rotator cuff
defect13 or a
relative contraindication (such as medical comorbidities) to a revision
arthroplasty. We would also consider this approach in a patient in whom
reimplantation of a glenoid component is expected to be exceptionally
difficult, such as an obese individual with a nonmodular humeral component.
Relative contraindications include the presence of a metal-backed glenoid
component, which requires a portal large enough for removal, and severe bone
loss that compromises the osseous rim of the glenoid such that containment of
the humeral head in the glenoid is not likely to be possible following removal
of the glenoid component. ?
Torchia ME, Cofield RH, Settergren CR.
Total shoulder arthroplasty with the Neer prosthesis: long-term results.
J Shoulder Elbow Surg.1997;6:
495-505.6495
1997
[PubMed][CrossRef]
Wirth MA, Rockwood CA Jr. Complications
of shoulder arthroplasty. Clin Orthop.1994;307:
47-69.30747
1994
[PubMed]
Cofield RH, Edgerton BC. Total shoulder
arthroplasty: complications and revision surgery. Instr Course
Lect.1990;39:
449-62.39449
1990
Bell SN, Gschwend N. Clinical experience
with total arthroplasty and hemiarthroplasty of the shoulder using the Neer
prosthesis. Int Orthop.1986;10:
217-22.10217
1986
[PubMed][CrossRef]
Boyd AD Jr, Thomas WH, Scott RD, Sledge
CB, Thornhill TS. Total shoulder arthroplasty versus hemiarthroplasty.
Indications for glenoid resurfacing. J Arthroplasty.1990;5:
329-36.5329
1990
[PubMed][CrossRef]
Field LD, Dines DM, Zabinski SJ, Warren
RF. Hemiarthroplasty of the shoulder for rotator cuff arthropathy. J
Shoulder Elbow Surg.1997;6:
18-23.618
1997
[CrossRef]
Groh GI, Badwey TM, Rockwood CA Jr.
Treatment of cysts of the acromioclavicular joint with shoulder
hemiarthroplasty. J Bone Joint Surg Am.1993;75:
1790-4.751790
1993
[PubMed]
Jonsson E, Brattstrom M, Lidgren L.
Evaluation of the rheumatoid shoulder function after hemiarthroplasty and
arthrodesis. Scand J Rheumatol.1988;17:
17-26.1717
1988
[PubMed][CrossRef]
Kay SP, Amstutz HC. Shoulder
hemiarthroplasty at UCLA. Clin Orthop.1988;228:
42-8.22842
1988
[PubMed]
Levine WN, Djurasovic M, Glasson JM,
Pollock RG, Flatow EL, Bigliani LU. Hemiarthroplasty for glenohumeral
osteoarthritis: results correlated to degree of glenoid wear. J
Shoulder Elbow Surg.1997;6:
449-54.6449
1997
[CrossRef]
Marmor L. Hemiarthroplasty for the
rheumatoid shoulder joint. Clin Orthop.1977;122:
201-3.122201
1977
[PubMed]
Williams GR Jr, Rockwood CA Jr.
Hemiarthroplasty in rotator cuff-deficient shoulders. J Shoulder Elbow
Surg.1996;5:
362-7.5362
1996
[CrossRef]
Franklin JL, Barrett WP, Jackins SE,
Matsen FA 3rd. Glenoid loosening in total shoulder arthroplasty. Association
with rotator cuff deficiency. J Arthroplasty.1988;3:
39-46.339
1988
[PubMed][CrossRef]
Hersch JC, Dines DM. Arthroscopy for
failed shoulder arthroplasty. Arthroscopy.2000;16:
606-12.16606
2000
[PubMed][CrossRef]
Petersen SA, Hawkins RJ. Revision of
failed total shoulder arthroplasty. Orthop Clin North Am.1998;29:
519-33.29519
1998
[PubMed][CrossRef]
Raab GE, Jobe CM, Williams PA, Dai QG.
Damage to cobalt-chromium surfaces during arthroscopy of total knee
replacements. J Bone Joint Surg Am.2001;83:
46-52.8346
2001
[PubMed]
Bonutti PM, Hawkins RJ, Saddemi S.
Arthroscopic assessment of glenoid component loosening after total shoulder
arthroplasty. Arthroscopy.1993;9:
272-6.9272
1993
[PubMed][CrossRef]