Revision shoulder arthroplasty and complications

Issue: December 2004

ByTom R. Norris, MD

Tom R. Norris, MD,
is co-director for the San Francisco Shoulder, Elbow and Hand Fellowship at California Pacific Medical Center, is in private practice, and has a voluntary academic appointment at UCSF, San Francisco.

In 2003, 25,000 shoulder arthroplasties were performed in the United States, whereas in the early 1990s, fewer than 10,000 procedures were performed annually. Approximately 85% of shoulder arthroplasties are performed by surgeons who complete one to three arthroplasties each year.¹ Surgeons argue that the number of complications requiring revision increases as the number of arthroplasties performed increases and a large pool of recipients build. Others believe that low volume is associated with more complications.² Often, multiple reasons for revision in a given case make classification problematic.

History of arthroplasty

Frank E. Stinchfield, MD, said, “The art of total hip arthroplasty will not be in doing the primary cases; the skill will be in the revisions.”³ In the beginning, the age restriction for hip arthroplasty was 65 years until a 16-year-old wheelchair bound patient presented with juvenile rheumatoid arthritis. She could walk after her bilateral total hip arthroplasties. The age restriction was then lifted. The same reasoning applies to shoulder arthroplasty where pathology dictates the need for treatment.

Considerations

Many principles from the hip apply to the shoulder, even as the reverse prosthesis is added to protocols. Pain and pathology are prime indicators for any surgery. Indications for shoulder revision surgery include glenoid arthritis, periprosthetic humeral fracture, glenoid loosening, osteolysis, wear, capsular contracture, proximal humerus fracture complications, instability and a deficient or nonfunctional rotator cuff. The most common cause for revision of a humeral head arthroplasty is glenoid arthritis caused by inadequate centering, sizing and stiffness or malpostion. Revision may also be due to a nonanatomic prosthesis (Figure).

Figure

Glenoid arthrosis caused by inadequate centering, sizing and stiffness or malposition (A) and non-anatomic prosthesis (B).

Figures courtesy of Tom R. Norris, MD.

Surgeons debate the indications for glenoid resurfacing in shoulder arthroplasty. Although the literature supports glenoid resurfacing in most patients, considerations must be made. A high rate of glenoid loosening has been noted in patients with rotator cuff deficiencies, including patients with rheumatoid arthritis, and in those with cuff tear arthropathy due to the eccentric forces placed on the superior glenoid rim.

Total shoulder arthroplasty vs. humeral head replacement

Levine and colleagues⁴studied 31 shoulders for glenohumeral osteoarthritis after undergoing hemiarthroplasty. Glenoid surface wear was evaluated. Postoperative results showed 23 shoulders with satisfactory results and eight shoulders with unsatisfactory results. Outcome correlated most significantly with the status of posterior glenoid wear. On the basis of these results, hemiarthroplasty can be an effective treatment for primary and secondary arthritis but should be reserved for patients with a concentric glenoid, which affords a better fulcrum for glenohumeral motion. At later follow-up of five to seven years, even shoulders that were doing well early on with humeral head replacements continued to deteriorate. Results with the humeral head replacement with concentric glenoids which had 80% good results at two years had 63% satisfactory results by five years. This is why biological resurfacing is emerging as an option for the younger patient with osteoarthritis.

Total shoulder arthroplasty vs. humeral head replacement should also be considered. In a study by Iannotti and colleagues,⁵ 128 shoulders in 118 patients with primary osteoarthritis who had been followed for a mean of 46 months were evaluated. On the basis of that data, researchers recommended the use of a glenoid component in shoulders with glenoid erosion. Humeral head subluxation was associated with a less favorable result regardless of the type of shoulder arthroplasty and must be considered in preoperative planning and counseling. Severe loss of the passive range of motion preoperatively was associated with a decreased passive range of motion postoperatively. A repairable tear of the supraspinatus tendon is not a contraindication to the use of a glenoid component.

A glenoid component is preferred if there is enough bone stock. Placing the glenoid is more difficult in a revision situation. A surgeon must reshape, graft or resurface with biological substitutes or polyethylene. Gartsman and colleagues⁶performed a prospective study showing that total shoulder arthroplasty provided superior pain relief compared with hemiarthroplasty in patients who had glenohumeral osteoarthritis, but total shoulder arthroplasty was associated with an increased cost per patient based on slightly increased operating room time and component cost. By contrast, the cost of revision surgery to add the glenoid resurfacing later is considerably higher, and it may well be more economical to spend a little more with a total shoulder arthroplasty as a primary procedure.

Edwards and colleagues⁷ performed a multicenter retrospective study and found patients undergoing total shoulder arthoplasty had significantly better outcome scores compared to patients undergoing humeral head replacement. Total shoulder arthroplasty provided better scores for pain, mobility and activity than hemiarthroplasty. Fifty-six percent of total shoulder arthroplasties had a radiolucent line around the glenoid component. Researchers also concluded that total shoulder arthroplasty provides results superior to those of hemiarthroplasty in primary osteoarthritis.

Periprosthetic fractures

The incidence of periprosthetic fracture during or after shoulder arthroplasty is 1% to 3% of all shoulder arthroplasties⁸ Literature on this subject is limited — the largest series is by Campbell and colleagues,⁹ who reviewed 21 periprosthetic humeral fractures retrospectively. Periprosthetic fracture risk factors include osteopenia and avoidable technical factors such as overzealous reaming and impaction, excessive external rotation and an overly large prosthesis stem. Soft tissues, which increase technical demand of exposure, and postoperative trauma are also risk factors for periprosthetic fractures. Treatment options include nonoperative procedures, implant-sparing techniques and supplemental fixation or a longer stem prosthesis that traverses the fracture site by two to three cortical diameters.

Various nonoperative treatment options are available. Nonoperative treatment may take months to heal periprosthetic fractures and rehabilitation is prolonged. Most fractures treated conservatively heal in seven months, on average.¹⁰ Improved stability with long-stem prosthesis is the best option in intraoperative fractures at the distal prosthesis. Postoperative fractures may be different when choosing to change the prosthesis. A surgeon may take down the subscapularis. Removal of cemented repair may require shaft episiotomy or an Ultradrive ultrasound device (Biomet, Warsaw, Ind), but still may be the best choice.

A surgeon may perform implant-sparing techniques such as open reduction internal fixation with plate construct and cables or with long cortical allograft bone plates and cables. Screw purchase in osteopenic bone can be problematic. Cables can cut through the thin cortical bone. Newer locking plates may offer more secure fixation to lessen the nonunion risks of traditional plates.

Conversion to a long-stem prosthesis with a supplemental fixation is a desirable construct option. Advantages of a longer prosthesis include no need for secure screw purchase, bending and torsion loads are better tolerated and the intramedullary location confers a biomechanical advantage.

Capsular contracture

Stiffness is a significant component of many failed arthroplasties. Stiffness can be due to prosthetic malposition, oversizing of the humeral component or inadequate rehabilitation. Treatment requires defining and addressing each of these factors. With the development of the third-generation prostheses based on normal humeral anatomy, the head sizes can now be essentially anatomic. Then, it becomes a matter of balancing the soft tissues with releases or plications to maximize motion while preserving stability, which is preferable to over- or undersizing components. In revision surgery, a malpositioned component may need to be resized or repositioned. If the complication is inadequate rehabilitation, then releases with minimum disruption of the cuff are the goal. A manipulation may be considered but risks humeral fracture or unwanted cuff dehiscence. An arthroscopic release would be ideal if damage to the surface of the prosthetic components can be avoided. It is unfortunate that stiffness may preclude safe passage of metal cannulas and instruments. Scratching or gouging of the cobalt chrome or polyethylene will substantially shorten the time an implant will function.

An alternative approach has been to open the rotator interval through the deltopectoral approach. This facilitates subacromial and deltohumeral extracapsular releases, followed by intracapsular releases around the glenoid. The anterior capsular structures are first released. Using a sponge or plastic retractor to protect the articular head surface, the humeral component can be pushed anterior-inferiorly to expose and release any tight capsule posteriorly until a circumferential capsular release has been achieved. Because the subscapularis remains intact and the exposure is through a generous opening in the rotator interval, motion can be started one day after surgery without risk of cuff dehiscence.

The most significant factors associated with poor results are tuberosity malposition and migration in the treatment of acute four-part displaced fractures. Prophylactically, tuberosity dehiscence may be avoided or decreased if the unpolished holes in the fins of many prosthetic components are avoided. These stress risors easily rupture suture fixation and allow the tuberosity to displace. Circumferential sutures around the polished upper humeral stem of specifically developed fracture prostheses have decreased this complication from 50% to 25%. Once a displaced greater tuberosity is recognized, it should be repaired early. After six months, this is not retrievable. A surgeon, then, is left with a conversion to a reverse prosthesis to bypass the cuff tuberosity combination that is otherwise irreparable.

Glenoid loosening

Radiolucent lines are common but do not necessarily predict a loose component. Progression of radiolucent lines correlates with a decrease in function and an increase in pain. Definite signs of loosening include circumferential radiolucent lines more than 1.5 mm and a shift in position of the glenoid component. As experience with hylamer and more rapid osteolysis has shown, the bone loss around a loose glenoid component is often worse than can be appreciated by radiographs. Oteolysis is essentially silent until a near catastrophic event is about to occur. The clinical picture is often one with a progressive increase in the radiolucent line around the glenoid and is accompanied by an increase in discomfort and gradual decrease in active motion. A surgeon may not be doing the patient a favor by waiting to perform the revision. With revision, replacement of the glenoid gives the best result, but if too much bone has eroded, then grafting the defect is preferred. Walch and Edwards have demonstrated the advantages of a tricortical iliac crest bone graft with good incorporation and minimal subsidence. In the few patients who still have significant pain, I have been able to sucessfully address this with reimplanting a glenoid six months to one year later. With contained defects, autografts or allografts reoptions, whereas if the anterior or posterior glenoid wall is missing, the tricortical iliac graft is preferred.

Reverse prosthesis

The latest technique in complex shoulder revision is the Aequalis reverse prosthesis (TORNIER, St Ismier, France). Walch and colleagues studied 54 patients undergoing revision arthroplasties from 1995 to 2003. A surgeon can help patients who have a nonfunctional rotator cuff or a dislocation that could not be fixed with a subscapularis repair or pectoralis transfer. With the new reverse prosthesis, a surgeon has a solution for difficult fracture problems and failures with humeral head replacement for cuff tear arthropathy when there is insufficient rotator cuff.

The reverse prosthesis tensions the deltoid and provides a stable fulcrum for the deltoid to elevate the arm. Therefore, the prosthetic device bypasses or substitutes in part for the deficient cuff. Normally, the rotator cuff centers the head in the glenoid. With increasingly larger cuff tears, this function is lost. The deltoid may be ineffective with superior humeral head migration or escape. The newer reversed prosthetic designs have the advantage of placing the center of rotation on the face of the scapula rather than in the center of the humeral head. This accounts for the stability of the uncemented glenoid sphere on the scapula when compared to the high rate of loosening of previous designs from 30 years ago. In older designs, the center of rotation was more lateral and had higher forces on the frail scapula. The reverse prosthesis has been used for 18 years in Europe and recently has been approved by the Food and Drug Administration for use in the United States.

Summary

To prevent fracturing the humerus, surgeons should avoid excessive torque on the humerus during reaming and perform releases before reaming the humerus or glenoid. Periprosthetic fractures are best salvaged with longer stemmed implants. Allograft cortical struts and cables provide immediate bone support and aid in load dispersal. Glenoid osteolysis is best treated with component revision, although bone grafting is an option when bone stock is not adequate for a glenoid component. With similar bone loss when converting to a reverse prosthesis, revision can often be performed in one stage with the baseplate holding the intercalary bone graft with fixation to the remaining scapula.

The most common humeral head replacement revision is glenoid arthrosis. Surgeons must not perform humeral head replacement alone for patients with significant posterior wear of the glenoid. Surgeons should avoid stress risor with prosthetic fins for tuberosity fixation in proximal humerus fractures. Complications are an inevitable part of working with new technology, but careful indications and technique and appropriate use of the technology will continue to decrease complication rates and improve patient outcomes.

References

Hasan S, Leith J, Smith KL, Matsen III FA. The distribution of shoulder replacements among surgeons and hospitals is significantly different than that of hip or knee replacements. J Shoulder Elbow Surg. 2003;12:164-169.
Hammond JW, Queale WS, Kim TK, McFarland EG. Surgeon experience and clinical and economic outcomes for shoulder arthroplasty. J Bone Joint Surg Am. 2003;85:2318-2324.
Personal communication.
Levine WN, Djurasovic M, Glasson JM, et al. Hemiarthroplasty for glenohumeral osteoarthritis: Results correlated to degree of glenoid wear. J Shoulder Elbow Surg. 1997;6:449-454.
Iannotti JP, Norris TR. Influence of preoperative factors on outcome of shoulder arthroplasty for glenohumeral osteoarthritis. J Bone Joint Surg Am. 2003;85:251-258.
Gartsman GM, Roddey TS, Hammerman SM. Shoulder arthroplasty with or without resurfacing of the glenoid in patients who have osteoarthritis. J Bone Joint Surg Am. 2000;82:26-34.
Edwards TB, Kadakia NR, Boulahia A, et al. A comparison of hemiarthroplasty and total shoulder arthroplasty in the treatment of primary glenohumeral osteoarthritis: Results of a multicenter study. J Shoulder Elbow Surg. 2003;12:207-213.
Williams GR Jr, Iannotti JP. Management of periprosthetic fractures: The shoulder. J Arthroplasty. 2002;17:14-16.
Campbell JT, Moore RS, Iannotti JP, Norris TR, Williams GR. Periprosthetic humeral fractures: Mechanisms of fracture and treatment options. J Shoulder Elbow Surg. 1998;7:406-413.
Kligman M, Roffman M. Humeral fracture following shoulder arthroplasty. Orthopedics. 1999;22:511-513.