 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1999 The Journal of Bone and Joint Surgery, Inc. Impaction Allograft Bone-Grafting for Revision Total Elbow Arthroplasty. A Case Report* , BIRMINGHAM, ALABAMAInvestigation performed at the Division of Orthopedic Surgery, University of Alabama at Birmingham, Birmingham
A failed total elbow arthroplasty associated with extensive bone resorption as well as cortical expansion and thinning is a difficult reconstructive problem. Treatment methods such as elbow arthrodesis, interposition arthroplasty, and resection arthroplasty either result in a grossly unstable elbow or are technically difficult to perform5,14-16. When a failed elbow arthroplasty is associated with a limited area of bone loss, a standard long-stem semiconstrained implant can bypass the area of bone deficiency2,4,9,11,14,16. However, the bone deficiency may be so extensive that it is not possible to bypass the area of bone loss with a standard stem. If a standard-length implant is used in the presence of extensive bone loss, severe shortening of the extremity may result. Additionally, the proximal tip of the humeral implant would be placed in close proximity to the humeral head, which could, in the future, preclude the use of a shoulder prosthesis, if needed. Therefore, in instances of major bone loss, custom elbow implants, massive allografts, or implant-allograft composites may be necessary1-6,9,10,13,15,16,21. In the present report, I describe the use of a technique involving the impaction of cancellous bone graft to compensate for inadequate bone stock in a patient who had a failed total elbow arthroplasty associated with extensive bone loss. This method has been used to compensate for inadequate bone stock encountered in revision total hip and knee arthroplasties7,8,12,17-19.
A fifty-two-year-old right-hand-dominant man was seen because of a two-month history of progressive pain and swelling about the left elbow seventeen years after a left total elbow arthroplasty. The pain and swelling interfered with the patient's activities of daily living, but he was able to work as his occupation was sedentary in nature. The arthroplasty had been performed to treat a comminuted fracture of the elbow sustained in a motor-vehicle accident. A first-generation Coonrad-Morrey prosthesis had been used. (This prosthesis was designed in 1971 with a rigid hinge. In 1978, the design was changed to a semiconstrained model that allowed 5 to 10 degrees of play.) Physical examination revealed a tender prominence along the anterior aspect of the distal third of the humerus. The elbow was noted to be grossly unstable. A range-of-motion examination of the elbow demonstrated full extension (0 degrees), 130 degrees of flexion, 30 degrees of pronation, and 20 degrees of supination. Radiographs of the elbow revealed gross loosening of both the humeral and the ulnar component (Figs. 1-A and 1-B). The humeral component had penetrated the anterior cortex, and a palpable mass was present anteriorly over the junction of the distal and middle thirds of the humerus.
A revision elbow arthroplasty was performed through a posteromedial triceps-sparing Bryan-Morrey approach. The ulnar nerve was identified, protected, and anteriorly transposed. There was gross loosening of both the humeral and the ulnar component, which allowed for easy removal of the prosthesis. Cortical defects were noted in the proximal aspect of the ulna and the anterior aspect of the humerus. Loose bone fragments and remaining cement were removed from the humeral and ulnar canals. An extraperiosteal soft-tissue dissection of the proximal aspect of the ulna was performed, with release of the interosseous membrane and soft-tissue attachments to the proximal aspect of the ulna. The ulnar nerve was identified and protected. The anterior neurovascular structures were not dissected out of the anterior soft-tissue envelope. A similar circumferential, extraperiosteal soft-tissue dissection of the humeral shaft was performed. The medial and lateral intermuscular septa were released from the humerus. The radial nerve was not isolated; however, since it was not visualized, the nerve was assumed to have been elevated from the humeral shaft within a soft-tissue envelope. The distances from the proximal end of the ulnar shaft and the distal end of the humeral shaft to the cortical defects were measured. Reconstructive wire mesh (Zimmer, Warsaw, Indiana) made of wrought chromium-cobalt alloy was cut to the appropriate length to cover the cortical defects and to fit circumferentially around both the proximal part of the ulnar shaft and the distal part of the humeral shaft. The soft tissue was elevated off the surface of the bone with periosteal elevators during placement of the wire mesh; care was taken to ensure that the wire mesh was placed directly over the cortex so that no soft tissue was interposed between the wire mesh and the bone surface. Cerclage wires were placed approximately one inch (2.5 centimeters) apart to hold the wire mesh evenly along the cortical surface. An impaction cancellous allograft bone-grafting technique was used to reconstruct the distal aspect of the humerus and the proximal aspect of the ulna. Cancellous bone graft obtained from the iliac crest was combined with morseled cancellous allograft. A bone plug was placed in the humeral canal; the ulnar canal did not need a bone plug because of its small diameter. Cancellous bone graft was tightly packed into the humeral and ulnar canals. A third-generation Coonrad-Morrey semiconstrained elbow implant (Zimmer) was used for the revision arthroplasty. (This model was developed in 1981 and had a flange added to the semiconstrained model. The ulnar and humeral components had a partially bead-coated surface to provide for bone ingrowth, particularly behind the flange.) A standard six-inch (15.2-centimeter) humeral component and a custom extra-long-stem ulnar component were used. A trial humeral component and the custom ulnar component were placed in the center of the humeral and ulnar canals, respectively. With use of a bone impactor, cancellous bone graft was tightly packed around both components, which created new medullary canals. The final articulated humeral and ulnar components were cemented into the newly created medullary canals with use of tobramycin-impregnated methylmethacrylate. A tricortical bone graft from the iliac crest was placed posterior to the anterior humeral flange at the time of insertion of the final components. Closure of the wound included repair of the reflected triceps soft-tissue sleeve. No attempt was made to repair the soft tissues that had been reflected off the proximal aspect of the ulna or the distal aspect of the humerus. Postoperatively, the patient wore an above-the-elbow splint for five days, after which active flexion and extension exercises were begun. A protective, removable, above-the-elbow splint was used constantly (except when the patient was exercising or bathing) for six months and then was used intermittently, during stressful activities, for one year. The patient was instructed to permanently restrict lifting with the left upper extremity to five pounds (2.3 kilograms) or less. At the thirty-month follow-up visit, the patient reported infrequent episodes of mild pain in the elbow. The range of motion of the elbow was 0 degrees of extension to 110 degrees of flexion (Figs. 2-A and 2-B), 40 degrees of pronation, and 50 degrees of supination. The patient was able to perform sedentary activities, including the tasks associated with his occupation in the field of business. There were no signs of loosening of the components on radiographs made at this visit. At some point between the twenty-four and thirty-month follow-up visits, the split locking ring to the bolt connecting the humeral and ulnar components had loosened; the patient had been unaware of this event. The humeral and ulnar components remained well articulated at the latest follow-up evaluation (Figs. 3-A and 3-B).
A total elbow implant may fail because of infection or because of failure, instability, or loosening of the device5,14-16. Mechanical loosening of either the humeral or the ulnar component may be associated with fracture, bone resorption, or cortical expansion, perforation, or thinning9,14-16. When there is only a moderate amount of bone loss and cortical perforation, a standard long-stem implant may be used to bypass cortical defects. These implants may be supplemented with autogenous bone graft placed in the cortical defects, with or without internal fixation of the fracture14,15. Morrey and Bryan15 reported on a series of thirty-three revision total elbow arthroplasties. Moderate bone loss (only supracondylar bone remaining) or severe bone loss (only the humeral diaphysis remaining) was noted in twenty-nine of the thirty-three elbows. Most (twenty-nine) of the revision arthroplasties were performed with use of six different commercially available prostheses; custom implants were used in four elbows. After a minimum duration of follow-up of three years, eighteen (55 percent) of the elbows had a good result and fifteen (45 percent) had a poor result. Another revision procedure was performed in the fifteen elbows that had a poor result, and eight of them eventually had a good result. King et al.11 reported on a series of forty-one revision total elbow arthroplasties. Nineteen elbows had grade-II bone loss (preservation of the medial and lateral supracondylar columns), nine elbows had grade-III bone loss (absence of either the medial or the lateral supracondylar column), and thirteen elbows had grade-IV bone loss (absence of the entire distal end of the humerus). Twelve elbows had no olecranon process. In three elbows, the tip of the humeral component perforated the humeral shaft. There were seventeen intraoperative complications; most were cortical perforations or fractures that occurred during removal of the cement. The forty-one elbows were revised with use of a standard or long-stem humeral component, and all but one received a standard ulnar component; in the remaining elbow, a commercially available long-stem ulnar component was inserted. Internal fixation (with either a plate or cerclage wire) was used in addition to a long-stem component to bypass any fractures or cortical perforations. Autogenous bone-grafting was performed to supplement cortical defects and fractures. In two patients, an allogenic femoral head graft was used to reconstruct a deficient olecranon; one of these patients had a nonunion of the allograft. Three patients had an injury of the radial nerve: two injuries were secondary to extravasation of cement, and one occurred when cement was removed during the revision procedure. One patient had a partial injury of the ulnar nerve. After an average duration of follow-up of six years, thirty-eight of the forty-one patients were able to perform activities of daily living. Eight patients had an intraoperative or a postoperative complication that was treated with an additional procedure. In both of those series11,15, the bone defects consisted of either cortical perforations or the loss of some portion of the distal part of the humerus. In contrast, severe cortical thinning and expansion was noted in my patient. Revision of a failed total elbow arthroplasty associated with extensive cortical expansion, thinning, or loss, with or without perforations, can be difficult. The options for revising a failed prosthesis in an elbow with extensive bone loss include the use of custom elbow implants, massive allografts, or implant-allograft composites1-6,9,10,13,15,16,20,21. Problems that have been associated with custom elbow implants include high cost, a delay while the implant is being manufactured, a component that does not fit as well as desired, and a persistently high rate of loosening4,16. Figgie et al.4 reported on a series of twenty revision elbow arthroplasties in which custom-designed implants were used in patients who had severe loss of bone or soft tissue. Revision of a loose total elbow prosthesis was successful in only three of five patients. Failure of the revision implants was attributed to lack of metaphyseal bone support, loss of a periarticular soft-tissue envelope, malalignment of the implant, and latent infection. Dent et al.3 reported on twenty-six revision elbow arthroplasties, in twelve of which a custom implant was used because of varying amounts of bone loss and soft-tissue disruption. Four of the revision implants, including one custom implant, loosened. There have been several reports in which limbs with a failed total elbow prosthesis associated with massive bone loss have been salvaged successfully with use of allograft bone or implant-allograft composites as replacement1,6,10,13,16,20,21. However, bone resorption, fracture, nonunion, infection, production of a neuropathic joint, and loosening of the components remain potential problems10,13,16,20. Techniques involving impaction allograft bone-grafting have been used in revision total hip and total knee arthroplasties in patients with massive bone defects7,8,12,17-19. Simon et al.17 reported the use of such a technique in seventy-two revision total hip arthroplasties in patients who had severe loss of proximal femoral bone stock. In thirty-eight patients who had been followed for more than one year, a variable degree of graft incorporation was noted; the results were thought to be probably related to the efficiency of the initial packing of the graft and of the cementing. Gie et al.8 subsequently reported on a series of fifty-six hips that had been revised with an impaction allografting technique. There were two intraoperative fractures, both of which were treated with open reduction and fixation with a plate followed by immediate impaction grafting and cementing of the femoral stem. One case of intraoperative perforation by the femoral stem was treated similarly. Overall, after a duration of follow-up of between eighteen and forty-nine months, most patients had a satisfactory result. Nine of thirteen hips that were followed for more than three years showed no signs of radiolucency around the implant. Ullmark and Hovelius19 reported the use of an impaction allografting technique to revise three failed total knee arthroplasties. They reported good clinical and radiographic results after durations of follow-up of eighteen, twenty-one, and twenty-eight months. I am not aware of any previous reports of impaction allograft bone-grafting for revision of a failed total elbow arthroplasty in a patient with severe bone defects. This technique is particularly applicable to elbows with severe cortical thinning, expansion, resorption, or perforation, or a combination of these conditions. It should be noted that the technique requires that a shell of cortical bone remain in place so that wire mesh can be placed around it. The technique makes it possible to use standard-length implants. Cortical perforations or defects are commonly encountered during revision elbow arthroplasties15,16. Extrusion of bone cement through cortical perforations11 or extensive dissection of the humeral shaft20 can produce injuries of the radial nerve. The use of wire mesh and impacted allograft bone may prevent extrusion of cement. Although the radial nerve was not specifically dissected in our patient, the nerve should be identified if there is any question of its location during placement of the wire mesh. Intraoperative fractures can be supplemented with cortical strut-grafting or plate fixation and impaction bone-grafting. On the basis of the results in the patient described in the present report, it appears that the index operation provides a method of salvaging a limb with a failed elbow arthroplasty associated with extensive bone loss.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
  CiteULike  Connotea  Del.icio.us  Facebook  Technorati  Twitter What's this?
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Introduction
Case Report
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |