Kudo type-5 total elbow arthroplasty in mutilating rheumatoid arthritis

Upper limb Kudo type-5 total elbow arthroplasty in mutilating rheumatoid arthritis A 5- TO 11-YEAR FOLLOW-UP T. Mori, H. Kudo, K. Iwano, T. Juji From the National Hospital Organization Sagamihara Hospital, Kanagawa, Japan We studied 11 patients (14 elbows) with gross rheumatoid deformity of the elbow, treated by total arthroplasty using the Kudo type-5 unlinked prosthesis, and who were evaluated between five and 11 years after operation. Massive bone defects were augmented by autogenous bone grafts. There were no major complications such as infection, subluxation or loosening. In most elbows relief from pain and stability were achieved. The results, according to the Mayo Elbow Performance Score, were excellent in eight, good in five and fair in one. In most elbows there was minimal or no resorption of the grafted bone. There were no radiolucent lines around the stems of the cementless components. This study shows that even highly unstable rheumatoid elbows can be replaced successfully using an unlinked prosthesis, with augmentation by grafting for major defects of bone. T. Mori, MD, Head of Department, Orthopaedic Surgeon H. Kudo, MD, PhD, President Emeritus of Sagamihara Hospital K. Iwano, MD, PhD, Orthopaedic Surgeon T. Juji, MD, Chief Orthopaedic Surgeon Department of Orthopaedic Surgery National Hospital Organization, Sagamihara Hospital, 18-1 Sakuradai, Sagamihara City, Kanagawa Prefecture 228-8522, Japan. Correspondence should be sent to Dr T. Mori; e-mail: t-mori@sagamihara-hosp.gr.jp 2006 British Editorial Society of Bone and Joint Surgery doi:10.1302/0301-620x.88b7. 17356 $2.00 J Bone Joint Surg [Br] 2006;88-B:920-4. Received 28 October 2005; Accepted after revision 28 February 2006 In elbows severely involved by rheumatoid arthritis, developments in total joint replacement have led to reliable clinical results, including good pain relief, restoration of stability and improved joint movement. However, reconstruction of a very unstable elbow with marked bone loss remains a very difficult operation. In choosing an implant for such a procedure, the trend is towards a linked, semiconstrained prosthesis, as described by Morrey and Adams. 1 A total of seven of 58 elbows in their series had severe destruction of the joint with gross instability. They published details of further such arthroplasties, including two with fractures of the distal humerus in 1999. 2 In 1998 we described the use of an unlinked resurfacing implant in six elbows with a mutilans deformity. 3 The mean follow-up was 4.5 years (2 to 8) and all were satisfactory. Notwithstanding these good results, the same type of operation has not been performed elsewhere. The aim of this paper is to reinforce the value of the operation by presenting the longterm outcome in 14 elbows with a total arthroplasty using the Kudo type-5 prosthesis (Biomet UK Ltd, Bridgend, Wales). Patients and Methods Between 1993 and 1999, we carried out a total elbow arthroplasty on 120 patients (139 elbows) with severe rheumatoid arthritis using the Kudo type-5 prosthesis. The humeral component is made of cobalt-chromium alloy. A portion of the stem is porous-coated with a plasma spray of titanium alloy, thereby allowing cementless use. The ulnar component may be polyethylene alone or metal-backed with a porous-coated stem, the latter designed mainly for cementless use. In 1999, we reported a satisfactory clinical outcome in elbows with severe rheumatoid arthritis using this prosthesis at a mean follow-up of 3 years and 10 months (2 years, 6 months to 5 years, 6 months). 4 In the current series, all patients met the diagnostic criteria of the American Rheumatism Association 5 and had pre-operative radiological change of Larsen grade 3, 4, or 5. 6 A total of 22 elbows (18 patients) had grade 5 changes pre-operatively. Severe bone defects were filled with autogenous grafts (20 elbows, 16 patients) or acrylic cement. We describe mainly the clinical results of the elbows which had bone graft. Two patients had died within five years of operation and another four elbows (three patients) were lost to follow-up. However, all six elbows had shown good clinical and radiological results without complications. The remaining 11 patients (14 elbows) were the subject of this study (Table I). Their mean age at the time of surgery was 53.3 years (27 to 67) and the mean duration of disease was 19.7 years (12 to 30). The patients were followed up for a mean of 7 years and 7 months (5 years to 11 years, 3 months). The pre-operative radiographs showed an almost complete loss of the trochlea and vari- 920 THE JOURNAL OF BONE AND JOINT SURGERY

KUDO TYPE-5 TOTAL ELBOW ARTHROPLASTY IN MUTILATING RHEUMATOID ARTHRITIS 921 Table I. Patient details Case Gender Age (yrs) Duration of RA * (yrs) Side Larsen grade 6 Fixation 1-L F 59 27 L 5 Cementless 135 1-R F 62 30 R 5 Hybrid 91 2 F 56 22 R 5 humeral fracture Cemented 108 3 F 59 15 R 5 Cementless 110 4 F 53 12 R 5 Hybrid 96 5 F 58 13 L 5 Hybrid 132 6 F 42 22 L 5 Hybrid 60 7-L F 60 19 L 5 Hybrid 92 7-R F 62 22 R 5 Hybrid 63 8 F 52 22 L 5 humeral fracture Cemented 84 9 F 61 28 L 5 Hybrid 60 10-R F 27 15 R 5 Hybrid 100 10-L F 28 15 L 5 Hybrid 97 11 F 67 14 L 5 Hybrid 62 * RA, rheumatoid arthritis Follow-up (mths) Bone graft Bone graft The operation has been described previously. 3,4,7 All 14 elbows had major defects in the medial humeral condyle. A corticocancellous bone block from the ipsilateral iliac wing was used as a strut graft in 13 elbows and a cortical bone block from the distal ulna was used in the other (Fig. 1). In three elbows with major defects in the coronoid process, a bone block was used. The humeral component was not cemented in 12 elbows. The ulnar component was cemented in 12 elbows. In ten elbows a hybrid method was used; in two, both components were cemented, and in the remaining two, both were uncemented. Fig. 1 A full-thickness corticocancellous graft from the iliac wing is used to augment the gross bone defect at the medial humeral condyle as well as the defect at the coronoid process. able degrees of bone loss in the medial humeral condyle. There was also severe bone erosion in the trochlear notch and the radial head in most of the elbows. Two elbows had fractures of the humeral condyles due to minor trauma. Before operation all the elbows had variable degrees of instability with moderate to severe pain on movement; none had associated ulnar neuropathy. All the patients were unable to bring their hands to their face or head while holding an object. The elbows were assessed clinically by the Mayo Elbow Performance Score. 1 The position of the implant, the status of any radiolucent lines and degrees of resorption or union of the grafted bone were assessed on routine anteroposterior and lateral radiographs. Results The clinical results were assessed according to the Mayo Elbow Performance Score 1 and are shown in Table II. Before operation, 11 elbows had severe pain, nine had gross valgus-varus instability, and all were rated poor. At the latest follow-up the overall result was excellent in eight elbows, good in five and fair in one, with almost complete relief from pain in 11. The arc of flexion increased from a mean of 117.5 (100 to 140 ) pre-operatively to a mean of 134.3 (115 to 150 ) post-operatively (p < 0.001, paired t- test). Conversely, any flexion contracture worsened slightly, from a mean of 29.0 (0 to 50 ) to 36.4 (10 to 50 ) (p = 0.06, paired t-test). Rotation of the forearm also improved. Pronation increased from a mean of 43.6 (0 to 90 ) to 60 (0 to 90 ) (p < 0.01, paired t-test). Supination increased from a mean of 56.4 (0 to 90 ) to 79.3 (30 to 90 ) (p < 0.05, paired t-test). As for valgus-varus, 12 elbows had good stability at the last review. Two had slight valgusvarus instability but this did not limit function, which was universally improved. All patients could move their elbows comfortably when combing their hair or eating. The six who had worsening of the flexion contracture by 15 or more complained of awkwardness in daily activities for the first few months but this did not concern them significantly thereafter. VOL. 88-B, No. 7, JULY 2006

922 T. MORI, H. KUDO, K. IWANO, T. JUJI Table II. Elbow performance scores, using the Mayo system 1 Pre-operative Last follow-up Case Pain Movement Stability Function Total Result Pain Movement Stability Function Total Result 1-L 15 15 0 10 40 Poor 30 15 10 20 75 Good 1-R 15 20 0 10 45 Poor 45 20 10 20 95 Excellent 2 0 15 0 10 25 Poor 45 15 10 10 80 Good 3 15 15 5 10 45 Poor 45 15 10 20 90 Excellent 4 15 15 0 10 40 Poor 45 15 10 20 90 Excellent 5 15 15 5 10 45 Poor 45 15 10 15 85 Good 6 15 20 0 10 45 Poor 45 20 5 20 90 Excellent 7-L 15 20 0 10 45 Poor 30 20 5 15 70 Fair 7-R 15 20 0 10 45 Poor 30 20 10 15 75 Good 8 0 15 0 10 25 Poor 45 15 10 10 80 Good 9 15 15 0 10 40 Poor 45 20 10 15 90 Excellent 10-R 0 15 5 10 30 Poor 45 20 10 20 95 Excellent 10-L 0 15 5 10 30 Poor 45 15 10 20 90 Excellent 11 15 15 5 10 45 Poor 45 15 10 20 90 Excellent Fig. 2a Fig. 2b Fig. 2c Fig. 2d Case 1. A 59-year-old woman with rheumatoid arthritis had undergone left total elbow arthroplasty with a strut bone graft on the medial side of the humerus. She had a right total elbow arthroplasty three years later. Radiographs of the left elbow pre-operatively (a and b) and at 11 years (c and d); and of the right at eight years (e and f) show an excellent result on the left and a good result on the right. There are some spot welds around the humeral component without resorption of the grafted bone. Fig. 2e Fig. 2f THE JOURNAL OF BONE AND JOINT SURGERY

KUDO TYPE-5 TOTAL ELBOW ARTHROPLASTY IN MUTILATING RHEUMATOID ARTHRITIS 923 Fig. 3a Fig. 3b Fig. 3c Case 8. A 52-year-old woman. a) Pre-operative radiograph showing severe joint destruction (Larsen grade 5 6 ) and a fracture of the medial condyle. b) The fracture was treated at the time of a total elbow arthroplasty. c) Radiograph seven years after operation shows no loosening of components and union of the fracture. Radiological assessment. Radiographs were examined for the status of the grafted bone, the implant, migration, bone formation at the bone-metal interface, radiolucent lines at the bone-cement and bone-metal interfaces, and bone resorption and atrophy because of stress shielding. Incorporation of the grafted bone was assumed when the absence of collapse or sclerotic changes in the graft had been confirmed at the last review. Union between graft and host bone was assumed when no radiolucent line between them was evident. At the final assessment, the grafted bone was unchanged in size in 12 elbows, but two showed a degree of resorption. There were no radiolucent lines around the 12 cementless humeral stems. Endosteal bone formation (spot welds) was seen around these components, particularly in the porous-coated area, suggesting solid osseous integration (Fig. 2). In the two elbows with preoperative fractures of the humeral condyles, these healed in a good position (Fig. 3). Of the two elbows with cemented humeral components, one had a thin, non-progressive radiolucent line of 1 mm or less at the bone-cement interface without evidence of migration. There were no radiolucent lines around the two porous metal-backed cementless ulnar components. Of the two elbows in which the porous metal-backed ulnar component had been cemented, one had a non-progressive radiolucent line 1 mm in width around the stem. All ten elbows with allpolyethylene ulnar components had radiolucent lines 1 mm wide or less at the bone-cement interface without evidence of migration. There were no major complications. Discussion Even with recent advances in total elbow arthroplasty, reconstruction of an unstable elbow with severe bone loss is difficult. Because of the major bone defects and the instability, a semi-constrained, linked prosthesis is generally regarded as a better option than a less constrained, unlinked implant. Among various types of the former, good results are reported with the Coonrad-Morrey (Zimmer Inc, Warsaw, Indiana) 8 and GSB-3 prostheses (Zimmer GmbH, Winterthur, Switzerland). 9 However, it is a concern with these prostheses that the small area of contact between the components may result in wear or breakage of the polyethylene bushing and subsequent failure. They usually have a long stem, require cement and involve the sacrifice of much more bone from the lower humerus than do unlinked types. The use of these, combined with grafting, can increase bone stock. In the Kudo type-5 prosthesis (Biomet UK Ltd) the surface replacement design and the vacant space within its condyle are very suitable for grafting gross bone defects in the humerus. In this series of 14 elbows, structural augmentation by grafts was very effective in securing initial fixation and the grafted bone united easily with the host bone without collapse. We have no experience with the use of an allograft to fill the bone defect. In the two cases where the defects were augmented by cement only, and so excluded from this study, there was no fracture or displacement of the cement or loosening of the implants. We therefore believe that where the bone defect is minor, augmentation with cement alone might be a good option. VOL. 88-B, No. 7, JULY 2006

924 T. MORI, H. KUDO, K. IWANO, T. JUJI The reason why joint stability was remarkably good post-operatively is probably due to the prosthesis acting as a good spacer, thereby restoring tension to the soft tissues and the shape of the articulating surfaces might afford an intrinsic constraint. Although the medial collateral ligament was insufficient in all the elbows, no effort to reconstruct it was made. In spite of this, no medial instability was found. Conversely, the lateral ligament was always carefully preserved to prevent varus instability. It is important that the incised margins of the dorsal fascial layer on the lateral side of the olecranon are securely sutured under appropriate tension. There tended to be better stability in elbows with a residual flexion contracture of 30 or more. A flexion contracture increases the length of the joint and increases tension in the surrounding soft tissues. In this context, the functional disadvantage of a flexion contracture is balanced by a more stable joint. There was no definite loosening seen at a mean follow-up of 7 years and 7 months (5 years to 11 years, 3 months). A relevant factor for these good results may be that all patients, hitherto severely disabled, might have been very cautious when using the operated elbow in their daily activities to avoid a high level of force being generated in the joint. Despite our good results, we do not feel that they justify the general use of an unlinked prosthesis for a mutilans elbow. A high level of surgical expertise is necessary when using such a prosthesis in advanced rheumatoid arthritis. 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. References 1. Morrey BF, Adams RA. Semiconstrained arthroplasty for the treatment of rheumatoid arthritis of the elbow. J Bone Joint Surg [Am] 1992;74-A:479-90. 2. Ramsey ML, Adams RA, Morrey BF. Instability of the elbow treated with semiconstrained total elbow arthroplasty. J Bone Joint Surg [Am] 1999;81-A:38-47. 3. Kudo H. Non-constrained elbow arthroplasty for mutilans deformity in rheumatoid arthritis: a report of six cases. J Bone Joint Surg [Br] 1998;80-B:234-9. 4. Kudo H, Iwano K, Nishino J. Total elbow arthroplasty with use of a nonconstrained humeral component inserted without cement in patients who have rheumatoid arthritis. J Bone Joint Surg [Am] 1999;81-A:1268-80. 5. Arnett FC, Edworthy SM, Bloch DA, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315-24. 6. Larsen A, Dale K, Eek M. Radiographic evaluation of rheumatoid arthritis and related conditions by standard reference films. Acta Radiol Diagn (Stockh) 1977;18: 481-91. 7. Campbell WC. Arthroplasty of the elbow. Ann Surg 1922;76:615. 8. Gill DR, Morrey BF. The Coonrad-Morrey total elbow arthroplasty in patients who had rheumatoid arthritis: a ten to fifteen-year follow-up study. J Bone Joint Surg [Am] 1998;80-A:1327-35. 9. Kelly EW, Coghlan J, Bell S. Five- to thirteen-year follow-up of the GSB III total elbow arthroplasty. J Shoulder Elbow Surg 2004;13:434-40. THE JOURNAL OF BONE AND JOINT SURGERY