Abstract
Background:
Bone defects of the distal end of the humerus require complex reconstructions, for which standard prostheses may be insufficient. We investigated the outcomes of distal humeral reconstruction with use of a modular prosthesis.
Methods:
Fifty-three elbows in fifty-two patients underwent reconstruction with a modular prosthesis (twelve total humeral replacements and forty-one distal humeral replacements) after tumor resection (thirty-eight elbows) or because of massive joint degeneration (fifteen elbows). In the tumor group, twenty-three patients (twenty-four elbows) had metastatic disease and fourteen had a primary tumor. Degenerative defects of the distal end of the humerus were caused by pseudarthrosis (six elbows), prosthetic failure (five), trauma (two), osteomyelitis (one), and supracondylar fracture (one). The mean duration of follow-up for all patients was twenty-eight months (median, thirteen months; range, one to 219 months).
Results:
The mean Inglis-Pellicci score in the tumor group was 84 points, and the mean Musculoskeletal Tumor Society score was 78%. Patients with total humeral reconstruction had worse scores than those with distal humeral reconstruction. Twenty-four patients died of disease at a mean of thirteen months after surgery. Local tumor control was achieved in all patients. In the revision group, the mean Inglis-Pellicci score was 76 points. The Inglis-Pellicci score was significantly better for patients in the tumor group. Eight patients (15%) had a deep periprosthetic infection, requiring amputation in one patient (2%) and prosthetic removal in two patients (4%). Four patients (8%) had the implants revised for aseptic loosening.
Conclusions:
Modular prostheses of the distal end of the humerus provide a stable reconstruction of the elbow with satisfactory function and disease control in patients with a tumor, but careful patient selection is required when the prostheses are used for revision surgery in patients without a tumor.
Level of Evidence:
Therapeutic Level IV. See Instructions to Authors for a complete description of levels of evidence.
Bone defects of the distal end of the humerus compromise function of the elbow and the entire upper extremity. The causes of such defects include tumors, fractures, infection, joint deformities related to inflammatory diseases, and catastrophic failure following arthroplasty.
While the prevalence of neoplasms of the distal end of the humerus is low, contributing to 1% of all primary bone lesions1, complications related to arthroplasty of the elbow are frequently associated with extensive bone destruction2,3.
After wide tumor excision, reconstruction of the resulting defect with hemiarthroplasty, plating and cementing, or massive allograft is often difficult to achieve and is associated with high complication rates4-6. Therefore, modular tumor prostheses have become an accepted treatment of cancer around the distal end of the humerus7. Additionally, endoprosthetic replacement of the elbow has been described as a treatment option for degenerative disease and trauma8-10, with similar or even lower complication rates than open reduction and internal fixation3,11-13. After successful use of modular endoprostheses in the treatment of bone tumors of the distal end of the humerus, we extended our indications to the reconstruction of large defects secondary to prosthetic failure or joint degeneration.
The aim of this study was to report the results for patients treated with a modular prosthesis to reconstruct major bone defects of the distal end of the humerus caused either by tumor or by failed treatment of elbow joint degeneration. We hypothesized that patients without a tumor compared with those with a tumor would have equal or better results, as their treatment would require less soft-tissue removal than the resection of bone or soft-tissue tumors.
From June 1989 to May 2008, we treated fifty-two patients (fifty-three elbows) who had bone defects of the distal end of the humerus with modular humeral implants. Twenty-seven patients (52%) were female, and twenty-five (48%) were male. Thirty-seven patients (thirty-eight elbows; 71%) were treated for a bone tumor of the distal end of the humerus (the tumor group), and fifteen patients (29%) underwent reconstruction of the elbow because of posttraumatic joint degeneration or prosthetic failure (the revision group).
The mean age at the time of implantation was fifty-five years (median, sixty-three years; range, six to ninety years). The overall mean follow-up time was twenty-eight months (median, thirteen months; range, one to 219 months). This retrospective investigation was approved by our institutional review board.
Anteroposterior and lateral radiographs were made at every follow-up examination for both groups. Postoperative function and subjective data were assessed according to the Inglis-Pellicci scoring system14 for the evaluation of patients after total elbow replacement. In the tumor group, the Musculoskeletal Tumor Society (MSTS) score15 was calculated. All scores were evaluated at the time of the most recent follow-up by orthopaedic surgeons involved in the treatment of the patients.
The Inglis-Pellicci scoring system evaluates the elbow with respect to the following categories (with a maximum score assigned to each): pain (30 points), range of motion (20 points), function (20 points), muscle strength (10 points), flexion contracture (6 points), extension contracture (6 points), and forearm pronation and supination (4 points each), with a possible maximum score of 100 points. A score from 100 to 90 points indicates an excellent result; 89 to 80 points, a good result; 79 to 70 points, a fair result; 69 to 60 points, a poor result; and <60 points, a failure.
The MSTS score for the upper extremity assesses pain, function, emotional acceptance, hand positioning, dexterity, and lifting ability, with a maximum score of 5 points assigned to each category for a possible total score of 30 points. The final result is expressed as a percent, with 100% representing normal function.
Biplanar radiographs made during the most recent follow-up evaluation were independently evaluated by two orthopaedic surgeons (P.T.F. and R.S.). The International Society of Limb Salvage (ISOLS) score16 for radiographic evaluation of the implant was calculated by each investigator. The score documents six radiographic categories: bone remodeling (i.e., resorption of the prosthetic fixation area); interface (i.e., extent of radiolucency); anchorage (i.e., dislocation or fracture of prosthetic material); implant body (i.e., debonding or displacement of prosthetic components); implant articulation problems (i.e., extent of implant wear); and extracortical bone bridging (i.e., extent of bone bridging on to coatings whenever intended by the prosthetic design). Each of these categories is rated as excellent, good, fair, or poor according to the absence or presence of the criteria defined within the scoring sheet. The score does not provide numerical values.
Five patients in the tumor group underwent clinical and radiographic follow-up at an outside institution. The remaining forty-seven patients were followed at our institution. A table in the Appendix provides an overview of the demographic data.
In all patients, the Howmedica Modular Resection System (Stryker Orthopaedics, Mahwah, New Jersey) was used (Fig. 1). This is a fully constrained, hinged total elbow replacement with modular titanium components in 1.5-cm increments. The minimal humeral resection length is 6 cm. The stem length is 5 cm, and the diameter ranges from 9 to 12 mm in 1-mm increments. The humeral diaphyseal bone is prepared by flexible reamers; rasps are used for preparation of the ulnar component. The design does not provide a porous coating, allowing cemented and cementless implantation. Within our institution, cementless fixation of the humeral stem and ulnar component is only performed when solid press-fit anchorage can be achieved. All procedures were performed by nine surgeons, with several years of experience in musculoskeletal oncology and/or revision surgery.
Overview of the modular prosthetic system depicting all single components (left) and an example of a fully assembled total humeral replacement (right).
Tumor Group
In the tumor group, twenty-one patients (57%) were male and sixteen (43%) were female, with a mean age of fifty-four years (median, sixty-three years; range, six to ninety years). Fourteen patients (38%) were treated for primary lesions, and twenty-three (62%) were treated for metastatic bone disease. Primary tumors included osteosarcoma in six patients, soft-tissue sarcoma in three, giant-cell tumor in three, and Ewing sarcoma in two. Metastases occurred in twenty-three patients (twenty-four elbows), including seven patients with renal cell carcinoma, five with breast cancer, four with lung cancer, and seven with other cancers. The mean duration of follow-up was twenty-eight months (median, thirteen months; range, one to 219 months) for all patients, twelve months (median, ten months; range, one to forty-four months) for the twenty-four patients (65%) who died of disease, and fifty-four months (median, thirty-seven months; range, three to 218 months) for the thirteen patients (35%) who were alive at the time of this investigation.
In these thirty-seven patients with a tumor, eleven were managed with a total humeral implant and twenty-seven, with a distal humeral implant (including one patient who had bilateral reconstruction) on the basis of the extent of bone resection. Total humeral replacement was indicated whenever tumor extension required resection of the proximal humeral diaphyseal bone, making stem anchorage impossible. The indication for prosthetic reconstruction was wide tumor excision in all primary tumors. In metastatic disease, the indications for prosthetic reconstruction were restricted to pain control and improvement of elbow mobility. Pathological fracture was regarded as an indication for surgery when fracture fixation was impossible because of the extent of the lesion or when an intra-articular fracture was involved. In twenty-four (63%) of all thirty-eight implantations, the prosthesis was implanted at the first surgical procedure. Six patients (16%) underwent an attempted tumor resection prior to implantation of the modular prostheses at our institution. The reasons for secondary implantation were local recurrence of bone metastases in five patients with plating, and loosening of an intercalary prosthesis in one patient. Eight patients (22%) had the primary treatment performed at another institution; the procedures included intramedullary nailing of a pathological fracture in three patients, resection of a soft-tissue mass in three patients, plating in one patient, and implantation of a spacer in the proximal part of the humerus in one patient. Revision was indicated because of local tumor recurrence in all patients.
A lateral surgical approach was used in all patients with a tumor unless a previous biopsy tract necessitated another approach. This approach spared the extensor mechanism, whenever possible, and avoided resection of the triceps from the olecranon. In primary tumors, the goal was wide tumor excision with clear margins; in metastatic disease, the primary goal was marginal excision, without compromising neurovascular structures (Fig. 2).
Preoperative anteroposterior and lateral radiographs of a sixty-year-old man with bone metastasis from renal cell carcinoma, showing an osteolytic metastasis (upper row), and anteroposterior and lateral radiographs made two years after reconstruction of a wide tumor resection with implantation of a cementless modular prosthesis of the elbow (bottom row).
In six patients, excision of the total humerus was required to obtain wide margins for primary tumors, including one extra-articular resection of the shoulder (Fig. 3). Five patients with metastatic disease required total humeral resection. Neurovascular structures were preserved in all patients except for two in whom the radial nerve was resected to obtain clear margins. The ulnar nerve was transferred anteriorly. In all but nine patients (24%), the implant was cemented in place. Restoration of length was calculated from measurement of the resected specimen and was possible in every case. One child, who was six years old at the time of the operation, received an expandable modular prosthesis for stepwise elongation. Total humeral prostheses required proximal muscular reattachment. In all but two distal humeral prostheses, no special functional soft-tissue and muscle reconstruction was necessary, as the extensor mechanism was not compromised. In two patients with partial resection of the triceps, reconstruction was performed by a fascia lata autograft or an implant for tendon repair (LARS ligament; JK Orthomedic, Dollard-des-Ormeaux, Quebec, Canada) in one patient each17-19. To avoid shoulder instability or dislocation of the prosthetic head in total humeral implants, the LARS ligament was used for reattachment of rotator cuff tendons and soft tissues to the proximal part of the prostheses in three patients. In one patient, a fascia lata autograft was used to reattach the rotator cuff.
Preoperative anteroposterior and lateral radiographs of a sixty-four-year-old man with malignant fibrous histiocytoma of the right arm, requiring wide resection of the total humerus (upper row), and anteroposterior and lateral radiographs made two years after implantation of a cementless modular total humeral prosthesis (bottom row).
Rehabilitation included passive elbow motion for two weeks after the procedure, followed by two to four weeks of active motion based on the extent of the resected bone and soft tissues. Adjuvant chemotherapy was administered whenever indicated by the primary disease. Twenty-one patients (57%) received chemotherapy, and three patients (8%) were treated by immunotherapy. Twenty-two patients (59%) underwent radiation therapy of the distal end of the humerus, which was done preoperatively for ten patients and postoperatively for twelve patients.
Revision Group
The revision group comprised fifteen patients, including four men and eleven women. The mean age at the time of surgery was fifty-nine years (median, sixty-five years; range, thirty-one to eighty-four years). The mean duration of follow-up was twenty-eight months (median, eighteen months; range, one to 188 months). Two patients died within twelve weeks after the operation because of unrelated causes. All patients had severe destruction of the elbow joint and the distal end of the humerus from nononcological causes. All patients had at least one (eight patients) or more (seven patients) previous operations. The indication for revision was based on elbow pain, joint instability, or implant loosening. Six patients had a nonunion after a distal humeral fracture treated by open reduction and internal fixation followed by bone loss due to long-standing implant loosening. Five patients were treated for complications related to a semiconstrained elbow prosthesis, including loosening in three and removal for infection in two. Two patients had a posttraumatic ankylosis, which was the indication for distal humeral resection, and one patient was treated for posttraumatic osteomyelitis. The remaining patient sustained a periprosthetic fracture of the distal end of the humerus nine years after resection of a low-grade chondrosarcoma of the proximal part of the humerus. The distal bone fragments were removed, and the preexisting proximal humeral component was connected to an elbow module to form a total humeral prosthesis. All other prostheses in the revision group were distal humeral reconstructions (Fig. 4).
Preoperative anteroposterior and lateral radiographs of a forty-two-year-old man with posttraumatic osteoarthritis and deformity of the left elbow (upper row), and anteroposterior and lateral radiographs made two and a half years after the reconstruction of the elbow joint with a cementless modular prosthesis (bottom row).
The surgical technique corresponded to the procedure described for patients in the tumor group, except that bone resection was performed subperiosteally and soft tissues were spared. The components were inserted with cement in eight patients and without cement in seven patients. Rehabilitation was identical to that in the tumor group.
Statistical Methods
Statistical analysis was performed with use of the two-tailed Student t test for comparison of means for continuous variables; the Fisher exact test was applied for categorical variables. Prosthetic survival was defined as the time until first revision of the prosthesis for any cause in all patients and was estimated according to Kaplan-Meier analysis. Survival was compared by the log-rank test. A power analysis was calculated for survival difference. Multivariate analysis was performed with use of a Cox regression model to identify independent risk factors of prosthetic failure. A p value of <0.05 was regarded as significant.
Source of Funding
No external funding was used to perform this investigation.
Thirty-four patients (thirty-five elbows; 66%) had no complications. Eighteen elbows (34%) required revision, with multiple revisions in five elbows (9%). Fourteen elbows (26%) had major complications related to the prosthesis, and four (8%) had minor complications, which left the prosthesis unaffected. Minor complications included superficial wound dehiscence, which healed after secondary closure, in two elbows (4%) in the tumor group and nerve palsies (one radial and one ulnar), which resolved after surgical decompression and neurolysis, in two elbows (4%) in the revision group. There were seven major complications in each group, representing 18% of the elbows in the tumor group and 47% in the revision group (p = 0.046) (Table I). The median prosthetic survival in both groups was thirty-six months (Fig. 5). Corresponding survival rates were 82% at twelve months, 62% at twenty-four months, 44% at thirty-six months, and 35% at forty-eight months. At twelve months, the prosthetic survival rate was 80% in the tumor group and 89% in the revision group; at twenty-four months, the prosthetic survival rate was 72% in the tumor group compared with 51% in the revision group. The two groups revealed no significant difference in the log-rank test (p = 0.410), and multivariate Cox regression models were unable to identify independent risk factors for infection or aseptic loosening. Due to the sample size, however, there was only a power of 0.6 to detect a 22% difference in prosthetic survival.
Kaplan-Meier estimation of prosthetic survival rate (upper graph) for all patients and for the tumor and revision groups (bottom graph), with the first revision of the implant as the end point. There was no significant difference between the groups (p = 0.410).
Tumor Group
In the tumor group, six patients (16%), three of whom had radiation therapy, developed deep infection at a mean of eleven months (median eight months; range, two to thirty-one months) after surgery. Three patients had a total humeral prosthesis, and three had a distal humeral reconstruction; the infection rates for total and distal humeral replacements were 27% and 11%, respectively. Three patients with early infection within three months after surgery subsequently underwent one-stage revision, during which the prosthesis was exchanged to a new implant. In all three patients, the infection resolved. Two patients had two-stage revisions. During the first stage, the prostheses were replaced by an antibiotic-impregnated cement spacer. The patient with an expandable prosthesis had successful infection control and underwent reimplantation of a total humeral prosthesis five months later. The second patient underwent removal of the spacer twenty days after implantation without further joint replacement. A final patient with an infection underwent shoulder disarticulation because of poor health from a metastatic soft-tissue sarcoma and died in the early postoperative phase.
One patient (3%) in the tumor group had aseptic loosening of the humeral component nine months postoperatively and underwent successful humeral stem revision.
Twenty-four patients (65%) died of the disease at a mean of thirteen months (median ten months; range, one to forty-four months) after the operation. Of those who died, nineteen (79%) were treated for metastatic disease. Twenty-three patients (62%) had generalized metastases at the time of presentation. Thirteen patients (35%), including four with osteosarcoma, three with giant-cell tumor, two with Ewing sarcoma, and four with metastatic disease, were alive at the time of the present investigation. All patients were without signs of disease at the time of the latest follow-up at a mean of forty-four months (median, twenty-three months; range, three to 218 months) after the operation.
In fourteen patients (38%) with primary bone tumors, three developed metastases at two, fourteen, and forty-six months, respectively, after the operation.
In all patients in the tumor group, local disease control was achieved with no local recurrence around the elbow. Resection margins were rated as marginal in fifteen patients (41%) and wide in twenty-two patients (59%). Two patients (5%) with metastatic renal cell carcinoma developed a further lesion around the scapula of the involved arm and underwent partial scapulectomy and conservative radiation, respectively.
Revision Group
In the revision group, two patients had a deep infection develop fourteen and twenty-four months after the operation, respectively. Both had had several previous operations. One patient was successfully treated by one-stage revision, and the other patient underwent prosthetic removal, but declined further surgery.
Three patients showed aseptic loosening of the humeral stem at seventeen, thirty-six, and thirty-eight months, respectively, after implantation. No patient had ulnar implant loosening. All patients had successful revision to a larger stem or underwent further bone resection to find adequate bone stock for refixation. Additionally, two patients underwent revision and refixation of their implant for traumatic periprosthetic fracture of the humerus.
Functional Results
Clinical outcomes were measured with the Inglis-Pellicci scoring system in both groups and with the MSTS scoring system in the tumor group. Seven patients (13%), including five in the tumor group who were not available and two who had an amputation, were excluded, leaving forty-five patients with forty-six joints (87%) available for complete scoring. All scores were obtained during the most recent follow-up evaluation. Two elbows that required prosthetic removal because of infection were rated as an overall failure in all scores. Table II indicates the results of clinical elbow function and outcomes assessed on the basis of the Inglis-Pellicci scores for both treatment groups. No significant difference was detected between the groups with respect to range of motion (p = 0.852), but the tumor group showed a better overall outcome on the basis of the Inglis-Pellicci score (p = 0.048).
The mean MSTS score (and standard deviation) for the tumor group was 23 ± 3 points (range, 17 to 29 points), representing 78% ± 11% (range, 57% to 97%) of normal function. The mean outcome for the categories evaluated were 5 ± 1 points for pain, 3 ± 1 for function, 4 ± 1 for emotional acceptance, 5 ± 1 for hand positioning, 5 ± 1 for dexterity, and 3 ± 1 for lifting ability. The mean MSTS score was 81% ± 10% (range, 67% to 97%) for patients with a distal humeral replacement of the elbow compared with 70% ± 9% (range, 57% to 87%) for those with a total humeral prosthesis.
Radiographic Results
Radiographic outcome was evaluated according to ISOLS criteria for all patients who underwent clinical scoring. In the revision group, radiographs of two patients, made at an outside institution, that did not meet requirements to be rated with the ISOLS scoring system and two patients with an amputation were excluded, so radiographs of forty-two prostheses were available for scoring. At least thirty-three patients (79%) had results rated as excellent in every category, and no more than two patients (5%) had results rated as poor in any category.
Minimum Two-Year Follow-up
Sixteen patients (31%) were available for follow-up of more than twenty-four months. The mean duration of follow-up was eighty months (mean, sixty-three months; range, twenty-six to 219 months) for the tumor group and sixty-one months (mean, forty months; range, twenty-four to 188 months) for the revision group.
Major complications occurred in ten patients (six in the tumor group and four in the revision group), including four patients who had aseptic loosening (one in the tumor group and three in the revision group) and six patients who had infection around an implant (five in the tumor group and one in the revision group). The patients in the tumor group had a mean of 0.7 revision per patient compared with 1.2 surgical procedures in the revision group.
The mean elbow motion was 100° ± 20° (range, 80° to 140°), with a lack of extension of 12° ± 10° (range, 0 to 30°) and flexion of 113° ± 19° (range, 90° to 140°). The mean Inglis-Pellicci score was 81 ± 13 points (range, 54 to 98 points). The results, according to the score, were excellent in four, good in five, fair in four, poor in one, and considered a failure in two patients. The mean Inglis-Pellicci score in the tumor group was 86 ± 10 points (range, 71 to 98 points) compared with 74 ± 14 points (range, 54 to 91 points) in the revision group. The ISOLS score revealed excellent or good results in all categories for fifteen patients and a failure for one patient.
Modular elbow prostheses can be used for reconstruction of bone defects of the distal end of the humerus after tumor resection or for massive bone defects after failed reconstructive surgery. Elbows in which the modular implant was used following failed elbow surgery for massive defects had worse outcomes, suggesting that previous surgery, extensive scarring, a history of infection, and poor bone stock can severely affect the results.
There are many limitations to this report. First, the patient cohort is small. This is related to the investigation of a rarely indicated treatment, limiting the power of our study. Second, the high percentage of patients dying of the underlying oncological disease considerably reduced follow-up periods as in other reports with a high number of patients with metastatic disease1. This fact also accounts for reduced survival functions and increasing declines over time in the Kaplan-Meier analysis as these patients were methodically censored. With metastatic disease accounting for 63% (twenty-four) of the thirty-eight elbows in the tumor group and for 56% of all fifty-three elbows, treatment often focused on local disease control, which was achieved in all patients. A minimum two-year follow-up was observed in sixteen patients (31%). Third, because of the retrospective design, we were unable to obtain standardized radiographs or perform blinded clinical examinations, causing a potential bias. This underlines the need for further multicenter trials and database collaborations.
As tumors of the distal end of the humerus are rare, most reports are case series, documenting diverse treatment options. There are, however, several reports about modular prosthetic reconstruction7,20,21 (see Appendix), all of which have a similar rate of complications and revisions of approximately 23%20, 40%21, and 50%7. The functional outcome in all series was satisfactory, with a Toronto Extremity Salvage Score (TESS) of 73% in the Birmingham (England) experience21 and an equal TESS score of 73% in the report from their colleagues in Stanmore (England)7. The latter group reported an MSTS score of 76%. In these series, the number of patients with metastatic disease at the time of treatment was high. Additionally, Hanna et al.7 reported rather high rates of local recurrence (22%) and concluded that complete excision of the tumor is extremely important and may improve long-term survival. Other authors1,22 have used standard total elbow prostheses for reconstruction after tumor resection. Sperling et al.1 reported excellent or good results in nine of thirteen patients, with local recurrence in three patients with metastatic disease. However, in an update and extension of their data seven years later22, they modified their results to a 25% local recurrence rate and a 20% revision rate. Functional outcome showed a mean of 75 of 100 points in the Mayo Elbow Performance Score. Weber at al.4 investigated the use of segmental elbow reconstruction by either prosthetic replacement or allograft-prosthesis composite for total humeral replacement and reported a mean MSTS score of 77%. However, the revision rate due to early complications (35%) and late complications (30%) was high. The rate of local recurrence was 26%. This high revision rate is similar to that reported in the investigation of allograft-prosthesis composite reconstructions of the elbow in ten patients by Renfree et al.5. A mean of 2.2 revisions per patient was necessary to obtain a 79% rate of allograft-host union. Osteochondral and total elbow allografts have also been described for limb salvage23. In a mainly oncological group of nineteen patients, including sixteen who had hemiarticular defects and three who had total joint allografts, six had complications, with a high likelihood of failure or secondary degenerative changes.
All of these reports indicate a relatively high rate of complications associated with segmental bone reconstruction methods around the elbow. We observed major complications in 26% of our patients. Due to extensive surgical resection, we avoided local relapse even in patients with metastatic lesions and function remained satisfactory with a mean MSTS score of 78%. The high rate of patients with metastatic disease emphasizes the role of a surgical procedure with fast recovery and successful tumor control. We recommend modular prostheses for reconstruction after tumor removal to allow adequate resection of the affected bone without compromising the functional result.
There are numerous reports about the treatment of complications after reconstructive surgery around the elbow and total elbow replacement2,3,10,12,23-27. However, reports about methods to reconstruct major bone defects that result from failure of these prostheses are rare. Mansat et al. described the Mayo experience with allograft-prosthesis composite reconstructions for such situations3. Seven of thirteen patients had complications, with revision in five. Deep infection occurred in four joints; three of these required implant removal. Seven patients showed excellent or good results. Strut graft augmentation in the proximal part of the ulna around loosened components had a complication rate of 38% in a series of twenty-one patients with elbow arthroplasty2.
In our revision group, seven of fifteen patients required further surgery. The predominant complication was aseptic loosening occurring in three of fifteen patients. Clinical evaluation showed excellent or good results on the Inglis-Pellicci score in the majority of patients However, with a high revision rate, modular prosthetic reconstruction of the humerus is not indicated in the setting of osteoporotic bone or a history of infection.
The complication rate of both treatment groups (tumor and revision) was similar to previously published results. No significant difference was detected between the groups with regard to prosthetic survival; however, the tumor group had a lower complication rate and better functional outcome (as assessed with the Inglis-Pellicci score). Patients with a total humeral prosthesis had worse MSTS scores than those with a distal humeral prosthesis mainly because of shoulder instability. The difference in outcome between the tumor and revision groups may be due to the scar tissue resulting from multiple surgical procedures, limiting elbow motion and function in the revision group.
The most severe complication in this series was infection. The overall infection rate was 15% (eight of fifty-three elbows), resulting in removal of the prosthesis in two patients and amputation in another patient. Among the patients who had a tumor, the infection rate was higher for total humeral replacements compared with distal humeral reconstructions, suggesting that infection may depend on implant size. This finding was underlined by a report that found no infections in a series of patients treated with standard elbow prostheses for tumor reconstruction22. Also, radiation therapy must be factored into the increased infection rate in oncology patients28. Our primary management of early infections (within three months after index surgery) consisted of one-stage revision29. All three patients in whom this procedure was used had control of the infection. However, the indication for a one-stage revision depends on the time of onset and extent of soft-tissue involvement. Therefore, two-stage revision is recommended for patients with more severe or late infection30. Cheung et al. reported that infection remained uncontrolled in 28% (eight) of twenty-nine patients with an infection around a standard elbow replacement31, indicating that infection control remains elusive. We have adapted our antibiotic regimen to prolonged prophylaxis over five days. Additionally, silver-coated modular prostheses of the femur and tibia have been shown to reduce infection rates32, and could provide improvements for humeral reconstructions as well.
In conclusion, modular prostheses of the elbow are a viable treatment option for reconstruction of bone defects after tumor resection around the distal end of the humerus. Functional results are satisfying, while oncological results in terms of local tumor control are excellent. Infection remains the major complication and accounts for most revisions. Modular prosthetic reconstruction for bone defects of the elbow due to joint degeneration results in relatively good functional results despite a high revision rate and should be avoided in patients with a history of infection or osteoporotic bone.
Tables showing the demographic details and indications in the tumor and revision groups as well as an overview of reports on segmental defect reconstructions around the elbow are available with the online version of this article at jbjs.org.
Sperling
JW;
Pritchard
DJ;
Morrey
BF. Total elbow arthroplasty after resection of tumors at the elbow. Clin Orthop Relat Res.
1999;367:256-61.[CrossRef][PubMed]
Kamineni
S;
Morrey
BF. Proximal ulnar reconstruction with strut allograft in revision total elbow arthroplasty. J Bone Joint Surg Am.
2004;86:1223-9.[PubMed]
Mansat
P;
Adams
RA;
Morrey
BF. Allograft-prosthesis composite for revision of catastrophic failure of total elbow arthroplasty. J Bone Joint Surg Am.
2004;86:724-35.[PubMed]
Weber
KL;
Lin
PP;
Yasko
AW. Complex segmental elbow reconstruction after tumor resection. Clin Orthop Relat Res.
2003;415:31-44.[CrossRef][PubMed]
Renfree
KJ;
Dell
PC;
Kozin
SH;
Wright
TW. Total elbow arthroplasty with massive composite allografts. J Shoulder Elbow Surg.
2004;13:313-21.[CrossRef][PubMed]
Kharrazi
FD;
Busfield
BT;
Khorshad
DS;
Hornicek
FJ;
Mankin
HJ. Osteoarticular and total elbow allograft reconstruction with severe bone loss. Clin Orthop Relat Res.
2008;466:205-9.[CrossRef][PubMed]
Hanna
SA;
David
LA;
Aston
WJ;
Gikas
PD;
Blunn
GW;
Cannon
SR;
Briggs
TW. Endoprosthetic replacement of the distal humerus following resection of bone tumours. J Bone Joint Surg Br.
2007;89:1498-503.[CrossRef][PubMed]
Gschwend
N. Present state-of-the-art in elbow arthroplasty. Acta Orthop Belg.
2002;68:100-17.[PubMed]
Garcia
JA;
Mykula
R;
Stanley
D. Complex fractures of the distal humerus in the elderly. The role of total elbow replacement as primary treatment. J Bone Joint Surg Br.
2002;84:812-6.[CrossRef][PubMed]
Moro
JK;
King
GJ. Total elbow arthroplasty in the treatment of posttraumatic conditions of the elbow. Clin Orthop Relat Res.
2000;370:102-14.[CrossRef][PubMed]
Loehr
JF;
Gschwend
N;
Simmen
BR;
Katzer
A. [Endoprosthetic surgery of the elbow]. Orthopade.
2003;32:717-22. .[CrossRef][PubMed]
Athwal
GS;
Morrey
BF. Revision total elbow arthroplasty for prosthetic fractures. J Bone Joint Surg Am.
2006;88:2017-26.[CrossRef][PubMed]
Robinson
CM;
Hill
RM;
Jacobs
N;
Dall
G;
Court-Brown
CM. Adult distal humeral metaphyseal fractures: epidemiology and results of treatment. J Orthop Trauma.
2003;17:38-47.[CrossRef][PubMed]
Inglis
AE;
Pellicci
PM. Total elbow replacement. J Bone Joint Surg Am.
1980;62:1252-8.[PubMed]
Enneking
WF;
Dunham
W;
Gebhardt
MC;
Malawar
M;
Pritchard
DJ. A system for the functional evaluation of reconstructive procedures after surgical treatment of tumors of the musculoskeletal system. Clin Orthop Relat Res.
1993;286:241-6.[PubMed]
Glasser
D;
Langlais
F. The ISOLS radiological implants evaluation system. In: Langlais
F;
Tomeno
B, eds. Limb salvage: major reconstructions in oncologic and nontumoral conditions. Berlin-Heidelberg, New York: Springer; 1991. p 23-31.
Dominkus
M;
Sabeti
M;
Toma
C;
Abdolvahab
F;
Trieb
K;
Kotz
RI. Reconstructing the extensor apparatus with a new polyester ligament. Clin Orthop Relat Res.
2006;453:328-34.[CrossRef][PubMed]
Dominkus
M;
Sabeti
M;
Kotz
R. [Functional tendon repair in orthopedic tumor surgery]. Orthopade.
2005;34:556-9. .[CrossRef][PubMed]
Trieb
K;
Blahovec
H;
Brand
G;
Sabeti
M;
Dominkus
M;
Kotz
R. In vivo and in vitro cellular ingrowth into a new generation of artificial ligaments. Eur Surg Res.
2004;36:148-51.[CrossRef][PubMed]
Ross
AC;
Sneath
RS;
Scales
JT. Endoprosthetic replacement of the humerus and elbow joint. J Bone Joint Surg Br.
1987;69:652-5.[PubMed]
Kulkarni
A;
Fiorenza
F;
Grimer
RJ;
Carter
SR;
Tillman
RM. The results of endoprosthetic replacement for tumours of the distal humerus. J Bone Joint Surg Br.
2003;85:240-3.[CrossRef][PubMed]
Athwal
GS;
Chin
PY;
Adams
RA;
Morrey
BF. Coonrad-Morrey total elbow arthroplasty for tumours of the distal humerus and elbow. J Bone Joint Surg Br.
2005;87:1369-74.[CrossRef][PubMed]
Inglis
AE. Revision surgery following a failed total elbow arthroplasty. Clin Orthop Relat Res.
1982;170:213-8.[PubMed]
Figgie
HE
3rd;
Inglis
AE;
Ranawat
CS;
Rosenberg
GM. Results of total elbow arthroplasty as a salvage procedure for failed elbow reconstructive operations. Clin Orthop Relat Res.
1987;219:185-93.[PubMed]
Sanchez-Sotelo
J;
O'Driscoll
S;
Morrey
BF. Periprosthetic humeral fractures after total elbow arthroplasty: treatment with implant revision and strut allograft augmentation. J Bone Joint Surg Am.
2002;84:1642-50.[CrossRef][PubMed]
Cil
A;
Veillette
CJ;
Sanchez-Sotelo
J;
Morrey
BF. Linked elbow replacement: a salvage procedure for distal humeral nonunion. J Bone Joint Surg Am.
2008;90:1939-50.[CrossRef][PubMed]
Peden
JP;
Morrey
BF. Total elbow replacement for the management of the ankylosed or fused elbow. J Bone Joint Surg Br.
2008;90:1198-204.[CrossRef][PubMed]
Jeys
LM;
Luscombe
JS;
Grimer
RJ;
Abudu
A;
Tillman
RM;
Carter
SR. The risks and benefits of radiotherapy with massive endoprosthetic replacement. J Bone Joint Surg Br.
2007;89:1352-5.[CrossRef][PubMed]
Holzer
G;
Windhager
R;
Kotz
R. One-stage revision surgery for infected megaprostheses. J Bone Joint Surg Br.
1997;79:31-5.[CrossRef][PubMed]
Jeys
LM;
Grimer
RJ;
Carter
SR;
Tillman
RM. Periprosthetic infection in patients treated for an orthopaedic oncological condition. J Bone Joint Surg Am.
2005;87:842-9.[CrossRef][PubMed]
Cheung
EV;
Adams
RA;
Morrey
BF. Reimplantation of a total elbow prosthesis following resection arthroplasty for infection. J Bone Joint Surg Am.
2008;90:589-94.[CrossRef][PubMed]
Hardes
J;
von Eiff
C;
Streitbuerger
A;
Balke
M;
Budny
T;
Henrichs
MP;
Hauschild
G;
Ahrens
H. Reduction of periprosthetic infection with silver-coated megaprostheses in patients with bone sarcoma. J Surg Oncol.
2010;101:389-95. [PubMed]