Secondary Logo

Journal Logo


What Is the Implant Survivorship and Functional Outcome After Total Humeral Replacement in Patients with Primary Bone Tumors?

Schneider, Kristian Nikolaus MD1; Bröking, Jan Niklas MD1; Gosheger, Georg MD1; Lübben, Timo MD1; Hardes, Jendrik MD1; Schorn, Dominik MD1; Smolle, Maria Anna MD2; Theil, Christoph MD1; Andreou, Dimosthenis MD1

Author Information
Clinical Orthopaedics and Related Research: August 2021 - Volume 479 - Issue 8 - p 1754-1764
doi: 10.1097/CORR.0000000000001677



The humerus is the third most common site affected by bone sarcomas [2]. In locally advanced tumors, patients with skip metastases or pathologic fractures, and in those who have undergone contaminating intralesional surgery, surgical treatment can involve resection of the entire humerus to achieve a wide margin [21]. The most common reconstructive option in limb-sparing procedures is endoprosthetic replacement using modular megaprostheses, which are widely available and allow for varying reconstruction lengths in 1- to 2-cm increments, different fixation options, and early functional rehabilitation [8, 9, 13, 21]. The main downside, however, is a fairly high risk of revision surgery over time, especially considering the improving survival rates of patients with extremity bone sarcomas [6, 8, 19].

Because of the rarity of total humeral replacement, implant survivorship and functional outcomes have been investigated in only a few previous studies; these studies were limited by patient heterogeneity, inconsistent indications for the prosthesis (including using total humeral replacement as a revision implant), and patients with bone metastases, which makes it difficult to apply the findings from these studies to patients with primary bone tumors [13, 15, 21]. Furthermore, previous studies did not investigate potential factors that may have been associated with an increased risk of prosthetic revision surgery, although these factors might greatly affect prosthetic survival and may be important when advising a patient and planning potential adjuvant treatment [16, 19, 21]. In addition, functional outcome has been mainly evaluated using the 1993 version of the Musculoskeletal Tumor Society (MSTS) score [4], which although widely used in orthopaedic oncology [13, 15, 16, 20], only crudely addresses shoulder and elbow function.

Therefore, we decided to perform this study in patients with primary bone tumors undergoing primary total humeral replacement with a single modular system and attempted to answer the following questions: (1) What is the revision-free implant and overall limb survivorship after total humerus replacement? (2) What factors are associated with implant revision surgery? (3) What is the functional outcome of the procedure as determined by the Musculoskeletal Tumor Society (MSTS) score and the American Shoulder and Elbow Surgeons (ASES) score?

Patients and Methods

Between August 1999 and December 2018, 666 patients underwent megaprosthetic reconstruction after resection of a primary malignant or locally aggressive/rarely metastasizing tumor of the long bones with a single modular system (MUTARS system, Implantcast GmbH) at our department of a tertiary university hospital (Fig. 1). During that time, we generally performed total humeral replacement in all patients with a locally advanced sarcoma, in patients with pathological fractures, in patients with skip metastases, or in patients with previous intralesional contaminating surgery, where no sufficient bone stock for a stable implant fixation for a single joint megaprosthetic replacement of the proximal or distal humerus was available. We performed no biological reconstructions or reconstructions with allograft-prosthetic composites. In patients, where the tumor extension meant that no wide surgical margins could be achieved or a functional limb could be preserved, we performed an amputation, but since our database was created for the documentation of the MUTARS prosthesis, it did not include the number of patients treated with amputation. Based on that, 5% (33 of 666) of patients were eligible; two others were excluded because they received a custom-made three-dimensionally (3-D)-printed hemiprosthesis as the off-the-shelf implant was too large for the existing anatomy in these very young patients, leaving 5% (31 of 666) of patients for our retrospective analysis. Of these, 18 of 31 patients died of their disease, 3 of 31 live overseas, and 1 of 31 could not be contacted, leaving 9 of 31 patients with functional outcome data at last follow-up (Fig. 1). Of the four patients we were not able to evaluate for functional follow-up, two had their resection more than 5 years ago but have not be seen in the last 5 years. Up to that point, they had had an uneventful orthopaedic and oncological follow-up of over 5 years after their respective surgeries in 2008 and 2009. Both patients were discharged from regular oncological follow-up by their respective oncologists and wished to only return to our institution in case of new problems. As of January 2006, silver-coated implants were available and used in all patients from that date onward, and beginning in 2010, we used a reverse-shoulder proximal humerus implant in patients in whom the axillary nerve could be preserved. The resections and reconstructions were performed by a total of 10 different consultant orthopaedic oncologists over the years. We used no other megaprosthetic implant system or biological reconstruction for osteoarticular reconstructions of the upper extremity like fibular interpositions or allografts during the study period.

Fig. 1:
Study flow diagram.

Multimodal treatment (chemotherapy and/or radiotherapy) was administered, when necessary, based on the recommendations of the local tumor board. Patients 65 years and younger with osteosarcoma and high-grade spindle cell sarcomas of bone generally underwent pre- and postoperative chemotherapy, according to the at the time active protocols of the Cooperative Osteosarcoma Study Group (COSS-96, EURAMOS-1, EURO-B.O.S.S.). Patients with Ewing sarcoma underwent pre- and postoperative chemotherapy according to the active protocols of the Cooperative Ewing Sarcoma Study Group (EURO-E.W.I.N.G 99, Ewing 2008) at that time. Radiotherapy was generally recommended in patients with Ewing sarcoma who had a poor response to induction chemotherapy and patients with high-grade spindle cell sarcomas who had previous contaminating surgery after a pathological fracture.

Study Population and Surgical Treatment

The study population consisted of 31 patients (19 men) with a median BMI of 24 kg/m2 (interquartile range 20 to 28 kg/m2). Thirty patients had a primary high-grade bone sarcoma, and one patient had a recurrent giant cell tumor (Table 1). Anatomic location of the tumor was the proximal humerus in 20 patients and the humerus diaphysis in the remaining 11 patients. Although 11 patients had pathological fractures, the indication for total humeral resection was only given due to the fracture in six patients. In the other five patients, the fracture alone could have been treated with a proximal/distal humerus replacement but other factors (a skip lesion, local tumor extent, previous intralesional surgery) led to a total humeral resection (Table 1). Median (IQR) age at the time of surgery was 15 years (14 to 25 years). Median follow-up for all patients was 42 months (21 to 75 months), for all surviving patients, it was 75 months (50 to 122 months), and for all surviving patients with functional follow-up it was 75 months (50 to 148 months). The overall probability of patient survival was 66% (95% CI 49% to 83%) after 2 years and 44% (95% CI 24% to 61%) after 5 years.

Table 1. - Patient demographics and oncologic and surgical details (n = 31 patients)
Variable % (n)
% Men 61 (19)
Tumor entity
 Osteosarcoma 68 (21)
 Ewing sarcoma 10 (3)
 Leiomyosarcoma 6 (2)
 Sarcoma not otherwise specified 6 (2)
 Chondrosarcoma 3 (1)
 Angiosarcoma 3 (1)
 Recurrent giant cell tumor 3 (1)
Indication for total humeral resection
 Pathologic fracture 19 (6)
 Skip lesion 10 (3)
 Local tumor extent 68 (21)
 Previous intralesional surgery 3 (1)
Pathologic fracture 35 (11)
Distant metastases at diagnosis 43a (13)
Secondary distant metastases developed during follow-up 53a (16)
Extraarticular resection 39 (12)
Adjuvant treatment 83a (25)
 Radiotherapy 20a (6)
  Preoperative 3a (1)
  Postoperative 17a (5)
 Chemotherapy 87a (26)
  Preoperative 73a (22)
  Postoperative 70a (21)
Reverse shoulder arthroplasty 10 (3)
Ulnar stem fixation
 Cemented 10 (3)
 Cementless 90 (28)
Silver-coated implants 68 (21)
aOne patient with recurrent giant cell tumor was excluded from the descriptive analysis of metastases and adjuvant treatment.

The MUTARS system can be used with a cemented or cementless ulnar stem depending on local bone quality, with cementless implant fixation being the preferred approach. It was possible to preserve the axillary nerve in 65% (20 of 31) of all patients, the radial nerve was preserved in 84% (26 of 31) of all patients, and the musculocutaneous nerve in all but one patient. The deltoid muscle was largely preserved in 58% (18 of 31) of all patients and partially resected in the remaining patients. If the rotator cuff muscles could be at least partially preserved, they were meticulously attached to an attachment tube. All patients underwent a planned wide resection and had histological negative margins. The physeal status as determined on plain radiographs and MRI showed at least partially open physes in 14 patients at the proximal humerus and in four patients at the distal humerus. No patient had self-reported shoulder dysfunction before the oncological treatment. Beginning in 2010, we started using a reverse proximal humerus component in patients in whom the axillary nerve could be preserved and sufficient glenoid bone stock was available for anchoring the glenoid component (3 of 13). No plastic surgery procedures like flaps for soft tissue covering were performed in our cohort.

The patient with the recurrent giant cell tumor was included in the analyses of implant survival and complications but was excluded from all oncological analyses. Patients who died of their disease or underwent revision surgery were included, irrespective of the follow-up period. For all other patients, the minimum follow-up was 25 months. Eighteen patients died of their disease after a median (IQR) of 22 months (13 to 49 months). Both patients who underwent subsequent amputation died of their disease and were not included in the functional analyses.

Data Collection

Data regarding patient demographics, tumor characteristics, surgical and oncologic treatment, endoprosthetic complications and their subsequent treatment, the presence of metastases at the time of diagnosis, and development of local recurrence and secondary metastases during follow-up were retrospectively obtained from the patients’ medical records by four authors (KNS, JNB, CT, DA). In survivors, the functional outcome was determined at the last follow-up using the 1993 MSTS score and ASES scores [4, 17].

Description of Complications and Assessment of Functional Outcome

Endoprosthetic complications (soft tissue complications, aseptic loosening, structural complications, infection, and tumor progression) were obtained from patients’ records [8]. We determined the functional outcome in surviving patients—including those who underwent revision surgery, except for the two patients who underwent subsequent amputation and died in the meantime—using two standardized scoring systems. The MSTS score is a physician-reported scoring system developed to evaluate the function of patients with sarcoma undergoing limb-sparing surgery, with a numerical value (0 to 5) assigned to each of the following six categories for the upper extremity: pain, function, emotional acceptance, hand positioning, dexterity, and lifting ability [4]. The final score is classified as excellent (≥ 23 points), good (15-22 points), fair (8-14 points), and poor (< 8 points) [14]. The ASES score is a patient- and physician-reported scoring system developed to assess functional outcome after shoulder and elbow surgery and is weighted for pain (50%) and function (50%) [17]. Although pain is assessed on a VAS, function is determined using 10 separate questions, each scored on an ordinal scale from 0 to 3 [12, 17, 22]. The maximum score is 100 points, with lower scores indicating greater pain and disability. We opted for the ASES score, which is widely used in shoulder and elbow surgery, as it very specifically focuses on shoulder function after surgery, rather than the more general Toronto Extremity Salvage Score (TESS) score, which also addresses functional outcome after resection of tumors of the forearm and the hand, to allow a comparison of the functional outcome of total humerus replacement with that of endoprosthetic shoulder replacement caused by trauma or arthritis.

Primary and Secondary Study Outcomes

Our primary goal was to evaluate the revision-free implant survival after total humeral replacement by evaluating and assessing postoperative complications and revision surgeries, with death as a competing risk. Our secondary goal was to evaluate the following factors for possible association with implant revision surgery: age, BMI, reconstruction length, duration of surgery, extraarticular resection, pathological fracture, previous intralesional surgery, (neo-)adjuvant radio- and chemotherapy, and metastatic disease and to determine the functional follow-up in survivors using standardized scoring systems.

Ethical Approval

Ethical approval for this study was obtained from our local ethics committee (Ethik-Kommission Westfalen-Lippe, reference number 2019-526-f-S).

Statistical Analysis

The statistical analysis was performed using SPSS 25.0 (IBM Corp). The duration of follow-up and time to event (revision surgery or death) were calculated from the date of surgery to the date of the respective event or the last documented contact with the patient as of October 2020. Overall patient survival probabilities were calculated using the Kaplan-Meier method. Competing risk regression analyses of implant survivorship were performed with Stata Version 15.1 for Mac (StataCorp) using the package stcrreg. One- and 5-year implant survival and limb survival were calculated with the Stata add-on package stcrprep. The endpoint was any reoperation related to the total humeral replacement and results are given as survival at 1 and 5 years. Subhazard ratios, corresponding 95% confidence intervals and p values are provided. Depending on the distribution of data, as determined by the Kolmogorov-Smirnov test, nonparametric analyses were performed with the Mann-Whitney U-test, and parametric comparisons were made using a t-test. All p values are two-sided; a p value < 0.05 was considered significant.


Revision-free and Overall Limb Survivorship

Using a competing risk analysis, revision-free implant survivorship at 1 year was 77% (95% CI 58% to 89%) and 74% (95% CI 55% to 86%) at 5 years, with an overall limb survivorship after 1 and 5 years, of 93% (95% CI 75% to 98%), respectively. Nine patients underwent a revision procedure after a median (IQR) of 9 months (2 to 13 months). Sixteen percent (5 of 31) of patients had infections, 6% (2 of 31) had local recurrences, and 6% (2 of 31) had soft tissue complications. Of the five patients with infection, three underwent one-stage revision and two underwent two-stage exchange with temporary implantation of an antibiotic-impregnated cement spacer. Of the two patients with soft tissue complications, one patient underwent wound revision because of aseptic wound dehiscence, while the second patient had limited joint function with symptomatic, subjective instability and underwent a modified Bateman procedure (trapezius transfer with the acromial portion at the deltoid insertion) to improve function 6 years after the initial surgery [5]. The two patients with local recurrences underwent amputation 9 and 12 months postoperatively. Limb salvage was achieved in all remaining patients.

Factors Associated with Implant Revision Surgery

Patients who underwent extraarticular resection of the shoulder had lower revision-free implant survivorship than patients with intraarticular resections (50% [95% CI 21% to 74%] versus 89% [95% CI 64% to 97%] after 5 years (subhazard ratio for extraarticular resections 4.4 (95% CI 1.2 to 16.5); p = 0.03). With the numbers of patients available for our analysis, we could not detect a difference in revision-free survivorship at 5 years between patients who underwent postoperative radiotherapy (40% [95% CI 5% to 75%]) and patients who did not (80% [95% CI 58% to 91%]; p = 0.09), or between patients who had metastatic disease from their tumor at the time of diagnosis (54% [95% CI 25% to 76%]) and patients who did not (88% [95% CI 61% to 97%]; p = 0.06). The other factors we evaluated had no influence on the probability implant revision surgery (Table 2). Additionally, we found no differences in median age (16 versus 15 years; p = 0.53), median reconstruction length (190 versus 310 mm; p = 0.32), or the median duration of the initial surgery (239 versus 272 minutes; p = 0.57) between patients who underwent revision for an implant complication and patients who did not.

Table 2. - Clinical factors that were associated with implant revision procedures (n = 31 patients)
Characteristic Patients, % (n) Implant survival in % at 5 years 95% CI (range) Subhazard ratio (SHR) 95% CI for SHR (range) p value
No metastases at diagnosis 58 (18) 88 61-97
Distant metastases at diagnosis 42 (13) 54 25-76 3.55 0.95-13.27 0.06
No metastases during follow-up 48 (15) 87 56-97
Secondary metastases developed during follow-up 10 (3) 100 N/A 2.35 0.63-8-76 0.21
No radiotherapy 81 (25) 80 58-91
Radiotherapy 19 (6) 50 11-80 2.23 0.62-8.14 0.22
Radiotherapy preoperatively 3 (1) 100 N/A N/A N/A N/A
Radiotherapy postoperatively 16 (5) 40 5-75 2.95 0.83-10.5 0.09
No chemotherapy 16 (5) 80 20-97
Chemotherapy 84 (26) 73 52-86 1.5 0.17-12.7 0.72
No pathologic fracture 65 (20) 75 50-89
Pathologic fracture 35 (11) 73 37-90 0.88 0.23-3.4 0.86
No prior intralesional procedure 97 (30) 76 56-88
Prior intralesional procedure 3 (1) 50 1-91 2.14 0.36-12.8 0.4
BMI ≤ 25 kg/m2 29 (9) 83 48-96
BMI > 25 kg/m2 36 (11) 89 43-98 0.5 0.55-4.6 0.54

Functional Outcome Scores

For the 9 of 13 patients available for functional testing, the median (IQR) MSTS score was 87% (67% to 92%), with 6 of 9 patients achieving an excellent result and 3 of 9 patients obtaining a good result. The median ASES score was 83 (63 to 89) of 100 points; higher scores represent better function. Although all patients had excellent MSTS scores for pain, emotional acceptance, and manual dexterity, some patients reported limitations in hand positioning, lifting ability, and function in occupational and recreational activities (Table 3). For the ASES score, excellent results were achieved for pain while deficits were reported regarding overhead activities, such as lifting 10 pounds above shoulder level, throwing a ball overhand, and reaching a high shelf (Table 4). With the numbers we had, we could not detect a difference in the functional outcome between patients who underwent reverse shoulder replacements and those who underwent anatomic shoulder replacements (median ASES score 83 versus 82; p > 0.99; median MSTS score 88% versus 77%; p = 0.67). The low number of evaluated patients precluded further meaningful statistical analyses regarding function.

Table 3. - Functional assessment of the nine surviving patients using the MSTS score
MSTS score by patient Pain Function Emotional acceptance Hand positioning Manual dexterity Lifting ability Total score
1 5 4 5 4 5 4 27
2 5 4 5 4 5 4 27
3 5 3 5 3 5 5 26
4 5 5 5 4 5 5 29
5 5 4 5 4 5 5 28
6 4 3 4 3 4 3 21
7 5 3 5 3 4 3 23
8 5 2 4 2 4 2 19
9 4 1 4 1 4 1 15

Table 4. - Functional assessment of the nine surviving patients using the ASES score
ASES score by patient Pain Is it difficult for you to put on a coat? Is it difficult for you to wash your back/do up your bra? Is it difficult for you to comb your hair? Is it difficult for you to lift 10 lbs. (4.5 kg) above your shoulder? Is it difficult for you to do your usual work?
1 0 2 0 2 0 3
2 0 3 1 3 2 3
3 0 3 1 3 1 3
4 0 3 1 2 3 3
5 0 3 1 2 2 3
6 1 1 0 0 0 2
7 0 1 0 0 1 0
8 0 3 1 2 1 3
9 1 2 0 0 0 0
ASES score by patient Is it difficult for you to sleep on the affected side? Is it difficult for you manage toileting? Is it difficult for you to reach a high shelf? Is it difficult for you to throw a ball overhand? Is it difficult for you to do your usual sport/leisure activity? Total scorea
1 3 2 1 0 2 75
2 3 3 2 2 2 90
3 3 3 1 2 3 88
4 3 3 2 3 3 93
5 3 3 2 2 2 88
6 2 1 0 0 3 60
7 2 1 1 2 1 65
8 3 3 1 1 2 83
9 1 1 0 0 0 52
aThe total ASES score is a maximum of 100 points, with pain and function each weighted at 50%; the responses to the 10 questions regarding function are each scored with an ordinal scale from 0 to 3 (maximum raw score: 30 points); we opted to include the total ASES score to provide insight on the high/low scores, depending on patients’ individual performance in each subcategory.


Complete humeral resection followed by megaprosthetic total humeral replacement is one of the few limb-sparing surgical treatments in patients with locally advanced bone sarcomas, pathologic fractures, skip metastases, or after previous contaminating intralesional surgery. These massive prostheses are at a greater risk for complications than single-joint megaprostheses, especially given improving patient survival; therefore, identifying potential risk factors for revision surgery is important to improve the long-term survival of these reconstructions [19]. Our analysis showed a high probability of implant revisions in the first year postoperatively, followed by a much lower risk later on, and we identified extraarticular shoulder resections as an important risk factor for complications and subsequent revision surgery. Although only 42% (13 of 31) of our patients survived with a median follow-up of 75 months (IQR 50 to 122), those who survived maintained a good-to-excellent functional outcome.


We acknowledge several limitations of our study, many of which are associated with its retrospective design. Our results are based on patients’ records and may be subject to selection bias. As such, the frequencies of our reported implant revisions may only be a low-end estimate because patients might have sought treatment elsewhere. Indeed, two of our surviving patients were not seen at our department in the last 5 years, leading to a potential transfer bias. However, considering the rarity of this procedure, it appears unlikely that patients undergo surgery at nonspecialized sarcoma centers, limiting the impact of this factor. Inevitably, some new developments in our megaprosthetic reconstruction system during the study period such as silver coating, which we started using regularly in 2006, caused for some inhomogeneity in our cohort. On the other hand, we did not use biological reconstructions or allograft-prosthetic composites in our department after total humerus replacement during the study period, and the total humerus replacement was the only reconstruction after limb-sparing surgery we performed. Given the retrospective study design, we cannot exclude the possibility of assessment bias because the treating surgeons decided which patients got revised and which did not, and that might influence our results. However, during the study period, patients were not assigned to one specific consultant at our department but were treated by a team of experienced orthopaedic oncologists, limiting the possible impact of this bias.

Our functional results were only available in 9 of 13 survivors; three patients live overseas and one patient could not be contacted, which may have inflated the reported functional scores. However, we repeatedly tried to contact all surviving patients through several methods, to avoid missing patients who were dissatisfied with their treatment. Finally, we were not able to identify potential preoperative dysfunctions, as patient-reported outcome measures were only obtained postoperatively. Thus, postoperative functional results must be interpreted with caution as the individual baseline outcome score is not known.

Revision-free and Overall Limb Survivorship

The probability of revision-free implant survivorship in our cohort was only 77% in the first year but remained at the same level in the following years, with an overall limb survivorship of 93% at 1 and 5 years. We were unable to find studies on total humeral replacement in oncological patients estimating implant and limb survivorship using the competing-risk approach we adopted in this study. Therefore, we could not compare our finding to those of previous studies, but we believe that our findings can be used as a reference value for future studies on total humeral replacement; these studies should use the competing-risk approach given the proportion of patients dying of their disease. Until such studies are available and considering the local tumor extent the candidates for total humeral replacement present with, we believe that megaprosthetic replacement of the total humerus is a good reconstruction method; however, patients should be counseled on the high risk for revision surgery in the first postoperative year. The most common complication in our cohort was periprosthetic infection, which occurred in 16% (5 of 31) of patients. A comparable infection rate of 13% was reported by a study of a more heterogeneous cohort of 34 patients with total humeral replacements [21]; 29 of these procedures were performed as primary reconstructions for bone sarcomas, metastatic carcinomas, or locally aggressive tumors and five were performed as a salvage procedure for previous reconstructive techniques. A study of 20 patients who underwent total humeral replacement after resection of a primary bone sarcoma reported no infections [16], and a smaller study of 11 patients with primary bone tumors and patients with aggressive fibromatosis observed no infections [15]. A possible reason for this discrepancy is that the patients in those two studies appeared to have undergone a different oncological management. For instance, no patients in these two studies underwent postoperative radiotherapy compared with 17% of the patients in our cohort, although both studies also reported a higher local recurrence rate (18% and 26%) than in our analysis (6%). Similar to the results of our study, a study of 20 patients with total humeral replacement after resection of bone sarcomas and metastases observed two infectious complications [13]. Both patients undergoing revision surgery had received postoperative radiation treatment [13].

Factors Associated with Implant Revision Surgery

The only risk factor for the implant complications we identified in our cohort was extraarticular shoulder resection. More than half of our patients who underwent an extraarticular tumor resection underwent revision surgery in the first year postoperatively. To our knowledge, no study has investigated the impact of extraarticular tumor resection on implant revision in total humeral replacement because extraarticular resections were deemed an exclusion criterion in previous studies [13, 15, 16, 21]. These procedures may be considered necessary in patients with tumors contaminating the joint and have been associated with a higher risk of implant complications in megaprosthetic reconstructions [1, 7]. A study of 54 patients undergoing extraarticular tumor resections of the shoulder followed by proximal humeral replacement found that 35% of the patients underwent revision surgery for complications, with soft tissue conditions being the most common reason for revision [1]. Poor soft tissue coverage after extraarticular tumor resections and megaprosthetic reconstructions has been identified as a main reason for the higher rate of implant complications these procedures are associated with [1, 7]. Therefore, we believe that future studies should evaluate whether the use of soft tissue coverage with plastic reconstructive surgery in patients undergoing extraarticular shoulder resections followed by total humeral replacement might decrease the risk of early revisions. With the numbers of patients available for our analysis, we found no differences in revision-free survivorship between patients who underwent postoperative radiotherapy and patients who did not. However, postoperative radiotherapy has been shown to be associated with a higher risk for implant revision in large cohorts of oncological patients undergoing megaprosthetic replacements in general [10, 11, 19], so we believe that future, larger studies should reevaluate the effect of postoperative radiotherapy on implant survival in patients undergoing total humeral replacement.

Functional Outcome Scores

The nine patients available for functional follow-up had good-to-excellent functional outcome scores. These results are comparable to or possibly slightly better than those previously reported in the evidence, with MSTS scores ranging from 72% to 83% [3, 13, 15, 21]. Our reported functional deficits in lifting abilities are also comparable with those reported by previous studies [3, 13, 21]. Furthermore, we also used the ASES score to analyze shoulder function more specifically, although future studies on megaprosthetic reconstruction should evaluate and ultimately validate other scoring systems to assess patient functional outcome [22]. One small study evaluated the functional outcome of 14 patients undergoing intraarticular resection for malignancies of the proximal humerus and subsequent megaprosthetic reconstruction and reported a similar mean ASES score of 85 [18]. Furthermore, the patients in that study displayed similar functional deficits when reaching a high shelf, lifting 10 pounds, and throwing a ball overhand, while nearly excellent score results were achieved for pain [18]. Despite variations in functional deficits, our analysis identified a patient who underwent total humeral replacement at the age of 16 after a pathological diaphyseal humerus fracture from a high-grade osteosarcoma (Fig. 2) with an excellent postoperative functional outcome noted at 10 years of follow-up, when the patient was 26 years old (see Video; Supplemental Digital Content,, although it should be noted that this outcome is not representative for most patients.

Fig. 2:
A-C (A) An AP radiograph and (B) MRI show a pathologic humeral shaft fracture in a 16-year-old boy with locally advanced high-grade osteosarcoma who presented at 10 years’ follow-up with an excellent functional outcome (see Videoa; Supplemental Digital Content, and (C) no radiographic signs of implant complications. aThe patient was 26 years of age when the video was filmed and has provided written consent for its publication.


Total humeral replacement with a modular megaprosthesis after resection of locally advanced bone tumors appears to be associated with a good functional outcome in patients who do not die of their tumor, which in our study was approximately one-third of those who were treated with a resection and total humerus prosthesis. However, the probability of early prosthetic revision surgery is high and appears to be higher in patients undergoing extra-articular resections, who should be counseled accordingly. Still, our results suggest that if the prosthesis survives the first year, the further risk for revision appears to be low. Future studies should re-examine the impact of postoperative radiotherapy on implant survival in a larger cohort—probably in a multicenter setting, given the rarity of this procedure—and evaluate whether the use of soft tissue coverage with plastic reconstructive surgery might decrease the risk of early revisions, especially in patients undergoing extraarticular resections.


We thank all our patients and their families for their continuous support of our work and acknowledge the physicians, nurses, and support staff who contributed to the treatment of the patients included in this study.


1. Angelini A, Mavrogenis AF, Trovarelli G, et al. Extra-articular shoulder resections: outcomes of 54 patients. J Shoulder Elbow Surg. 2017;26:e337-e345.
2. Below C. Dahlins Bone Tumors: General Aspects And Data On 11 087 Cases. 5th ed. Lippincott-Raven; 1996.
3. Bernthal NM, Upfill-Brown A, Burke ZDC, et al. Long term outcomes of total humeral replacement for oncological reconstructions: a single institution experience. J Surg Oncol. 2020;122:778-786.
4. 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):241-246.
5. Gosheger G, Hardes J, Ahrens H, Gebert C, Winkelman W. Endoprosthetic replacement of the humerus combined with trapezius and latissimus dorsi transfer: a report of three patients. Arch Orthop Trauma Surg. 2005;125:62-65.
6. Grimer RJ, Aydin BK, Wafa H, et al. Very long-term outcomes after endoprosthetic replacement for malignant tumours of bone. Bone Joint J. 2016;98-B:857-864.
7. Hardes J, Henrichs M-P, Gosheger G, et al. Endoprosthetic replacement after extra-articular resection of bone and soft-tissue tumours around the knee. Bone Joint J. 2013;95:1425-1431.
8. Henderson ER, Groundland JS, Pala E, et al. Failure mode classification for tumor endoprostheses: retrospective review of five institutions and a literature review. J Bone Joint Surg Am. 2011;93:418-429.
9. Henrichs M-P, Krebs J, Gosheger G, et al. Modular tumor endoprostheses in surgical palliation of long-bone metastases: a reduction in tumor burden and a durable reconstruction. World J Surg Oncol. 2014;12:330.
10. Jeys LM, Luscombe JS, Grimer RJ, Abudu A, Tillman RM, Carter SR. The risks and benefits of radiotherapy with massive endoprosthetic replacement. Bone Joint J. 2007;89-B:1352-1355.
11. Kapoor SK, Thiyam R. Management of infection following reconstruction in bone tumors. J Clin Orthop Trauma. 2015;6:244-251.
12. Kocher MS, Horan MP, Briggs KK, et al. Reliability, validity, and responsiveness of the American Shoulder and Elbow Surgeons subjective shoulder scale in patients with shoulder instability, rotator cuff disease, and glenohumeral arthritis. J Bone Joint Surg Am. 2005;87:2006-2011.
13. Kotwal S, Moon B, Lin P, Satcher R Jr, Lewis V. Total humeral endoprosthetic replacement following excision of malignant bone tumors. Sarcoma. 2016;2016:6318060.
14. Mohler DG, Chiu R, McCall DA, Avedian RS. Curettage and cryosurgery for low-grade cartilage tumors is associated with low recurrence and high function. Clin Orthop Relat Res. 2010;468:2765-2773.
15. Natarajan M, Sameer M, Kunal D, Balasubramanian N. Custom-made endoprosthetic total humerus reconstruction for musculoskeletal tumours. Int Orthop. 2012;36:125-129.
16. Puri A, Gulia A. The results of total humeral replacement following excision for primary bone tumour. J Bone Joint Surg Br. 2012;94:1277-1281.
17. Richards RR, An KN, LU Bigliani, et al. A standardized method for the assessment of shoulder function. J Shoulder Elbow Surg. 1994;3:347-352.
18. Tang X, Guo W, Yang R, Tang S, Tau J. Synthetic mesh improves shoulder function after intraarticular resection and prosthetic replacement of proximal humerus. Clin Orthop Relat Res. 2015;473:1464-1471.
19. Theil C, Röder J, Gosheger G, et al. What is the likelihood that tumor endoprostheses will experience a second complication after first revision in patients with primary malignant bone tumors and what are potential risk factors? Clin Orthop Relat Res. 2019;477:2705-2714.
20. Tunn PU, Pomraenke D, Goerling U, Hohenberger P. Functional outcome after endoprosthetic limb-salvage therapy of primary bone tumours--a comparative analysis using the MSTS score, the TESS and the RNL index. Int Orthop. 2008;32:619-625.
21. Wafa H, Reddy K, Grimer R, et al. Does total humeral endoprosthetic replacement provide reliable reconstruction with preservation of a useful extremity? Clin Orthop Relat Res. 2015;473:917-925.
22. Wylie JD, Beckmann JT, Granger E, Tashjian RZ. Functional outcomes assessment in shoulder surgery. World J Orthop. 2014;5:623-633.

Supplemental Digital Content

© 2021 by the Association of Bone and Joint Surgeons