- Email: [email protected]
Functional and oncological outcomes after limb-salvage surgery for primary sarcomas of the upper limb
Journal of Plastic, Reconstructive & Aesthetic Surgery (2008) 61, 382e387
Functional and oncological outcomes after limb-salvage surgery for primary sarcomas of the upper limb* E.H.C. Wright a,*, S. Gwilym b, C.L.M.H. Gibbons c, P. Critchley d, H.P. Giele d a
Royal Berkshire Hospital, Reading, Berkshire, Reading, UK John Radcliffe Hospital, Headley Way, Headington, Oxford, UK c Nuffield Orthopaedic Centre, Headington, Oxford, UK d Radcliffe Infirmary, Oxford, UK b
Received 13 March 2006; accepted 23 January 2007
KEYWORDS Upper limb; Salvage; Sarcoma; Functional; Outcome
Summary The surgical treatment of upper limb sarcoma poses an oncological and reconstructive challenge. Limb-salvage surgery aims to balance adequate excision margins for disease control and preservation of all important structures to retain maximum function. Reported here is an assessment of the functional and oncological outcomes of limb salvage surgery for primary sarcoma of the upper limb and limb girdle in 72 patients referred to a specialist musculoskeletal tumour unit over 9 years. All patients underwent excision of the sarcoma with reconstruction and adjuvant treatment as needed. Functional outcome was assessed using the Toronto Extremity Salvage Score (TESS) after discharge from hospital. The upper limb sarcomas treated by limb-salvage surgery achieved planned margins of excision in 85% of cases with primary surgery. This increased to 100% with re-excision, resulting in local recurrence in 15% and survival of 75% among those at 5 years or more after surgery, while retaining good to excellent function (TESS mean of 87 out of 100). A total of 38 patients completed TESS questionnaires and, as a single group, had a mean TESS of 87. Patient age, anatomical site of tumour and adjuvant treatment made no significant difference to TESS. Liposarcomas had a significantly better TESS than leiomyosarcomas, chondrosarcomas and malignant peripheral nerve sheath tumours. Oncological outcome was assessed in terms of excision margins achieved, local recurrence, re-excision and disease-free survival. Eight patients died of disease, all with high-grade primary tumours. Fourteen had local recurrence, four low-grade disease and the remaining high-grade disease. All four low-grade recurrences were successfully re-excised, as were five of the high-grade recurrences. The remaining five died of disease, with
*
Presented in part as a podium presentation at the British Orthopaedic Oncological Society Meeting, Oxford, 2005. * Corresponding author. Address: 74 Elliotts Way, Caversham, Reading, RG4 8BF, UK. E-mail address: [email protected] (E.H.C. Wright).
1748-6815/$ - see front matter ª 2007 Published by Elsevier Ltd on behalf of British Association of Plastic, Reconstructive and Aesthetic Surgeons. doi:10.1016/j.bjps.2007.01.080
Functional and oncological outcomes after limb-salvage surgery for primary sarcomas of the upper limb
383
or without further recurrences. Consistently good functional results as measured by TESS were reported by responders, but high-grade disease and early recurrence were identified as predictors of recurrence and death from disease, respectively. ª 2007 Published by Elsevier Ltd on behalf of British Association of Plastic, Reconstructive and Aesthetic Surgeons.
Before 1980, bone and soft-tissue sarcomas of the extremities were treated almost exclusively with amputation in order to achieve adequate margins to control disease.1 This was particularly true for the upper limb where the combination of multiple neurovascular structures and small volumes of tissue made adequate tissue margins difficult to achieve. Wide local excision has a greater risk of local recurrence, whereas amputation has significant morbidity because of loss of function and disfigurement. No statistically significant survival difference was found between amputation and limb salvage.2 Limb-salvage surgery aims to strike a balance between achieving sufficient margins in order to control local disease, and preserving critical neurovascular and musculoskeletal structures in order to retain maximal function of the affected limb.3 Until recently, no study had shown a statistically significant improvement in functional outcome in limb salvage compared with ablative surgery for upper extremity sarcoma. The functional benefits of limb-salvage procedures for sarcomas in the lower limb, however, have been demonstrated.4
Aim The aim of this study was to evaluate the functional and oncological outcomes of limb-salvage surgery for primary sarcomas of the upper limb and shoulder girdle.
Patients and methods Patients from the Oxford Bone and Soft-tissue Tumour Registry, who had undergone surgery for soft-tissue or bone tumours of the upper limb and shoulder girdle between 1997 and 2005 at the Nuffield Orthopaedic Centre, Oxford, and the Radcliffe Infirmary, Oxford, were selected for inclusion in the study. Approval was received from the Oxfordshire Research and Ethics Committee A02-068. Operation notes and pathology reports of all patients were consulted. The inclusion criteria were primary soft-tissue or bone sarcoma arising in the upper limb or shoulder girdle; the exclusion criteria were treatment with primary amputation; disease not amenable to surgery; and inadvertent treatment of an undiagnosed sarcoma at a referring hospital followed by amputation. All patients were seen and assessed in a multidisciplinary clinic. Initial investigations included ultrasound-guided biopsy of the tumour, whole-limb magnetic resonance imaging, and plain films of the tumour and limb. Staging investigations included a computed tomography scan of the chest, isotope bone scans and blood chemistry. Surgery was carried out by surgeons with an interest in sarcoma surgery and limb reconstruction. Multidisciplinary
procedures typically involved patients with extensive involvement of neurovascular structures or potentially complex soft-tissue reconstructions. Where low-grade disease was diagnosed close to functionally important structures, a planned marginal excision was carried out. All surgical specimens were reviewed by a single musculoskeletal pathologist experienced in examining sarcoma specimens, and panelled for further opinions when necessary. The tumours were graded according to the guidelines of the Musculoskeletal Tumour Society. Margins were reported as clear if a layer of normal tissue was included at all resection margins, marginal if the excision involved, but did not breach, the pseudocapsule encasing the tumour, and positive if the abnormal cells abutted the resection margins. Patients were reviewed at a multi-disciplinary meeting (including an oncology team specialising in the treatment of sarcomas) and adjuvant treatment planned. After discharge, the patients were then reassessed with the Toronto Extremity Salvage Score (TESS) questionnaire by post. TESS is a well-validated, disease-specific self-assessment5 questionnaire completed by the patient after either upper or lower limb-salvage surgery for musculoskeletal tumours.6 The patient grades 29 activities of daily living using a five-point Likert-type scale from 1, impossible to do, to 5, not at all difficult. At the end of the questionnaire, the patient grades the overall difficulty of all activities undertaken in the previous week using the same scale, and their overall level of disability from 1, completely disabled, to 5, not disabled at all. The patient’s score for each daily living activity is added to form a raw score. Activities that the patients deem not applicable to them can be marked as such on the form, and the maximum possible score is adjusted accordingly. The TESS is then calculated as:
100
Raw score-minimum score ZTESS Maximum score
Thus, a better function, as perceived by the patient, gives a higher TESS. Patients who had undergone primary limb salvage and then amputation for disease control were not assessed with the TESS questionnaire, but were included in the study, as they represent one possible outcome of limb-salvage surgery. The patients’ demographics, presentation, operation details, sarcoma type, grade, margins and metastases status and TESS were entered onto the database. Postal reminders were sent to non-responders. Oncological outcome was assessed in terms of excision margins achieved, local recurrence, re-excision and disease-free survival
384
E.H.C. Wright et al.
Results This study spanned a period of 9 years, reflecting the relative rarity of upper limb sarcomas amenable to limbsalvage surgery. The mean follow-up period was 34 months (3e96 months), with 26 patients followed up to 5 years or longer. Seventy-two patients were included (37 men and 35 women), with no significant difference in mean age (55.7 [men] and 57.9 [women] years). Their anatomical distribution was 28 of the shoulder girdle, 23 of the arm, 22 of the forearm and one of the hand Table 1. At presentation, all had a mass with or without loss of function, one had lymph-node involvement alone, and one had lymph-node and lung metastases. A total of seven patients had been treated elsewhere before being referred either for excision of residual disease or for definitive treatment of a previously undiagnosed sarcoma Table 2. Excision margins were inadequate in 11 patients who then underwent re-excision, and clear margins were achieved in all cases. None of the patients had local disease recurrence, but one patient did develop pulmonary metastases. Fifty-nine surviving patients suitable for TESS assessment were surveyed, and 38 questionnaires were returned completed, three were returned blank (a 69% overall response, with 64% completed). The mean TESS score for the group was 87 (range 36e 100). Overall, 22 patients rated their daily activities over the previous week as ‘not at all difficult’, and none as ‘impossible’. Twenty-one patients felt that they were ‘not at all disabled’ and none felt that they were ‘completely disabled’ Tables 3, 4 and 5. In the study, 14 patients had local disease recurrence. Twelve had clear, one marginal and one positive margins at initial surgery. Of these 14 patients, nine (64%) had highgrade disease and one other had low-grade disease, with vascular invasion that recurred as high-grade. Only four (29%) had consistently low-grade disease. This is in contrast
Table 1
Discussion
Summary of Sarcomas by Type
Bone-derived (n Z 29)
Soft tissue-derived (n Z 43)
to the recurrence-free group, in which 13 (22%) had highgrade and 45 (78%) had low-grade disease. In the local recurrence group, three were too old for adjuvant treatments and one could not receive radiotherapy, as the lymphangiosarcoma had arisen within a radiotherapy field for previous breast cancer. This patient also had the only positive margin at primary surgery, owing to the sarcoma involving the circumference of the arm and the axilla, and had a suspicious nodule on chest computed tomography before surgery. The marginal excision for the malignant peripheral nerve sheath tumour was planned as such to preserve function. Eleven of the recurrent disease group underwent reexcision, and clear margins were achieved in all. Two underwent amputation, one for recurrent Ewing’s sarcoma, who is currently disease-free, and the other for a highgrade rhabdomyosarcoma, who died of pulmonary metastatic disease. Of the 11 who underwent re-excision, three had recurrences, all high-grade disease, and two with vascular invasion. These were excised completely in two cases and incompletely excised in one. All three had recurrences that were operated upon, with two completely excised and one incompletely excised (who subsequently died of spinal metastatic disease). Of the two with clear margins, one is disease-free under follow-up, and one died of pulmonary metastatic disease after another complete excision. Overall, the mortality rate from disease in the group with local recurrences was 36%. Of the patients to be followed up to 5 years or more after surgery, 20 of the 26 were alive without disease (75% 5-year survival). Three patients in the group who did not develop local recurrences (3/58 [5%]) died of disease, two with pulmonary metastases and one with pulmonary and spinal metastases. All three of these patients had high-grade disease. Three other patients died of unrelated comorbid disease without recurrences or metastases.
Sarcomas
Number of cases
Chondrosarcoma Malignant fibrous histiocytoma Fibrosarcoma Osteosarcoma Ewing’s sarcoma Liposarcoma Leiomyosarcoma Malignant peripheral nerve sheath tumours Dermatofibrosarcoma Pleomorphic sarcoma Synovial sarcoma Lymphangiosarcoma Spindle cell sarcoma Rhabdomyosarcoma Haemangioendothelioma
16 cases 5 cases 4 cases 3 cases 1 case 17 cases 10 cases 7 cases 2 2 1 1 1 1 1
cases cases case case case case case
Amputation is best practice in wide excision of extremity sarcoma with adequate margin. However, the functional and psychological disability of amputation is high. If it is possible to excise the tumour preserving the function of the limb, then this would be the treatment of choice as long as the survival of the patient was not compromised. Other studies have shown no deterioration in survival when limbpreserving surgery is carried out. As expected, the local recurrence rate is increased,7 yet local recurrence does not affect metastasis and survival,8 provided that the local recurrence is adequately treated. Studies of the lower limb study have reported successful excision of the tumour in the region of 85%,9 local recurrence rates of 11%,10 survival of 40e50% at 5 years, and remaining function have been shown in the lower limb to be an American Musculoskeletal Tumour Society score of 37e100%. We report similar results for upper limb sarcomas, with planned margins of excision achieved in 85% of cases with primary surgery, increasing to 100% with re-excision, local recurrence at 15%, and survival at 75% among those at 5 years or more after surgery.
Functional and oncological outcomes after limb-salvage surgery for primary sarcomas of the upper limb Table 2
385
Surgical resection, reconstruction and adjuvant treatment by sarcoma type
Sarcoma type
Resection
n
Reconstruction
n
Adjuvant therapy
Bone derived (n Z 29)
Excision Proximal humeral resection Partial scapulectomy
11 6 4
Direct closure Endoprosthesis Ipsilateral latissimus dorsi flap Iliac bone graft Vascularised fibular graft Plate Latissimus dorsi free flap
17 4 3
Chemotherapy Radiotherapy Both
2 1 2
Total
5
Direct closure Tendon transfers Local flap Radial forearm flap Lateral arm free flap Groin free flap Split skin graft
26 2 3 1 1 1 2
Soft-tissue derived (n Z 43)
Curettage Excision of distal radius Excision of distal ulna
4 2 1
Tikhoff-Linberg extra-articular scapular resection Wide local excision Partial scapulectomy Axillary clearance Excision of common flexor origin Clavicular osteotomy
1 39 2 2 1
In our study, we found that successful excision of the sarcoma with the planned margin could be achieved in 61 out of 72 patients with primary surgery. Re-excision in the 11 cases of inadequate margins achieved clear margins in all cases, and there were no local recurrences in this group. Table 3 Toronto Extremity Salvage Score by age, tumour type, anatomical site, adjuvant given and recurrence Group
N
Mean TESS (þ/Standard Deviation)
Overall Aged 65 years Aged < 65 years Bone-derived sarcoma Chondrosarcoma Malignant fibrous histiocytoma/ fibrosarcoma Soft tissue-derived sarcoma Liposarcoma Leiomyosarcoma Malignant peripheral nerve sheath tumours High-grade disease Low-grade disease Anatomical site: shoulder Anatomical site: arm Anatomical site: forearm Adjuvant given: none Adjuvant given: Chemotherapy Adjuvant given: radiotherapy Adjuvant given: chemotherapy þ radiotherapy No recurrence Recurrence treated surgically
38 12 26 18 10 6
87 80 90 85 79 92
(þ/ (þ/ (þ/ (þ/ (þ/ (þ/
18) 23) 14) 20) 24) 12)
20 11 4 5
89 96 78 81
(þ/ (þ/ (þ/ (þ/
16) 6) 17) 24)
8 30 16 12 10 28 1 8 1
82 88 90 81 87 87 94 87 96
(þ/ (þ/ (þ/ (þ/ (þ/ (þ/ () (þ/ ()
22) 16) 16) 21) 18) 18)
32 6
90 (þ/ 14) 73 (þ/ 24)
TESS, Toronto Extremity Salvage Score.
20)
1
2 1 1 1
n
Chemotherapy Radiotherapy Both
1 10 5
Total
16
This is significantly better than other studies in which inadequate margins have been reported in up to half of cases.11 In one case, an incomplete excision was planned in a woman with metastases at presentation. This case shows that there is a sub-group of patients with limited survival due to untreatable metastatic disease for whom there is no advantage in achieving wide clearance by amputation. The local recurrence rate was 14 out of 72, which is in keeping with the local recurrence rates reported in other studies. When a local recurrence did occur, further local recurrences, despite re-excision, occurred in three out of 14 cases. Five patients with local recurrence later died of pulmonary metastases. Pulmonary metastases were only present in one case at preoperative staging (the case of the lymphangiosarcoma that had a planned incomplete excision), but it is impossible to determine if metastases were present, but undetectable at the time of staging (up to 85% of patients may have micro-metastases at presentation12) in the other four cases, or related to the local recurrence. Amputation was carried out in two patients with local recurrence (3% amputation rate), on the basis of absent distant metastases and inability to perform a further adequate wide excision while preserving a functional limb. Of these two patients, one is alive and disease-free and the other died of pulmonary metastases. Amputation as a secondary procedure after failed limb-salvage has been reported as about 9%;13 however, in a study looking specifically at treatment of local recurrences after limb-salvage surgery, a 30% amputation rate has been reported.14 The real aim of this study was to determine whether upper limb function was satisfactorily preserved in this limb-sparing surgery. This has been found in lower limb sarcomas,4 although the lower limb has less functional complexity than the upper limb. The TESS reported by our patients showed that those people who underwent limbsparing surgery had a good functional outcome. No difference was found at the anatomical level at which this surgery was performed. No significant difference was found
386
E.H.C. Wright et al.
Table 4 Local recurrences by tumour type, grade, initial excision margins, adjuvant therapy given and subsequent management Grade and tumour
Margin
Time to recurrence
Adjuvant
Re-excision
Further recurrence
High-grade synovial
Clear
24 months
Nil- too old
Clear
No
Low-grade malignant Clear fibrous histiocytoma Low grade Clear leiomyosarcoma vascular invasion positive
120 months Chemotherapy Clear and radiotherapy 8 months; Radiotherapy Clear high grade and immunotherapy
High-grade leiomyosarcoma High-grade rhabdomyosarcoma High-grade dermatofibrosarcoma
Clear
3 months
Clear
4 months
Clear
12 months
Chemotherapy and radiotherapy
Clear
Clear
12 months
Radiotherapy
Clear
Clear
36 months
Nil; too old
Clear
Clear
300 months Radiotherapy
Clear
Clear
4 months
Radiotherapy
Positive
Clear
18 months
Amputation
Positive
2 month
Chemotherapy and radiotherapy Nil (preventative radiotherapy for breast cancer on the same site) Nil; too old
Low-grade liposarcoma High grade leiomyosarcoma Low-grade fibrosarcoma High-grade malignant fibrous histiocytoma vascular invasion positive High-grade Ewing’s sarcoma High-grade lymphangiosarcoma
High-grade malignant peripheral nerve sheath tumours Low-grade chondrosarcomas
Marginal 3 months
24 months
Chemotherapy and radiotherapy Chemotherapy
Radiotherapy
in TESS between people over the age of 65 years and people under the age of 65 years. No difference was found in functional scores between patients with bone or soft-tissue sarcoma, indicating either Table 5 Time to recurrence of disease by grade of tumour and disease-free survival Patient group
N
Time to recurrence þ/e standard deviation
Low-grade disease High-grade disease Survivors or died of unrelated disease Non-survivors who died of disease
4 10 9
114þ/133 months 11þ/11 months 96þ/122 months
5
12þ/9 months
Clear Amputation
Died of chest metastases No Alive; no evidence of disease Yes, 3 months- Died of chest fully excised, metastases further recurrence fully excised Nil Alive; no evidence of disease No Died of chest metastases Yes 12 months- Died of spinal incompletely metastases excised No Alive; no evidence of disease No Alive; no evidence of disease Nil Alive; no evidence of disease 12 monthsCurrently in F/U, fully excised no recurrence
No
No, widespread NA metastases
Clear
Outcome
Nil
TikoffeLinberg Nil resection Clear
Alive; no evidence of disease Died of chest metastases
Died of unrelated disease Alive; no evidence of disease
group benefit from limb salvage surgery and reconstruction. Nor was there any difference in functional scores seen in patients receiving radiotherapy, suggesting that radiotherapy does not preclude limb-salvage surgery, and its benefit as an adjunct in limb-salvage surgery is well-documented.15 Surgical re-excision of recurrent disease did not adversely affect TESS compared with the group without recurrent disease, despite the distorted anatomy and altered anatomical landmarks frequently found in a previously operated, and often irradiated, site. Disease recurrence occurred more frequently in highgrade disease, within 2 years of excision, both of these factors predicting mortality from disease and metastasis. In conclusion, this study has shown that the principles of limb-salvage surgery are applicable to a wide range of tumour-types and grades, to all patient age groups and anatomical sites with good functional results, comparable to those in literature. Recurrences of disease can be
Functional and oncological outcomes after limb-salvage surgery for primary sarcomas of the upper limb successfully treated by re-excision. Early recurrence of high-grade disease carries a worse prognosis, especially if adjuvant therapy cannot be given for whatever reason.
References 1. Springfield DS. Introduction to limb-salvage surgery for sarcomas. Orthop Clin North Am 1991;22:1e5. 2. Rosenberg SA, Tepper J, Glatstein E, et al. The treatment of soft tissue sarcomas of the extremities: prospective randomised evaluation of (1) limb-sparing surgery plus radiation therapy compared with amputation and (2) the role of adjuvant therapy. Ann Surg 1982;196:305e15. 3. Gitelis S. Limb salvage for appendicular tumours. Curr Opin Orthop 1991;2:811e6. 4. Renard AJ, Veth PV, Schreuder HWB. Function and complications after ablative and limb-salvage therapy in lower extremity sarcoma of bone. J Surg Oncol 2000;73:198e205. 5. Davis AM, Bell RS, Bradley EM, et al. Evaluating functional outcome in patients with lower extremity sarcoma. Clin Orthop Relat Res 1999;358:90e100. 6. Davis AM, Wright JG, Williams JI, et al. Development of a measure of physical function for patients with bone and soft-tissue sarcoma. Qual Life Res 1996;5:508e16.
387
7. Creighton JJ, Clayton AP, Mindell ER. Primary malignant tumours of the upper extremity: retrospective analysis of one hundred and twenty-six cases. J Hand Surg 1985;10A: 805e14. 8. Brennan MF, Casper ES, Harrison LB. The role of multimodality therapy in soft-tissue sarcoma. Ann Surg 1991;214:328e38. 9. Kumta SM, Cheng JCY, Li CK. Scope and limitations of limbsparing surgery in childhood sarcomas. J Pediatr Orthop 2002;22:244e8. 10. Simon MA, Aschilman MA, Thomas N. Limb-salvage treatment versus amputation for osteosarcoma of the distal end of the femur. J Bone and Joint Surg 1986;68A:1331e7. 11. Collin C, Hadjdu SI, Godbold J. Localised operable soft tissue sarcoma of the upper extremity. Ann Surg 1987;205:331e9. 12. Weis LD. The success of limb-salvage surgery in the adolescent patient with osteogenic sarcoma. Adolesc Med 1999;10:451e8. 13. Bray PW, Bell RS, Bowen CVA. Limb salvage surgery and adjuvant radiotherapy for soft tissue sarcomas of the forearm and hand. J Hand Surg 1997;22A:495e503. 14. Catton C, Davis A, Bell R. Soft tissue sarcoma of the extremity. Limb salvage after failure of combined conservative therapy. Radiother Oncol 1996;41:209e14. 15. Karakousis CP, Proimkis C, Walsh DL. Primary soft tissue sarcoma of the extremities in adults. Br J Surg 1995;82: 1208e12.
Functional and oncological outcomes after limb-salvage surgery for primary sarcomas of the upper limb* E.H.C. Wright a,*, S. Gwilym b, C.L.M.H. Gibbons c, P. Critchley d, H.P. Giele d a
Royal Berkshire Hospital, Reading, Berkshire, Reading, UK John Radcliffe Hospital, Headley Way, Headington, Oxford, UK c Nuffield Orthopaedic Centre, Headington, Oxford, UK d Radcliffe Infirmary, Oxford, UK b
Received 13 March 2006; accepted 23 January 2007
KEYWORDS Upper limb; Salvage; Sarcoma; Functional; Outcome
Summary The surgical treatment of upper limb sarcoma poses an oncological and reconstructive challenge. Limb-salvage surgery aims to balance adequate excision margins for disease control and preservation of all important structures to retain maximum function. Reported here is an assessment of the functional and oncological outcomes of limb salvage surgery for primary sarcoma of the upper limb and limb girdle in 72 patients referred to a specialist musculoskeletal tumour unit over 9 years. All patients underwent excision of the sarcoma with reconstruction and adjuvant treatment as needed. Functional outcome was assessed using the Toronto Extremity Salvage Score (TESS) after discharge from hospital. The upper limb sarcomas treated by limb-salvage surgery achieved planned margins of excision in 85% of cases with primary surgery. This increased to 100% with re-excision, resulting in local recurrence in 15% and survival of 75% among those at 5 years or more after surgery, while retaining good to excellent function (TESS mean of 87 out of 100). A total of 38 patients completed TESS questionnaires and, as a single group, had a mean TESS of 87. Patient age, anatomical site of tumour and adjuvant treatment made no significant difference to TESS. Liposarcomas had a significantly better TESS than leiomyosarcomas, chondrosarcomas and malignant peripheral nerve sheath tumours. Oncological outcome was assessed in terms of excision margins achieved, local recurrence, re-excision and disease-free survival. Eight patients died of disease, all with high-grade primary tumours. Fourteen had local recurrence, four low-grade disease and the remaining high-grade disease. All four low-grade recurrences were successfully re-excised, as were five of the high-grade recurrences. The remaining five died of disease, with
*
Presented in part as a podium presentation at the British Orthopaedic Oncological Society Meeting, Oxford, 2005. * Corresponding author. Address: 74 Elliotts Way, Caversham, Reading, RG4 8BF, UK. E-mail address: [email protected] (E.H.C. Wright).
1748-6815/$ - see front matter ª 2007 Published by Elsevier Ltd on behalf of British Association of Plastic, Reconstructive and Aesthetic Surgeons. doi:10.1016/j.bjps.2007.01.080
Functional and oncological outcomes after limb-salvage surgery for primary sarcomas of the upper limb
383
or without further recurrences. Consistently good functional results as measured by TESS were reported by responders, but high-grade disease and early recurrence were identified as predictors of recurrence and death from disease, respectively. ª 2007 Published by Elsevier Ltd on behalf of British Association of Plastic, Reconstructive and Aesthetic Surgeons.
Before 1980, bone and soft-tissue sarcomas of the extremities were treated almost exclusively with amputation in order to achieve adequate margins to control disease.1 This was particularly true for the upper limb where the combination of multiple neurovascular structures and small volumes of tissue made adequate tissue margins difficult to achieve. Wide local excision has a greater risk of local recurrence, whereas amputation has significant morbidity because of loss of function and disfigurement. No statistically significant survival difference was found between amputation and limb salvage.2 Limb-salvage surgery aims to strike a balance between achieving sufficient margins in order to control local disease, and preserving critical neurovascular and musculoskeletal structures in order to retain maximal function of the affected limb.3 Until recently, no study had shown a statistically significant improvement in functional outcome in limb salvage compared with ablative surgery for upper extremity sarcoma. The functional benefits of limb-salvage procedures for sarcomas in the lower limb, however, have been demonstrated.4
Aim The aim of this study was to evaluate the functional and oncological outcomes of limb-salvage surgery for primary sarcomas of the upper limb and shoulder girdle.
Patients and methods Patients from the Oxford Bone and Soft-tissue Tumour Registry, who had undergone surgery for soft-tissue or bone tumours of the upper limb and shoulder girdle between 1997 and 2005 at the Nuffield Orthopaedic Centre, Oxford, and the Radcliffe Infirmary, Oxford, were selected for inclusion in the study. Approval was received from the Oxfordshire Research and Ethics Committee A02-068. Operation notes and pathology reports of all patients were consulted. The inclusion criteria were primary soft-tissue or bone sarcoma arising in the upper limb or shoulder girdle; the exclusion criteria were treatment with primary amputation; disease not amenable to surgery; and inadvertent treatment of an undiagnosed sarcoma at a referring hospital followed by amputation. All patients were seen and assessed in a multidisciplinary clinic. Initial investigations included ultrasound-guided biopsy of the tumour, whole-limb magnetic resonance imaging, and plain films of the tumour and limb. Staging investigations included a computed tomography scan of the chest, isotope bone scans and blood chemistry. Surgery was carried out by surgeons with an interest in sarcoma surgery and limb reconstruction. Multidisciplinary
procedures typically involved patients with extensive involvement of neurovascular structures or potentially complex soft-tissue reconstructions. Where low-grade disease was diagnosed close to functionally important structures, a planned marginal excision was carried out. All surgical specimens were reviewed by a single musculoskeletal pathologist experienced in examining sarcoma specimens, and panelled for further opinions when necessary. The tumours were graded according to the guidelines of the Musculoskeletal Tumour Society. Margins were reported as clear if a layer of normal tissue was included at all resection margins, marginal if the excision involved, but did not breach, the pseudocapsule encasing the tumour, and positive if the abnormal cells abutted the resection margins. Patients were reviewed at a multi-disciplinary meeting (including an oncology team specialising in the treatment of sarcomas) and adjuvant treatment planned. After discharge, the patients were then reassessed with the Toronto Extremity Salvage Score (TESS) questionnaire by post. TESS is a well-validated, disease-specific self-assessment5 questionnaire completed by the patient after either upper or lower limb-salvage surgery for musculoskeletal tumours.6 The patient grades 29 activities of daily living using a five-point Likert-type scale from 1, impossible to do, to 5, not at all difficult. At the end of the questionnaire, the patient grades the overall difficulty of all activities undertaken in the previous week using the same scale, and their overall level of disability from 1, completely disabled, to 5, not disabled at all. The patient’s score for each daily living activity is added to form a raw score. Activities that the patients deem not applicable to them can be marked as such on the form, and the maximum possible score is adjusted accordingly. The TESS is then calculated as:
100
Raw score-minimum score ZTESS Maximum score
Thus, a better function, as perceived by the patient, gives a higher TESS. Patients who had undergone primary limb salvage and then amputation for disease control were not assessed with the TESS questionnaire, but were included in the study, as they represent one possible outcome of limb-salvage surgery. The patients’ demographics, presentation, operation details, sarcoma type, grade, margins and metastases status and TESS were entered onto the database. Postal reminders were sent to non-responders. Oncological outcome was assessed in terms of excision margins achieved, local recurrence, re-excision and disease-free survival
384
E.H.C. Wright et al.
Results This study spanned a period of 9 years, reflecting the relative rarity of upper limb sarcomas amenable to limbsalvage surgery. The mean follow-up period was 34 months (3e96 months), with 26 patients followed up to 5 years or longer. Seventy-two patients were included (37 men and 35 women), with no significant difference in mean age (55.7 [men] and 57.9 [women] years). Their anatomical distribution was 28 of the shoulder girdle, 23 of the arm, 22 of the forearm and one of the hand Table 1. At presentation, all had a mass with or without loss of function, one had lymph-node involvement alone, and one had lymph-node and lung metastases. A total of seven patients had been treated elsewhere before being referred either for excision of residual disease or for definitive treatment of a previously undiagnosed sarcoma Table 2. Excision margins were inadequate in 11 patients who then underwent re-excision, and clear margins were achieved in all cases. None of the patients had local disease recurrence, but one patient did develop pulmonary metastases. Fifty-nine surviving patients suitable for TESS assessment were surveyed, and 38 questionnaires were returned completed, three were returned blank (a 69% overall response, with 64% completed). The mean TESS score for the group was 87 (range 36e 100). Overall, 22 patients rated their daily activities over the previous week as ‘not at all difficult’, and none as ‘impossible’. Twenty-one patients felt that they were ‘not at all disabled’ and none felt that they were ‘completely disabled’ Tables 3, 4 and 5. In the study, 14 patients had local disease recurrence. Twelve had clear, one marginal and one positive margins at initial surgery. Of these 14 patients, nine (64%) had highgrade disease and one other had low-grade disease, with vascular invasion that recurred as high-grade. Only four (29%) had consistently low-grade disease. This is in contrast
Table 1
Discussion
Summary of Sarcomas by Type
Bone-derived (n Z 29)
Soft tissue-derived (n Z 43)
to the recurrence-free group, in which 13 (22%) had highgrade and 45 (78%) had low-grade disease. In the local recurrence group, three were too old for adjuvant treatments and one could not receive radiotherapy, as the lymphangiosarcoma had arisen within a radiotherapy field for previous breast cancer. This patient also had the only positive margin at primary surgery, owing to the sarcoma involving the circumference of the arm and the axilla, and had a suspicious nodule on chest computed tomography before surgery. The marginal excision for the malignant peripheral nerve sheath tumour was planned as such to preserve function. Eleven of the recurrent disease group underwent reexcision, and clear margins were achieved in all. Two underwent amputation, one for recurrent Ewing’s sarcoma, who is currently disease-free, and the other for a highgrade rhabdomyosarcoma, who died of pulmonary metastatic disease. Of the 11 who underwent re-excision, three had recurrences, all high-grade disease, and two with vascular invasion. These were excised completely in two cases and incompletely excised in one. All three had recurrences that were operated upon, with two completely excised and one incompletely excised (who subsequently died of spinal metastatic disease). Of the two with clear margins, one is disease-free under follow-up, and one died of pulmonary metastatic disease after another complete excision. Overall, the mortality rate from disease in the group with local recurrences was 36%. Of the patients to be followed up to 5 years or more after surgery, 20 of the 26 were alive without disease (75% 5-year survival). Three patients in the group who did not develop local recurrences (3/58 [5%]) died of disease, two with pulmonary metastases and one with pulmonary and spinal metastases. All three of these patients had high-grade disease. Three other patients died of unrelated comorbid disease without recurrences or metastases.
Sarcomas
Number of cases
Chondrosarcoma Malignant fibrous histiocytoma Fibrosarcoma Osteosarcoma Ewing’s sarcoma Liposarcoma Leiomyosarcoma Malignant peripheral nerve sheath tumours Dermatofibrosarcoma Pleomorphic sarcoma Synovial sarcoma Lymphangiosarcoma Spindle cell sarcoma Rhabdomyosarcoma Haemangioendothelioma
16 cases 5 cases 4 cases 3 cases 1 case 17 cases 10 cases 7 cases 2 2 1 1 1 1 1
cases cases case case case case case
Amputation is best practice in wide excision of extremity sarcoma with adequate margin. However, the functional and psychological disability of amputation is high. If it is possible to excise the tumour preserving the function of the limb, then this would be the treatment of choice as long as the survival of the patient was not compromised. Other studies have shown no deterioration in survival when limbpreserving surgery is carried out. As expected, the local recurrence rate is increased,7 yet local recurrence does not affect metastasis and survival,8 provided that the local recurrence is adequately treated. Studies of the lower limb study have reported successful excision of the tumour in the region of 85%,9 local recurrence rates of 11%,10 survival of 40e50% at 5 years, and remaining function have been shown in the lower limb to be an American Musculoskeletal Tumour Society score of 37e100%. We report similar results for upper limb sarcomas, with planned margins of excision achieved in 85% of cases with primary surgery, increasing to 100% with re-excision, local recurrence at 15%, and survival at 75% among those at 5 years or more after surgery.
Functional and oncological outcomes after limb-salvage surgery for primary sarcomas of the upper limb Table 2
385
Surgical resection, reconstruction and adjuvant treatment by sarcoma type
Sarcoma type
Resection
n
Reconstruction
n
Adjuvant therapy
Bone derived (n Z 29)
Excision Proximal humeral resection Partial scapulectomy
11 6 4
Direct closure Endoprosthesis Ipsilateral latissimus dorsi flap Iliac bone graft Vascularised fibular graft Plate Latissimus dorsi free flap
17 4 3
Chemotherapy Radiotherapy Both
2 1 2
Total
5
Direct closure Tendon transfers Local flap Radial forearm flap Lateral arm free flap Groin free flap Split skin graft
26 2 3 1 1 1 2
Soft-tissue derived (n Z 43)
Curettage Excision of distal radius Excision of distal ulna
4 2 1
Tikhoff-Linberg extra-articular scapular resection Wide local excision Partial scapulectomy Axillary clearance Excision of common flexor origin Clavicular osteotomy
1 39 2 2 1
In our study, we found that successful excision of the sarcoma with the planned margin could be achieved in 61 out of 72 patients with primary surgery. Re-excision in the 11 cases of inadequate margins achieved clear margins in all cases, and there were no local recurrences in this group. Table 3 Toronto Extremity Salvage Score by age, tumour type, anatomical site, adjuvant given and recurrence Group
N
Mean TESS (þ/Standard Deviation)
Overall Aged 65 years Aged < 65 years Bone-derived sarcoma Chondrosarcoma Malignant fibrous histiocytoma/ fibrosarcoma Soft tissue-derived sarcoma Liposarcoma Leiomyosarcoma Malignant peripheral nerve sheath tumours High-grade disease Low-grade disease Anatomical site: shoulder Anatomical site: arm Anatomical site: forearm Adjuvant given: none Adjuvant given: Chemotherapy Adjuvant given: radiotherapy Adjuvant given: chemotherapy þ radiotherapy No recurrence Recurrence treated surgically
38 12 26 18 10 6
87 80 90 85 79 92
(þ/ (þ/ (þ/ (þ/ (þ/ (þ/
18) 23) 14) 20) 24) 12)
20 11 4 5
89 96 78 81
(þ/ (þ/ (þ/ (þ/
16) 6) 17) 24)
8 30 16 12 10 28 1 8 1
82 88 90 81 87 87 94 87 96
(þ/ (þ/ (þ/ (þ/ (þ/ (þ/ () (þ/ ()
22) 16) 16) 21) 18) 18)
32 6
90 (þ/ 14) 73 (þ/ 24)
TESS, Toronto Extremity Salvage Score.
20)
1
2 1 1 1
n
Chemotherapy Radiotherapy Both
1 10 5
Total
16
This is significantly better than other studies in which inadequate margins have been reported in up to half of cases.11 In one case, an incomplete excision was planned in a woman with metastases at presentation. This case shows that there is a sub-group of patients with limited survival due to untreatable metastatic disease for whom there is no advantage in achieving wide clearance by amputation. The local recurrence rate was 14 out of 72, which is in keeping with the local recurrence rates reported in other studies. When a local recurrence did occur, further local recurrences, despite re-excision, occurred in three out of 14 cases. Five patients with local recurrence later died of pulmonary metastases. Pulmonary metastases were only present in one case at preoperative staging (the case of the lymphangiosarcoma that had a planned incomplete excision), but it is impossible to determine if metastases were present, but undetectable at the time of staging (up to 85% of patients may have micro-metastases at presentation12) in the other four cases, or related to the local recurrence. Amputation was carried out in two patients with local recurrence (3% amputation rate), on the basis of absent distant metastases and inability to perform a further adequate wide excision while preserving a functional limb. Of these two patients, one is alive and disease-free and the other died of pulmonary metastases. Amputation as a secondary procedure after failed limb-salvage has been reported as about 9%;13 however, in a study looking specifically at treatment of local recurrences after limb-salvage surgery, a 30% amputation rate has been reported.14 The real aim of this study was to determine whether upper limb function was satisfactorily preserved in this limb-sparing surgery. This has been found in lower limb sarcomas,4 although the lower limb has less functional complexity than the upper limb. The TESS reported by our patients showed that those people who underwent limbsparing surgery had a good functional outcome. No difference was found at the anatomical level at which this surgery was performed. No significant difference was found
386
E.H.C. Wright et al.
Table 4 Local recurrences by tumour type, grade, initial excision margins, adjuvant therapy given and subsequent management Grade and tumour
Margin
Time to recurrence
Adjuvant
Re-excision
Further recurrence
High-grade synovial
Clear
24 months
Nil- too old
Clear
No
Low-grade malignant Clear fibrous histiocytoma Low grade Clear leiomyosarcoma vascular invasion positive
120 months Chemotherapy Clear and radiotherapy 8 months; Radiotherapy Clear high grade and immunotherapy
High-grade leiomyosarcoma High-grade rhabdomyosarcoma High-grade dermatofibrosarcoma
Clear
3 months
Clear
4 months
Clear
12 months
Chemotherapy and radiotherapy
Clear
Clear
12 months
Radiotherapy
Clear
Clear
36 months
Nil; too old
Clear
Clear
300 months Radiotherapy
Clear
Clear
4 months
Radiotherapy
Positive
Clear
18 months
Amputation
Positive
2 month
Chemotherapy and radiotherapy Nil (preventative radiotherapy for breast cancer on the same site) Nil; too old
Low-grade liposarcoma High grade leiomyosarcoma Low-grade fibrosarcoma High-grade malignant fibrous histiocytoma vascular invasion positive High-grade Ewing’s sarcoma High-grade lymphangiosarcoma
High-grade malignant peripheral nerve sheath tumours Low-grade chondrosarcomas
Marginal 3 months
24 months
Chemotherapy and radiotherapy Chemotherapy
Radiotherapy
in TESS between people over the age of 65 years and people under the age of 65 years. No difference was found in functional scores between patients with bone or soft-tissue sarcoma, indicating either Table 5 Time to recurrence of disease by grade of tumour and disease-free survival Patient group
N
Time to recurrence þ/e standard deviation
Low-grade disease High-grade disease Survivors or died of unrelated disease Non-survivors who died of disease
4 10 9
114þ/133 months 11þ/11 months 96þ/122 months
5
12þ/9 months
Clear Amputation
Died of chest metastases No Alive; no evidence of disease Yes, 3 months- Died of chest fully excised, metastases further recurrence fully excised Nil Alive; no evidence of disease No Died of chest metastases Yes 12 months- Died of spinal incompletely metastases excised No Alive; no evidence of disease No Alive; no evidence of disease Nil Alive; no evidence of disease 12 monthsCurrently in F/U, fully excised no recurrence
No
No, widespread NA metastases
Clear
Outcome
Nil
TikoffeLinberg Nil resection Clear
Alive; no evidence of disease Died of chest metastases
Died of unrelated disease Alive; no evidence of disease
group benefit from limb salvage surgery and reconstruction. Nor was there any difference in functional scores seen in patients receiving radiotherapy, suggesting that radiotherapy does not preclude limb-salvage surgery, and its benefit as an adjunct in limb-salvage surgery is well-documented.15 Surgical re-excision of recurrent disease did not adversely affect TESS compared with the group without recurrent disease, despite the distorted anatomy and altered anatomical landmarks frequently found in a previously operated, and often irradiated, site. Disease recurrence occurred more frequently in highgrade disease, within 2 years of excision, both of these factors predicting mortality from disease and metastasis. In conclusion, this study has shown that the principles of limb-salvage surgery are applicable to a wide range of tumour-types and grades, to all patient age groups and anatomical sites with good functional results, comparable to those in literature. Recurrences of disease can be
Functional and oncological outcomes after limb-salvage surgery for primary sarcomas of the upper limb successfully treated by re-excision. Early recurrence of high-grade disease carries a worse prognosis, especially if adjuvant therapy cannot be given for whatever reason.
References 1. Springfield DS. Introduction to limb-salvage surgery for sarcomas. Orthop Clin North Am 1991;22:1e5. 2. Rosenberg SA, Tepper J, Glatstein E, et al. The treatment of soft tissue sarcomas of the extremities: prospective randomised evaluation of (1) limb-sparing surgery plus radiation therapy compared with amputation and (2) the role of adjuvant therapy. Ann Surg 1982;196:305e15. 3. Gitelis S. Limb salvage for appendicular tumours. Curr Opin Orthop 1991;2:811e6. 4. Renard AJ, Veth PV, Schreuder HWB. Function and complications after ablative and limb-salvage therapy in lower extremity sarcoma of bone. J Surg Oncol 2000;73:198e205. 5. Davis AM, Bell RS, Bradley EM, et al. Evaluating functional outcome in patients with lower extremity sarcoma. Clin Orthop Relat Res 1999;358:90e100. 6. Davis AM, Wright JG, Williams JI, et al. Development of a measure of physical function for patients with bone and soft-tissue sarcoma. Qual Life Res 1996;5:508e16.
387
7. Creighton JJ, Clayton AP, Mindell ER. Primary malignant tumours of the upper extremity: retrospective analysis of one hundred and twenty-six cases. J Hand Surg 1985;10A: 805e14. 8. Brennan MF, Casper ES, Harrison LB. The role of multimodality therapy in soft-tissue sarcoma. Ann Surg 1991;214:328e38. 9. Kumta SM, Cheng JCY, Li CK. Scope and limitations of limbsparing surgery in childhood sarcomas. J Pediatr Orthop 2002;22:244e8. 10. Simon MA, Aschilman MA, Thomas N. Limb-salvage treatment versus amputation for osteosarcoma of the distal end of the femur. J Bone and Joint Surg 1986;68A:1331e7. 11. Collin C, Hadjdu SI, Godbold J. Localised operable soft tissue sarcoma of the upper extremity. Ann Surg 1987;205:331e9. 12. Weis LD. The success of limb-salvage surgery in the adolescent patient with osteogenic sarcoma. Adolesc Med 1999;10:451e8. 13. Bray PW, Bell RS, Bowen CVA. Limb salvage surgery and adjuvant radiotherapy for soft tissue sarcomas of the forearm and hand. J Hand Surg 1997;22A:495e503. 14. Catton C, Davis A, Bell R. Soft tissue sarcoma of the extremity. Limb salvage after failure of combined conservative therapy. Radiother Oncol 1996;41:209e14. 15. Karakousis CP, Proimkis C, Walsh DL. Primary soft tissue sarcoma of the extremities in adults. Br J Surg 1995;82: 1208e12.