Of the 26 patients who had primary tumor resections, neuro-vascular structures were intimately close to the tumor in nine patients (34.6%) and embedded in the tumor in three patients (11.5%). In one patient both popliteal vessels were surrounded by the tumor and in two patients the peroneal nerve was embedded in the tumor. One patient had segmental resection of the peroneal nerve and one patient had resection of both popliteal vessels and reconstruction of the popliteal artery with a saphenous vein graft. One patient with extensive metastatic disease had the peroneal nerve preserved despite its embedment in the tumor for palliation. Fourteen patients (53.8%) had negative surgical margins (greater than 1 cm), six (23.07%) had close surgical margins (less than 1 cm), and six (23%) had positive surgical margins.20,25,30,35,36 Five of the six positive margins involved either the adventitia or the nerve sheath. One patient with positive surgical margins who presented with metastatic disease had palliative surgery; her popliteal nerve, which was embedded in the tumor, was preserved (Table 3).
Eight of the 29 patients were treated with neoadjuvant and adjuvant therapy, two patients were treated with only adjuvant chemotherapy, and one patient received one cycle of neoadjuvant chemotherapy before it was stopped because of life-threatening side effects. Seventeen patients received postoperative radiation therapy. We did not treat any of our patients with preoperative radiation therapy. Patients were followed for a minimum of 2 years or until death (range, 4-192 months; median, 79 months).
Preoperative evaluation of all patients included a complete history and physical examination with an emphasis on the neurologic examination of the ipsilateral lower extremity and assessment of arterial pulses. Imaging studies included plain radiography, MRI, angiography of the involved limb (Fig 1), and occasionally computed tomography (CT) (Fig 1). Patients had chest CT to rule out distant metastasis. Before institution of definitive treatment, tissue diagnosis was obtained using core needle biopsy to minimize local contamination. The biopsy site was determined by the surgeon (MMM) and the interventional radiologist (JJ) who performed the biopsy.
For popliteal tumor resection, we placed the patient in the prone position and both lower limbs were draped. The contralateral leg was prepared for saphenous vein harvesting for arterial reconstruction if the popliteal artery was resected. An S-shaped incision was made, crossing from proximal-medial to distal-lateral at the level of the knee. The medial proximal arm of the incision facilitated identification of the popliteal vessels as they exit the adductor hiatus, and the lateral distal arm enabled easy exposure of the peroneal nerve, posterior to the fibular head. In addition, by making the distal arm of the incision lateral we avoided damaging the greater saphenous vein, which runs on the medial aspect of the leg (Fig 2).
The two main vessels responsible for venous drainage of the leg are the popliteal vein and the greater saphenous vein. During tumor resection, excision of the popliteal vein may be unavoidable. Therefore, care should be taken not to damage the greater saphenous vein. Ligation of the popliteal vein and ipsilateral greater saphenous veins may lead to severe venous insufficiency.
The thin and friable popliteal fascia lies in close proximity to the neurovascular bundle (especially the peroneal nerve, which lies just deep to the popliteal fascia at the level of the fibular head), making it a critical landmark. Subcutaneous flaps were made to identify the popliteal fascia. The landmarks and various structures of the popliteal fossa often can be palpated through the fascia, which then was incised accordingly. Failure to realize the dissection is underneath the fascia and only a few millimeters separate the blade from the vessels and nerves of the popliteal fossa can easily result in injury to those structures.
The initial step of a popliteal fossa resection was exposure and identification of the neurovascular bundle. This allowed mobilization of the vulnerable structures before resection. In the popliteal space the nerves usually are found posterior to the tumor mass and the vessels usually are anterior to it. Mobilization usually was accomplished by exposing the structures in the distal thigh and proximal leg and following them to the popliteal fossa. The inferior genicular vessels pull the popliteal artery toward the joint capsule and usually must be ligated to allow its mobilization. We routinely detach the origin of the medial and lateral hamstrings and both heads of the gastrocnemius to achieve a wide exposure. An intraoperative Doppler ultrasound device can be helpful if the popliteal vessels are difficult to expose (Fig 2).
After mobilizing the neurovascular bundle, the tumor was resected with a cuff of normal tissue if possible. Not uncommonly, the vessels, the nerves, or both, were in close proximity to the tumor mass or adherent to its pseudocapsule. In such cases, the structures were dissected free and the nerve sheath and adventitia removed and examined with frozen sections to determine the surgical margins. Vessels and nerves of the popliteal fossa were resected only if they were embedded in the tumor mass. The popliteal artery then could be reconstructed with a saphenous vein graft taken from the contralateral leg. We consider reconstructing the popliteal vein unnecessary because the ipsilateral saphenous vein can compensate for its loss.18 Nerve involvement or vascular involvement is not an indication for primary amputation if these structures can be adequately resected.1,2,13,15,18,29
After the tumor is resected, the two heads of the gastrocnemius were sutured to each other and to the hamstring muscles to form a uniform muscle layer that covered the popliteal space (Fig 3). This wound closure technique minimized deep wound infection by forming a muscular barrier between the skin incision and the popliteal space.
After surgery, the patient's extremity was placed in a long posterior splint in 15° to 30° knee flexion to relieve tension on the neurovascular bundle and skin incision. Physiotherapy for muscle strengthening and range of motion was not started until the skin incision was completely healed, typically after 10 to 14 days.
Use of radiation therapy, chemotherapy, or both was based on the characteristics of the tumor, the patient, and the type of surgical margins. Treatment was at the discretion of the attending oncologist and was not standardized to a specific protocol. Treatment for patients with low-grade soft tissue sarcomas depended on the surgical margins.7 If wide negative surgical margins were achieved, patients were simply followed closely. If tumor cells were close to or involved the surgical margins patients were treated with adjuvant external beam radiation. Regardless of the resected margins, all patients with intermediate-grade tumors received postoperative radiation therapy. Patients presenting with high-grade soft tissue tumors usually were treated with neoadjuvant and adjuvant chemotherapy. Contraindications for chemotherapy included advanced age, severe co-morbidities, and a maximum tumor diameter less than 5 cm. Chemotherapy was not administered when the oncologist believed the potential risks of treatment outweighed its possible benefits. Patients with high-grade soft tissue sarcomas generally were treated with adjuvant radiation therapy. Exceptions included patients whose tumor necrosis rate exceeded 90% after being treated with neoadjuvant chemotherapy and whose surgical margins were negative, and patients who had primary amputation with negative margins.14
For the first 2 years, patients were evaluated every 3 months by physical examination, MRI of the operated limb, and chest CT. Patients were evaluated semiannually for an additional 3 years and annually thereafter. Two orthopaedic oncologists (TP, JB) reviewed and analyzed the clinical records, imaging studies, and operative reports for details regarding the clinical presentation of popliteal sarcomas, tumor size, surgical margins, the occurrence of local recurrence and distant metastasis, the limb salvage rate, and postoperative morbidity and complications following limb-sparing resections.
We used the Mann-Whitney nonparametric statistical test to compare the sizes of the lesions between the high-grade tumors and the low- and intermediate-grade tumors. A p value < 0.05 was considered significant.
The rate of local recurrence was surprisingly low (Table 4). Three patients had local recurrences (10.3%) develop; all had high-grade soft tissue sarcomas. One of these patients first presented to our outpatient clinic with local recurrence after initially being treated at another institution. He had a second recurrence after his tumor was re-resected at our institution. In two patients, the local recurrence developed within 1 year of surgery. The third patient presented with local recurrence 5 years after her primary amputation (Table 3). All patients with local recurrences were treated with additional resection and adjuvant radiation therapy. Two patients who had local recurrences were tumor-free 4 and 6 years after their second resection, respectively. The third patient had two reresections and an above-knee amputation after multiple recurrences, and eventually died from metastatic disease.
The rate of distant metastasis was unexpectedly low and most patients were alive at the time this study was conducted. Of the 16 patients with high-grade tumors four (25%) had distant metastasis develop. Two patients were diagnosed with metastatic disease during preoperative staging and the other two were diagnosed with metastatic disease after resection of their primary lesions. None of the patients with low-grade and intermediate-grade soft tissue sarcomas had metastatic disease. Overall, 26 of the 29 patients were still alive. Two patients died from pulmonary metastases and one patient, who had resection of low-grade soft tissue sarcoma, died 3 years postoperatively from unrelated reasons. Of the 16 patients who presented with high-grade soft tissue sarcomas, 14 (87.5%) were still alive at the time this study was conducted. Two of these patients had distant metastasis without any sign of local recurrence and 12 (85.7%) had no signs of local or metastatic disease.
The overall limb salvage rate was 86.2% (Table 4). Of the 29 study patients, three had primary above-knee amputations, and one patient first treated with two local resections after multiple recurrences had an above-knee amputation. The limb-salvage rate for patients with high-grade soft tissue sarcomas was 75%. The limb-salvage rate for patients with high-grade soft tissue sarcomas who were alive at the time this study was conducted and who had no sign of local or systemic disease also was 75%.
Patients with popliteal sarcomas often presented with atypical symptoms (Table 2), high-grade lesions were smaller (p = 0.023) than low- and intermediate-grade lesions, and the time to diagnosis was relatively short. Of the nine patients (31%) who presented with painful masses, five had high-grade soft tissue sarcomas, one had an intermediate-grade soft tissue sarcoma, and three had low-grade soft tissue sarcomas. The larger tumors usually protruded to the posterior aspect of the thigh, but none expended distally into the posterior leg. Six patients reported having an asymptomatic mass in the popliteal fossa for more than 2 years (range, 2.5-10 years) that enlarged, became painful, or both, shortly before they first presented to our outpatient clinic. Of these six patients, three were diagnosed with low-grade lesions, two with intermediate-grade lesions, and one with high-grade soft tissue sarcoma. The median time to diagnosis from the onset of patients' symptoms was 2.5 months (range, 0.5-12 months).
The postoperative complications and morbidity rates were low. Of the 26 patients who had primary resections six had superficial surgical wound dehiscence, four of whom had wound breakdown during the course of adjuvant radiation therapy. All six patients were treated with systemic antibiotics and two patients had surgical débridement. All surgical wounds healed. Five patients had common peroneal nerve palsies postoperatively, one of whom had additional paresthesias in the distribution of the tibial nerve. In four of these patients, the neurologic deficiencies were secondary to neuropraxia and eventually resolved. The remaining patient sustained permanent peroneal palsy from segmental nerve resection. Other postoperative complications, each affecting one patient, were deep vein thrombosis of the superficial femoral vein and radiation neuritis.
The patients' knee range of motion (ROM) often was preserved or only mildly compromised. Fourteen of the 26 patients (53.8%) who had primary resections had full ROM of the knee, 12 patients (46.1%) had mild to moderate limitations in knee flexion (120°-90°), and four patients (15.3%) had mild flexion contractures (5°-15°). No patient in our series required knee manipulation or contracture release.
Treatment outcome and clinical presentation of popliteal sarcomas were not extensively studied.8,23,37 Because of their extracompartmental location and proximity to neurovascular structures, their resection is considered challenging and their prognosis is regarded as inferior to intracompartmental extremity soft tissue sarcomas.8,12,22,28 In our series of 29 consecutive patients with popliteal sarcomas, the local recurrence rate and metastatic rate were low and limb-salvage rate was high. In addition, the clinical presentations of the tumors often were unusual, and the morbidity and complication rates after limb salvage resections were low.
Our study has several limitations. It is a noncontrolled study of a relatively small heterogeneous population and not histopathologically uniform; the minimum followup was only 2 years. However, to our knowledge, this is the largest reported series of popliteal soft tissue sarcomas in the English literature. Because most local and systemic recurrences of soft tissue sarcomas occur during the first 2 postoperative years, and because our median followup was more than 6 years, we believe the rates of local recurrence and systemic disease would increase only slightly during a longer followup.32 An additional limitation of our study is that patients were not uniformly treated with neoadjuvant and adjuvant therapy. Conversely, patients' treatment was based on a constructed approach which took into consideration the characteristics of the tumor, the patient, and the type of surgical margins. This limitation precludes making strong conclusions regarding the overall efficacy of neo-adjuvant and adjuvant therapy; however, our therapeutic approach, and specifically the value of postoperative radiotherapy in limiting local recurrence, can be evaluated. Finally, we did not have functional scores available given the extended time frame of the study and its retrospective nature. Nonetheless, some generalizations may be made about other aspects of outcome.
The rates of local recurrence and distant metastasis were unexpectedly low. None of the patients who presented with low- and intermediate-grade tumors had local recurrence or metastatic disease, three patients (10.3%) with high-grade soft tissue sarcomas had local recurrences, and four patients had distant metastasis develop. Our results are comparable to the treatment results of patients with soft tissue sarcomas in other anatomic locations, and are similar to results of other studies of soft tissue sarcomas of the flexor fossae (axilla, epitrochlear fossa, ante-cubital fossa, inguinal space, and popliteal fossa), where patients also were treated with chemotherapy and radio-therapy.8,23,32,37
Several studies have emphasized the importance of negative surgical margins in the excision of soft tissue sarcomas of the extremities.6,17,26,34 In an extracompart-mental space such as the popliteal fossa, where major vessels and nerves traverse, performing radical resections is impossible and wide resections often are difficult to achieve.12,23 When the popliteal vessels or sciatic nerve are in close proximity to the tumor, we routinely remove the adventitia or the nerve sheath respectively to achieve close or only microscopically positive surgical margins. Accordingly, the surgical margins were positive in 23% of resections and close in 23%. Adjuvant radiation therapy, which has been reported to achieve acceptable local control in the presence of close or microscopically positive surgical margins, probably played an important role in limiting local recurrences in our series.20,36 Similarly, Eilber et al8 reported one local recurrence in 16 patients with high-grade soft tissue sarcomas of the popliteal fossa treated with neoadjuvant chemotherapy and radiotherapy. Philippe et al23 reported no local recurrences in eight children with popliteal sarcomas who were treated with tumor resections and adjuvant therapy.
The overall limb salvage rate was 86.2%. The indications for amputations were multiple recurrences of an especially aggressive tumor, and heavily scarred and damaged skin and soft tissues overlying the popliteal fossa secondary to previous surgery and radiotherapy. Others have reported comparable results. Eilber et al8 reported on 16 patients with high-grade popliteal soft tissue sarcomas. Only two of their patients had amputations. Philippe et al23 reported on 11 children with soft tissue sarcomas of the popliteal fossa, and only three had amputations.
Close proximity of the tumor to neurovascular structures or even neurovascular involvement were not an indication for amputation. Although the sciatic nerve, its major branches, and the popliteal vessels were often in close proximity to the tumor mass, only in 11.5% were they embedded or surrounded by the tumor. This is a typical growth pattern of sarcomatous tumors, which unlike carcinomas, displace anatomic structures instead of infiltrating them.32 Consequently, by dissecting the adventitia and nerve sheath we usually were able to preserve the vessels and nerves with close or microscopically positive surgical margins. When the popliteal vessels were embedded in the tumor and were resected, the popliteal artery was reconstructed with a contralateral greater saphenous vein. This is in agreement with previous studies where vascular involvement was treated by resection and reconstruction instead of primary amputation.15,18,29 Reconstruction of the popliteal vein is not necessary because the ipsilateral saphenous vein adequately compensates for its loss. When the peroneal, tibial, or sciatic nerves were embedded in the tumor they were resected en bloc with the tumor mass instead of performing primary amputation, similar to methods reported by others.1,2,13 Bickels et al1 reported on 15 patients who had resections of the sciatic nerve. Most of their patients achieved good functional results and none had pressure sores develop. Primary or graft repair of the nerve during the same operation is controversial because of the questionable negative effect of adjuvant radiotherapy on nerve healing.19,24,27
On presentation, high-grade tumors were smaller than low- and intermediate-grade tumors. Although this is not contradictory to the accepted opinion that small size (< 5 cm) is a good prognostic sign for high-grade sarcomas, tumor size is not a recognized characteristic of a specific histologic grade.16,26,37 Patients' presenting symptoms were atypical. Only 55% of the patients presented with nonpainful asymptomatic masses and 31% presented with locally painful masses. This clinical finding is in contrast to the classic presentation of soft tissue sarcomas as non-painful, growing masses.3 Only one patient was diagnosed with synovial sarcoma, which is considered the typical pain-causing soft tissue tumor.3 We hypothesize the increased incidence of localized pain resulted partially from local compression of the tibial or peroneal nerves. The majority of painful masses were high-grade tumors, despite being smaller than low- and intermediate-grade tumors, presumably because their high growth rate did not allow adaptation of the nearby neural structures and surrounding tissues.
Considering tumors were often in close proximity to neurovascular structures and most patients were treated with postoperative radiation, the morbidity and complication rates after limb-sparing resections were unexpectedly low. We consider our routine wound closure technique of tenodesing the heads of the gastrocnemius to each other and to the hamstring muscles, important in preventing deep surgical wound breakdown because it creates a muscle barrier between the popliteal space and the skin and subcutaneous tissues. Another reason for the low complication rate might be that no preoperative radiotherapy was administered. O'Sullivan et al21 conducted a randomized trial comparing preoperative with postoperative radiation therapy and found more wound complications in the preoperative radiation group. Other have supported this finding.4,5 Unfortunately, we were not able to find any reports in the English literature on the morbidity and complications after resections of soft tissue sarcomas from the popliteal fossa.
We analyzed the treatment outcomes and clinical presentation of popliteal sarcomas. We found the presenting symptoms of popliteal sarcomas were often atypical, and high-grade lesions were more likely smaller than low- and intermediate-grade lesions. By using a wide anatomic approach coupled with careful surgical technique, and by individually tailoring perioperative treatment, limb-salvage procedures were accomplished in most patients, with low residual morbidity and good local and systemic control.
We thank Dr. James Jelinek for performing all the CT-guided needle biopsies.
1. Bickels J, Wittig JC, Kollender Y, Kellar-Graney K, Malawer MM, Meller I. Sciatic nerve resection: is that truly an indication for amputation? Clin Orthop Relat Res
2. Brooks AD, Gold JS, Graham D, Boland P, Lewis JJ, Brennan MF, Healey JH. Resection of the sciatic, peroneal, or tibial nerves: assessment of functional status. Ann Surg Oncol
3. Campanacci M. Bone and Soft-Tissue Tumors
. 1st ed. New York, NY: Springer-Verlag; 1990.
4. Cheng EY, Dusenbery KE, Winters MR, Thompson RC. Soft tissue sarcomas: preoperative versus postoperative radiotherapy. J Surg Oncol
5. Chmell MJ, Schwartz HS. Analysis of variables affecting wound healing after musculoskeletal sarcoma resections. J Surg Oncol
6. Collin C, Goldbold J, Hadju S, Brennan M. Localized extremity soft-tissue sarcoma: an analysis of factors affecting survival. J Clin Oncol
7. Choong PFM, Petersen IA, Nascimento AG, Sim FH. Is radiotherapy important for low-grade soft tissue sarcoma of the extremity? Clin Orthop Relat Res
8. Eilber FC, Eckardt JJ, Rosen G, Nelson SD, Selch M, Eilber FR. Large, deep, high-grade extremity sarcomas: treating tumors of the flexor fossa. Surg Oncol
9. Eilber FR, Eckardt JJ, Rosen G, Fu YS, Seeger LL, Selch MT. Neoadjuvant chemotherapy and radiotherapy in the multidisciplinary management of soft-tissue sarcomas of the extremity. Surg Oncol Clin N Am
10. Eilber FR, Huth JF, Mirra J, Rosen G. Progress in the recognition and treatment of soft-tissue sarcomas. Cancer
11. Eilber FR, Morton DL, Eckhardt JJ, Grant T, Weisenburger T. Limb salvage for skeletal and soft-tissue sarcomas: multidisciplinary pre-operative therapy. Cancer
12. Enneking WF, Spanier SS, Malawer MM. The effect of the anatomic setting on the results of surgical procedures for soft parts sarcoma of the thigh. Cancer
13. Fuchs B, Davis AM, Wunder JS, Bell RS, Masri BA, Isler M, Turcotte R, Rock MG. Sciatic nerve resection in the thigh: a functional evaluation. Clin Orthop Relat Res
14. Henshaw RM, Priebat DA, Perry DJ, Shmookler BM, Malawer MM. Survival after induction chemotherapy and surgical resection for high-grade soft-tissue sarcoma: is radiation necessary? Ann Surg Oncol
15. Hohenberger P, Allenberg JR, Schlag PM, Reichardt P. Results of surgery and multimodal therapy for patients with soft tissue sarcoma invading to vascular structures. Cancer
16. Li GH, Li JQ, Cai YH, Huang M. Surgical management of soft-tissue sarcomas, with an analysis of 313 cases. Semin Surg Oncol
17. Markhede G, Angervall L, Stener B. A multivariant analysis of the prognosis after surgical treatment of malignant soft-tissue tumors. Cancer
18. Matsushita M, Kuzuya A, Mano N, Nishikimi N, Sakurai T, Nimura Y, Sugiura H. Sequelae after limb-sparing surgery with major vascular resection for tumor of the lower extremity. J Vasc Surg
19. Myckatyn TM, Brenner M, Mackinnon SE, Chao CK, Hunter DA, Hussussian CJ. Effect of external beam radiation in the rat tibial nerve after crush, transection and repair, or nerve isograft paradigms. Laryngoscope
20. McKee MD. Liu DF, Brooks JJ, Gibbs JF, Driscoll DL, Kraybill WG. The prognostic significance of margin width for extremity and trunk sarcoma. J Surg Oncol
21. O'Sullivan B, Davis AM, Turcotte R, Bell R, Catton C, Chabot P, Wunder J, Kandel R, Goddard K, Sadura A, Pater J, Zee B. Preoperative versus postoperative radiotherapy in soft-tissue sarcoma of the limbs: a randomized trial. Lancet
22. Peabody TD, Simon MA. Principles of staging of soft-tissue sarcomas. Clin Orthop Relat Res
23. Philippe PG, Rao BN, Rogers DA, Fontanesi J, Bowman L, Parham D, Shapiro D, Lobe TE, Fleming I. Sarcoma of the flexor fossa in children: is amputation necessary? J Pediatr Surg
24. Pillsbury HC, Fisch U. Extratemporal facial nerve grafting and radiotherapy. Arch Otolaryngol
25. Pisters PW, Leung DH, Woodruff J, Shi W, Brennan MF. Analysis of prognostic factors in 1,041 patients with localized soft tissue sarcomas of the extremities. J Clin Oncol
26. Potter DA, Kinsella T, Glastein E, Wesley R, White DE, Seipp CA, Chang AE, Lack EE, Costa J, Rosenberg SA. High-grade soft-tissue sarcomas of the extremities. Cancer
27. Reddy PG, Arden RL, Mathog RH. Facial nerve rehabilitation after radical parotidectomy. Laryngoscope
28. Rooser BO, Attwell R, Berg NO, Rydholm A. Prognostication on soft-tissue sarcoma: a model with four risk factors. Cancer
29. Schwarzbach MH, Hormann Y, Hinz U, Bernd L, Willeke F, Mechtersheimer G, Bockler D, Schumacher H, Herfarth C, Buchler MW, Allenberg JR. Results of limb-sparing surgery with vascular replacement for soft tissue sarcoma in the lower extremity. J Vasc Surg
30. Shiu MH, Collin C, Hilaris BS, Nori D, Manolatos S, Anderson LL, Hajdu SI, Lane JM, Hopfan S, Brennan MF. Limb preservation and tumor control in the treatment of popliteal and antecubital soft tissue sarcomas. Cancer
31. Singer S, Corson JM, Gonin R, Labow B, Eberlein TJ. Prognostic factors predictive of survival and local recurrence for extremity soft tissue sarcoma. Ann Surg
32. Sondak VK, Chang AE. Clinical Evaluation and Treatment of Soft Tissue Tumors. In: Weiss SW, Goldblum JR, eds. Soft Tissue Tumors
. 4th ed. St Louis, MO: Mosby, 2001;21-44.
33. Stotter A, Fallowfield M, Mott A, Fisher C, Westbury G. Role of compartmental resection for soft-tissue sarcoma of the limb and limb girdle. Br J Surg
34. Tanabe KK, Pollock RE, Ellis LM, Murphy A, Sherman N, Romsdahl MM. Influence of surgical margins on outcome in patients with preoperatively irradiated extremity soft tissue sarcomas. Cancer
35. Vraa S, Keller J, Nielsen OS, Jurik AG, Jensen OM. Soft-tissue sarcoma of the thigh: surgical margin influences local recurrence but not survival in 152 patients. Acta Orthop Scand
36. Yang JC, Chang AE, Baker AR, Sindelar WF, Danforth DN, Topalian SL, DeLaney T, Glatstein E, Steinberg SM, Merino MJ, Rosenberg SA. Randomized prospective study of the benefit of adjuvant radiation therapy in the treatment of soft tissue sarcomas of the extremity. J Clin Oncol
© 2007 Lippincott Williams & Wilkins, Inc.
37. Yang RS, Lane JM, Eilber FR, Dorey FJ, Al-Shaikh R, Schumacher LY, Rosen G, Forscher CA, Eckhardt JJ. High grade soft-tissue sarcoma of the flexor fossae. Cancer