Results of Surgery and Multimodal Therapy for Patients with Soft Tissue Sarcoma Invading to Vascular Structures
 

Taken from: Cancer 1999;85:396-408. © 1999 American Cancer Society.
Presented at the third annual meeting of the Connective Tissue Oncology Society (CTOS), 
Milan, Italy, November 6-8, 1997.
 

Peter Hohenberger, M.D., Ph.D. 1  Jens R. Allenberg, M.D., Ph.D. 2 Peter M. Schlag, M.D., Ph.D. 1 Peter Reichardt, M.D.3 Address for reprints:  Peter Hohenberger, M.D., Ph.D. Robert-Rossle Hospital and Tumor Institute,  Medizinische Fakultat Charite,  The Humboldt University of Berlin,  Lindenberger Weg 80, D-13125 Berlin, Germany.

1 Division of Surgery and Surgical Oncology, Robert Rossle Hospital and Tumor Institute, Charite,  Humboldt University of Berlin, Berlin, Germany. 2 Section of Vascular Surgery, Department of Surgery, University of Heidelberg, Heidelberg, Germany. 3 Division of Hematology, Oncology and Tumorimmunology, Robert Rossle Hospital and Tumor Institute, Charite, Humboldt University of Berlin, Berlin, Germany.

BACKGROUND. The aim of this study was to analyze the impact of resection and reconstruction of major vessels on the limb salvage rate, local disease free survival, and overall survival for patients with soft tissue sarcomas invading to neurovascular bundles.

METHODS. Twenty patients were treated in a 7-year period by one surgical team. Preoperative therapy consisted of isolated limb perfusion (n = 6), systemic chemotherapy (n = 4), systemic chemotherapy combined with regional hyperthermia (n = 2), and external beam irradiation (n = 1). All patients underwent resection of the sarcoma monobloc together with the neurovascular bundle invaded. Vessels were replaced by an autologous vein transplant or an allograft, and, in six patients, a myocutaneous flap or skin graft had to be used for soft tissue coverage.

RESULTS. Histologic examination revealed negative histologic margins (RO-resection) and infiltration of the neurovascular bundle in all patients. In four patients, a local recurrence was observed, and, in three of them, reresection with negative margins was achieved. The mean local recurrence free survival was 54 months 
(confidence interval [CI], 42-66 months), and the mean overall survival was 48 months (CI, 32-57 months). Limb salvage was achieved in 19 of 20 patients. Eleven patients developed distant metastases after a mean survival time of 30 months. 

CONCLUSIONS. Extended sarcoma resection, including vessel replacement after preoperative multimodal therapy, provides long term local control and limb salvage. Amputation of extremity sarcoma can hardly be justified, even in cases of tumor invasion to neurovascular bundles. However, efforts to achieve better control over systemic spread are required for long term disease free survival. Cancer 1999;85:396-408. © 1999 American Cancer Society.

KEYWORDS: sarcoma, neurovascular invasion, multimodal therapy, resection, ves- 
sel replacement

Soft tissue carcinoma (STS) is a very uncommon disease accounting for less than 1% of all newly diagnosed adult cancers in Europe and the United States annually.¹ The preservation of the limb has become increasingly possible in lesions affecting the extremities due to improved surgical skill and multimodality therapy.^2,3 Limb-sparing surgery was found to result in equal survival but higher rates of local recurrence than amputation.^3,4
A recent analysis of prognostic factors obtained from prospec-tively collected data on more than 1000 patients revealed microscopically positive margins and recurrent disease at presentation as significant independent adverse prognostic factors for local recurrence.⁵ For distant recurrence and disease specific survival, independent adverse factors were deep location of the sarcoma and high grade.^5,6
High grade sarcomas are characterized by an increased rate of infiltration to neurovascular bundles and a high rate of distant metastases.^7.8 The attachment of tumors to major vascular structures with adequate margins of clearance difficult to achieve has to be regarded as a major limitation of effective surgical therapy. Previously, amputation had been the adequate treatment, but the problem of systemic spread could not be solved by this procedure either. Vessel-invasive tumors often present as recurrent disease, resulting in positive margins after tumor resection. Both adverse prognostic factors further contribute to the poor outcome.
To save the limb despite the vessel-invasive tumors, limited resections combined with brachyther-apy to the tumor bed have been applied successfully,^9,10 However, as a sole surgical treatment, radical resection, including removal of the vessels infiltrated by tumor, with consecutive reconstruction using an allograft or an autologous vein bypass was inaugurated more than 15 years ago.¹¹
Twenty patients underwent extended soft tissue sarcoma resection, including vascular surgical reconstruction, by one surgical team during a 7-year period. Preoperatively, the majority of the patients were treated by combined modality therapy intended to down-stage the tumor and facilitate resection. We analyzed the further course of our patients with respect to limb-salvage rate, local disease free survival, and overall survival.

PATIENTS AND METHODS
Twenty patients (14 males and 6 females; median age, 47 years; range, 22- 65 years; see Table 1) underwent resection of soft tissue sarcoma of the extremities with curative intent, including vascular reconstructions of the arteries and/or veins resected due to tumor infiltration of the neurovascular bundle. Five patients were treated consecutively from 1989 to 1992 at the Department of Surgery, University of Heidelberg. Subsequently, another 15 patients were operated on at the Division of Surgery and Surgical Oncology, Robert-Rossle Hospital, Humboldt University of Berlin.

Previous Treatment
Ten patients were treated for the primary tumor, and five patients were referred to for recurrent disease (patients 6, 9, 14, 15, and 18; see Table 2). Of these, three patients had been treated previously by radiotherapy (54-78 grays [Gy]; patients 14, 15, and 18) and presented with an in-field recurrence. Another five patients were referred for residual disease after an attempt to resect the primary tumor (patients 1, 4, 5, 11, and 16). Fifteen tumors were located extracom-partmentally.

Angiogram
Angiogram detected stenosis of the artery in five patients, with subtotal occlusion in another patient. The artery was displaced by the tumor in eight patients and was encased in another patient (Fig. 1). Thrombosis or occlusion of a major vein in close vicinity to the tumor was observed in seven patients. Three patients showed stenosis, another patient showed direct invasion to the vein wall, and, in two patients, the vein was displaced by the tumor.

Multimodality Treatment
Ten patients were treated by surgical resection alone, and another ten patients underwent multimodal treatment. Isolated hyperthermic limb perfusion (ILP) with rh tumor necrosis factor a (rhTNFa) and melphalan12 was performed in five patients. Of these, two patients had systemic chemotherapy^{13, 14} with a complete remission of lung metastases but only partial remission of the primary tumor. Two patients (9 and 20) had systemic neoadjuvant chemotherapy for their tumors combined with regional hyperthermia15 in a third patient (18). Another patient (17) received preoperative radiotherapy of 66 Gy. Another patient (13) underwent postoperative radiotherapy (64 Gy) combined with five cycles of doxorubicin plus ifosfamide participating in a multicenter trial (CWS-E protocol; coordinator, W. Hartlapp, M.D., Osnabriick, Germany). All studies in which the patients participated were approved by the local ethics committee of the Humboldt University of Berlin, and the patients provided written informed consent.

Surgical Procedures
The indication to resect the vessels was due either to the angiographic findings, the location of the sarcoma (e.g., Hunter's canal), or fixation of the tumor to the neurovascular bundle. Infiltration to nerves was present in five patients (femoral or tibial nerve), and the nerve had to be sacrificed totally or partly. In another five patients, predominantly those with lipo-sarcomas, tumor growth was adjacent to major nerves that could be salvaged by using the epineurium as a dissection plane (epineurectomy). Intracompartmen-tal lesions were resected by compartmental excision. In extracompartmental lesions, a "wide -excision" that included the surrounding muscle group was performed. Vascular resection was performed within the anatomical segments of the vessels as far as possible from the point of infiltration by the sarcoma (Table 1). Resection specimens were delivered to the pathologist immediately without any fixation.

FIGURE 1. 
(a) Extracompartmental malignant fibrous histiocytoma surrounding the common femoral artery with invasion to artery and vein on magnetic resonance imaging (MRI; case 13). (b) Corresponding angiogram shows the impression of the superficial femoral artery and vein en bloc with that segment of the common, superficial, and deep femoral artery and vein. (c) Reconstruction of the arterial supply to the leg by interposing an autologous contralateral saphenous vein graft from the external iliac to the superficial femoral artery with reimplantation of the main stem of the deep femoral artery to the vein graft after muscle group resection. Due to maintenance of the greater saphenous vein, no venous reconstruction was required, (d) Histology section demonstrates invasion of the arterial wall by sarcoma.
(Please click on images to enlarge)
 

a	b	c	d

TABLE 1 
Characteristics, Postoperative Course, and Current Status of Patients

No.	Age/sex	Postoperative course	Status
1	58/M	Bleeding, revision for hematoma day 1, wound necrosis	DWD 48 mo, lung metastases
2	62/M	Wound and graft infection, omentoplasty, venous graft occluded at 6 mo with persistent infection, explantation of both prostheses, iliacopopliteal bypass	DWD 66 mo, lung and soft tissue metastases
3	46/F	Deep vein thrombosis, revision for hematoma at 2 mo	DWD 14 mo, lung, bone, brain metastases
4	56/F	Revision for bleeding day 1, delayed wound healing	DWD 27 mo, lung metastases, tuberculosis
5	54/M	Uneventful	ANED 69 mo
6	62/M	Uneventful	AWD 39 mo, lung metastases, systemic chemotherapy
7	65/M	Died day 1, myocardial infarction
8	22/F	Occlusion of arterial graft resulting in amputation day 19	ANED 44 mo
9	31/M	Uneventful	DWD 10 mo, lung metastases
10	28/M	Uneventful	DWD 35 mo, recurrent lung metastases,
11	55/M	Uneventful, vein replacement occluded at 3 mo without symptoms	DWD 27 mo, lung metastasectomy at 22 mo after systemic chemotherapy
12	49/M	Uneventful	ANED 34 mo, lung metastasectomy at 24 mo after systemic chemotherapy
13	28/M	Uneventful, lymphedema Grade I-II	AWD 27 mo, lung metastasectomy at 23 mo after systemic chemotherapy
14	56/M	Uneventful	ANED 31 mo
15	35/M	Successful thrombectomy arterial and venous graft day 6	ANED 23 mo
16	29/F	Uneventful	AWD 21 mo, lung metastases with systemic chemotherapy
17	55/M	Lymphatic fistula closure at 10 weeks	ANED 20 mo
18	36/F	Uneventful, lymphedema Grade I, stent for arterial graft stenosis at 15 mo	ANED 20 mo
19	29/F	Uneventful	ANED 7 mo
20	25/M	Uneventful	ANED 7 mo

Reconstructive procedures: Arterial vessels
Nineteen reconstructions for resected arterial segments had to be performed (Table 2). Femoropopliteal reconstruction was the most common single type (n = 9: contralateral saphenous vein, n = 4; ePTFE 6 mm or 8 mm [Goretex], n = 5). In three patients, the reconstruction ended up above-knee; in six patients, the distal anastomosis was at the below-knee level. In six patients, the reconstruction started from the external iliac artery to the femoral or popliteal artery (ring-walled ePTFE, 8 mm, Gore-tex) positioned through the obturator foramen in two cases. One reconstruction required a vein graft to reimplant the major branch of the deep femoral artery (Fig. 1). In another patient (9), a bypass was inserted from the common iliac crossing over to the common femoral artery. In three patients with primary tumors of the popliteal fossa, saphenous vein grafts from the supragenual to the infragenual popliteal artery were^the method of choice.

Venous reconstructions
In 11 patients a reconstruction of resected vein segments also was required. This was achieved by using an allograft in four patients, by using of an autologous contralateral saphenous vein graft in six patients, and by direct suturing in another patient. Nine reconstructions started at the below-knee level and ended up either above the knee (n = 4) or at the femoral level (n = 5). In another patient, a bypass from the femoral to the external iliac vein was implanted through the obturator foramen.

Soft tissue coverage
Four patients required myocutaneous flaps to cover the area of vascular reconstruction due the extent of resection and to provide sufficient nutrient blood supply (patients 6, 14, 15, and 18). We used the rectus abdominis, latissimus dorsi, and parascapular flap. Two patients required a split thickness skin graft.

Follow-up
All patients were followed routinely by computed tomography scan or magnetic resonance imaging (MRI) of the area of the primary tumor and by bidimensional X-ray of the chest every 3 months in the first and second postoperative years and at 6 months intervals thereafter. All data were collected prospectively.

Statistics
Survival analysis was performed with the method of Kaplan and Maier¹⁶ by using the SPSS program (Mi-croSoft Windows release 6.1; SPSS Inc., Chicago, IL). Survival data were calculated from the start of treatment. Operative mortality was included in the analysis of survival and was excluded from calculating survival rates for patients who were free from local or distant relapse.
 

TABLE 2 
Tumor Type and Grade, Location and Previous Therapy, Extent of Tumor Resection, and Vascular Reconstruction

No.	Grade/type	Localization and history of prior treatment	Preoperative treatment, surgical procedure (muscles and nerves resected), vessel repair, and plastic reconstructive measures
1	Gl/leiomyosarcoma	Hunter's canal, residual tumor after R2 resection	Adductor group and vastus intermedius m., superficial femoral artery and vein (AG)
2	G3/liposaicoma	Quadriceps and adductor m.	Quadriceps and adductor compartment, femoral nerve, superficial femoral artery and vein (AG)
3	G3/leiomyosarcoma	Upper thigh, femoral vein occlusion	Adductor group and quadriceps, external iliac and superficial femoral artery and vein (AG)
4	G2/MFH	Groin and adductor group, residual after R2 resection	Adductor group resection, external iliac artery plus common femoral vein (direct suture)
5	G3/MFH	Groin, residual after Rl resection	Abdominal wall and groin, external iliac and common femoral artery (AG), rectus abdominis myocutaneous flap
6	G3/leiomyosarcoma	Popliteal fossa, third recurrence	TNF-ILP; gastrocnemius m., tibial nerve, popliteal artery and vein (VG), myocutaneous parascapular flap
7	G3/mesenchymoma	Hunter's canal, popliteal artery occlusion	Adductor group, epineurectomy sciatic nerve, superficial femoral artery and vein (AG)
8	G3/alveolar type	Hunter's canal	Adductor group, vastus, sartorius, and gracilis m., superficial femoral artery and vein (VG)
9	G3/embryonal type	Small pelvis and groin, recurrence iliac vein occlusion, systemic chemotherapy	Tumor resection plus vessels, common iliac to common femoral artery and vein (AG)
10	G2/leiomyosarcoma	Hunter's canal/popliteal fossa, systemic chemotherapy for lung metastasis, lung metastasectomy	TNF-ILP; adductor and flexor group, superficial femoral artery and vein (VG)
11	G2-3/MFH	Hunter's canal residual after R2 resection	Adductor group, superficial femoral artery and vein (VG)
12	G3/leiomyosarcoma	Flexor group lower limb	Systemic chemotherapy, TNF-ILP; soleus/gastrocnemius m., popliteal and anterior tibial artery and vein (vein patch)
13	G3/MFH	Junction common femoral artery and vein, extracompartmental	Sartorius, rectus, gracilis and adductor group, external iliac to superficial femoral artery and vein (VG), reimplantation deep femoral artery
14	Gl-3/liposarcoma	Popliteal fossa second recurrence fast neutron RT	TNF-ILP; gastrocnemius m., epineurectomy tibial and peroneal nerve, popliteal artery and vein (VG), myocutaneous latissimus dorsi flap
15	Gl/liposarcoma	Groin, second recurrence, fast neutron RT	Abdominal wall, groin dissection, femoral nerve, external iliac and common femoral artery and vein (AG), myocutaneous rectus abdominis flap
16	G3/alveolar type	Thigh and Hunter's canal, residual after R2 resection	Quadriceps compartmentectomy, superficial femoral artery and vein (VG)
17	G3/MFH	Thigh	Preoperative RT (66 Gy), quadriceps compartmentectomy, superficial femoral artery and vein (AG)
18	G3/synovial	Groin, third recurrence, neutron RT, iliac vein occlusion	Systemic chemotherapy and HT; abdominal wall resection and groin dissection, femoral nerve, external iliac to superficial femoral artery and vein (VG), myocutaneous rectus abdominis flap
19	G3/liposarcoma	Thigh and Hunter's canal	TNF-ILP; flexor and adductor group thigh, superficial femoral artery and vein (VG)
20	G3/MFH	Popliteal fossa, infiltration to bone	Gastrocnemius, flexor muscles, proximal tibia, peroneal nerve, endoprosthesis, popliteal artery and vein (VG)

MFH: malignant fibrous histiocytoma; TNF-ILP: isolated, hyperthemic limb perfusion with rhu tumor necrosis factor alpha; HT: regional radiofrequence hyperthermia; m: muscle; VG: vein graft, autologous, contralateral saphenous vein; AG: aliograft (Goretex, 6 mm or 8 mm); RT: radiotherapy. 
 

RESULTS 

Pathologic Findings
Histologic typing of the tumors identified malignant fibrous histiocytoma in six patients, leiomyosarcoma in five patients, and liposarcoma in four patients, whereas alveolar type, synovial sarcoma, mesenchymoma, and embryonic type accounted for the tumors in the remaining five patients (Table 2). The degree of tumor differentiation was determined according to Trojani et al.¹⁷ and was classified as well differentiated in 2 patients, moderately differentiated in 3 patients, and poorly differentiated in another 15 patients.
In all patients, resection resulted in a radical procedure with negative margins (RO-resection) that were confirmed by the pathologist's report. Three resection specimens showed no viable tumor cells after multi-modality therapy.
Invasion of the neurovascular bundle was confirmed histologically in all patients. Invasion of the wall of the resected vein was detected histologically in nine patients, and specimens from two other patients showed that the vein was surrounded by tumor without invasion. A vein thrombosis was documented in four patients. Infiltration to the arterial wall was observed in resection specimens from nine patients: at the adventitial level in four patients, at the muscular layer in another four patients, and to the intima in one patient. Three other patients showed tumor infiltration to a perivascular scar following previous operations. The tumor in one patient was classified according to its origin from the vessel wall (vascular leiomyosarcoma; patient 1). A tumorous infiltration to nerves was confirmed in seven patients.

Mortality and Morbidity
One patient (7) died postoperatively due to occlusion of the anterior branch of the left coronary artery (autopsy finding). Three patients suffered from major complications (see Table 1). Amputation as a result of graft failure was required for patient 8. She presented with occlusion of an autologous femoropopliteal venous graft 6 hours postoperatively that was managed successfully by thrombectomy and intraarterial rh-TPA lysis. Reocclusion occurred on postoperative day 17. She had started with her contraceptive pill without informing the nursing staff. Thrombectomy and intraarterial lysis could not reverse ischemia of the lower limb and forced above-knee amputation. In patient 2, wound infection led to occlusion of the prosthesis replacing the femoral vein. After explantation of the prostheses, arterial blood supply to the lower leg was maintained by an extraanatomical iliacopopliteal (be-low-knee) bypass. Thrombectomy of both arterial and venous grafts was required for patient 15, but his further course was uneventful.
Minor revisions for complications such as postoperative bleeding or lymphatic fistula were required in Jour patients. The overall wound complication rate was 4 of 19 patients (21%).

Overall and Disease Free Survival
The overall median survival of our patients was 48 months (range, 7+ to 69 months; 95% confidence interval [CI], 24-72 months), with a mean survival of 45 months (95% CI, 32-57 months; Fig. 2). Seven patients remained completely free from recurrence with
a median follow-up of 29 months (7-69 months). Currently, another two patients are free from disease at 31 and 34 months postoperatively. Both have undergone resection of recurrent tumor with curative intent and have been free from disease for more than 12 months. In one of them, a solitary lung metastasis was resected 24 months after operation, whereas another patient had resection of a local recurrence that was detected 15 months postoperatively.
Eight patients died 10-66 months after onset of treatment. Reasons for death were progressive lung metastases in seven patients and a caverna from tuberculosis following systemic chemotherapy for lung metastases in another patient.

Local Tumor Relapse
Four patients (4, 10, 11, and 14) developed a local recurrence at 15, 18, 20, and 27 months postoperatively. The mean survival time free from locoregional recurrence was 54 months (95% CI, 42-66 months;
Fig. 3), the median survival has not yet been reached. The 5-year local relapse free survival rate was 68.3%. In three patients (10, 11, and 14), the recurrence was amenable to a second RO-resection with no further local recurrence. However, two patients developed systemic metastases prior to their local relapse (patients 10 and 11). Two patients of those who developed locally recurrent disease had been treated previously for a recurrent (patient 14) or an incompletely resected (patient 4) primary tumor.

Distant Metastases
Eight patients are free from distant metastases 7-69 months after surgery. Another 11 patients developed distant metastases, and seven of them have died. The median survival time free from distant recurrence is 24 months (95% CI, 13-34 months), and the mean survival time is 30 months (95% CI, 18-43 months;
Fig. 4).
In all patients who developed distant recurrences, the lung was the site of metastases. The patients were treated by systemic chemotherapy, and four of them underwent resection of their lung lesions. Other sites that were involved by metastatic spread were bone (n = 1), skin (n = 2), and brain (n = 1).
 

DISCUSSION
Limb-saving surgery is the treatment of choice for soft tissue sarcoma of the extremities.¹⁸ Amputation as an initial treatment can be justified only in cases of infiltration of joints or intractable exulceration and pain. In locally advanced tumors, combined modality approaches, such as pre- or postresection radiotherapy and preoperative systemic or regional chemotherapy, are recommended for limb salvage.^19-23 Hyperthermic ILP provides another effective treatment to induce regression of initially irresectable sarcomas.^24,25 However, even if tumor shrinkage by preoperative therapy can be achieved surgically, adequate resection with clear margins remains a challenging problem.
The patient's prognosis is influenced predominantly by the degree of differentiation of the primary tumor and an adequate surgical resection margin. Microscopic residual disease induces local recurrence and impairs overall survival.^26-30 Especially in recurrent disease, the extent of tumor infiltration into surrounding tissue often cannot be determined definitely. Thus, decisions about adequate surgical margins seem particularly difficult in tumors close to or invading major vessels and nerves. Even within the past few years, such a condition was regarded as cause for amputation.³¹ However, resection and reconstruction of these vessels is feasible and may allow limb salvage without compromise in surgical radicality.
Our policy during the past years was to extend resections for locally advanced sarcomas into neighboring structures to obtain safe margins of clearance. We used preoperative combined-modality therapies, like hyperthermic ILP with rhTNFa and melphalan,³² or systemic chemotherapy combined with regional radiofrequency hyperthermia,^14,15 or hyperfraction-ated radiotherapy to down-stage the tumor in order to facilitate resection. In this setting, vascular surgical techniques are aimed at the en bloc resection of the neurovascular bundle involved. A violation of the tumor border can be avoided, and safe resection margins can be obtained. The necessity to resect arterial and venous vessels is obvious due to tumor involvement of their common fascial sheath; thus, subadven-titial dissection during tumor excision may be inadequate.
In 1977, Fortner et al.¹¹ were the first to describe monobloc soft tissue and vascular resection with arterial and venous reconstruction for soft tissue sarcoma of the thigh in six patients as a measure to avoid lymphedema or postthrombotic syndrome in contrast to earlier patients in whom no reconstruction had been performed. Later, Impararato et al.³³ treated 13 patients for soft tissue and bone sarcomas with en bloc resection of the tumor with surrounding muscle groups, including major vessels and nerves. In all patients, the artery was replaced by an autologous sa-phenous vein graft from uninvolved extremities. Venous reconstructions were performed in three patients in whom the deep femoral vein had to be resected. Another study²⁴ reported on 21 patients with soft tissue sarcoma of the lower limb invading major vessels, which were resected along with surrounding muscles and replaced with vascular grafts. With the exception of one patient, all patients were resected without preoperative treatment.
We achieved a radical resection (negative margins, RO) in all patients. Consequently, only one patient (13; Fig. 1) underwent postoperative radiotherapy, because he took part in a study combining adjuvant chemotherapy for high grade lesions with radiotherapy for all extracompartmental tumors greater than 5 cm. Four patients had local recurrences. Two of them were noted among the group of patients who were referred initially for recurrent or residual tumor. Three recurrent tumors could be excised with curative intention by minor resections. Another patient did not require any treatment of the local relapse due to disseminated systemic tumor spread. In the series by Karakousis et al.,³⁴ a single local recurrence occurred in 12 patients with histologically uninvolved margins who receiving no adjuvant irradiation. Of the nine patients who received adjuvant irradiation, two developed local recurrence, and one required amputation. In another study of seven patients who were treated with monobloc sarcoma and vessel resection, local recurrence was observed in two patients, requiring disarticulation.¹¹ The low recurrence rate in our study with no amputation required for local relapse may have been due to the intensive multimodality treatment concept, which resulted in a histologically complete remission in resected specimens from three patients.
 
 

FIGURE 2. 
Overall survival following surgical treatment of soft tissue sarcoma invading to neurovascular bundles (Kaplan-Meier estimates).

FIGURE 3. 
Local recurrence free survival following surgical treatment of soft tissue sarcoma invading to neurovascular bundles (Kaplan-Meier estimates).

FIGURE 4. Survival free from distant metastases after treatment of soft tissue sarcoma invading to neurovascular bundles (Kaplan-Meier estimates).

Wound healing problems and postoperative bleeding occurred during our initial experience, probably due to excessive anticoagulation in view of the vessel grafts. Three of the first four patients developed wound healing complications due to hematoma formation and inadequate muscle coverage of the reconstruction. In our series, the wound infection rate was 21%, whereas other authors reported a rate of 24%.³⁴ In patients who have been treated more recently, the use of free or pedicled myocutaneous flaps has helped us to avoid these problems. The majority of patients experienced a completely uneventful course.
It has been argued that, after resection of the external iliac vein or superficial femoral vein, a venous bypass would be mandatory to avoid lymphedema.¹¹ From our experience, the capacity of the saphenous vein and its accompanying lymphatics is sufficient to drain the venous system if it can be preserved without compromising radicality. It should to be kept in mind that harvesting of the contralateral saphenous vein also may damage the lymphatic bundle, contributing to edema formation. With the exception of two patients, there was no incapacitating edema in any of the extremities. Mild edema could be treated very successfully with elastic support. Two patients suffered from lymphedema at 24 months and 16 months post-operatively (cases 13 and 18), with a circumference discrepancy of 3 cm that might have been due to postoperative irradiation in patient 13. The rate of postoperative lymphedema formation in our series seems to be lower than that reported after combined modality that included high dose radiotherapy despite resection of the lymphatics.³⁵
If the saphenous vein is absent or must be resected for radicality reasons together with the superficial femoral vein, then it is mandatory to restore the venous drainage of the limb. However, the use of allografts to replace the vein is hampered by a patency rate at 6 months of less than 50%.³⁶ One series on venous grafts reported that only two grafts stayed open for 3 and 4 months, and there were two infections, one of which necessitated amputation.³³ On the other hand, the formation of collateral vessels develops rapidly. Once occlusion of the venous bypass occurs more than 4 weeks after the operation, no attempt for salvage is required.
Two patients with major complications required multiple surgeries (patients 2 and 8). In one patient, occlusion of the arterial reconstruction resulted in amputation-a course that also has been reported in other series.³³ Autologous vein grafts in reversed position always should be preferred to allografts due to a decreased risk of wound infection, less risk in case of hematoma development, and less interference of an-ticoagulation policy with additional oncologic treatment. Muscle coverage of the entire vascular reconstruction is considered critical for maintaining their integrity. Soft tissue defects in the groin or thigh can be covered by local muscle flaps. Free microvascular flaps may be advisable if large popliteal resections expose vessel grafts.^37,38 We used the latissimus dorsi
and parascapular flap in a second procedure about 1 week after initial resection and vascular replacement. A two-step procedure was chosen so that there would be no risk of a free myocutaneous flap due to an occlusion of the arterial or venous implant. However, an immediate free tissue transfer may lead to a shorter hospital stay.³⁹
Thus, the perioperative morbidity accompanying vessel resection and reconstruction seems to be acceptable in relation to the underlying disease, to the extent of the operative procedure, and to the operative procedure itself. The competitive approach combines vessel-preserving surgery with high dose radiotherapy. The wound infection rate of patients undergoing mul-timodality treatment of sarcoma including intraoper-ative or postoperative brachytherapy or external beam radiation seems to be slightly greater (38-48%).^35,40-43 Furthermore, a rate of 9% of radiation neuritis was reported.21
However, brachytherapy provided excellent local control of advanced soft tissue sarcomas. Randomized trials reported that brachytherapy significantly improved local control in high grade lesions but demonstrated no impact in patients with low grade tumors.44 This is particularly helpful in the group of vessel-invasive tumors. In a study of soft tissue sarcomas invading to neurovascular bundles, the in-field local relapse rate was 21%.²¹ The local recurrence rate in series with vessel resection ranged from 14% to 20%,^11,33,34 which also was true in this series. Thus, the local control rates of aggressive surgery or "limited" surgery combined with extensive radiotherapy seem to be comparable. Preoperative down-staging by, e.g., systemic chemotherapy or ILP followed by extended surgery versus limited surgery with intra- and postoperative radiotherapy are competitive approaches in patients with sarcomas invading to neurovascular bundles.
The decision whether or not to resect vessels in close proximity to soft tissue sarcoma depends on how exactly a true invasion by tumor can be assessed pre-or intraoperatively. Recent developments in MRI techniques seem to be promising for providing more accurate information on the distance between the border of the tumor and vessels or nerves.⁴⁵ However, until such reliable data are available, vessel resection is recommended if there remains doubt whether a safe margin of clearance can be obtained by dissecting perivascular or perineural tissue.
Nerve involvement in vessel-invasive tumors often creates a problem due to infiltration of the common fascial sheath. In five patients, we had to sacrifice major nerves (femoral, tibial, and peroneal). Nerve preservation often is not a feasible option for radicality reasons. However, an insensate foreleg, e.g., after resection of the sciatic nerve, creates a significant problem if the saphenous nerve does not contribute enough to the sensibility of the foot. Limb salvage by resection of vessels and major nerves that leaves behind an extremity without tactile sensations is not advisable. In these cases, amputation still may be the treatment of choice.
During the further course of their disease, in none of our patients was local tumor relapse a major problem. However, distant metastases developed in 11 patients (55%), and another two patients already had lung metastases preoperatively. This high incidence is interrelated to the proportion of high grade tumors, their deep location, and often their size.^6,46 The rate of distant metastases in high grade soft tissue sarcoma was reported up to 45%.^5,7,8,47 In another report on tumors invading to neurovascular bundles, it was 30%.²¹ Lung metastases developed predominantly, but there were also other distant metastases to the pelvis, brain, and spine.
Consequently, the major cause of death in this group of patients remained metastatic disease but not local recurrence.⁴⁸ The median survival in this series was 48 months, which is a marked improvement compared with previous data reporting 26 months³³ that have been confirmed by other authors.³⁴ The survival rate of the extended surgery approach also seems to be higher than the 32% reported for patients undergoing tumor resection alone without preoperative multimodal therapy.⁸ Thus, our greatest challenge is to improve control of distant disease. Despite excellent local control rates following limb-sparing procedures, greater than 50% of patients ultimately will die of their disease.⁴⁹
In many instances, only the availability of vascular surgical techniques allows radical surgery for primary or recurrent sarcoma. The limits of resectability, however, are extended in such a way that they necessitate plastic reconstruction of large soft tissue defects. There is a growing need for a plastic surgeons to join the medical team in order to treat soft tissue sarcoma successfully.
In conclusion, this study has demonstrated that the involvement of major vessels by soft tissue sarcomas of the extremities is not an indication for amputation by itself. After en bloc resection of major vessels, the local recurrence and 5-year survival rates parallel those of patients with soft tissue sarcomas that do not require vessel resection. Long term local relapse free survival could be provided after preoperative combined modality therapy followed by extended surgery. Vascular resection and reconstruction as well as plastic reconstructive measures enabled radical resections with negative margins in all patients and confirmed three histologically complete tumor remissions.
This aggressive surgical approach is in competition with more limited resections combined with in-tra- and postoperative radiotherapy. The morbidity data of the more surgical approach seem to be favorable and make irradiation techniques available in cases of local recurrence. The limitation for long term disease free survival, however, is distant metastases. Therefore, new systemic therapies must be developed to control disseminated tumor growth.
 

REFERENCES
 

1. Boring CC, Squires TS, Tong T, Montgomery S. Cancer statistics, 1994. CA Cancer J din 1994;44:7-26.
2. Karakousis CP, Emrich LJ, Rao U, Khalil M. Limb salvage in soft tissue sarcomas with selective combination of modalities. EurJ Surg Oncol 1991;17:71-80.
3. Williard WC, Hajdu SI, Casper ES, Brennan MF. Comparison of amputation with limb-sparing operations for adult soft tissue sarcoma of the extremity. Ann Surg 1992;215:269-75.
4. Rosenberg SA, Tepper J, Glatstein E, Costa J, Baker A, Brennan MF, et al. The treatment of soft-tissue sarcoma of the extremities. Prospective randomized evaluation of limb-sparing surgery plus radiation therapy compared with amputation and the role of adjuvant chemotherapy. Ann Surg 1982:305
5. 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 1996;14:1679-89.
6. Coindre JM, Terrier P, Bui NB, Bonichon F, Collin F, Le Doussal V, et al. Prognostic factors in adult patients with locally controlled soft tissue sarcoma. A study of 546 patients from the French Federation of Cancer Centers Sarcoma Group. J Clin Oncol 1996;14:869-77.
7. Rooser B, Berg NO, Ranstam J, Rydholm A, Willen H. Prediction of survival in patients with high-grade soft tissue sarcoma. Int Orthop 1990; 14:199-204.
8. Ruka W, Emrich LJ, Driscoll DL, Karakousis CP. Prognostic significance of lymph node metastasis and bone, major wessel, or nerve involvement in adults with high-grade soft tissue sarcomas. Cancer 1988:62:999-1006.
9. Shiu MH, Hilaris BS, Harrison LB, Brennan MF. Brachyther-apy and function-saving resection of soft tissue sarcoma arising in the limb. Int J Radiat Oncol Biol Phys 1991;21:1485-92.
10. Hug EB, Spiro IJ, Cole DJ, Suit HD. Combined surgery and radiotherapy for conservative management of soft tissue sarcomas. Rec Res Cancer Res 1995; 138:47-55.
11. Fortner JG, Kim DK, Shiu MH. Limb-preserving vascular surgery for malignant tumors of the extremities. Arch Surg 1977; 112:391-4.
12. Eggermont AM, Schraffordt Koops H, Klausner JM, Kroon BB, Schlag PM, Lienard D, et al. Isolated limb perfusion with tumor necrosis factor and melphalan for limb salvage in 186 patients with locally advanced soft tissue extremity sarcomas. Ann Surg 1996;224:756-65.
13. Reichardt P, Lentzsch S, Hohenberger P, Dorken B. Dose intensive treatment with ifosfamide, epirubicin, and filgras-tim for patients with metastatic or locally advanced soft tissue sarcoma. ] Clin Oncol 1998;16:1438-43
14. Reichardt P, Tilgner J, Mapara MY, Hohenberger P, Dorken B. Dose-intensive chemotherapy with or without stem cell support for adult patients with advanced soft tissue sarcoma. ProcASCO 1997;16:A1791.
15. Issels RD, Bosse D, Abdel Rahman S, Starck M, Panzer M, Jauch KW, et al. Preoperative systemic etoposide/ifosfamide/doxorubicin chemotherapy combined with regional hyperthermia in high-risk sarcoma: a pilot study. Cancer Chemother Pharmacol 1993;31(Suppl 2):S233-7.
16. Kaplan EL, Maier P. Nonparametric estimation from incomplete observations. } Med Statist Assoc 1958;53:457-81.
17. Trojani M, Contesso G, Coindre JM, et al. Soft tissue sarcoma of adults; study of pathological prognostic variables and definition of a histopathological grading system. Int J Cancer 1984:33:37-42.
18. Williard WC, Collin C, Casper ES, Hajdu SI, Brennan MF. The changing role of amputation for soft tissue sarcoma of the extremity in adults. Surg Gynecol Obstet 1992;175:389-96.
19. Engel CJ, Eilber FR, Rosen G, Selch MT, Fu YS. Preoperative chemotherapy for soft tissue sarcomas of the extremities: the experience at the University of California, Los Angeles. Cancer Treat Res 1993;67:135-41.
20. Suit HD, Spiro I. Role of radiation in the management of adult patients with sarcoma of soft tissue. Semin Surg Oncol 1994; 10:347-56.
21. Zeiefsky MJ, Nori D, Shiu MH, Brennan MF. Limb salvage in soft tissue sarcomas involving neurovascular structures using combined surgical resection and brachytherapy. Int J Radiat Oncol Biol Phys 1990;19:913-8.
22. Fein DA, Lee WR, Lanciano RM, Corn BW, Herbert SH, Hanlon AL, et al. Management of extremity soft tissue sarcomas with limb-sparing surgery and postoperative irradiation: do total dose, overall treatment time, and the surgery-radiotherapy interval impact on local control? Int J Radiat Oncol Biol Phys 1995:32:969-76.
23. Wanebo HJ, Temple WJ, Popp MB, Constable W, Aron B, Cunningham SL. Preoperative regional therapy for extremity sarcoma. A tricenter update. Cancer 1995;75:2299-306.
24. Lienard D, Ewalenko P, Delmotte JJ, Renard N, Lejeune FJ. High-dose recombinant tumor necrosis factor alpha in combination with interferon gamma and melphalan in isolation perfusion of the limbs for melanoma and sarcoma. J Clin Oncol 1992;10:52-60.
25. Rossi CR, Vecchiato A, Foletto M, Nitti D, Ninfo V, Fornasi-ero A, et al. Phase II study on neoadjuvant hyperthermic-antiblastic perfusion with doxorubicin in patients with intermediate or high grade limb sarcomas. Cancer 1994;73:2140-6.
26. Choong PF, Gustafson P, Rydholm A. Size and timing of local recurrence predicts metastasis in soft tissue. Acta Or-thop Scand 1995;66:147-52.
27. Emrich LJ, Ruka W, Driscoll DL, Karakousis CP. The effect of local recurrence on survival time in adult high-grade soft tissue sarcomas. / Clin Epidemiol 1989:42:105-10.
28. Heslin MJ, Gaynor JJ, Newman E, Wolf RF, Woodruff J, Casper ES, et al. Effect of perioperative blood transfusion on recurrence and survival in 232 primary high-grade extremity sarcoma patients. Ann Surg Oncol 1994; 1:189-97.
29. Herbert SH, Corn BW, Solin LJ, Lanciano RM, Schultz DJ, McKenna WG, et al. Limb-preserving treatment for soft tissue sarcomas of the extremities. The significance of surgical margins. Cancer 1993;72:1230-8.
30. Lewis JJ, Leung D, Heslin M, Woodruff JM, Brennan MF. Association of local recurrence with subsequent survival in extremity soft tissue sarcoma. J Clin Oncol 1997; 15:646-52.
31. Lawrence WD, Donegan W, Natarajan N, Mettlin C, Beart R, Winchester DP. Adult soft tissue sarcoma. Ann Surg 1987;205:349-59.
32. Eggermont AM, Schraffordt Koops H, Lienard D, Kroon BB, van Geel AN, Hoekstra HJ, et al. Isolated limb perfusion with high dose tumor necrosis factor alpha in combination with interferon gamma and melphalan for irresectable extremity soft tissue sarcomas: a multicenter trial. J Clin Oncol 1996;14:2653-65.
33. Imparato A, Roses D, Francis K, Lewis M. Major vascular reconstructions for limb salvage in patients with soft tissue and skeletal sarcomas of the extremities. Surg Gynecol Obstet 1978;147:891-6.
34. Karakousis CP, Karmpaliotis C, Driscoll DL. Major vessel resection during limb-preserving surgery for soft tissue sarcoma. World J Surg 1996:20:345-50.
35. Stinson SF, DeLaney TF, Greenberg ], Yang JC, Lamport MH, Hicks JE, et al. Acute and long-term effects on limb function of combined modality limb sparing therapy for extremity soft tissue sarcoma. Int J Radiat Oncol Biol Phys 1991;21:1493-9.
36. Okadome K, Muto Y, Eguchi H, Kusaba A, Sugimachi K. Venous reconstruction for iliofemoral venous occlusion facilitated by temporary arteriovenous shunt. Long-term results in nine patients. Arch Surg 1989; 124:957-60.
37. Steinau HU, Hussmann J, Hebebrand D. Rekonstruktion-smoglichkeiten nach erweiterter Resektion maligner Weich-teilgeschwulste [Reconstructive possibilities after extensive resection of malignant soft tissue tumors]. Chirurg 1993;64:517-26.
38. Evans GR, Black JJ, Robb GL, Baldwin BJ, Kroll SS, Miller MJ, et al. Adjuvant therapy: the effects on microvascular lower extremity reconstruction. Ann Plast Surg 1997;39:141-4.
39. Reece GP, Schusterman MA, Pollock RE, Kroll SS, Miller MJ, Baldwin BJ, et al. Immediate versus delayed free-tissue transfer salvage of the lower extremity in soft tissue sarcoma patients. Ann Surg Oncol 1994; 1:11-7.
40. Engin K, Leeper DB, Cater JR, Thistlethwaite AJ, Tupchong L, McFarlane JD. Extracellular pH distribution in human tumours. Int} Hyperthermia 1995;11:211-6.
41. Dalton RR, Lanciano RM, Hoffman JP, Eisenberg BL. Wound complications after resection and immediate postoperative brachytherapy in the management of soft-tissue sarcomas. Ann Surg Oncol 1996;3:51-6.
42. Arbeit JM, Hilaris BS, Brennan MF. Wound complications in the multimodality treatment of extremity and superficial truncal sarcomas. / Clin Oncol 1987;5:480-8.
43. Alekhteyar KM, Leung DH, Brennan MF, Harrison LB. The effect of combined external beam radiotherapy and brachytherapy on local control and wound complications in patients with high-grade soft tissue sarcomas of the extremity with positive microscopic margin. Int J Radiat Biol 1996:36:321-4.
44. Pisters PW, Harrison LB, Leung DH, Woodruff JM, Casper ES, Brennan MF. Long-term results of a prospective randomized trial of adjuvant brachytherapy in soft tissue sarcoma. / Clin Oncol 1996;14:859-68.
45. Panicek DM, Go SD, Healey JH, Leung DH, Brennan MF, Lewis JJ. Soft tissue sarcoma involving bone or neurovascu-lar structures: MR imaging prognositc factors. Radiology 1997;205:871-5
46. Heise HW, Myers MH, Russell WO, Suit HD, Enzinger FM, Ed-mondson JH, et al. Recurrence-free survival time for surgically treated soft tissue sarcoma patients. Cancer 1986;57:172-7.
47. Potter DA, Kinsella T, Glatstein E, Wesley R, White D, Seipp C, et al. High-grade soft tissue sarcomas of the extremities. Cancer 1986:58:190-205.
48. Heslin MJ, Woodruff J, Brennan MF. Prognostic significance of a positive microscopic margin in high-risk extremity soft tissue sarcoma: implications for management. / Clin Oncol 1996;14:473-8.
49. Spiro IJ, Rosenberg AE, Springfield D, Suit H. Combined surgery and radiation therapy for limb preservation in soft tissue sarcoma of the extremity. Cancer Invest 1995; 13:86-95.