Venous reconstructions in lower limbs associated with resection of malignancies
Article Outline
Background
Patients with tumors in the limbs who undergo surgical treatment may have involvement of major vessels. Major arteries are always reconstructed for limb salvage. Major veins may be reconstructed to avoid the onset of venous hypertension signs and symptoms. The objective of this study was to analyze the results from surgical treatment of a sample of patients who underwent lower limb venous reconstructions associated with the resection of malignant tumors.
Methods
Follow-up was performed of 17 patients with malignant tumors involving major vessels in the lower limbs. The median length of follow-up was 22 months. Venous reconstruction concomitant to arterial reconstruction was performed in 15 patients, and an isolated venous reconstruction was performed in 2 patients. The venous substitutes used were the contralateral long saphenous vein (n = 12), expanded polytetrafluoroethylene prosthesis (n = 3), and Dacron prosthesis (n = 2).
Results
Vascular complications occurred in seven patients: three occlusions of the venous graft, edema in seven patients, and one rupture of the arterial graft. The primary 2- and 5-year patency rates of venous reconstructions were 79.3% and 79.3%, respectively. Nonvascular complications occurred in six patients: neurological deficit (n = 3), partial necrosis of the flap (n = 2), wound infection (n = 1), hematoma (n = 1), and enteric fistula (n = 1). Eight patients were still alive and disease free, although one of them underwent above-knee amputation as a result of local disease recurrence. One patient experienced regional disease recurrence and is undergoing chemotherapy. Eight patients died due to pulmonary metastases. The 2- and 5-year overall survival rates were 58.6% and 42.4%, respectively. The 2- and 5-year thrombosis-free survival rates were 51.9% and 35.2%, respectively.
Conclusions
Lower limb venous reconstructions associated with tumor resection in this study gave good functional results, although the prognosis for these patients had been unfavorable. The saphenous vein is a suitable substitute.
Up to the 1970s, patients with neoplasia in the extremities with involvement of the vascular bundle frequently had to undergo amputation of the limb. Progress in surgical procedures and the advent of multidisciplinary approaches have allowed tumor resection to be performed with an adequate safety margin, thus salvaging the limb1, 2, 3 with good functional ability.
When the main artery is resected with the tumor mass, there is no doubt regarding the need for its reconstruction, because of the risk of severe ischemia and limb loss. However, there is no consensus in relation to venous reconstruction, regarding either the need to do it or the support measures to be used. The biggest sample published contains 15 patients,4 although there are no objective data (imaging examinations) for analyzing the patency of the reconstructions.
The objective of this study was to analyze the results from surgical treatment of a large sample of patients who underwent lower limb venous reconstruction, considering the primary patency, vascular and nonvascular complications, tumor recurrence, and patient survival.
Materials and methods
Seventeen patients who underwent operations because of malignant tumors that had involved major arteries and/or veins in the lower limbs were followed up from September 1997 to January 2006. The length of follow-up ranged from 2 to 96 months, with a median of 22 months. Thirteen patients were women and four were men, with ages ranging from 15 to 91 years (median, 38 years). With regard to prior treatment for the tumor, six patients had not undergone any type of therapy. Separate or comprehensive previous treatments had included surgery, chemotherapy, and radiotherapy. Seven patients had already undergone operations, seven had had chemotherapy, and six had had radiotherapy. The sites for the neoplasia were soft tissue in the thigh (n = 8), pelvis (n = 2), fibula (n = 2), femur (n = 2), inguinal nodes (n = 2), and vulva (n = 1).
Upon clinical vascular examination, five patients presented with alterations: four patients presented edema at the impaired limb, and one of them did not present popliteal and tibial arterial pulses. Another patient also did not present popliteal and tibial arterial pulses.
With regard to complementary examinations, all patients underwent computerized tomography or nuclear magnetic resonance, in which the tumoral mass was always shown to be lacking a cleavage plane or agglomerating the resected vessels. The patients who had femoral-popliteal obstruction underwent lower limb magnetic resonance angiography, which demonstrated obstruction of the superficial femoral artery and refilling of the popliteal and tibial arteries, in addition to the absence of parietal irregularities in these segments.
En-bloc resection of the vessels involved by the tumor was chosen by the oncology surgeon on the basis of the operative findings, which brought out evidence of agglomeration or adherence to the tumoral mass. The procedure was indicated in the cases for which vascular resection would be appropriate from an oncologic point of view and there were no other areas of tumor that were not resectable or where the resectability would be in doubt.
Venous reconstruction was performed routinely, except when the stump distal to the tumoral bloc was occluded or when the vascular resection was performed in a bundle that was considered nonessential. Arterial reconstruction was always performed.
Venous and arterial reconstructions were performed concomitantly in 15 patients: femoral-popliteal (n = 7), external iliac–common femoral (n = 3), external iliac–superficial femoral (n = 2), common iliac–external iliac (n = 1), common femoral–popliteal (n = 1), and popliteal–posterior tibial (n = 1). An isolated venous reconstruction was performed in two patients: femoral–external iliac (n = 1) and popliteal-femoral (n = 1).
The vascular substitutes used for the venous reconstructions were the contralateral long saphenous vein (n = 12), nonringed expanded polytetrafluoroethylene prosthesis (n = 3), and Dacron prosthesis (DuPont, Wilmington, Del; n = 2). For the patients who underwent arterial and venous reconstructions, the same substitute was used for both grafts. In the cases in which the long saphenous vein was used, the arterial grafts were constructed by using a reversed vein.
Tissue coverage of resected areas and associated vascular grafts was provided by local muscle/skin flaps. Prophylactic antibiotic therapy and vacuum drains were used in all cases. No reconstruction of nerves was performed in the patients who underwent combined resection.
The histologic types encountered were synovial sarcoma (n = 4), osteosarcoma (n = 3), fibrohistiocytoma (n = 2), leiomyosarcoma (n = 2), malignant melanoma (n = 2), squamous cell carcinoma (n = 1), fibrosarcoma (n = 1), hemangiopericytoma (n = 1), and liposarcoma (n = 1).
During the hospital stay, prophylaxis for deep vein thrombosis, consisting of low-molecular-weight heparin, was administered to all patients. After discharge from the hospital, no drugs aimed at prolonging the patency of the grafts were administered, such as antiplatelet aggregation drugs, anticoagulants, or peripheral vasodilators.
In Table I, the characteristics of the operated patients are reported. The follow-up of the patients was performed by means of outpatient returns, at which a clinical examination and duplex surveillance were performed to study the patency of the grafts. Duplex mapping was performed every 6 months during the first year after the operation and annually thereafter or upon suspicion of a complication regarding the vascular grafts (occlusion).
Table I. Epidemiologic characteristics, site, histology type, impaired vessels, vascular substitute, and vascular and nonvascular complications
| Patient No. | Sex/age (y) | Site | Type | Impaired vessels | Substitute | Vascular | Nonvascular |
|---|---|---|---|---|---|---|---|
| 1 | F/29 | Thigh | Fibrohistiocytoma | Fem v | SV | — | — |
| 2 | F/91 | Vulva | Squamous cell carcinoma | Fem v | PTFE | — | Flap necrosis |
| 3 | F/16 | Fibula | Osteosarcoma | Popliteal v + a | SV/SV | — | Neurologic deficit |
| 4 | M/24 | Femur | Osteosarcoma | Fem v + sup fem a | SV/SV | — | — |
| 5 | M/23 | Thigh | Fibrosarcoma | Fem v + sup fem a | SV/SV | Edema | — |
| 6 | F/15 | Thigh | Synovial sarcoma | Fem v + sup fem a | SV/SV | — | — |
| 7 | F/39 | Thigh | Hemangiopericytoma | Fem v + com fem a | SV/SV | Edema | — |
| 8 | F/65 | Inguinal nodes | Malignant melanoma | Fem v + com fem a | Dacron/Dacron | Venous occlusion + edema | Flap necrosis + neurologic deficit |
| 9 | F/59 | Inguinal nodes | Malignant melanoma | Fem v + com fem a | PTFE/PTFE | Venous occlusion + edema | Enteric fistula + neurologic deficit |
| 10 | F/23 | Thigh | Synovial sarcoma | Fem v + sup fem a | SV/SV | Arterial rupture + edema | Wound infection |
| 11 | F/27 | Thigh | Fibrohistiocytoma | Fem v + sup fem a | SV/SV | — | — |
| 12 | F/62 | Fibula | Synovial sarcoma | Popliteal v + a | SV/SV | — | — |
| 13 | F/55 | Pelvis | Leiomyosarcoma | Ext iliac v + a | PTFE/PTFE | — | — |
| 14 | M/52 | Pelvis | Liposarcoma | Ext iliac v + a | Dacron/Dacron | Venous occlusion + edema | — |
| 15 | M/42 | Thigh | Synovial sarcoma | Fem v + sup fem a | SV/SV | Edema | Hematoma |
| 16 | F/21 | Femur | Osteosarcoma | Fem v + sup fem a | SV/SV | — | — |
| 17 | F/38 | Thigh | Leiomyosarcoma | Fem v + sup fem a | SV/SV | — | — |
Results
Eight patients did not experience any type of complication during the follow-up. Vascular complications occurred in seven patients. Three patients had venous graft occlusions. All of these cases presented with an abrupt onset of swelling in the affected limb, and duplex mapping confirmed the diagnosis. The occlusions occurred at 2, 3, and 8 months after surgery. In all cases, the substitute used was prosthetic, and the patients were treated by means of systemic anticoagulation in addition to elastic support. No patient progressed with hyperpigmentation, eczema, or venous claudication in the lower limb. The primary 2- and 5-year patency rates of venous reconstructions were 79.3% and 79.3%, respectively (Fig 1).
Fig 1.
Primary patency of venous reconstructions.
One patient who underwent venous and arterial reconstructions progressed with infection of the operative wound and was readmitted to the hospital to receive endovenous systemic antibiotic therapy but experienced rupture of the arterial graft on the 21st day after the operation. During exploratory surgery, the location of the rupture was identified, and this graft was ligated with Prolene 5-0 (Ethicon, Johnson & Johnson, São José dos Campos, São Paulo, Brazil). It was not possible to identify the venous graft, which had a deeper location, but postoperative duplex mapping showed that the venous graft was patent and continued to be so throughout the follow-up. This patient had had previous femoral-popliteal obstruction, and, despite the vascular ligature, there was no severe ischemia of the limb. All remaining arterial grafts were patent. The primary 2- and 5-year patency rates of arterial reconstructions were 93.1% and 93.1%, respectively.
Limb edema was the most frequent vascular complication, and this occurred in seven patients, of whom four had previously experienced some degree of edema. All these patients were treated on an outpatient basis by means of manual lymphatic drainage sessions together with inelastic bandaging during the first 3 weeks. After this, the manual drainage was associated with elastic compression, and, after the condition stabilized, only the elastic support was used. Only one patient, who already had substantial preoperative lymphedema, did not improve.
Nonvascular complications were detected in six patients: neurologic deficit (n = 3), partial necrosis of the flap (n = 2), wound infection (n = 1), hematoma (n = 1), and enteric fistula (n = 1). From a functional point of view, four had moderate functional limitation in the operated limb. One patient already had significant lymphedema in the affected limb, and it worsened after the operation. In another three patients, the limitation was caused by combined resection of nerves (fibular in one case and femoral in two cases), and one required an orthosis to be able to walk.
There was no death during the intraoperative or hospitalization periods. Eight patients were still alive and disease free after a median of 48 months of follow-up, although one of these underwent above-knee amputation because of local tumor recurrence after 11 months. One patient had regional disease recurrence and is undergoing chemotherapy after 96 months. Eight patients died as a result of pulmonary metastases after a median of 20.5 months. The 2- and 5-year overall survival rates were 58.6% and 42.4%, respectively (Fig 2). The 2- and 5-year thrombosis-free survival rates were 51.9% and 35.2%, respectively (Fig 3).
Fig 2.
Overall survival of patients treated with en-bloc vessel resection and reconstruction in the lower limbs.
Fig 3.
Thrombosis-free survival of patients treated with en-bloc vessel resection and reconstruction in the lower limbs.
Discussion
The treatment of patients with malignant neoplasia in the lower limbs has been continually improved over the last 30 years. Because of the development of imaging examinations (with the possibility of early diagnosis), effective adjuvant therapies, and better operative techniques, the concept of tumor irresectability is becoming increasingly restricted, such that vascular invasion no longer represents an obstacle to resection of these tumors. Among the indications for primary amputation of a limb, the following can be considered: patients with large infected tumors and patients with tumors that are complex because of the involvement of the vascular-nervous bundle and the bone. These patients are candidates for extensive resection, although it is impossible to accomplish adequate coverage for the bleeding areas.
Considering the high risk of severe ischemia and limb loss after arterial ligature, reconstruction of the artery is always indicated. In cases of neoplasm resection, in addition to the trunk vessels, a large part of the collateral circulation is resected for oncologic purposes, thus corroborating the need for the arterial reconstruction.5, 6, 7, 8, 9, 10, 11, 12
In our sample, there was only one arterial graft complication, which consisted of rupture consequent to wound infection, despite the use of an autologous substitute. This patient had previously had femoral-popliteal obstruction, and there was no ischemia of the limb, despite the ligature of the graft.
Venous reconstructions associated with the resection of neoplasia are less prominent in the literature than arterial reconstructions. Sometimes the samples of venous reconstruction do not have follow-up data.2, 4, 6, 12 Venous ligature does not interfere with the preservation of the limb,1, 2, 13 and therefore reconstructions are not always performed. In addition to this, after ligature, some authors have reported achieving satisfactory control over the edema during its acute period via limb elevation and elastic support.2, 5, 7 Nonetheless, over the long term, a patient submitted to venous resection without reconstruction may experience severe edema, claudication, hyperpigmentation, and eczema on the affected limb.13 In such patients, the deficit in the lymphatic drainage8 is another significant factor that generates edema.
A variety of venous substitutes have been described (Table II). The use of a prosthetic substitute has been predominant,4, 5, 7, 8, 14 although the best patency rates have been obtained from samples in which autologous substitutes prevailed.9, 10, 11 Moreover, infection of the vascular graft has been more associated with prosthetic substitutes.4, 5, 8 For these reasons, we gave preference to autologous substitutes, such that the contralateral long saphenous vein was used in 12 of the 17 reconstructions. In the segments where there was some degree of caliber discrepancy, end-to-side anastomoses were performed with good graft functioning. In the cases in which there was great caliber discrepancy with the saphenous vein, or in its absence (previous saphenectomy), we decided on prosthetic substitution. One technical alternative in these cases would be the use of a panel graft or spiral graft of autologous superficial veins, such as the long saphenous vein itself or the short saphenous vein and veins from the upper limbs. Other autologous substitutes described have been the femoral vein12 and axillary vein,1 although we believe that resection of veins in the deep vein system may cause morbidity in that limb.
Table II. Venous reconstructions
| Author | n | Substitutes | Patency |
|---|---|---|---|
Fortner1 (1977) | 3 | Saphenous 2/axillary 1 | ? |
| Imparato2(1978) | 3 | Saphenous | ? |
| Nambisan5(1987) | 6 | PTFE | 33% |
| Steed6(1987) | 1 | Saphenous | ? |
| Karakousis7(1989) | 9 | PTFE | 0% |
| Kawai8(1996) | 7 | PTFE 5/saphenous 2 | 14% |
| Koperna9(1996) | 13 | Saphenous 8/PTFE 4/Dacron 1 | 77% |
| Karakousis4(1996) | 15 | PTFE 14/saphenous 1 | ? |
| Hohenberger10(1999) | 10 | Saphenous 6/PTFE 4 | 72% |
| Bonardelli11(2000) | 5 | Saphenous 3/transposition 2 | 100% |
| Leggon12(2001) | 8 | Saphenous 5/femoral 2/Dacron 1 | ? |
| Schwarzbach14(2005) | 12 | PTFE 10/saphenous 2 | 58% |
| Present study | 17 | Saphenous 12/PTFE 3/Dacron 2 | 82% |
The patency rates for venous reconstructions are lower than for arterial ones.11, 12, 14 Some authors have used adjuvant measures such as the construction of an arteriovenous fistula8 or oral anticoagulation treatment,8, 9 with the aim of improving the patency rates for venous reconstructions, although without success. The arterial patency can be detected by means of clinical examination, although complementary imaging examinations are needed for ascertaining whether venous patency has been achieved. Some authors have performed phlebography5, 7, 8 as a component of the follow-up, but we prefer duplex mapping9, 11 because of its noninvasive nature and high accuracy.
In our sample, there were no cases of occlusion of the venous graft with an autologous substitute, whereas occlusion occurred in three of the five cases of prosthetic grafts. Nevertheless, these occlusions occurred after an average of 4 months and, therefore, without increasing morbidity during the acute postoperative period.
Edema in the affected limb was the most frequent vascular complication1, 2, 4, 5, 7 and was seen in seven patients. This complication had lymphatic and venous components and was caused by the extensive vascular resection during the operation that was needed for obtaining oncologic safety margins. This was, in some cases, associated with occlusion of the graft. Clinical treatment was satisfactory in all of these patients, with the exception of one case in which there was substantial preoperative lymphedema.
Over the period of this study, 106 patients with soft tissue sarcomas in the lower extremities underwent operations in our institution. Ten of them underwent primary amputation of the limb. The length of follow-up for the patients who underwent primary amputation ranged from 1 to 96 months, with a median of 13.5 months. The estimated 2- and 5-year survival rates were, respectively, 30% and 30%. Although the group of patients who underwent vascular reconstruction was not absolutely comparable with the group of patients who underwent primary amputation (the first group also contained patients with other malignant neoplasias, as well as soft tissue sarcomas), the survival rates (58.6% and 42.4%) presented were better than those for the group of amputees, with the advantage of having preserved the limb.
The occurrence of a functional deficit in the limb was expected and discussed during the preoperative period with some patients; however, no one accepted primary amputation as the first alternative of treatment. In patients for whom a functional deficit in the limb is expected, which is principally caused by combined resection of nerves, primary amputation is becoming less and less accepted, because of the emergence of orthoses together with the difficulty in accepting loss of the limb.
Conclusions
Lower limb venous reconstructions associated with tumor resection in this study gave good functional results, although the prognosis for these patients had been unfavorable. The saphenous vein is a suitable substitute.
Author contributions
References
- . Limb-preserving vascular surgery for malignant tumors of the lower extremity. Arch Surg. 1977;112:391–394
- . Major vascular reconstruction for limb salvage in patients with soft tissue and skeletal sarcomas of the extremities. Surg Gynecol Obstet. 1978;147:892–896
- . Is amputation necessary for sarcomas? (A seven-year experience with limb salvage). Ann Surg. 1980;192:431–438
- . Major vessel resection during limb-preserving surgery for soft tissue sarcomas. World J Surg. 1996;20:345–350
- . Vascular reconstruction for limb salvage in soft tissue sarcomas. Surgery. 1987;101:668–677
- . Limb sparing operations for sarcomas of the extremities involving critical arterial circulation. Surg Gynecol Obstet. 1987;164:493–498
- . Soft-tissue sarcomas of the proximal lower extremity. Arch Surg. 1989;124:1297–1300
- . Vascular reconstruction in limb salvage operations for soft tissue tumors of the extremities. Clin Orthop. 1996;332:215–222
- Vascular reconstruction for limb salvage in sarcoma of the lower extremity. Arch Surg. 1996;131:1103–1107
- . Results of surgery and multimodal therapy for patients with soft tissue sarcoma invading to vascular structures. Cancer. 1999;85:396–408
- Limb salvage in lower-extremity sarcomas and technical details about vascular reconstruction. J Orthop Sci. 2000;5:555–560
- . Limb salvage surgery with vascular reconstruction. Clin Orthop. 2001;387:207–216
- Sequelae after limb-sparing surgery with major vascular resection for tumor of the lower extremity. J Vasc Surg. 2001;33:694–699
- Results of limb-sparing surgery with vascular replacement for soft tissue sarcoma in the lower extremity. J Vasc Surg. 2005;42:88–97
Competition of interest: none.
PII: S0741-5214(06)01306-1
doi:10.1016/j.jvs.2006.07.033
© 2006 The Society for Vascular Surgery. Published by Elsevier Inc. All rights reserved.
