Outcomes of Planned Marginal and Wide Resection of Sarcomas Associated with Major Vascular Structures in Extremities

Abstract

Bone and soft tissue sarcomas of lower and upper extremities may sometimes be in close contact with neurovascular structures. In such cases, it is controversial that whether en bloc resection and vascular reconstruction to reach wider surgical margins or planned marginal resection with the help of adjuvant therapies should be preferred. This study aimed to determine surgical and oncological outcomes of planned marginal and wide resection of extremity sarcomas that are associated with major vascular structures in the extremities. The collected database of 54 patients treated by the same orthopedic and vascular surgeon for primary or locally recurrent soft and bone tissue sarcoma of extremities was retrospectively reviewed. Eligible subjects for this study were patients diagnosed with upper and lower extremity soft and bone tissue sarcomas that encased a maximum of 50% of the circumference of the major vascular structures, requiring limb-sparing resection. When microscopic positive (19 patients, 33.9%) and negative cases’ (35 patients, 66.1%) surgical margins were compared, local recurrence, metastasis, amputation, and tumor type (soft/bone) parameters showed no statistically significant difference. When metastatic and non-metastatic patients were compared, it was shown that bone tumors metastasized more than soft tissue tumors (p = 0.001). However, there was no difference between metastasis and amputation, histopathology, grade, nerve involvement, surgical margins, or local recurrences. The mean survival was 1460.6 ± 137.4 days, and the 6-year mortality was 87.5%. Anesthetic and surgical complication rates may be higher since en bloc resection surgeries of large tumors with vascular reconstructions take a very long time. Therefore, we suggest marginal resection with sub-adventitial dissection in those locations and wide resection at other areas according to the surgeon’s experiences about safe margin with the contribution of radiotherapy.

Background

Bone and soft tissue malignant tumors located in the extremities are rare. Currently, with the use of adjuvant chemotherapy, radiation protocols, or both, as well as advanced imaging techniques, 90–95% of patients with primary malignant bone and soft tissue tumors involving the extremities can be treated safely using wide resection and limb salvage surgery with a low risk of recurrence and the same disease-free survival rate as amputation surgery [1–5]. The goal of limb-sparing surgery is to perform a wide resection without compromising the limb function and to achieve acceptable complication rates [2, 4]. Limb-sparing surgical techniques are as follows: marginal resection, wide resection, and en-bloc (compartmental) resection [1, 2, 6]. Different safe surgical margins are defined in the literature: 3–4-cm margins for bone tumors and 1–3-cm margins for soft tissue sarcomas are accepted as safe [4, 7, 8]. For intramuscular locations, 1 cm of muscle is generally considered an appropriate wide margin [9].

Large bone and soft tissue sarcomas of the lower and upper extremities may sometimes be in close contact with neurovascular structures. In such cases, whether en bloc resection and vascular reconstruction to reach wider surgical margins or planned marginal resection with the help of adjuvant therapies should be preferred is controversial. Vascular infiltration by sarcomas necessitating vessel resection is suspected preoperatively if more than 50% of the circumference of the vessel is encased by the tumor. If the tumor is found abutting the neurovascular bundle, excision of the tumor along with the peri-adventitial tissues is sufficient [10, 11].

Non‐muscular margins, which depend on the tumoral side considered to be adequate (i.e., tumor rarely invades), include fascia, perineurium, periosteum, and vascular adventitia [9]. Wide resection can be achieved in most of the malign tumors associated with main vascular structures, and in some selected patient groups, en bloc resection is preferred in limb- sparing surgery [12, 13]. Arterial or arteriovenous reconstruction (autologous or synthetic graft) is made to maintain blood flow after en bloc resection [12–14]. Soft tissue sarcomas (STS) surrounded by a plane of normal tissue can be dissected from major blood vessels. By longitudinally splitting the adventitia opposite the tumor, a rim of normal tissue is preserved in the vessel-tumor interface [15].

The aim of this study was to determine surgical and oncological outcomes of marginal and wide resection extremity sarcomas that are associated with major vascular structures in the extremities.

Methods

In this study, consecutive patients in whom bone and soft tissue sarcomas adjacent to major arterial or venous blood vessels in the lower and upper extremity, were treated by limb-sparing surgery between May 2010 and September 2015 were described.

A collected database of 54 patients treated by the same orthopedic and vascular surgeon for a primary or locally recurrent bone and soft tissue sarcoma of the extremities was reviewed retrospectively.

Eligible subjects for this study were patients diagnosed with upper and lower extremity bone and soft tissue sarcomas that encased a maximum of 50% of the circumference of a major vascular structure, requiring limb-sparing resection.

The tumors requiring segmental resection of vessels with vascular reconstructions and tumors which did not show any close contact with vascular structures were excluded.

Feydy et al. suggest that MR imaging classification for these types of patients is defined as follows: 0, fat between tumor and vessel; 1, no fat; 2, partial encasement (< 50%); and 3, complete encasement [16]. According to tumor invasion to the neurovascular bundle, all cases were classified as “partial encasement” (types 1 and 2).

Except for two cases, all sarcomas were stage IIb tumors according to Enneking classification (stage III) [17].

All patients were informed about the types of treatment and the corresponding surgical procedure. Patients were treated according to the ethical standards of the Declaration of Helsinki and were invited to read, understand, and sign the written informed consent form. With a multidisciplinary approach, all operations were planned after preoperative approval by a council, including an orthopedist, a cardiovascular surgeon, a radiotherapist, a medical oncologist, a radiologist, and a pathologist. It was decided with the same multidisciplinary team whether or not neoadjuvant/adjuvant treatment would be given after biopsy and postoperative final pathology results. The surgical method of epineural and sub- adventitial fine dissection with marginal resection was applied where the neurovascular bundle was adjacent to the tumor. For all the regions unrelated to neurovascular structures, wide resection with a minimum of 3-cm safe margins for bones and soft tissues was performed (Fig. 1).

Fig. 1.

Right cruris liposarcoma: A X-ray images, B CT-angiography images, C intraoperative photograph, D macroscopic view of the mass

Besides MR images, all patients were assessed with Doppler ultrasonography (USG) to determine the relationship of tumor vessels and vascular pathologies, such as thrombosis and stenosis. According to the relationship of vascular structures with tumors, cases were classified as follows: type I (both arterial and venous involvement), type II (arterial involvement), and type III (venous involvement) [12]. According to vascular involvement, all cases were classified as type I.

Prophylactic antibiotic therapy (2 gr cephalosporin preoperative and 3gr/day postoperative) was administered until the suction drains were removed. During the perioperative period, low fractionated heparin was given and maintained with low-molecular- weight heparin in the postoperative period.

Patients were evaluated statistically according to their age, sex, localization, histopathology, types of tumor (primary or recurrent), the intervention of preoperative embolization and vena cava filter, surgical margins, tumor grade (low or high grade), local recurrence, metastasis, amputation, postoperative complications (edema, infection, skin problems, neurologic deficit), nerve involvement, extremity reconstruction techniques, neoadjuvant chemotherapy and radiotherapy, adjuvant chemotherapy and radiotherapy, and survival.

Statistical Analysis

Data was coded and recorded in SPSS for Windows 17.0.0 version. Chi-square and Fisher’s exact tests were used to compare the distribution of categorical variables between groups. Students’ t-tests or Mann Whitney U tests were used to compare parametric and nonparametric variables in two independent groups, while one-way ANOVA or Kruskal Wallis tests were used to compare more than two independent groups. Survival analysis was performed with the Kaplan–Meier method, and the survival of subgroups was compared by the Log Rank (Mantel-Cox) method. In all tests, p < 0.05 was accepted as statistically significant.

Results

There were 28 males (51.9%) and 26 females (48.1%) with a mean age of 39.1 ± 20.6 (median: 36.5) years. Of the tumors, 38 (70.4%) were located at the proximal lower extremity, 5 (9.2%) were located at the distal lower extremity, and 11 (20.4%) were at the upper extremity. There were 42 (77.8%) primary and 12 (22,2%) recurrent tumors. According to the Musculoskeletal Tumor Society (MSTS) classification, 36 (66.7%) of them were high grade and 18 (333%) low-grade tumors.

Histopathology of the tumor specimens revealed the following: 7 (13.0%) angiosarcomas, 2 (3.7%) malignant giant cell tumors of bone, 4 (7.4%) Ewing sarcomas, 4 (7.4%) chondrosarcomas, 6 (11.1%) liposarcomas, 10 (18.5%) malignant mesenchymal tumors, 12 (22.2%) osteosarcomas, 3 (5.6%) pleomorphic sarcomas, and 6 (11.1%) synovial sarcomas. Of the tumors 33 (61.1%) were soft tissue, and 21 (38.9%) were bone tumors. Two patients had distant metastasis (lung) at admission.

To minimize the intraoperative bleeding, 12 (22.2%) tumors with major feeding arteries were embolized 24–36 h before surgery. In addition, 5 (9.3%) vena cava filters were administered to patients with a high pulmonary thromboembolic risk.

Five (9.3%) major nerve involvements were determined as intraoperative, and despite epineural dissection, postoperative neurological deficit occurred in 3 (5.6%) patients Fig. 1.

Neoadjuvant and adjuvant chemotherapy protocols were administered to 16 (29.6%) patients (Ewing and Osteosarcoma patients). Neoadjuvant radiotherapy was given to 4 (7.4%) patients who had big malignant mesenchymal tumors to reduce the tumor size, and adjuvant chemotherapy was used postoperatively because of the microscopic surgical margin positivities. A total of 19 (35.1%) patients with microscopic positive surgical margins, including these four patients, received postoperative radiotherapy.

Postoperative pathological assessment of the surgical margins was “negative” in 35 patients (66.9%) and “microscopic-positive” (< 1 mm) in 19 (35.1%) patients. All microscopic positivities were only visible on the dissection side of the neurovascular bundle. In the other parts of the specimen, the surgical margins of bone and soft tissues were reached at a minimum of 3 cm in width.

Endoprosthetic and allograft reconstructions were made to 22 (40.7%) and 3 (5.6%) patients, who had bone sarcomas and soft tissue sarcomas invading the bone, respectively.

Postoperative complications were as follows: superficial infection in 10 patients (18.5%), skin necrosis which required surgical revision in 2 patients (3.7%), and venous edema requiring compression therapy in 8 patients (14.8%). Deep venous thrombosis was not seen as a complication in any patient at follow-up.

At a median follow-up of 36 months (range: 12–72 months), 34 patients were alive without evidence of disease, and 20 died. Eleven (20.4%) patients had local recurrences at mean.

55.9 ± 43.4 weeks (median: 48, 6–134). Wide resection could be achieved in 3 (5.6%) of the recurrent tumors. Interestingly, none of the recurrences had a relationship with main vascular structures. In the follow-up period, lung metastasis was detected in 17 (31.5%) patients, including two patients with lung metastasis at the time of diagnosis. Six (11.1%) amputations were made because of recurrence, and one was made because of deep chronic periprosthetic infection. Vascular ischemia was not seen in any case.

When the cases with microscopic positive and negative surgical margins were compared, local recurrence, metastasis, amputation, and tumor type (soft tissue/bone) parameters showed no statistically significant difference (Table 1).

Table 1.

The histopathological evaluation of the tumors and comparison of patients with positive and negative surgical margins

	Surgical margin (-) n = 35 (%66.1)	Surgical margin ( +) n = 19 (%33.9)	p
Local recurrence	9 (%24.3)	2 (%11.8)	0.470
Distant metastasis	13 (%35.1)	4 (%23.5)	0.532
Amputation	5 (%13.5)	2 (%11.8)	1.000
Nerve involvement	2 (%5.4)	3 (%17.6)	0.311
Soft tissue tumor Bone tumor	22 (%59.5) 15 (%40.5)	11 (%64.7) 6 (%35.3)	0.713

When metastatic and non-metastatic patients were compared, it was shown that bone tumors metastasize more than soft tissue tumors in our cohort (p = 0.001). However, there was no difference between metastasis and amputation, histopathology, grade, nerve involvement, surgical margins, or local recurrences (Table 2).

Table 2.

Relationship between metastasis and tumor type, grade, tumor origin, nerve involvement, amputation, surgical margins, and local recurrence

	Metastasis (-) n = 37 (%68.5)	Metastasis ( +) n = 17 (%31.5)	p
Primary tumor Recurrent tumor	27 (%73.0) 10 (%27.0)	15 (%35.7) 2 (%11.8)	0.300
High grade Low grade	25 (%67.6) 12 (%32.4)	11 (%64.7) 6 (%35.3)	0.836
Soft tissue tumor Bone tumor	28 (%75.7) 9 (%24.3)	5 (%29.4) 12 (%70.6)	0.001*
Nerve involvement	3 (%8.1)	2 (%11.8)	0.645
Amputation	4 (%10.8)	3 (%17.6)	0.665
Surgical margin ( +)	15 (%40.5)	4 (%23.5)	0.224
Local recurrence	6 (%16.2)	5 (%29.4)	0.293

^*In this series, statistically significant metastases were seen in bone tumors compared to soft tissue tumors

The mean survival was 1460.6 ± 137.4 days, while the overall survival was 48.15%. According to the Kaplan–Meier analysis, 1-year mortality was 14.81%, 2-year mortality was 40.74%, 3-year mortality was 54.16%, 4-year mortality was 64.28%, 5-year mortality was 80%, and 6-year mortality was 87.5% (Graphic 1).

Graphic 1.

Cumulative survival rates

Discussion

In sarcoma surgery, post-resection margins are classified as intralesional, marginal, wide, and radical by Enneking [17]. With the development of limb-sparing surgery and adjuvant therapies, surgical margins were shortly referred to as “positive” and “negative.” If the surgeon and the pathologist see gross tumor residue at the margin of surgical resection, this is called a “positive” surgical margin, but it is called “microscopic positivity” if a tumor is detected at the surgical margins only after histopathological examination under microscopy.

The term “planned surgical margin positivity” is a type of resection performed with the support of adjuvant therapies in order to protect one or more critical structures (vessels, nerves) by epineural and sub-adventitial dissection [11]. In this manner, preoperative planning plays a key role in tumor surgery as in all areas of orthopedics. Gerrand et al. showed a significant difference in the local recurrence rates between preoperative planned microscopic positive surgical marginal resection with the support of neoadjuvant radiotherapy and unplanned surgical excisions with positive surgical margins [18].

Arterial replacement is mandatory in limb-sparing surgery of malignant tumors related to vascular structures, especially in total encasement cases. Reconstruction of the artery can be done with grafts (autologous, saphenous, femoral vein; prosthetic materials, Dacron, polytetrafluorethylen, etc.) [11, 19]. In this report, vascular involvement is described as primary vascular origin or infiltrating, encasing, or affecting the tumor. In all cases, vascular reconstruction was made in all vessels (vein, arteria, or both) that touch tumors regardless of whether the vessels are primarily resectable or not. When compared with non-reconstructed cases, complications (hematoma, deep venous thrombosis, wound infection, skin and soft tissue necrosis, or severe limb edema) are higher in the reconstruction group in these reports [12, 13, 20]. Ghert et al. found high complication rates in vascular-reconstructed cases (severe limb edema 87%, deep venous thrombosis 26%, amputation rate related with local recurrences, and local factors 16%). Ghert’s reports revealed that more soft tissue resection was made in the reconstructed vascular group and more defects were occurred. However, muscle flap coverage surgeries could not prevent complications [13]. A study about en bloc resection of tumors and comparison of arterial and arteriovenous reconstruction revealed that the surgical time required for vascular reconstruction was long, attributable to the need for a flap transfer to cover the large defect after wide resection. Also, a high rate of postoperative complications has been reported. In addition, two groups were similar in respect to functional scores and postoperative edema. However, local complications were high in both groups, and muscle flap coverage surgeries did not prevent complications [14]. Amputation rates were higher (16% versus 3%), and there was no significant difference in terms of local recurrences and metastasis, suggesting that vascular reconstruction can allow limb preservation only in appropriately selected cases.

Tumors rarely invade the fascia, perineurium, periosteum, and vascular adventitia, which are the extra muscular borders adjacent to the tumor, and these margins are considered safe surgical margins [9]. However, nowadays, as limb-sparing surgeries are preferred more than amputation, the use of vascular surgery techniques and surgical procedures in bone and soft tissue sarcomas associated with vascular structures is increasing [2]. Considering the information so far, “planned surgical margin positivity” may lead to acceptable clinical and oncological results in soft tissue sarcomas adjacent to neurovascular structures and bone sarcomas with soft tissue components. Nevertheless, the clinical results of this approach have not been adequately studied in the literature. Performing “marginal sub-adventitial primary dissection” instead of “en bloc resection and vessel reconstruction” in appropriate cases and locations may reduce the complication rates due to tissue defect and circulatory failure by sacrificing less soft tissue.

This study evaluated the surgical and oncological results of a technique applied by the orthopedic oncology and vascular surgery team. In this study, bone sarcomas and soft tissue sarcomas adjacent to soft tissue neurovascular structures were operated with the “planned surgical margin positivity approach.” There was no significant difference between cases with positive and negative surgical margins in terms of local recurrence, metastasis, and amputation. These results show that a solution can be found that will not increase the risk of local recurrence and that the morbidity associated with en bloc resection can be reduced with good preoperative planning and a multidisciplinary approach in an extremity sarcoma with a certain proximity to the neurovascular structures. Preoperative planning methods are becoming more common with the support of technology and are very helpful in cases that are thought to be treated with planned surgical margin positivity. As a recent example, Yun et al. developed a new preoperative planning method for sarcoma resections that can assist surgeons in creating a surgery plan with an accurate three-dimensional surgical margin beforehand by using a patient- specific virtual model through 3D software manipulation of DICOM images [21].

However, although surgical margins were negative in most patients in this study, long- term survival outcomes were worse than in the literature. Because our institute is a tertiary-care cancer center, the majority of cases are externally referred. Furthermore, most of our cases are in the advanced stage and have large tumor volumes (mean 20 cm, range 15–35 cm) and recurring diseases operated in other clinics. In large patients’ series of non-vascular origin soft tissue sarcomas, size and grade of the tumor were the most important factors affecting long-term outcomes [1, 22]. The review of 3442 patients showed that the most predicting factor of metastasis and tumor-related death is its grade [1]. Another study revealed that the tumor size and depth is an important predictive factor of metastasis-free (MFS) and overall survival [6]. The other factor that affects the survival rates in this report is the shorter follow-up time compared with other studies in the literature.

Another important issue in sarcomas is imaging techniques in preoperative planning. Magnetic resonance (MR) imaging is now regarded as the optimal investigation method in evaluating most primary musculoskeletal tumors [16, 23]. MR imaging had a sensitivity of 64%, a specificity of 95%, a positive predictive value of 88%, a negative predictive value of 83%, and an accuracy of 84% in the detection of vascular invasion on the basis of findings of partial or total encasement [16]. While most radiological reports insist on the relationship between tumor and arterial structures, venous involvement is underestimated. From the aspect of vascular surgery, we think that the optimal evaluation method is Doppler ultrasound (USG). Research made by vascular surgeons suggest that Doppler USG must be used with standard oncological orthopedic investigation methods in limb-sparing surgery of adjacent vascular tumors [3, 12]. With Doppler USG, the main arterial and venous structures adjacent to the tumor and pathologies of these structures can be evaluated. Information about partial or total tumoral encasement, tumoral compression, venous thrombosis (intraluminal or secondary to the compression), arterial and venous collaterals, and contralateral vena saphena magna caliber (for bypass surgery) can be obtained with this noninvasive method. The chance of preoperative planning of the vascular surgeon with an accompanying radiologist during Doppler USG is another advantage to make optimal surgical intervention intraoperatively. Investigating all arterial and venous collaterals, vascular ligation can be made safely by a vascular surgeon. The effective use of Doppler USG limits the use of angiography. Although all our patients were evaluated with contrast-enhanced MR, Doppler USG performed in the presence of a vascular surgeon improved our preoperative surgical planning. Angiography as an interventional diagnostic procedure was used by our team only when we planned embolization of a feeding artery and implantation of a vena cava filter. The effects of preoperative embolization and vena cava filter application on perioperative bleeding and pulmonary embolism were not compared because of the small number of interventions.

The role of radiotherapy in the planned surgical margin positivity approach is very important. Surgical margin positivity is associated with a high rate of poor results in the conventional approach. There are many reports suggesting that positive (inadequate) surgical margin is the most important risk factor that increases local recurrence in the surgical treatment of sarcomas (9, 25–27). Although postoperative radiotherapy has been used for positive marginal resections in some reports, this conclusion has not changed (28–31). In terms of the influence of preoperative radiotherapy on local recurrence, the results are slightly more confusing. In a series of 132 patients, Sadoski compared the negative and microscopically positive surgical margins of soft tissue tumor patients who received preoperative radiotherapy and reported 5-year local controls of 97% and 82%, respectively (31). The outcomes of Sadoski suggest that, although less than 1 mm, the negative surgical margin (microscopically positive) is crucial for local control and the increase in surgical margin width does not cause an extra increase in local control in soft tissue sarcomas. Similarly, Dagan et al. in the series of 317 soft tissue sarcoma patients showed that, in the presence of 50 Gy neoadjuvant radiotherapy, according to the Enneking classification of tumors resulting in marginal resection, does not change local recurrence and amputation-free survival (AFS) compared to wide or radical resection [32]. Therefore, it may be possible to achieve local control by reaching a negative surgical margin even in very narrow width, especially in tumors that are susceptible to vital structures such as vessels and nerves with preoperative radiotherapy intervention. Although four of our patients received neoadjuvant and 19 received adjuvant radiotherapy, local recurrence rates of our 19 patients who had “planned positive microscopic surgical margins” were not significantly higher than cases with negative surgical margins (p = 0.470). Also, recurrences of our 11 cases were not located near the vascular structures dissected in the initial surgery, and vascular surgery intervention was not required in their re-resection. This may be due to the good preparation of the field close to the neurovascular structures in preoperative and postoperative radiotherapy planning or the fact that adventitia, periosteum, and epineurium are really high-quality barriers.

This study had some limitations. It is known that bone sarcomas, and especially soft tissue sarcomas, are quite heterogeneous in themselves and there are many types and subtypes that require different clinical approaches and treatment protocols and differ from each other in prognosis. However, there is no definitive information on whether bone sarcomas and soft tissue sarcomas generally have a better or worse prognosis than the other. In this study, both bone sarcomas with soft tissue components and soft tissue sarcomas were included in order to obtain statistically significant data. There is a need for further multicenter, large-scale, long- term studies to make assessments in more homogeneous groups. Another potential limitation of this study is that it has a retrospective design and includes data from a single center. However, it should also be mentioned that an orthopedist and cardiovascular surgeon participated in all cases and all interventions were performed by the experienced team of authors of this study.

Conclusions

Surgery for en bloc resection with additional reconstructions in vascular-related sarcomas is an option with high morbidity and complication rates. Despite some limitations, this study shows that surgery can be performed in selected cases with the contribution of neoadjuvant radiotherapy, with planned surgical margin positivity obtained by sub-adventitial dissection, without causing deterioration in local control.

Author Contribution

AFK, İG, and RÖ made the study design. İG and RÖ made the data collection and interpretation. AFK, İG, RÖ, and BŞG made manuscript writing. AFK, İG, RÖ, and BŞG did the final approval of the manuscript. All authors read and approved the final manuscript.

Declarations

Competing Interests

The authors declare no competing interests.