Adjuvant therapy. Group 1 had had previous surgery; group 2, no previous surgery.
Patient status. Group 1 had had previous surgery; group 2, no previous surgery. CDF indicates continuously disease free; NED, no evidence of disease; DOC, death of other causes; and DOD, death of disease.
Flugstad DL, Wilke CP, McNutt MA, Welk RA, Hart MJ, McQuinn WC. Importance of Surgical Resection in the Successful Management of Soft Tissue Sarcoma. Arch Surg. 1999;134(8):856-862. doi:10.1001/archsurg.134.8.856
A systemic disease-free state necessitates a local disease-free state. This cannot be accomplished without a properly performed resection by an experienced surgical team. Successful local management of soft tissue sarcoma (STS) may lead to improved disease-free survival. An STS treatment protocol using wide local excision followed by radiation therapy is effective in achieving local tumor control and survival similar to that of multiple-modality regimens, but with lower morbidity.
Retrospective cohort review (August 1, 1987, to May 6, 1998).
Referral to a single musculoskeletal oncologic surgeon, with surgery performed at a tertiary care medical center in a large urban area.
Ninety patients with STS of the trunk or extremities.
Preoperative evaluation included surveillance computed tomographic scan of the chest, magnetic resonance imaging of primary site to assess tumor extent and to plan the surgical approach, and angiography if vascular bypass was proposed. Wide local excision of tumor was performed, with concomitant vascular bypass and/or complex plastic reconstruction as needed. Postoperative radiation therapy was given in most patients. Adjuvant chemotherapy was used selectively.
Main Outcome Measures
Morbidity, local recurrence rates, and survival.
Histologically negative margins were obtained in 89 (99%) of 90 patients; 86 (96%) remained free of local disease at follow-up. Five patients died of systemic metastatic disease.
Excellent local control obtained with aggressive, appropriate surgery followed by radiation therapy in most patients and chemotherapy in only selective high-risk patients leads to excellent survival, with low morbidity and good functional outcome.
SURVIVAL IS OF paramount importance in the management of soft tissue sarcoma (STS). Disease-free survival is not possible without a local disease-free state. The evolution of various modalities of oncologic treatment, especially chemotherapy, has significantly improved the survival in the management of many malignant neoplasms. Surgery for STS is the only modality that alone can produce a local disease-free state. Local or marginal excision alone leads to unacceptably high local failure rates of 70% to 90%; even radical limb-sparing operations have a 25% recurrence with surgery alone.1 The use of adjuvant radiation therapy has allowed for less radical surgery and thus more functional outcomes even for patients presenting with recurrences.2 Radiation therapy in the treatment of patients with more than small isolated microscopically positive margins or who have had the pseudocapsule violated during surgery is likely to be unsuccessful. The use of adjuvant chemotherapy remains controversial in STS except perhaps for high-risk patients, ie, those with high-grade histological findings and tumor masses greater than 10 cm. Chemotherapy as the sole modality of treatment in a patient with STS serves only a palliative role at best. It is our hypothesis that a systemic disease-free state first necessitates a local disease-free state. A local disease-free state requires an appropriate, aggressive operation, meticulously performed with adequate margins by an experienced oncologic surgeon who understands the biology of STS and is able, by that experience and knowledge, to obtain appropriate curative margins. In some cases, when the lesion is low grade, small, and perhaps superficial, local control can be accomplished with surgery alone. However, due to adjacent microsatellites of tumor, the close or marginal margins needed to maximize function or to preserve major nerves or other vital structures require the use of adjuvant radiation therapy. Different adjuvant protocols have been used, including preoperative intra-arterial local chemotherapy, preoperative and postoperative radiation therapy, and various chemotherapy regimens.1,3- 5 Radiation has become the mainstay of adjuvant therapy. Timing of perioperative radiation therapy remains a topic of controversy, but postoperative radiation therapy has been shown to lead to fewer wound complications, and this has been our experience.
Thus, we have developed a treatment protocol for patients with STS using wide local excision. We defined wide local excision as a complete excision of the tumor and most of its reactive zone with no intralesional pseudocapsule violation. Excision of bone, vessels, or nerves was performed only if necessitated by major tumor juxtaposition or encasement. Reconstruction with bone grafts, oncologic implants, vascular bypass grafts, muscle flaps, or skin grafts was required in some cases. If a major neurovascular structure was juxtaposed to the tumor pseudocapsule, a careful marginal dissection was performed along that narrow field, with a marginal margin in that region and wider margins being accomplished in the remainder of the resection to minimize any residual microscopic tumor colonies. Postoperative radiation therapy thus served to obliterate any retained microprojections or micrometastases of tumor away from the pseudocapsule in the operative field. Adjuvant chemotherapy was considered after local control was obtained in selective patients with high-risk histological findings or large tumors (>10 cm). In addition, careful local and systemic monitoring of disease status with immediate intervention using other aggressive modalities in the face of systemic recurrences, such as wedge resections for lung metastases and aggressive chemotherapy, was performed when indicated. With these methods, we have obtained very favorable local disease-free states, which in turn has led to favorable systemic disease-free states and thus survival.
Medical records of 103 consecutive patients who underwent surgery for STS by a single musculoskeletal oncologic surgeon (D.L.F.) from August 1, 1987, to May 6, 1998, were reviewed. Thirteen patients were unavailable for follow-up, leaving 90 patients for complete review. Of these, 34 (38%) presented having had previous surgery (group 1), whereas 56 (62%) presented without previous surgery (group 2). Extra-abdominal desmoid and fibromatosis were not considered STS and thus were not included in this series.
Patient preoperative characteristics reviewed include age, sex, lesion location, duration of symptoms, lesion size on examination, biopsy type and result, results of magnetic resonance imaging (MRI) including tumor association with critical structures, results of computed tomographic (CT) scan of chest, and chemotherapy, radiation, or surgery before presentation. Results of histological examination of biopsy specimens obtained before presentation were reviewed by a pathologist with expertise in STS (M.A.M.). Intraoperative variables reviewed were procedure, including degree of resection and need for vascular bypass grafts and/or plastic closure with muscle flaps; final pathological results and lesion size; whether negative margins were obtained; and any complications. Postoperative variables reviewed include use of adjuvant radiation or chemotherapy, any additional surgery, date and nature of disease recurrence, complications, and patient status at follow-up. Follow-up was obtained by medical record review and telephone interview when appropriate.
The review included 41 female and 49 male patients with a mean age of 50.1 years (range, 17-92 years). Because of our impression that previous surgery might imply a less favorable outcome, the patients were divided into the 2 groups. Four group 1 patients had undergone multiple attempts at resection. Properly performed needle core biopsies done before referral were not counted as previous surgery. Tumor location included forearm or hand (n=5), upper arm or shoulder (n=12), axilla (n=3), thorax (n=9), abdomen (n=1), groin crease (n=2), spermatic cord (n=2), buttock or hip (n=6), thigh (n=37), popliteal fossa (n=5), leg (n=7), and foot (n=1).
Lesion size ranged from not palpable in 10 patients (11%) secondary to previously contaminated excisions to 32×26 cm on initial presentation. Once the diagnosis of STS was confirmed by results of biopsy or review of outside histological findings, staging was completed with a CT scan of the chest. A preoperative bone scan provides little if any benefit, as metastases to bone are very uncommon in STS and invasion of bone is rare. Also, bone scan uptake poorly predicts bone invasion if present. We used MRI scans to guide all aspects of planned treatment, including biopsy, and operative planning for the surgeon.6 Ideally, MRI of the site is obtained before biopsy.7 The lesions were then staged according to the Enneking sarcoma staging system as advocated and adopted by the Musculoskeletal Tumor Society.8 The distribution of the histological subtypes were similar in both groups, with malignant fibrous histiocytoma and liposarcoma being the first and second most common.
Group 1 included patients with the unfavorable pathological report of an STS from an inadvertent excision of what was thought to be a benign lesion and patients with a tumor recurrence after previous surgery elsewhere. All patients in this group underwent surgical excision. Figure 1 demonstrates the adjuvant therapy received by both groups. Nineteen group 1 patients (56%) received postoperative radiation therapy alone after our surgery, whereas 6 (18%) received adjuvant chemotherapy with radiation therapy for high-risk lesions as previously defined. Forty-three group 2 patients (77%) received radiation therapy alone postoperatively, whereas 7 (12%) received chemotherapy and radiation therapy. Nine patients in group 1 and 6 in group 2 received no adjuvant treatment due to lack of outlined indications or lack of compliance by the patient (1 patient in group 1 and 4 patients in group 2). No patient in either group received chemotherapy alone as adjuvant therapy.
Table 1 demonstrates the primary (index) procedure performed. In group 1, lesions in 3 patients were deemed unresectable at presentation, and these patients underwent amputation as their index procedure. No amputations were performed as the index procedure in group 2 patients, implying that previous surgical procedures can produce a situation where limb salvage is not possible. Three patients (3%) required intraoperative vascular reconstruction with bypass grafts (1 patient in group 1 and 2 patients in group 2). A complex plastic reconstruction or closure was needed in 22 patients (24%) (7 in group 1 and 15 in group 2). Thus, there did not appear to be an increased need for vascular reconstruction or plastic intervention with previous surgery.
Table 2 demonstrates patient distribution by tumor grade and size based on results of final pathological examination. Size of margin around the pseudocapsule varied according to the size of the tumor and could vary substantially within the same specimen due to marginally juxtaposed critical structures. The intraoperative goal was to minimize the number of residual microscopic colonies, satellites, or microprojections of neoplasm left within the tumor bed by leaving a minimum of 1 cm of muscular or subcutaneous margin when we could, thus preserving function and decreasing the risk for local recurrence. Minimal margins bordering critical neurovascular structures were accepted. Encased structures were excised en bloc and reconstructed. Larger tumors tended to require closer marginal margins. Previous surgical procedures also required larger skin and muscle resections, thus affecting the functional outcome, when such resections obviously would not have been needed if the lesion had not been violated before presentation to us. Of the 90 resections, 89 (99%) had negative margins on final pathological report. No reexcision was performed in the patient with a microscopically focal positive margin; with adjuvant radiation therapy, she remained disease free at 2 years postoperatively.
There were no intraoperative or early postoperative deaths. Medical complications included 1 case of Clostridium difficile colitis, 1 cerebrovascular accident 3 weeks after surgery, and 1 episode of postoperative chest pain treated medically. Minor skin problems were common, whereas significant skin breakdown or ulceration during adjuvant radiation treatment necessitating skin grafting occurred in 5 patients (15%) in group 1 and in 5 (9%) in group 2. Wound dehiscence occurred in 1 patient (3%) in group 1 and 3 patients (5%) in group 2, necessitating muscle flaps, all with good results. A deep infection developed in 1 patient in group 2; this was treated with incision and drainage and free-flap coverage, with a good result.
At follow-up, 28 (82%) of the group 1 patients and 40 (71%) of the group 2 patients were continually disease free. Six group 2 patients underwent pulmonary wedge resections for lung metastases, 4 of whom are alive (follow-up of 2, 3, 4, and 6 years) with no evidence of disease. One patient is alive with recurrent pulmonary metastases, and another has died of the disease. This tends to support the effectiveness of aggressive management of metastatic disease whenever feasible. Local recurrence necessitating amputation for salvage of disease status occurred in 1 patient in group 1 and 2 patients in group 2 (3%). Two of these patients were alive with no evidence of disease at 2- and 3-year follow-up; the third died of metastatic disease. Figure 2 illustrates disease status at follow-up. Seventy-nine (88%) of the 90 patients were alive with no evidence of disease at an average follow-up of 4.4 years (range, 1-10 years). Thirty (88%) of the group 1 patients and 49 (88%) of the group 2 patients were alive without disease, showing that with appropriate and aggressive surgery in the face of previous significant violation of the primary tumor, survival can be preserved. Disease-free and overall survival were excellent in both groups, tending to support our hypothesis that local disease control positively affects survival. We had expected that local surgery before presentation would have an unfavorable effect on disease-free status and complication rate.9,10 Not measured by this study is the fact that greater amounts of what would have been healthy, unviolated tissue had to be removed in these contaminated cases. Frequently, incisions were unfavorably placed. Nevertheless, based on our results, there was no significant difference in continually disease-free survival, no evidence of disease, and overall survival between groups. Cosmesis and objective functional outcomes were not assessed, and perhaps would demonstrate differences between both groups in these variables.
Curative, yet functional, outcome is the goal in the overall management of STS. Surgical management is by far the most important factor in accomplishing a local disease-free state and maximizing function. Previous studies have shown margins to be a strong, independent factor in local control, distant disease prevention, and prognosis.4,11,12 Wide local excision in our treatment protocol meant a margin to include as much of the tumor's reactive zone as was technically possible, accepting marginal margins along narrow strips of the pseudocapsule if this meant preserving a major peripheral nerve. If this nerve was encased, however, it was excised. Excision of other major structures, such as bone and vessels, was performed if necessitated by large areas of juxtaposition or encasement, and if judged necessary by the morbidity of the bone or vascular reconstruction. Expendable encased nerves can be excised (eg, superficial nerves and even major nerves, such as the radial, peroneal, or femoral, could be thought of as expendable if the alternative is amputation). However, one has to debate the usefulness of a lower extremity in which the entire sciatic nerve or an upper extremity in which 2 to 3 major nerves are excised. Thus, certain neurovascular and/or bone encasements produce, in our opinion, a relative indication for primary amputation.
Due to the growth pattern of these tumors, the longitudinal extent should have a more generous margin than the transverse. Compartment resections should be performed only if the tumor involves the entire compartment; otherwise, functional status is severely impaired. This was required in only 1 patient in this series. Arterial bypass grafts were performed when the tumor encased a major artery. The vein was not reconstructed. It is important to include the entire length of any previous incision, biopsy tract, or drain tract with the definitive resection. This has implications for the planning of any biopsy or excision of any soft tissue mass.7,13,14 A transverse incision should never be used for a biopsy, and neurovascular structures should never be exposed during a biopsy.7,13 The biopsy should be performed by the surgeon who will excise the neoplasm to ensure its proper placement.13 Incisions should always be longitudinal on a limb, and drains, if used, should be brought out in line with and close to the incision.13 A previous tumor-contaminated resection or poorly placed biopsy greatly complicates the definitive resection, necessitating more tissue resection than would otherwise have been needed.
As we stated, the size of the margin can vary considerably from patient to patient and even within the same specimen. The size of the STS frequently dictates the feasibility of obtaining wide margins. A massive lesion means more marginal margins. However, within such a specimen, margins can vary from 1 to 2 mm to 2 to 3 cm. Pathological reports often mention the closest margin. This is misleading if only 1% of the surface area of the resection has a margin of 1 mm, and the rest has margins of 2 to 3 cm. Our goal in performing limb-sparing procedures is to minimize the number of residual microscopic colonies, satellites, or microprojections of neoplasm past the pseudocapsule and to preserve as much functional tissue as possible. Thus, we believe it is reasonable to accept a 1- to 2-mm margin to dissect free a major nerve as long as larger margins are obtained on the remainder of the specimen. Of critical importance, of course, is to avoid any violation of the tumor pseudocapsule, as spill of gross tumor cells into the tumor bed, even with adjuvant therapy, is likely to result in a local recurrence. A very tedious and cautious dissection is usually required to accomplish the desired resection. A 99% negative margin rate was achieved in our series, and the 1 patient with a positive margin remains disease free.
Our study did not investigate the effectiveness of radiation therapy, as every patient for whom it was believed to be indicated received or was offered this modality. We do not question the usefulness of adjuvant radiation therapy with the type of resections and margins obtained in our study. In fact, we must rely on it to sterilize the tumor bed of the microcolonies of residual tumor we know must be there in the margins we have described. One cannot and must not rely on radiation therapy to salvage major or multiple positive margins or violation of the pseudocapsule at the time of surgery. Thus, adjuvant radiation therapy is judged as essential in this patient group except in patients with small, low-grade lesions with large margins in locations that can be monitored easily. Radiation therapy was begun 4 to 6 weeks after surgery unless wound problems necessitated a delay. With this approach, wound healing, even in these large and difficult resections, is usually rapid and with fewer wound complications than after preoperative radiation therapy. Major wound problems developed in 10 (18%) of the group 2 patients with this approach, compared with a reported 17% to 37% rate after preoperative radiation therapy.15 All of these patients, although with delayed onset of radiation therapy, had a healed wound, a functional extremity, and no evidence of local recurrence at follow-up. In a large study comparing preoperative vs postoperative radiation therapy, Cheng et al16 reported that there was no difference in the success of local control or disease-free survival. Our study tends to support these findings, with a local recurrence rate of 3% compared with studies showing 3% to 8% and 18% to 33% local failure rates in patients with negative and positive margins, respectively, who received preoperative radiation therapy.5,17 Although other studies have shown that presentation with a local recurrence was a significant independent risk factor for further local failure, we did not observe that, probably due to the small numbers of local recurrences and the prompt recognition and treatment of the recurrence.9,10
Too few patients received chemotherapy, and our survival (79 of 90) was too high to demonstrate any evidence to support or refute its use in an adjuvant fashion. Patients with recognized metastatic disease underwent immediate evaluation by a medical oncologist; chemotherapy was used as the principle modality, with surgical resection of the foci if feasible. Adjuvant chemotherapy was discussed with all patients by the medical oncologist and recommended in those with high-risk histological findings (eg, high histological grade or tumor with aggressive biological behavior such as synovial sarcoma) or T3 lesions (>10 cm) and in younger patients (<60 years of age). In patients with low-grade lesions, chemotherapy was not recommended. Intermediate histological-grade tumors of less than 10 cm were individualized; the decision was made by the oncologist and the patient after being presented with the potential risks and benefits of adjuvant chemotherapy with their lesion.
In summary, our present management of a high-risk mass (>5 cm and subfascial) is to image the lesion with MRI. If it is not homogeneous fat, a tru-cut or small incisional biopsy is performed. If the biopsy results confirm an STS, staging is completed with a CT of the chest. We determine the expected need for a vascular graft or a major plastic reconstruction preoperatively and obtain appropriate consultations.
A wide local excision, as defined, is performed. Marginal margins along neurovascular structures are accepted. Large areas of bone exposure require resection of bone with reconstruction. Encasement of vessels, nerves, certain organs, or bone necessitate their excision with reconstruction. Primary amputation is offered in cases with multiple major nerve, vessel, and bone encasement, because although the resection may be feasible technically, the function and/or the durability of such an insensate limb would probably be less than that of a prosthesis. Once the wound is healed, radiation therapy is begun, usually 4 to 6 weeks postoperatively. Adjuvant chemotherapy is individualized as discussed and initiated after radiation therapy has been completed. Patients are observed closely with serial MRI scans of the site and CT scans of the chest. Early recognition of metastatic disease is important and, if approached aggressively with surgery and chemotherapy, can lead to long-term survival of the patient.
We believe that a consistent multidisciplinary approach used from the moment of patient presentation will have a major impact on patient outcome.4 Successful, properly performed surgery is the priority in STS management; without it, local control and thus a disease-free state cannot be obtained. An increased chance of a single-step, margin-negative resection is possible with careful preoperative planning and with performance by a surgical team experienced in the management of STS.
Presented at the 70th Annual Session of the Pacific Coast Surgical Association, San Jose de Cabo, Baja California Sur, February 14, 1999.
Reprints: Daniel L. Flugstad, MD, Musculoskeletal Oncology, 1145 Broadway, Seattle, WA 98122 (e-mail: email@example.com).
Peter Bostick, MD, Santa Monica, Calif: I am going to actually read the comments by Dr Morton. He states: Contrary to the authors' assertion that limb salvage for STS began during the last decade, we first proposed 23 years ago that amputation could be avoided in STSs by limb salvage using a multidisciplinary approach of chemotherapy, surgery, and radiation therapy in over 90% of patients. This was published in the Annals of Surgery in 1996, volume 184. Even though the authors did not acknowledge my earlier work, I must agree these are excellent results for a large series of STSs from a single institution.
The low surgical complication rate, high limb-salvage rate, and low incidence of local recurrence indicate a high level of surgical skill and very competent postoperative radiation therapy. In fact, the major comment that I would have is that the results appeared too good when compared to other series. One would expect only a 65% to 75% 5-year survival from this mixture of various stages of STSs. Yet the authors state that 89% of their patients are alive and disease free at an average of 4.4 years of follow-up. However, the authors have excluded from their analysis 13% of the total who were lost to follow-up. This is not acceptable statistical methodology, because it must be assumed that many of those patients who were lost to follow-up are dead of their disease. I would suggest that the paper would be improved by input from a statistician who could assist the authors with a more correct Kaplan-Meier type of analysis of disease free and overall survival according to the standard staging for STSs according to the AJCC staging system.
In closing, I would like to ask if the authors believe their results are superior to those of other series, and, if so, what aspect of their approach makes the difference?
Steven Parks, MD, Fresno, Calif: Normally when we stage STSs, we talk about size and tumor grade. You did have a slide that showed that. Normally you cure these people if [the tumors] are small and low grade by just excising. We don't really know which patients fell into which category and got which therapy. It would be a better paper if those things were itemized so that we know that the large tumors were the ones that got radiation, even though they got wide excision, and that the high-grade tumors were the ones that got radiation, even though they got wide excision. Wide excision of the small, low-grade tumor should cure them anyway.
Herman Vargas, MD, Torrance, Calif: Is your 89% disease-free survival because of your patient selection? Number 2, grade and size of the tumor are obviously some of the factors that allow us to define and select therapies for our patients and will be primary determinants of the survival. I think it would be important to hear what the correlation of the size of the tumor and the grade of the tumor is with the outcome.
Lawrence Wagman, MD, Duarte, Calif: Throughout the presentation, we were told that the patient had a good functional outcome, but I didn't see the documentation for that, and I think that is a key element in proving this point.
Frederick Eilber, MD, Los Angeles, Calif: If I had one of these tumors, I would go up to the Swedish Hospital. These are the best survival results I have ever heard. The survival rate, as the other discussants mentioned, is primarily determined by size and grade rather than local control. Surgery and radiation are local treatment methods, and there are lots of ways to treat these tumors locally. Almost any large center in the world can get 95% limb salvage with a low local recurrence rate. But death rate is a different problem, and that is size and grade. Just applying surgery and radiation is not going to change the outcome for the patients who have large, high-grade tumors. So if this was a study, it was a study that is only addressing local treatments. So my comment is, first of all, you have done very well in local control, and my question is, do you plan a study to improve survival?
Finally, these patients had extremity and trunk lesions, and if you take these tumors and put them in a different place (retroperitoneum, head and neck), it is a totally different problem with a much higher local failure. So if the study is going to address the disease, it should not ignore these anatomic sites.
William F. Heer, MD, San Francisco, Calif: I applaud the authors and their results. Donald Morton has once again achieved excellent results. I too agree that staging would be desirable and more appropriate. My question is, how do you manage the tumor that comes within your 1- or 2-cm margin of long bone or joint, and what special recommendations would you make to avoid some type of more crippling operation?
James E. Goodnight, Jr, MD, PhD, Sacramento, Calif: A histology that we cannot cure is neurofibrosarcoma or malignant schwannoma. The slide with the different histologies blew by me. If you would comment on that particular histology, I would appreciate it.
Dr Hart: I would like to note that this is the work of Dan Flugstad, an oncologic orthopedic surgeon at Swedish Medical Center. The importance and elegance of this information is Dr Flugstad's surgical efforts. I will return to that later. I would like to thank Dr Bostick for pinch-hitting for Dr Morton, and I would like to apologize to Dr Morton in absentia for failing to cite his early pioneering efforts and successes in this area. In terms of the lost-to-follow-up patients, yes, that is indeed a problem with this study. As Dr Wilke presented, 34 of these patients had previous surgery and were referred from elsewhere. The 56 patients who were the primary surgery patients also come from a widely disparate area. In the LA area, you can accumulate a large series just coming from southern California; in Seattle, if you accumulate a big series, they are from Montana, Wyoming, Alaska, every place you can think of, and acquiring the follow-up on those patients can be somewhat difficult. I concur with the comments that our statistical analysis is weak. Unfortunately, just in looking at this group of patients, it is a very heterogeneous group of patients, different ages, different tumor types, different sizes, different locations; thus, we have not given it rigorous statistical analysis. We don't think that our results are superior to other groups, but rather the results are equal to other groups with what I think is a very meticulous surgical technique followed by radiation therapy alone. Dr Parks and Dr Vargas asked questions concerning grade and size, and although it was indeed difficult to see on the slides, it will be in the printed paper; I can tell you that 33% of the total patient group had grade 2 tumors; 36% had grade 3 or high-grade tumors. Thirty-four percent had T2 or 5- to 10-cm tumors, and 33% had T3 or greater than 10-cm tumors. So this was indeed a high-risk group of patients. The patients not receiving postoperative radiation were those who had smaller, lower-grade tumors and/or were elderly or infirm patients in whom surgery alone was adequate for their treatment.
Dr Wagman questioned functional outcome; this was very much a consideration in the initial treatment in that patients who had, for instance, sciatic nerve involvement did not get en bloc resection of their sciatic nerve. They are better candidates for amputation rather than having an insensate limb. Likewise, tumors that traverse 2 compartments and would require a 2-compartment resection were better candidates for amputation, and, although there were no amputations performed in the primary group, in the population that we deemed group 1, that is, patients who had been operated on previously, the primary procedure for a number of those patients was amputation.
When you start asking about functional outcome and you ask an orthopedic surgeon about functional outcome, you start getting information about degrees of motion rather than can they actually walk. As best I can tell you, there was major limb impairment in approximately 7% of these patients postoperatively.
Dr Eilber, again this is a very heterogeneous group of patients. I think if we started breaking it down too much further into histological types and sizes, it would be difficult to get statistically valid information.
Dr Heer asked about close margins. Indeed, a major part of the preoperative planning in these patients is to determine how to achieve appropriate margins. Three and one-half percent of these patients had major vascular reconstruction with vein grafts because major arterial structures were encased by tumor, veins were sacrificed for venous encasement, bone grafts were utilized as needed; thus there was a major effort to restore function and at the same time do a cancer operation.
Dr Goodnight's question about schwannomas: yes, we did have schwannomas in this group, but at this moment I cannot break out the specific numbers.
Again, as a final remark, I think last year we all found Dr Heald's presentation at this meeting on total mesorectal excision interesting, innovative, and exciting as a surgical procedure, not because it is a bigger operation, but because it is a better operation that is based on pathology, anatomy, and function. I think what Dr Dan Flugstad has attempted to do in this group of patients is similar; the operation gives a better result in a disease process where local control is essential but preservation of function is equally of paramount importance.