Abstract
Sixty-five patients who had been referred to our unit for additional management after an unplanned excision of a soft-tissue sarcoma of an extremity at another institution were studied retrospectively to determine the prevalence of residual tumor and to identify factors that predict which patients will have a tumor following such an excision.Unplanned excision was defined as excisional biopsy or unplanned resection of the lesion without benefit of preoperative imaging and without regard for the necessity to resect the lesion with a margin of normal tissue. In each patient, histological evaluation of the specimen removed at the unplanned excision had demonstrated positive resection margins, but postoperative physical examination on our unit revealed no gross evidence of residual tumor and no tumor was identified on cross-sectional imaging of the local site. Patients who had evidence of residual disease on physical examination or on imaging were thought to have definite evidence of sarcoma at the site of the operative wound and were therefore excluded from the study.After multidisciplinary consultation, all patients had a repeat resection at our cancer center. Extensive pathological sampling of the specimen from this second procedure was carried out, with sections obtained at mean intervals of 1.2 ± 0.7 centimeters. Nodules initially thought to indicate disease were identified grossly in twenty-seven (42 per cent) of the sixty-five patients, but histological evaluation confirmed the presence of tumor in only sixteen (59 per cent). Histological evidence of sarcoma was identified in seven additional patients in whom gross nodules were not apparent in the specimen. Thus, sarcoma was identified in a total of twenty-three (35 per cent) of the sixty-five patients.The mean duration of follow-up was forty-six months (range, twenty-four to eighty months; median, thirty-nine months). The margins of the second resection were positive in nine (39 per cent) of the twenty-three patients who had residual sarcoma. Five (22 per cent) of the twenty-three had a local recurrence. Four of the five patients who had a local recurrence had positive margins on repeat resection. This rate of local recurrence (five of twenty-three patients) was significantly higher than that in the remainder of our patients who had a soft-tissue sarcoma of an extremity (sixteen [7 per cent] of 227) (p = 0.03).There was no association between the detection of sarcoma at the second procedure and the initial size or grade of the tumor, the use of irradiation preoperatively, or the interval between the initial, unplanned excision and referral to our cancer center. These data indicate that it is not possible to predict which patients will have residual tumor at the site of the operative wound. Therefore, it is prudent to advise repeat excision for all patients who have had an unplanned excision of a soft-tissue sarcoma of an extremity. Unplanned excision complicates decision-making in the treatment of this disease and should be avoided.
The local treatment of a soft-tissue sarcoma in an extremity has progressed from radical operative intervention, including amputation, to the use of multidisciplinary treatment and a less extensive resection that emphasizes the preservation of function of the limb as well as control of the disease1,2,5,9. The availability of cross-sectional imaging with computerized tomography and magnetic resonance imaging has contributed to the development of better limb-sparing procedures. These imaging modalities provide the clinician with a more accurate assessment of the extent of the lesion and its relationship to the surrounding normal structures; such assessment is critical to the planning of limb salvage for patients who have a soft-tissue sarcoma.
A patient who has a soft-tissue mass of an extremity may be seen initially by a surgeon who has little familiarity with the principles of modern operative treatment of sarcoma. In this situation, the surgeon may proceed with an excisional biopsy or an unplanned resection of the lesion without benefit of preoperative imaging and without regard for the necessity to resect the sarcoma with a margin of normal tissue. Giuliano and Eilber characterized these procedures as "unplanned total excisions." In the current report, we use the term unplanned excision synonymously with the term unplanned resection, to include patients who have had an excisional biopsy or an unplanned resection without preoperative imaging and without regard for the necessity to resect the tumor with a margin of normal tissue. Frequently, both the surgeon and the patient are surprised when a diagnosis of sarcoma is made, and the patient may be referred to a cancer center for consideration of additional management at that time.
When the surgeon and radiation oncologist decide, in consultation, what type of additional treatment (if any) should be advised after an unplanned excision, they are handicapped by insufficient data. It is often difficult to assess accurately the initial pathological margins of the resection, and it is impossible to estimate the extent of tumor that remains at the site of an unplanned excision. When the residual tumor burden is small, it might be safe to manage the patient with irradiation alone. However, when there is a substantial burden of residual sarcoma in the wound, additional wide (if possible) resection of the previous operative site should be advised, possibly in conjunction with preoperative or postoperative irradiation.
Approximately one-quarter of new patients who are referred to our multidisciplinary soft-tissue sarcoma clinic have been sent for an opinion regarding additional management after an unplanned excision. Our protocol has been to advise most patients who have been managed with excision of a sarcoma and who probably have positive resection margins to have complete resection of the previous operative site, with or without adjuvant irradiation. After resection, we have carried out gross and histological evaluation of the second specimen to determine what proportion of patients have residual tumor and what factors might predict which patients will have a tumor at the operative site following an unplanned excision.
More than one-third of our patients have had residual tumor at the operative site, but it has been impossible to predict which patients would have this finding. Our observations indicate that resection (or repeat resection) at a musculoskeletal tumor center should be advised for patients who have had an unplanned excision of a soft-tissue sarcoma, and that the initial operation should be performed at a musculoskeletal tumor center.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
†University Musculoskeletal Oncology Unit, Suite 476 (S. N., A.D., J. L., J. W., and R. B.), and Department of Pathology (R. K.), Mount Sinai Hospital, 600 University Avenue, Toronto, Ontario M5G 1X5, Canada.
‡Department of Radiation Oncology, Princess Margaret Hospital, 600 Sherbourne Street, Toronto, Ontario M4X 1K9, Canada.
All patients who were entered into this study met six criteria: (1) they were first seen clinically on our unit between January 1986 and February 1993 after having had a recent excision of a soft-tissue sarcoma of an extremity, (2) the initial operative and pathology reports and the initial pathological material indicated positive resection margins at the time of the initial operation, (3) the initial physical examination on our unit revealed no evidence of a mass at the initial operative site, (4) neither computerized tomography (before 1989) nor magnetic resonance imaging of the initial operative site revealed any evidence of a mass, (5) a second resection was performed by one of us (R. B.), a surgical oncologist, and (6) histological assessment of the specimen obtained at the repeat resection was performed by another of us (R. K.), a musculoskeletal pathologist.
Several of these criteria for inclusion require further definition and explanation. Assessment of the margins of the first resection (done at the referring hospital) was carried out in a rigorous fashion. Personal discussions were frequently held between the surgical oncologist and the initial surgeon to clarify the operative report. There was similar consultation between the consulting and the initial pathologist. Whenever possible, histological slides of the material from the initial resection margins were reviewed by the surgeon and pathologist (if the orientation of the margins had been preserved after the unplanned resection), and in all instances the original material was reviewed to evaluate the histological diagnosis and grade. The final determination of a positive resection margin was based on previously reported criteria2.
We excluded patients who had a mass that was palpable on physical examination or that was evident on cross-sectional imaging. There is little question that patients who have had an extensive incisional biopsy or partial removal of a tumor have residual disease and need additional operative treatment. Our goal was to restrict the study to patients who had been referred after gross removal of the tumor with an excisional biopsy but who had a high risk of microscopic residual sarcoma.
Patients who had an unplanned excision and had negative histological margins also were not included in the current study. Additional management after unplanned excision was based on the extent of the margins that had been achieved. If it could be clearly documented that radical compartmental resection had been accomplished, observation was advised. If the initial resection margins were found to be free of disease histologically but the resection could not be characterized as radical, radiation was generally administered post-operatively. Since these patients were not treated operatively at our center, they are not included in this analysis.
Between 1986 and 1993, 292 patients who had a soft-tissue sarcoma of an extremity were managed on our unit; 137 of them had been referred after an initial operation that had been performed elsewhere. Of these 137 patients, sixty-eight were excluded from the study: sixty-four had a mass that was palpable on physical examination or evident on magnetic resonance imaging or computerized tomography, and insufficient data had been recorded for the remaining four. One of us (R. K.), a musculoskeletal pathologist, then reviewed the specimens and histological slides for the remaining sixty-nine patients. To characterize the adequacy of sampling of the specimen taken at the repeat resection, the longest dimension of the specimen was divided by the number of sections obtained. This number represented the mean distance between the samples obtained from the tumor. Four patients had relatively incomplete sampling of the resection specimen, and these patients were excluded. This left sixty-five patients for whom complete data were available.
Before the repeat resection, the cases of all sixty-five patients were discussed at a multidisciplinary conference attended by personnel from the radiology, pathology, medical, surgical, radiation oncology, and allied-health departments. These consultants decided, on the basis of the initial pathology and operative reports as well as the imaging studies performed after the unplanned excision, whether adequate treatment would necessitate irradiation as well as operative intervention (combined treatment). If they were certain that irradiation would be necessary because of the anatomical constraints to a wide, curative local resection, they next considered whether there would be a benefit from radiation therapy before the operation. Preoperative irradiation was recommended only if it was obvious that combined treatment would be needed and that at least one of three situations was applicable: (1) the location and extent of the initial procedure would make it technically difficult to provide optimum irradiation after the second excision; (2) dissection during the second operation would be along a major neurovascular bundle, with the possibility of leaving microscopic tumor on these critical structures; and (3) the surgeon expected that remote tissue flaps or skin grafts would be necessary during the second procedure. Preoperative irradiation consisted of a total of twenty-five fractions in five weeks, with two gray given daily.
When the patients had not received irradiation pre-operatively, the operative specimen was evaluated to determine whether there was any gross or histological evidence of sarcoma and, if so, whether it was close to the margins of the repeat resection. If there was residual disease within two centimeters of the margins, postoperative irradiation with sixty to sixty-six gray in two-gray fractions daily was generally recommended. The principles used in the planning and delivery of radiation therapy have been described in greater detail elsewhere9.
The primary objective of the current study was to determine, by histological examination, whether there was residual sarcoma in the specimen obtained at the repeat resection. To ensure that the specimen was processed satisfactorily, the surgeon always examined the specimen with the pathologist in the frozen-section suite of the operating room. After orientation of the specimen, the tissue was serially sectioned at approximately one-centimeter intervals and the number of sections was recorded. The previous operative site was identified, and the scar tissue was palpated and visually examined for evidence of nodules. Multiple blocks were prepared from the region of the previous resection and were evaluated by the pathologist.
The null hypothesis for this investigation was that there would be no variable (or variables) associated with the finding of residual sarcoma in the wound. For the purpose of this study, information with regard to the size of the initial sarcoma was obtained whenever possible from the pathology report of the referring institution. These data were available for forty-eight of the sixty-five patients. When this information was not available, we reviewed cross-sectional images of the initial lesion (available for only three patients) or, if there was no other data source (fourteen patients), we ultimately relied on the surgeon's and patient's best estimate of the size. Slides of material taken at the original operation were reviewed to determine the grade of the tumor (high or low) and the histological subtype.
The statistical methods included descriptive statistics and univariate logistic regression analysis (SAS-PC version 6.03; SAS Institute, Cary, North Carolina). The regression analysis was used to determine whether there was a significant relationship between the frequency of identification of sarcoma in the sample from the repeat resection and the initial grade of the sarcoma, the initial size of the tumor, the interval between the initial resection and the consultation by our multidisciplinary unit, and the preoperative use of irradiation.
The margins of the repeat resection were evaluated in the specimen obtained during that procedure, and the patients were followed to determine how many had a local recurrence. In order to compare these data with those for our remaining patients who had a soft-tissue sarcoma of an extremity, we subtracted the sixty-five patients in the study group from the total cohort of 292 patients, leaving 227 patients for comparison with the study group. The comparison group included 155 patients who had been managed only on our unit with needle or incisional biopsy before the resection and seventy-two patients who had had the initial operation elsewhere before being referred to our unit for definitive management. These seventy-two patients had had biopsies (ranging from needle biopsy to extensive attempts at excisional biopsy) before referral, but all had palpable disease (or disease evident on cross-sectional imaging) that indicated the extent of involvement by the tumor. The four patients who were excluded from the study because of inadequate sampling of the specimen obtained at the repeat resection were included in the comparison group.
The mean age of the thirty-four men and thirty-one women was fifty-four years. Twenty-five (38 per cent) of the sixty-five patients had a malignant fibrous histiocytoma; eighteen (28 per cent), a liposarcoma; eight (12 per cent), a leiomyosarcoma; seven (11 per cent), a malignant schwannoma; four (6 per cent), a malignant hemangiopericytoma; and one (2 per cent) each, an epithelioid sarcoma, a soft-tissue chondrosarcoma, and a primitive neuroectodermal tumor. Nine patients had a low-grade sarcoma and fifty-six, a high-grade sarcoma. The median interval between the initial operation and the definitive procedure was thirteen weeks. Twenty-nine patients received irradiation preoperatively and eleven received it postoperatively after the resection margins had been evaluated; the remaining twenty-five patients received no radiation therapy.
Samples from the specimen obtained at the repeat resection were taken at intervals of 1.2 ± 0.7 centimeters (mean and standard deviation). Gross nodules thought to be sarcoma were identified in the specimens from twenty-seven patients (42 per cent); however, histological evaluation confirmed the presence of disease in only sixteen. In the remaining eleven, only scar tissue was demonstrated. Residual sarcoma was identified on histological examination in seven additional patients, for a total rate of twenty-three (35 per cent) of the sixty-five patients. In nine of these twenty-three patients, the sarcoma was identified at the margins of the repeat resection. Additional resection was performed in three of these nine patients, and negative margins were finally achieved. In the remaining six patients, additional resection was thought to be inadvisable as it would have necessitated amputation or a major loss of function of the limb (for example, sacrifice of the sciatic nerve).
At the time of follow-up, after a mean of forty-six months (range, twenty-four to eighty months; median, thirty-nine months), five patients had a local recurrence and eleven had metastases. Four of the five local recurrences were in patients in whom the repeat resection had had positive margins, and all were in patients who had had detectable residual disease. Two patients who had a local recurrence were managed with an amputation. The rate of local recurrence in the twenty-three patients who had residual tumor (five [22 per cent] of twenty-three) was significantly higher than that in the comparison group (sixteen [7 per cent] of 227) (p = 0.03, Fisher exact test).
Logistic regression analysis revealed that the likelihood of finding sarcoma at the previous operative site was not associated with the size (p = 0.98) or grade (p = 0.38) of the sarcoma or with the interval between the initial resection and referral to our unit (p = 0.66) (continuous data). It was also not associated with preoperative use of irradiation (p = 0.16) (dichotomous data). This analysis failed to demonstrate any variable that could be used to predict which patients were most likely to have residual sarcoma at the operative site following an unplanned excision of a soft-tissue sarcoma of an extremity.
In the current study, at least one-third of the patients who had no detectable tumor on physical examination or cross-sectional imaging after an unplanned excision of a soft-tissue sarcoma of an extremity had residual sarcoma on histological examination of the specimen obtained at the repeat resection. All of our patients who had positive margins following an unplanned excision performed at another institution had a second resection, ensuring limited selection bias. Possible sources of error that might have resulted in underestimation of the prevalence of tumor include the adequacy of sampling of the specimen obtained at the repeat resection and the limitations of histological identification of sarcoma in operative scars, especially after preoperative irradiation.
The adequacy of sampling of the specimens was evaluated by determining the number of sites that were evaluated within the specimen. Samples from the specimen were taken at intervals averaging 1.2 centimeters, and the sampling was most extensive in the region of the previous operative scar. Thus, although it is possible that the proportion of patients who had residual disease was underestimated because of limitations in the evaluation of the specimens, the sampling was extensive. The analysis of the gross specimen immediately after the resection by both the surgeon and the musculoskeletal pathologist also helped to ensure that complete information regarding the previous operative treatment and irradiation was available4.
Even if sampling is adequate, histological identification of sarcoma within an operative wound is not always straightforward. Proliferation of fibrovascular repair tissue in a scar can resemble the cellularity found in a mesenchymal malignant lesion. This difficulty can be increased when the site has been irradiated before the resection. We and other authors have found that sarcoma treated with irradiation is often replaced by reparative fibroblasts, which may demonstrate atypia6. Cellular atypia in fibroblasts following irradiation might result in the overestimation of residual sarcoma. At present, there is no technique, other than the assessment of the specimen by an experienced musculoskeletal pathologist, that would be more effective for identifying microscopic deposits of residual sarcoma at a post-operative site. We are therefore reasonably certain that our estimate is accurate within the limits of current techniques.
Considering the frequency with which soft-tissue sarcomas are initially treated with excisional biopsy or an unplanned local resection, the relatively poor documentation in the literature of the likelihood of finding residual sarcoma at the site of the operative wound is surprising. Giuliano and Eilber reported on ninety patients who had been referred after an excisional biopsy; all had been managed with Adriamycin (doxorubicin) and radiation therapy before the repeat resection. Despite this intensive preoperative treatment, the authors found gross evidence of residual sarcoma in about one-half of the patients. The discrepancy between their findings and our lower estimates is probably due to our exclusion of patients who had disease that was detectable on physical examination or cross-sectional imaging. Of the 137 patients who were referred after an initial procedure done elsewhere, sixty-eight were excluded for that reason. Some of these patients may well have been included in the study of Giuliano and Eilber.
Peabody et al. found that seventy-four (43 per cent) of 172 consecutive patients referred to the same surgical oncologist between 1975 and 1990 had had a marginal excision before referral. Their data indicated that a subcutaneous tumor (originating superficial to the fascia of the limb) was more likely to be associated with a previous excision than a deep lesion (originating deep to the fascia of the limb). These authors found evidence of residual sarcoma in about 50 per cent of the patients who had been referred after a marginal resection. It is difficult to compare the findings of Peabody et al. with those in the current investigation because of differences between the two populations of patients. Only sixty-five (22 per cent) of our 292 patients were referred after marginal excision. This probably can be credited to education about the treatment of sarcoma that had changed the treatment practices of the referring surgeons by the time that our study began (in 1986). Peabody et al. did not state whether cross-sectional imaging was used to identify residual gross disease before repeat excision. As mentioned previously, evidence of residual gross disease was a criterion for exclusion in the current study and probably accounts for the somewhat lower prevalence of residual disease in our patients (35 compared with approximately 50 per cent).
The ability to detect residual disease with cross-sectional imaging methods was evaluated by Hudson et al. These authors reported on twenty-one patients who had been referred after an excisional biopsy of a sarcoma. Of thirteen patients who had no palpable tumor, eleven had no evidence of disease on computerized tomography; however, a tumor was detected in seven patients on reoperation. The remaining eight patients had a palpable lesion; it was shown to be tumor on reoperation in seven, but computerized tomography had failed to detect the residual palpable lesion in two patients and had underestimated the extent of the lesion in two others.
On the basis of the estimate that at least one-third of patients who have had an unplanned excision of a soft-tissue sarcoma have recognizable disease in the wound despite negative findings on physical examination and magnetic resonance imaging, it is disappointing that there was no significant association between the factors that were evaluated (the size and grade of the tumor, the interval between operations, and preoperative irradiation) and the presence of disease. We had hoped that the results of this study would permit us to stratify patients who are referred after an unplanned resection with regard to the risk of residual disease, in order to advise them better.
The high prevalence of positive margins on analysis of the specimen obtained at the repeat resection is of concern. Positive margins were identified in nine of our twenty-three patients who were recognized to have residual sarcoma. This rate is much higher than the usual rate of positive margins (less than 10 per cent) at our institution9. These poor oncological results are due to the difficulty of resecting a pre-existing operative field that is contaminated by sarcoma but lacks a central mass that can provide visual and tactile clues as to the extent of the disease. This situation is particularly difficult when critical neurovascular structures have been exposed during the initial resection of the gross tumor. Previous dissection of these structures eliminates the anatomical planes on which the surgeon relies when performing a limb-salvage procedure and necessitates the dissection of the nerve or vessel, or both, out of dense scar tissue that may be full of microscopic sarcoma. Alternatively, the nerve or vessel would have to be sacrificed to remove scar tissue that may contain no recognizable disease. The difficulty of achieving adequate margins of resection after an unplanned excisional biopsy is reflected in the fact that our twenty-three patients who had residual sarcoma had a significantly higher risk (p = 0.03) of local recurrence compared with our total population of patients who had a soft-tissue sarcoma in an extremity.
In summary, our data provide no assistance to surgeons in deciding whether disease is present in a scar, but they do re-emphasize the difficulty inherent in advising patients who have had an unplanned resection of a soft-tissue sarcoma of an extremity. Our findings should serve to remind the surgeon that all patients who have a subfascial mass should have cross-sectional imaging of the local site as well as incisional or needle biopsy before resection is attempted. Unplanned excision of a deep soft-tissue sarcoma of an extremity complicates additional treatment and should be avoided.
Bell, R. S.; O'Sullivan, B.; Davis, A.; Langer, F.; Cummings, B.; and |and |Fornasier, V. L.: Functional outcome in patients treated with surgery and irradiation for soft tissue tumours. J. Surg. Oncol.,48: 224-231, 1991.48224
1991
[PubMed][CrossRef]
Bell, R. S.; O'Sullivan, B.; Liu, F. F.; Powell, J.; Langer, F.; Fornasier, V. L.; Cummings, B.; Miceli, P. N.; Hawkins, N.; Quirt, I.; and |and |Warr, D.: The surgical margin in soft-tissue sarcoma. J. Bone and Joint Surg.,71-A: 370-375, March 1989.71-A370
1989
Giuliano, A. E., and |and |Eilber, F. R.: The rationale for planned reoperation after unplanned total excision of soft-tissue sarcomas. J. Clin. Oncol.,3: 1344-1348, 1985.31344
1985
[PubMed]
Hajdu, S. I.; Shiu, M. H.; and |and |Brennan, M. F.: The role of the pathologist in the management of soft tissue sarcomas. World J. Surg.,12: 326-331, 1988.12326
1988
[PubMed][CrossRef]
Herbert, S. H.; Corn, B. W.; Solin, L. J.; Lanciano, R. M.; Schultz, D. J.; McKenna, W. G.; and |and |Coia, L. R.: Limb-preserving treatment for soft tissue sarcomas of the extremities. The significance of surgical margins. Cancer,72: 1230-1238, 1993.721230
1993
[PubMed][CrossRef]
Hew, L.; Kandel, R.; Davis, A.; O'Sullivan, B.; Catton, C.; and |and |Bell, R. S.: Histological necrosis in soft tissue sarcoma following preoperative irradiation. J. Surg. Oncol.,57: 111-114, 1994.57111
1994
[PubMed][CrossRef]
Hudson, T. M.; Schakel, M.; and |and |Springfield, D. S.: Limitations of computed tomography following excisional biopsy of soft tissue sarcomas. Skel. Radiol.,13: 49-54, 1985.1349
1985
[CrossRef]
Peabody, T. D.; Monson, D.; Montag, A.; Schell, M. J.; Finn, H.; and |and |Simon, M. A.: A comparison of the prognoses for deep and subcutaneous sarcomas of the extremities. J. Bone and Joint Surg.,76-A: 1167-1173, Aug. 1994.76-A1167
1994
Wilson, A. N.; Davis, A.; Bell, R. S.; O'Sullivan, B.; Catton, C.; Madadi, F.; Kandel, R.; and |and |Fornasier, V. L.: Local control of soft tissue sarcoma of the extremity: the experience of a multidisciplinary sarcoma group with definitive surgery and radiotherapy. European J. Cancer,30A: 746-751, 1994.30A746
1994
[CrossRef]