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 percent) of the sixty-five patients, but histological evaluation confirmed the presence of tumor in only sixteen (59 percent). 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 percent) 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 percent) of the twenty-three patients who had residual sarcoma. Five (22 percent) 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 percent] 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.