Wound healing after preoperative radiation for sarcoma of soft tissues.

Morbidity from wound healing was retrospectively analyzed in a series of 202 consecutive patients with tumors of the soft tissue of the extremities, torso and head and neck region who were treated with preoperative irradiation and conservative operation at the Massachusetts General Hospital between January 1971 and June 1989. A radiation boost dose was given to 143 patients (71 percent) postoperatively. The overall wound complication rate was 37 percent. One patient died because of necrotizing fasciitis. In 33 instances (16.5 percent), secondary operation was necessary, including six patients (3 percent) who required amputation. The wounds in the remaining 40 patients (20 percent) were treated without operation. Multivariate analyses of the data showed the factors that were significantly associated with wound morbidity: tumor in the lower extremity (p < 0.001), increasing age (p = 0.004) and postoperative boost with interstitial implant (p = 0.016). Accelerated fractionation (BID, two fractions per day) reached borderline statistical significance (p = 0.074). Two other factors showed association with wound morbidity by univariate analysis, but not in multivariate model: high pathologic grade (p = 0.02) and estimated volume of resected specimen > or = 200 milliliters (p = 0.065). Patient gender, intercurrent disease (diabetes or hypertension), obesity, maximal tumor size, primary versus recurrent tumor, duration of bed rest postoperatively, dose of postoperative boost radiation, the use of postoperative boost, the use of adjuvant chemotherapy and year of treatment did not show significant importance for wound morbidity. When the severe wound complications (defined as requiring secondary operation and including the patient who died because of necrotizing fasciitis) were considered, among all analyzed variables, only localization of tumor in the lower extremity as a single factor was significant (p < 0.001). Techniques for managing the wound are considered which are judged likely to contribute to a decrease of the incidence of wound healing delays.