The charts of 423 patients with localized extremity soft-tissue sarcoma treated at our institution over a 10-year period (1968-1978) were reviewed. Data were subjected to both univariate and multivariate analysis, with independent variables in the multivariate analysis appearing in capital letters. Factors associated with an increased risk of local recurrence included the following: AGE greater than 53; PRESENTATION WITH RECURRENT DISEASE; HIGH TUMOR GRADE; positive regional nodes; TREATMENT BY LIMB-SPARING SURGERY (LSS); invasion of vital structures (LSS only); INADEQUATE MARGINS; and biopsy with delayed definitive resection. Survival was adversely affected by the following: AGE greater than 53; PAINFUL MASS; PROXIMAL SITE; SIZE greater than or equal to 10 cm; HIGH TUMOR GRADE; POSITIVE NODES; invasion of vital structures; TREATMENT BY AMPUTATION; INADEQUATE MARGINS; and local recurrence after treatment at our institution. Significant variations in both local recurrence and survival according to histopathology were also observed, with EMBRYONAL RHABDOMYOSARCOMA, ANGIOSARCOMA, and MALIGNANT PERIPHERAL NERVE TUMORS emerging as independent predictors of local recurrence. Using the Cox models for local recurrence and survival, patients were stratified into high-, intermediate-, and low-risk categories based on the presence or absence of each variable. Risk factor analysis should be part of the overall evaluation of each patient with extremity sarcoma.