BACKGROUND Local recurrence occurs in 10% to 20% of patients with extremity soft tissue sarcomas despite optimal treatment. The association of local recurrence with subsequent survival is controversial and conflicting. There is a need for a staging system to predict outcome in this subset of patients and also to plan optimal treatment, including adjuvant systemic therapy. METHODS Data collected from 110 patients with locally recurrent extremity soft tissue sarcomas were studied. The influence of clinical and pathologic factors on local recurrence, distant metastasis, and disease-specific survival were analyzed by univariate and multivariate techniques. RESULTS Of the 110 patients who presented with local recurrence, 92 had an isolated local recurrence and 18 had prior or concomitant distant metastases. The 5-year disease-specific survival for all patients was 63% and for those with isolated local recurrence, it was 69%. Histologic grade, malignant fibrous histiocytoma histology, pathologic margins, previous local recurrence, and prior radiotherapy were independent prognostic factors for subsequent local recurrence. Tumor size, histologic grade, and time to local recurrence were independent prognostic factors for distant metastasis and disease-specific survival. A prognostic index was calculated by assigning a score of 1 to 3 for each of the three independent prognostic factors for survival and added to give the prognostic index for each patient. As the prognostic index increased from 3 to 9, there was a progressive increase in the hazard ratios and a corresponding deterioration in survival. The patients were then categorized into three prognostic groups based on the hazard ratios for disease specific survival. The differences in the survival curves were highly statistically significant (P < .0001). CONCLUSIONS Tumor size, histologic grade, and time to local recurrence are the primary determinants of distant metastases and survival in locally recurrent extremity soft tissue sarcomas. The impact of local recurrence on survival varies considerably. The nature of the local recurrence, rather than its presence per se, is a more useful guide to prognosis. We propose a simple staging system based on size, grade, and time to recurrence that correlates extremely well with prognosis and may serve as a guide to make therapeutic decisions in patients with locally recurrent extremity soft tissue sarcomas.