Study design: Retrospective study. Objectives: To investigate prognostic factors of sacral chordomas and provide theoretical foundation for an improvement of continuous disease-free survival (CDFS). Methods: Thirty-six patients underwent initial operation for sacral chordoma between 1992 and 2007. Data regarding age, gender, tumor size, tumor location, and type of surgery, surgical margins, surrounding muscle invasion, radiation therapy, and recurrences were reviewed and analyzed statistically. Results: The average duration of follow-up was 74.4 months (range, 16-182 months). Sixteen patients developed local recurrences at the time of final follow-up. The average recurrence time was 30 months (range, 3-84 months). The 5-year and 10-year actuarial CDFS were 59.5 and 42%, respectively. CDFS was found statistically longer in patients whose tumor highest level at or below S3 compared with above S3 (P ¼ 0.047). Patients who had invasion into the surrounding muscle had a shorter CDFS than those without surrounding muscle invasion (P ¼ 0.014). Statistic analysis showed that the type of surgery was the most valuable indicator of CDFS (P ¼ 0.001). The local recurrence rate was statistically higher when an inadequate margin has been achieved (P ¼ 0.021). The Cox multivariate regression analysis showed that surrounding muscle invasion (P ¼ 0.024) was an independent adverse prognostic factor for CDFS, whereas incomplete excision (P ¼ 0.056) achieved borderline significance. Conclusions: The higher level of tumor involvement, invasion into the surrounding muscle, incomplete excision, and inadequate surgical margins are poor prognostic factors. Resecting the tumor completely with wide surgical margins may provide a better prognosis for these patients.