Purpose: The purpose of this study is to determine the usefulness of arterial embolization on sacral and pelvic giant cell tumor (GCT). Materials and Methods: We retrospectively reviewed the medical records of 9 patients who had undergone serial arterial embolization between December 1996 and May 2008. We analyzed the clinical outcomes and therapeutic responsiveness of arterial embolization on sacral and pelvic GCT. Results: Six of 9 cases showed progression of disease (PD) status, even if 5 cases showed PD status despite of additional treatments including surgery and radiation, implying that serial arterial embolization on sacral and pelvic GCT is not effective. Three of 9 cases showed stable disease (SD) or continuous disease free (CDF) status and we analyzed associated factors with these good responses for embolization by ${chi}^2$ test. The number of feeding vessels under six (p=0.048) and the number of collateral arterial supply under three (p=0.048) in the first angiogram showed significant relationships with good response for embolization, while remaining tumor staining by contrast after the first embolization and repeated embolization times were not significant. Conclusion: Although serial arterial embolization is not an effective modality on sacral and pelvic giant cell tumors, it may be a pilot modality under narrow indication of tumors with poor vascularity at first angiogram.