To clarify the validity of positron emission tomography using fluoro-2-deoxyglucose (FDG-PET) for the preoperative evaluation of endometrial cancer, we analyzed the preoperative FDG-PET images of both primary and metastatic lesions of 30 patients with endometrial cancer, and compared them with computed tomography (CT) and/or magnetic resonance imaging (MRI) images and the results of postoperative pathologic findings. As to the primary lesions, FDG-PET could easily identify the cancer, and the sensitivity was 96.7%, which tended to be higher than that of 83.3% by CT/MRI. As to the evaluation of retroperitoneal lymph node metastasis, FDG-PET could detect none of five cases of lymph node metastatic lesions of up to 0.6 cm in diameter but had higher specificity (100%) compared with CT/MRI (85.7%). The sensitivity of FDG-PET for detection of extrauterine lesions excluding retroperitoneal lymph nodes was 83.3% and was superior to that of CT/MRI (66.7%), although there was no difference in the specificity between the modalities (100%). The diagnostic ability of FDG-PET was limited if used alone, but FDG-PET gave additional information especially with regard to the extrauterine lesions whose significance could not be determined on CT/MRI. However, we also found that FDG-PET could not identify any lymph node metastasis less than 1 cm in diameter; therefore, a negative finding of lymph node metastasis on FDG-PET should not be interpreted as a reason for omitting retroperitoneal lymph node dissection for the precise surgical staging of endometrial cancer.