OBJECTIVE To study the clinical significance of pelvic and para-aortic lymph node sampling in endometrial carcinoma. METHODS Data were analyzed from 311 patients who received surgical treatment in our hospital during the period from January 1995 to December 2002. Among the patients, 197 underwent lymph node sampling or lymphadenectomy. The patients were divided into 2 groups based on the nature of their lymph node dissection, i.e. a) The sampling group included 114 patients with an extrafascial hysterectomy or modified radical hysterectomy plus pelvic or paraaortic lymph node sampling of the abdominal aorta; b) The dissection group, included 83 patients with a radical or modified radical hysterectomy plus systemic pelvic lymph node clearance or paraaortic lymph node dissection of the abdominal aorta.RESULTS The median of the sampling sites for lymph node removal was 5 in the sampling group, and the median of the lymph nodes removed was 15 per case. Lymph node metastasis was found in 8 cases. In the dissection group, the median of the cases for lymph node removal was 8, and the median of the lymph nodes removed was 27 per case. Lymph node metastasis was found in 6 cases. The 5-year survival rates were 90.2% and 90.9% in the 2 groups, respectively.CONCLUSION Lymph node sampling of endometrial cancer is a good way of precisely finding lymphatic metastases, and is suitable for surgical staging without causing immoderate surgical treatment and without affecting the survival rate.