To define a high-risk group for paraaortic (PA) lymph node metastasis among endometrial carcinoma patients. Materials and methods: Prognostic factors determining PA lymph node metastasis were defined. Using these factors in different combinations, 14 risk groups were formed. A patient with at least one of these factors was considered as high-risk for PA lymph node metastasis. Results: This study included 152 patients. Of these patients, 18 had tumors in the PA region. Lymphovascular space invasion (LVSI) and pelvic lymph node metastasis were independent prognostic factors for PA lymph node involvement. In the risk modeling system, pelvic lymph node metastasis was an important factor in predicting PA lymph node status, and in groups where this factor was included for risk modeling, PA lymph node involvement was significantly increased in high-risk patients. Best results were obtained with the risk group model (group 10) involving cell type, LVSI, serosal spread, adnexal metastasis, and pelvic lymph node as prognostic factors. In this group sensitivity was 94%, specificity was 53.7%, and negative and positive predictive values were 98.6% and 21.5%, respectively. According to this model, 52% of all patients were in the high-risk group. Conclusion: Group 10 seemed to include the guiding properties for a decision on paraaortic lymphadenectomy and it was possible to reduce unnecessary paraaortic lymphadenectomies by 50%.

To define a high-risk group for paraaortic (PA) lymph node metastasis among endometrial carcinoma patients. Materials and methods: Prognostic factors determining PA lymph node metastasis were defined. Using these factors in different combinations, 14 risk groups were formed. A patient with at least one of these factors was considered as high-risk for PA lymph node metastasis. Results: This study included 152 patients. Of these patients, 18 had tumors in the PA region. Lymphovascular space invasion (LVSI) and pelvic lymph node metastasis were independent prognostic factors for PA lymph node involvement. In the risk modeling system, pelvic lymph node metastasis was an important factor in predicting PA lymph node status, and in groups where this factor was included for risk modeling, PA lymph node involvement was significantly increased in high-risk patients. Best results were obtained with the risk group model (group 10) involving cell type, LVSI, serosal spread, adnexal metastasis, and pelvic lymph node as prognostic factors. In this group sensitivity was 94%, specificity was 53.7%, and negative and positive predictive values were 98.6% and 21.5%, respectively. According to this model, 52% of all patients were in the high-risk group. Conclusion: Group 10 seemed to include the guiding properties for a decision on paraaortic lymphadenectomy and it was possible to reduce unnecessary paraaortic lymphadenectomies by 50%.