Objective: The aim of this study was to evaluate selected clinical, histological and diagnostic factors as prognostics of time and location of endometrial cancer recurrence. We also attempted to design a more efficient follow-up protocols for patients with recurrent endometrial cancer., Methods: A retrospective review of 196 patients with recurrent endometrial cancer treated at the Department of Gynecologic Oncology, Maria Sklodowska-Curie Memorial Cancer Center, Warsaw, was done. Recurrence was analyzed by site, presence or absence of symptoms and methods of detection. We focused our attention on the impact of specific disease characteristics on pattern and risk of recurrence., Results: Vast majority of recurrences (88%) occurred within the first 3 years after the primary treatment, whereas 71% within the first two years after the completion of the treatment. More than half of the patients (55%) were symptomatic when relapse occurred, while no statistically significant differences were found in time to relapse between the symptomatic and the asymptomatic group. Disease-free time observed when vaginal relapse occurred was significantly shorter compared to distant sites (p = 0.049). Histological maturity of primary tumor seems to be a prognostic of disease-free time (p = 0.039). Also patients with histological type I and II seem to have different courses of the disease (p = 0.064). Among recurrent endometrial cancer cases, pelvic examination, abdominal or pelvic ultrasound and chest x-ray could detect 95% of relapses., Conclusions: A follow-up program in the first three years after the primary treatment of endometrial cancer is useful in detecting recurrent disease. There is no reason to use an intensive program of follow-up in patients with low risk primary disease. Pelvic examination, vaginal and abdominal ultrasound, chest x-ray are good and not expensive methods in relapse detection. There is no clinical justification for the routine use of the Pap smear in the follow-up of patients with endometrial cancer.