Background: To assess the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of intraoperative gross examination (IGE) of uterine specimens in determining deep myometrial invasion and cervical invasion compared to final histology. Materials and Methods: The clinical, surgical and histological data of all FIGO stage I-II endometrial cancer (EC) patients who had primary surgery were reviewed. Results of the IGE for myometrial invasion and cervical invasion were compared to the final histology. The sensitivity, specificity, PPV, NPV, and accuracy of the IGE in determining deep myometrial invasion and cervical invasion were calculated. Association between clinico-pathological factors and discrepancy between IGE and final histology in the determination of myometrial invasion was also assessed. A p-value of <0.05 was considered significant. Results: From January 2007 to December 2012, 179 patients diagnosed with clinical stage I-II endometrial cancer underwent surgical staging. The sensitivity and specificity of IGE in detecting deep myometrial invasion were 42.4% and 90.0%, respectively, and the PPV and NPV were 67.6% and 76.1%. The overall accuracy of IGE was 74.3%. The sensitivity and specificity of IGE in identifying cervical invasion were 28.6% and 97.5%, respectively, while the PPV and NPV were 60.0% and 91.1%. The overall accuracy of IGE was 89.4%. Conclusions: The sensitivity of IGE for detecting deep myometrial invasion and cervical invasion in early-stage EC is too low to be used alone. Alternative methods including intraoperative frozen section analysis, preoperative three dimensional ultrasound, and preoperative magnetic resonance imaging should be strongly considered.