Many gynecologic oncologists use intraoperative frozen section (IFS) assessment of histologic grade and depth of myometrial invasion (DOI) as predictors of final grade and stage in women with endometrial cancer. This clinical decision method has never been critically examined.

We retrospectively reviewed charts of patients with a preoperative diagnosis of uterine endometrioid adenocarcinoma and an intraoperative frozen section assessment of histologic grade and depth of myometrial invasion. Intraoperative predictors combining intraoperative frozen section assessment of histologic grade and depth of myometrial invasion were established and compared with final grade and surgical stage. We then modeled the risks of pelvic and paraaortic lymph node metastases for each predictor.

There were 129 patients who met inclusion criteria. Thirty-six patients had the IAG1 predictor; 17 (47%) were stage IAG1 on final pathology. Ten patients had the IAG2 predictor; 3 (30%) were stage IAG2 on final pathology. Thirty-four patients had the IBG1 predictor; 18 (53%) were stage IBG1 on final pathology. Forty-nine patients had the IBG2 predictor; 34 (69%) were stage IBG2 on final pathology.

Our decision models predict that the IAG1 predictor has a 1% risk of paraaortic and a 2% risk of pelvic lymph node metastases. The IAG2 and IBG1 predictors have a 2% risk of paraaortic and a 4% risk of pelvic lymph node metastases. The IBG2 predictor has a 2% risk of paraaortic and a 6% risk of pelvic lymph node metastases.

The combination of intraoperative frozen section analysis for histologic grade and depth of myometrial invasion does not correlate well with final pathologic grade and stage. Data from our models suggest a significant risk of lymph node spread even for patients with seemingly low-risk disease. Until better markers of lymphatic spread exist, we recommend complete surgical staging of all patients with endometrial cancer.