Diagnostic accuracy of intraoperative frozen section at radical abdominal trachelectomy for early-stage cervical cancer

Intraoperative frozen section examination is crucial for confirming the oncological safety of radical abdominal trachelectomy (RAT) for early-stage cervical cancer. This study evaluated the diagnostic accuracy of frozen section during RAT at our institution. We retrospectively identified patients with International Federation of Gynecology and Obstetrics 2008 stage IA1-IB1 (tumor size ≤2 cm) cervical cancer treated between 2002 and 2021. In performing RAT, frozen section analysis was routinely performed on uterine surgical margins and grossly enlarged lymph nodes, and was confirmed to be negative. Medical records were reviewed to compare the frozen section diagnoses with the final pathology. Among the 326 patients initially planned to undergo RAT, 298 (91.4%) underwent RAT, while 28 (8.6%) were converted to radical hysterectomy. The histological types were squamous cell carcinoma in 251 (77.0%) patients and adenocarcinoma in 67 (20.6%). Of 361 frozen section for surgical margins, 4 false negatives were identified. Discrepancies were due to freezing artifacts, staining quality on frozen sections, and slight differences in cross-sections between frozen and paraffin-embedded sections. Among 446 intraoperative lymph-node biopsies, one false-negative was recorded. Sensitivities of frozen section examination were 93.5% (58/62) for surgical margins and 94.1% (16/17) for lymph nodes. Lymph-node metastases were identified in systematic lymphadenectomy specimens of 21/326 (6.4%) patients planned to undergo RAT, with 10/21 (47.6%) detected intraoperatively. Lymph-node metastases were found in 7/112 (6.3%) patients without lymph-node biopsy. Frozen section examination during RAT provides satisfactory diagnostic performance, although biopsy of grossly enlarged lymph nodes is an unreliable method.