Utility of intraoperative frozen sections of thyroid tissue in the age of molecular testing

Objectives Our study aims to examine the correlation between preoperative ultrasound‐guided fine‐needle aspiration and intraoperative frozen section and examine the clinical benefit of frozen section in the context of the latest national guidelines on the management of differentiated thyroid cancer. Study Design A retrospective review of thyroid frozen section from 2012 to2017 at one institution. Setting Tertiary care centre. Participants/Main Outcome Patient demographics, fine‐needle aspiration results, molecular testing results, frozen section diagnosis (classified as benign, indeterminate, or malignant), final pathologic diagnosis, initial planned surgery, actual surgery performed, need for additional surgery and complications were recorded. Complications included hematoma formation, hypocalcaemia (requiring readmission, symptomatic, or >24‐hour stay post op) and recurrent or superior laryngeal nerve damage. Results 728 total patients had an intraoperative frozen section performed. A Thy 4/Bethesda V USGFNA diagnosis (n = 55) significantly correlated with a clinically important intraoperative frozen section (n = 17, P  < .01). Intraoperative management was changed by the frozen section 53 times (7.2%). Molecular testing was sent on 92 USGFNA specimens, 80 of which were deemed "suspicious." Of the 49 patients whose management was upstaged intraoperatively, 29 (59%) would not necessitate a completion thyroidectomy under the latest UK and ATA guidelines based on final pathology. Conclusion Intraoperative frozen sections rarely alter the pre‐surgical plan and indeed may result in expanded surgery that could have been avoided based on latest UK and US guidelines. Molecular testing of indeterminate fine‐needle aspiration results does not appear to predict meaningful intraoperative frozen section results.