Importance: The role of intraoperative frozen section (iFS) in thyroid surgery remains controversial and continues to evolve. The utility of iFS in nodules >4cm has not been established.
Objective: To determine if the routine use of iFS in patients with nodules >4cm with non-malignant cytology undergoing a thyroid lobectomy results in cost savings.
Design: A decision-tree model of thyroid lobectomy with iFS was created. A retrospective cohort study was performed to derive parameters, which were also obtained from the literature. Costs were estimated from 2014 Medicare, Bureau of Labor Statistics, and the Nationwide Inpatient Sample data.
Setting: Single Academic Center
Participants: Consecutive sample of 48 patients with thyroid nodules >4cm and non-malignant cytology undergoing thyroid lobectomy in which iFS was performed between 2009-2015.
Intervention: Modeled Thyroid lobectomy with and without iFS
Main Outcome and Measure: Overall cost savings per case with iFS.
Results: In our cohort the overall malignancy rate was 25% and 33% of these malignancies were identified intraoperatively. The specificity and positive predictive value of iFS were 100%. The negative predictive value and sensitivity were 91% and 80%, respectively. When applying the malignancy rates obtained from our cohort, performing routine iFS was the less costly scenario, resulting in a savings of $486/case. When the rate of malignancy identified on iFS was adjusted, obtaining iFS remained the less costly scenario as long as the rate of malignancies identified on iFS exceeded 12%. If patients with follicular lesions on cytology were excluded, 50% of malignancies were identified intraoperatively, resulting in a savings of $768/case.
Conclusions and Relevance: In patients with nodules >4cm undergoing a diagnostic lobectomy, routine use of iFS would result in decreased health care utilization. Additional cost savings would be obtained if iFS was avoided in patients with follicular lesions.