Adequacy Criteria for Thyroid Fine-Needle Aspiration in the Era of the Bethesda Reporting System

Abstract

Purpose: There is a lack of consensus and data validating lower cell counts for sample adequacy of thyroid fine-needle aspiration (FNA). We investigated less stringent adequacy thresholds under the Bethesda System for Reporting Thyroid Cytopathology (TBSRTC) and evaluated malignancy risks for "nondiagnostic" nodules by ultrasound features. Materials and Methods: A total of 2459 nodules with initial FNA were included. We built 11 new adequacy criteria based on the number of cell groups and total follicular epithelial cells. Diagnostic performance of each criterion was compared with the original criterion using the general estimating equation. Nondiagnostic nodules under each criterion were categorized by the American College of Radiology Thyroid Imaging Reporting and Data System (ACR TIRADS), and malignancy rates of each ACR TIRADS category were compared. Results: Malignancy rates of nondiagnostic nodules were under 3.5% across all criteria, and showed no significant differences compared with the original. More than 40 cells, regardless of cell group, or more than three cell groups showed no significant difference in diagnostic accuracy and false negative rates compared with the original criterion. Malignancy rates of nondiagnostic nodules were above 28.6% for ACR TIRADS 5, and below 5% for ACR TIRADS 1 to 4, in all criteria. Conclusion: Less stringent thresholds for sample adequacy can show comparable diagnostic performances to the original criterion of the TBSRTC. Given their markedly higher malignancy rates, nondiagnostic ACR TIRADS 5 nodules may warrant more active management than lower category nodules.