Abstract
OBJECTIVE: Current Thyroid Imaging Reporting and Data Systems (TIRADS) exhibit considerable variability in size thresholds for fine-needle aspiration biopsy. This study harnesses the systematic variations among dissimilar TIRADS optimization strategies for biopsy selection. METHODS: The analysis focused on the discrepancies observed among the four widely utilized TIRADS systems: ACR-TIRADS, Kwak-TIRADS, C-TIRADS and EU-TIRADS. Subsequently, several methods derived from the combination of two TIRADS were constructed via serial testing. Last but not least, diagnostic performance was assessed through unnecessary biopsy rate (UBR), missed malignancy rate and the frequency of clinically significant missed diagnoses. RESULTS: A total of 699 nodules were included in the study. The accuracy for nodules consistently recommended for biopsy by the four TIRADS was merely 50.8%. Without elevating the risk of missed diagnoses, which could potentially influence prognosis as per the current literature, for eligible nodules recommended for biopsy by original TIRADS, incorporating another TIRADS in serial could further reduce the number of biopsies by 7.8-19.2%. CONCLUSIONS: Conspicuous disparities exist in biopsy guidelines among TIRADS systems, urging increased caution among healthcare providers, particularly when they are extensively applied in patient evaluations. As evidently demonstrated by our research findings, combining recommendations from two TIRADS systems could effectively and safely lessen UBRs. These findings also advocate for the integration of prognostic-impact assessment in developing novel biopsy optimization strategies.