Examining the Bethesda criteria risk stratification of thyroid nodules

The Bethesda criteria are proposed for appropriate stratification of malignancy risk in thyroid nodules, but controversy exists regarding their accuracy and reliability in decision making. Additionally, previous studies have suggested higher rates of both malignancy and false negative fine needle aspiration biopsy (FNA) associated with increasing nodule size. This study aims to determine the accuracy of ultrasound (US)-guided FNA using the current Bethesda criteria in surgical practice. We also aimed to investigate the relationship between nodule size and malignancy.

A retrospective analysis of US-guided FNAs by a single surgeon during a 4.5 year period. FNA results using Bethesda criteria were compared to final surgical pathology.

611 patients with thyroid nodules underwent US-guided FNA. FNA results in 375 subsequently excised thyroid nodules were recorded according to the Bethesda criteria: 192 (51%) benign, 65 (17%) atypia of unknown significance/follicular lesion of undetermined significance (AUS/FLUS), 42 (11%), suspicious for follicular neoplasm (SFN), 17 (5%) suspicious for malignancy (SM), 28 (8%) malignancy, and 31 (8%) non-diagnostic. Malignancy was confirmed by surgical pathology in 15%, 34%, 50%, 88%, 100%, and 39% of the above groups respectively. Sensitivity, specificity, and false-negative rate were 61%, 99%, and 15% respectively. No correlation existed between the size of nodules with indeterminate FNA results and malignancy rate (p=0.89), or size of nodules with non-diagnostic FNA and malignancy rate (p=0.50).

The current Bethesda risk stratification system underestimated malignancy rates in benign, indeterminate and non-diagnostic cytopathologic categories in our experience. There was no positive linear correlation between nodule size and malignancy rate in these cytopathologic categories.