RESUMO
ABSTRACT Objective Ultrasonography (US) is the best diagnostic tool for initial assessment of thyroid nodule. Recently, data reporting systems for thyroid lesions, such as the Thyroid Imaging Reporting and Data System (TI-RADS) and American Thyroid Association (ATA), which stratifies the risk for malignancy, have demonstrated good performance in differentiating malignant thyroid nodules. The purpose of this study is to determine the reliability of both data reporting systems in predicting thyroid malignancy in a tertiary care hospital. Materials and methods We evaluated 195 thyroid nodules using modified TI-RADS and ATA risk stratification. The results were compared to the cyto-pathology analysis. Histopathological results were available for 45 cases after surgery, which is considered the golden standard for diagnosis of thyroid cancer. Results When compared with cytological results, sensitivity, specificity, negative predictive value (NPV), and accuracy were 100, 61.1, 100, and 63%, respectively, for TI-RADS; and 100, 75, 100, and 76%, respectively, for ATA. When compared with histopathological results, sensitivity, specificity, NPV, and accuracy were 90, 51.4, 94.7, and 60% respectively, for TI-RADS; and 100, 60, 100, and 68%, respectively, for ATA. All patients with malignant nodules were classified in the categories 4 or 5 of TI-RADS and in the intermediate or high suspicion risk according to the ATA system. Conclusion Both TI-RADS and the ATA guidelines have high sensitivity and NPV for the diagnosis of thyroid carcinoma. These systems are feasible for clinical application, allowing to better select patients to undergo fine-needle aspiration biopsies.