Background: The accuracy of both the Thyroid Imaging Reporting and Data Systems (TIRADS) staging system and fine needle aspiration cytology (FNAC) for the diagnosis of malignant thyroid nodules remains controversial.
Objective: The aim of this study was to compare the utility of sonographic features using TIRADS criteria versus FNAC in the diagnosis of solitary and dominant thyroid nodules.
Methods: This cross-sectional study enrolled patients with solitary or dominant thyroid nodules of both sexes. Patient data were obtained from their medical records. Thyroid ultrasound characteristics, FNAC, and cell block slides were reviewed. The TIRADs approach and the Bethesda system were used to categorize thyroid lesions. The receiver operating characteristic curve was performed on all radiological and pathological findings.
Results: The study included 158 patients with solitary or dominant thyroid nodules. TIRADS was significantly associated with Bethesda diagnoses and the histopathological diagnosis of malignant thyroid nodules (p<0.001). At a cutoff of ≥3, the TIRADS showed significantly good discrimination between malignant and benign nodules (AUC=0.842, p<0.001). At a cutoff ≥II, Bethesda showed a significant fair power of diagnosis of malignant nodules (AUC=0.784, p<0.001). The overall accuracy of Bethesda was slightly higher compared to TIRADS. There was a significantly poor concordance between TIRADS and Bethesda classification systems (weighted kappa =0.186, 95% CI: 0.117 to 0.255, p<0.001).
Conclusion: In patients with solitary and dominant thyroid nodules, there is a poor diagnostic correlation between TIRADS and Bethesda classification systems. However, the overall accuracy of Bethesda was slightly higher than that of TIRADS.