
Diagnostic Performance of Thyroid Imaging Reporting and Data system (TIRADS) and Acoustic Radiation Force Impulse (ARFI) imaging in Detection of Malignant Thyroid Nodules: A Cross-sectional Study
TC01-TC06
Correspondence
Dr. Subathra Adithan,
Additional Professor, Department of Radiodiagnosis, Jawaharlal Institute of Postgraduate Medical Education and Research (JIPMER), Dhanvantari Nagar, Puducherry-605006, India.
E-mail: subathra26@gmail.com
Introduction: Nodular lesions of the thyroid gland are frequent findings on Ultrasonography (USG). The Thyroid Imaging Reporting and Data System (TI-RADS) scores, based on USG, followed by Fine Needle Aspiration Biopsy (FNAB), are traditionally employed to study these nodules. Different TI-RADS systems are available for the risk stratification of Thyroid Nodules (TN). Thyroid Elastography (TE) is a novel method for diagnosing nodules in recent times.
Aim: To determine the diagnostic accuracy of conventional USG TI-RADS scoring systems and TE using Acoustic Radiation Force Impulse (ARFI) in diagnosing malignant and benign TNs, taking FNAB as the gold standard.
Materials and Methods: This was a cross-sectional study conducted at Jawaharlal Institute of Postgraduate Medical Education and Research (JIPMER), Puducherry, India on a sample of 255 patients (with TNs >1 cm) who underwent thyroid USG (with colour Doppler) followed by TE using ARFI. Virtual Touch imaging (VTi) and VTq quantification (VTq) were performed during TE. TI-RADS scoring, as per the American College of Radiology (ACR) white paper and the Kwak system, along with Shear Wave Velocity (SWV), was recorded along with other qualitative parameters for all the nodules. All the nodules were subjected to either Fine Needle Aspiration Cytology (FNAC) or biopsy for final characterisation into benign or malignant. The Receiver Operating Characteristic (ROC) curve was plotted for the total score as per Kwak, the total number of points as per the ACR committee white paper, and mean SWV, with the best cut-off obtained for each. Sensitivity, specificity, Positive Predictive Value (PPV), Negative Predictive Value (NPV) and diagnostic accuracy were calculated as per these cut-offs.
Results: Out of the 255 nodules, 204 were benign (80%) and 51 (20%) were malignant, as per FNAC or biopsy. The mean age of the patients was 42.77±13.7 years; 49 (19.2%) were male patients and 206 (80.8%) were female patients. According to ACR TI-RADS, 19 benign and 45 malignant nodules were categorised as TI-RADS 5, while the Kwak TI-RADS system identified 22 malignant nodules as TI-RADS 5. A total of 194 (95.1%) benign nodules had a VTi grade ≤2, while 45 (88.2%) malignant nodules had a VTi grade of 3 or above. There was a significant difference (p-value <0.05) in the VTi grade and SWV of the nodules between groups. Based on the ROC curves, the best cut-off to differentiate benign and malignant nodules was 3.4 m/sec for mean SWV (sensitivity 88.2%, specificity 92.2%, PPV 73.77%, NPV 96.91%). The diagnostic accuracy of SWV was 0.91 (0.85-0.95), ACR TI-RADS was 0.90 (0.78-0.90), and Kwak TI-RADS was 0.95 (0.91-0.97).
Conclusion: ACR and Kwak TI-RADS scoring based on conventional thyroid USG, along with TE using ARFI, demonstrated good diagnostic accuracy in detecting malignant TNs. ARFI-based TE should be combined with conventional USG for better diagnostic performance and optimal management of nodules.