Diagnostic Performance of Thyroid Nodule Risk Stratification Systems

Abstract

Abstract The purpose of this study was to compare the diagnostic performance of 4 different ultrasound-based risk scoring systems for thyroid nodules (TNs). This study consecutively included 256 patients (mean age: 43.98 ± 12.94 years, min–max: 18–89 years; 225 females, 31 males) with 266 TNs. Each nodule was evaluated and classified according to the American Thyroid Association (ATA), American College of Radiology (ACR), European Thyroid Association, and Korean Thyroid Imaging Reporting and Data System (ACR-TIRADS, EU-TIRADS, and K-TIRADS, respectively) before performing ultrasound-guided fine-needle aspiration biopsy. Pathological results were reported according to the Bethesda system. Outcomes of the 4 classification systems were compared with respect to Bethesda results. Twenty-eight (10.5%) nodules had malignant cytology results. Diagnostic performances of the scoring systems were comparable with similar area under the curve values according to the reference standards of category 5 of each scoring system. The sensitivity and specificity values of these guidelines were as follows: ACR-TIRADS, 60.7% and 95.4%; EU-TIRADS, 71.4% and 93.3%; ATA-2015, 71.4% and 93.3%; and K-TIRADS, 67.9% and 93.3%. The biopsy rate of malignant nodules was 57.1% for K-TIRADS and ATA, whereas this value was 46.4% for ACR and EU-TIRADS. ACR-TIRADS had the lowest unnecessary biopsy rate (141 of 238 benign nodules, 46%). The diagnostic performance of 4 risk stratification systems appears to be comparable, as shown by similar sensitivity, specificity, and area under the curve values. However, the ACR-TIRADS had slightly higher accuracy and necessitated fewer unnecessary biopsies for benign nodules.