Single-center validation study of the American College of Radiology Thyroid Imaging Reporting and Data System in a Turkish adult population

Abstract

Objectives: This study aims to evaluate diagnostic performance of the American College of Radiology Thyroid Imaging Reportingand Data System (ACR-TIRADS) in the Turkish adult population.Patients and Methods: A total of 422 thyroid nodules of 368 patients (103 males, 265 females; mean age 46.43 years;range, 18 to 81 years) diagnosed by fine needle aspiration biopsy (FNAB) or surgical pathology were retrospectively analyzed. Thedecisions for FNABs were mostly made by the experts of our hospital without using the ACR-TIRADS. The observers assessedfive different aspects of thyroid nodules: echogenicity, shape, composition, margin, and echogenic foci. Nodules with 0 points werecategorized as TR1, 2 points as TR2, 3 points as TR3, 4-6 points as TR4, and ≥7 points as TR5. The sensitivity, specificity, anddiagnostic accuracy of ACR and the proportion of benign nodules and malignant nodules to be biopsied according to the ACRTIRADS were calculated.Results: The ACR-TIRADS identified 47 of 54 malignant nodules and identified 264 of 368 benign nodules yielding 73.6% diagnosticaccuracy. Using the ACR-TIRADS would have spared 64.1% of benign nodules from FNAB. Of 54 malignant nodules, 23 were to bebiopsied, 30 to be followed, and one malignant nodule was not planned to be biopsied or followed according to the ACR-TIRADSrecommendations.Conclusion: Despite its robustness in sparing a substantial number of benign nodules from FNAB, the use of ACR-TIRADS wouldprevent 55.5% of malignant nodules from FNAB. We suggest that FNAB threshold levels of the ACR-TIRADS should be extensivelyreviewed by the Turkish radiologists, endocrinologists, general surgeons, and ear, nose and throat surgeons before integrating thesystem to their daily practice.