Comparison of thyroid nodule FNA rates recommended by ACR TI-RADS, Kwak TI-RADS and ATA guidelines

Abstract

ObjectivesTo compare the differences among the thyroid imaging reporting and data system (TI-RADS) proposed by American College of Radiology (ACR TI-RADS), TI-RADS proposed by Kwak (Kwak TI-RADS), and American Thyroid Association (ATA) guidelines in the specificity, sensitivity and the unnecessary FNA rate (the UFR, the false positive rate) of recommended fine needle aspiration (FNA), and to observe the changes of the UFR in the KwakTI-RADS and ATA guidelines with the recommended FNA nodule size threshold. MethodsThe specificities, sensitivities and UFRs of recommended FNA in the ACR TI-RADS, ATA guidelines and Kwak TI-RADS were calculated and compared. The nodule sizes for recommended FNA of ATA guidelines and Kwak TI-RADS were systematically varied to establish new FNA thresholds. The specificities, sensitivities and UFRs of recommended FNA under the new models were calculated and compared to those in the ACR TI-RADS. ResultsFor all thyroid nodules, the UFRs in the ACR TI-RADS, ATA guidelines and Kwak TI-RADS were 26.3%, 47.4% and 40.0%, respectively. The UFR in the ACR TI-RADS was lower than that in the others, and the specificity of recommended FNA in the ACR TI-RADS(73.7%) was higher than that in the others (all P < 0.001), but the sensitivity of recommended FNA in the ACR TI-RADS(89.3%) was lower than that in the others (all P < 0.001). When nodule sizes threshold of the recommended FNA for ATA guidelines Intermediate Suspicion, Low Suspicion, and Very Low Suspicion, and Kwak TI-RADS grade 4b and 4a were gradually increased, the UFRs gradually decreased. ConclusionsThe UFRs of FNA recommended by the Kwak TI-RADS and ATA guidelines were higher than that of the ACR TI-RADS, and were affected by the recommended FNA nodule size threshold.