Can ACR TI-RADS predict the malignant risk of medullary thyroid cancer?

Abstract

Objectives

This study aimed to evaluate the diagnostic performance for medullary thyroid cancer (MTC) based on the 2017 Thyroid Imaging Reporting and Data System by the American College of Radiology (ACR TI-RADS) guideline, and the ability to recommend fine needle aspiration (FNA) for MTC.

Methods

Fifty-six MTCs were included, and 168 benign thyroid nodules (BTNs) and 168 papillary thyroid nodules (PTCs) were matched according to age. Ultrasound (US) features were reviewed according to ACR TI-RADS. US, clinical features and diagnostic performance of cytology of MTC, BTN and PTC were compared. Multivariate logistic regression analysis was performed to assess independent variables to predict MTC.

Results

Multivariate logistic regression showed that position, hypoechoic, AP/T ratio ≥ 0.9 and marked internal blood flow were independent predictors of MTC compared to BTN (P < 0.05) and nodule sizes, AP/T ratio < 1, smooth or ill-defined margin and marked internal blood flow were independent predictors of MTC compared to PTC (P < 0.05). The area under the receiver operating characteristic (ROC) curve (AUC) of MTC based on ACR TI-RADS was inferior to that of PTC (0.687 vs 0.823) (P < 0.001). The recommended rate of FNA for MTC and PTC was 55.4 and 88.7 % respectively. 8 of 14 MTCs with negative FNA results (Bethesda II) had abnormal calcitonin (Ctn) results.

Conclusions

Based on the ACR TI-RADS classification, the malignant risk features of MTC were intermediate between BTN and PTC. The diagnostic efficacy of MTC and FNA recommendation rate were inferior to PTC. Ctn examination would reduce the FNA missed diagnosis of MTC.