-Original-

Ultrasonography-Based Classification and Reporting System for the Malignant Risk of Thyroid Nodules

Ensheng Xue^1,2, Meijuan Zheng¹, Sufang Zhang¹, Liping Huang¹, Qingfu Qian¹ and Yunlin Huang¹

¹Department of Ultrasound, Fujian Medical University Union Hospital, Fuzhou, China
²Department of Imaging, Technology and Engineering School of Fujian Medical University, Fuzhou, China

Objective: The aim of this study was to test the predictive value of a Thyroid Imaging Reporting and Data System (TI-RADS) for malignant thyroid nodules.
Methods: Ultrasonographic data was examined for 910 thyroid nodules with histopathologically confirmed diagnoses. Nodules were placed into incomplete (category 0) or complete final categories (1, 2, 3a, 3b, 3c, 4, or 5) based on the presence and number of ultrasonographic features of malignancy, and the predictive value for the malignancy of nodules in categories 2-4 was assessed.
Results: The overall rate of malignancy among thyroid nodules included in the study was 59.34%. The rate of malignancy gradually increased according to TI-RADS categories as follows: category 2, 5.4%; category 3 (a-c), 36% to 92%; and category 4, 99.0%. When nodules of category 2 were counted as benign, the reliability of the TI-RADS classification for determining the risk of malignancy was as follows; sensitivity, 98.15%; specificity, 47.84%; positive predictive value, 73.31%; negative predictive value, 94.65%; and odds ratio, 48.61.
Conclusions: The TI-RADS classification used in this study is relatively simple and provides a reliable measure of the risk of malignancy of thyroid nodules.

J Nippon Med Sch 2017; 84: 118-124

Keywords
thyroid nodules, ultrasonography, Thyroid Imaging Reporting and Data System

Correspondence to
Ensheng Xue, Department of Ultrasound, Fujian Medical University Union Hospital, No. 29 Xinquan Road Gulou District, Fuzhou 350001, China
enshengxuecn@163.com

Received, September 14, 2015
Accepted, April 14, 2016