Try a new search

Format these results:

Searched for:



Total Results:


Proportion of Malignancy and Evaluation of Sonographic Features of Thyroid Nodules Classified As Highly Suspicious Using ACR TI-RADS Criteria

Hussain, Najia; Goldstein, Michael B; Zakher, Mariam; Katz, Douglas S; Brandler, Tamar C; Islam, Shahidul; Rothberger, Gary D
OBJECTIVES/OBJECTIVE:The reported malignancy rate of highly suspicious thyroid nodules based on the ACR TI-RADS criteria (TI-RADS category 5 [TR5]) varies widely. The objective of our study was to determine the rate of malignancy of TR5 nodules at our institution. We also aimed to determine the predictive values of individual sonographic features, as well as the correlation of total points assigned to a nodule and rate of malignancy. METHODS:Our single-institution retrospective study evaluated 450 TR5 nodules that had cytology results available, in 399 patients over a 1-year period. Sonographic features and total TI-RADS points were determined by the interpreting radiologist. Statistical analyses included logistic regression models to find factors associated with increased odds of malignancy, and computing sensitivity, specificity, positive and negative predictive values of various individual sonographic features. RESULTS:Of the 450 nodules, 95 (21.1%, 95% exact confidence interval 17.4-25.2%) were malignant. Each additional TI-RADS point increased the odds of malignancy (adjusted odds ratio 1.35, 95% confidence interval 1.13-1.60, P < .001). "Very hypoechoic" was the sonographic feature with the highest specificity and positive predictive value for malignancy (95.5 and 44.8%, respectively), while "punctate echogenic foci" had the lowest positive predictive value (20.0%). CONCLUSIONS:The rate of malignancy of TR5 nodules at our institution was 21.1%, which is lower than other malignancy rates reported in the literature. The total number of points assigned on the basis of the TI-RADS criteria was positively associated with malignancy, which indicates that TR5 should be viewed as a spectrum of risk.
PMID: 36106704
ISSN: 1550-9613
CID: 5336322


Ibrahim, Samer; Seshan, Nandini; Berges, Phillip; Hussain, Najia; Hussain, Kashif; Sharma, Deepali
ISSN: 0012-3692
CID: 5323372