Back to Search Start Over

TI-RADS Diagnostic Performance: Which Algorithm is Superior and How Elastography and 4D Vascularity Improve the Malignancy Risk Assessment

Authors :
Andreea Borlea
Florin Borcan
Ioan Sporea
Cristina Adriana Dehelean
Romeo Negrea
Laura Cotoi
Dana Stoian
Source :
Diagnostics, Vol 10, Iss 4, p 180 (2020)
Publication Year :
2020
Publisher :
MDPI AG, 2020.

Abstract

Given the increased prevalence of thyroid nodules in the general population (~50%), the real challenge resides in correctly recognizing the suspicious ones. This study proposes to compare four important Thyroid Imaging and Reporting Data Systems (TI-RADS) and evaluate the contribution of elastography and 4D Color Doppler assessment of vascularity in estimating the risk of malignancy. In the study, 133 nodules with histopathological examination were included. Of these, 35 (26.31%) proved to be malignant. All nodules were classified using the four selected systems and our proposed improved score. The American College of Radiology (ACR) and EU TI-RADS had good sensitivity (94.28%, 97.14%) and NPV (93.33%, 95.83%), but fairly poor specificity (31.81%, 23.46%) and PPV (35.48%, 31.19%), with an accuracy of 42.8% and 45.8%, respectively. Horvath TI-RADS had better accuracy of 66.9% and somewhat improved specificity (62.24%), but poorer sensitivity (80%). Russ’ French TI-RADS includes elastography in the risk assessment strategy. This classification proved superior in all aspects (Se: 91.42%, Sp:82.65%, NPV:96.42%, PPV:65.30%, and Acc of 84.96%). The mean strain ratio (SR) value for malignant lesions was 5.56, while the mean SR value for benign ones was significantly lower, 2.54 (p < 0.05). It also correlated well with the response variable: histopathological result (p < 0.001). Although, adding 4D vascularity to the French score generated a similar calculated accuracy and from a statistical point of view, the parameter itself proved beneficial for predicting the malignancy risk (p < 0.001) and may add important knowledge in uncertain situations. Advanced ultrasound techniques definitely improved the risk estimation and should be used more extensively.

Details

Language :
English
ISSN :
20754418
Volume :
10
Issue :
4
Database :
Directory of Open Access Journals
Journal :
Diagnostics
Publication Type :
Academic Journal
Accession number :
edsdoj.452bcaa855b34d0bb00eaf02b243710a
Document Type :
article
Full Text :
https://doi.org/10.3390/diagnostics10040180