Related Experiment Video
Updated: Jun 15, 2026

Substernal Thyroid Biopsy Using Endobronchial Ultrasound-guided Transbronchial Needle Aspiration
Published on: November 10, 2014
The Association between Subcategorization, Cytomorphological Features, and Ultrasonographic Characteristics with
Ahmet Kursat Soyer1, Aysegul Aksoy Altinboga2, Gokcen Nailer Ertuna3
1Department of Endocrinology and Metabolism, Ankara Bilkent City Hospital, Ankara, Turkey, ahmetkursatsoyer@gmail.com.
Introduction:
We aimed to investigate malignancy rates in atypia of undetermined significance (AUS) subcategories and their association with cytomorphologic and ultrasonographic features.
Methods:
A total of 201 thyroid nodules with AUS cytology that underwent surgical resection were analyzed, including 169 AUS-nuclear (AUS-N) and 32 AUS-other (AUS-O) nodules. Cytomorphological and ultrasonographic features, along with the American College of Radiology Thyroid Imaging Reporting and Data System (ACR-TIRADS) and the European Thyroid Imaging Reporting and Data System (EU-TIRADS) classifications, were analyzed to assess the association between malignancy and AUS subcategories.
Results:
The overall risk of malignancy (ROM) for AUS nodules was 19.4%, with a significantly higher rate observed in the AUS-N subgroup compared to AUS-O (21.9% vs. 6.3%, p = 0.04). A significantly higher ROM was observed in nodules with irregular margin, taller-than-wide shape, microcalcification, hypoechogenicity, and solid composition (OR = 9.63, 5.81, 3.33, 2.14, and 2.07, respectively). A statistically significant difference in ROM was observed across ACR-TIRADS and EU-TIRADS categories within the AUS nodules (p < 0.001 for both) and the AUS-N group (p = 0.001 and <0.001). A marked increase in ROM was observed with nuclear overlapping, pseudoinclusions, and enlargement (OR: 9.16, 4.47 and 2.80, respectively), while oncocytic atypia was associated with a reduced risk (OR: 0.44). In multivariate analysis, nuclear overlapping, pseudoinclusions, and sonographic irregular margins remained as independent predictors of malignancy (OR = 6.97, 6.09, and 5.79, respectively).
Conclusion:
To our knowledge, this is the first study to demonstrate a significant association between ACR-TIRADS classification and malignancy risk in the AUS-N subcategory.

