Please use this identifier to cite or link to this item:
Title: Reducing unnecessary thyroid fine needle aspirations using American College of Radiology's thyroid imaging reporting and data system: A 5-year retrospective audit.
Author: Hayes, Christopher
Shvarts, Yasmin
Sewgolam, R.
Nguyen, Tri
Ussher, Simon
Issue Date: 2022
Publication Title: Sonography
Volume: 9
Issue: 1
Start Page: 29
End Page: 32
Abstract: Background With the growing investigative imaging techniques available in medicine, increasing numbers of thyroid lesions are being identified, either symptomatically or as incidental findings. Ultrasonography remains a key first-line investigation into evaluating these lesions, and accurately stratifying whether further investigation or management is required. Methods This study retrospectively analysed a 5-year patient cohort (2013–2017) who underwent fine needle aspirate (FNA) investigation of thyroid lesions at a large regional public hospital in Victoria, Australia. We aimed to identify whether using the American College of Radiology's thyroid imaging reporting and data system (ACR TI-RADS) in grading the initial ultrasound (US) images would have resulted in a reduction in the number of patients undergoing invasive testing and if all malignant lesions would still have been captured. Results Application of TI-RADS criteria retrospectively to the audited population demonstrated a theoretical 56.16% (n = 164) reduction in FNAs with 0.00% of malignant nodules missed (n = 14). Sensitivity for referral to FNA testing based on ACR TI-RADS grading was high (100%) with a moderately low specificity (58.99%). Conclusion Further management of thyroid lesions guided by ACR TI-RADS grading on initial US assessment is recommended, due to significant potential reduction in FNAs performed without loss of sensitivity for malignant nodules.
Internal ID Number: 01889
Health Subject: ACR TI-RADS
Type: Journal Article
Appears in Collections:Research Output

Files in This Item:
There are no files associated with this item.

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.