Enhancing ACR TI-RADS Predictive Value: A Logistic Regression Approach With Size and Age to Personalized Fine Needle Aspiration Decision-Making
Year: 2026
Type: article
Study design: Observational Study
Abstract: Background: The American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS) stratifies thyroid nodules by malignancy risk and aims to reduce unnecessary fine-needle aspiration (FNA). We evaluated whether incorporating patient age and nodule size improves the positive predictive value for clinically relevant thyroid neoplasia. Methods: We retrospectively reviewed thyroid ultrasound examinations of patients age 18-75, first applying standard ACR TI-RADS criteria and then fitting logistic regression models using a composite predictor defined as TI-RADS category multiplied by nodule size in millimeters (TI-RADS × mm). Results: Of 446 eligible patients, 171 (38%) met standard ACR criteria for FNA. Fourteen (3.0%) had clinically relevant thyroid lesions (carcinoma or neoplasm) on surgical pathology. Standard ACR criteria for FNA define TI-RADS categories versus abrupt size thresholds. In this evaluation, logistic regression using TI-RADS × mm supported a continuous risk gradient. Age was not independently associated with malignancy risk but was associated with a higher frequency of benign nodules meeting biopsy criteria, yielding an inverse age-risk relationship among nodules meeting threshold criteria. We propose an exploratory model to consider a continuum of nodule size and age of the patient risk model could influence and individualized approach for FNA. Conclusions: TI-RADS × mm scores provide a continuous estimate of malignancy risk. While TI-RADS ≤2 is reassuring and TI-RADS 5 is strongly suspicious, the lower risk category of TI-RADS 3 may benefit from individualized, shared decision-making that considers patient age and nodule size, particularly in borderline cases.
Source: AACE Endocrinology and Diabetes
Language: English
Subfield: Endocrinology, Diabetes and Metabolism
Field: Medicine
Domain: Health Sciences
Open Access status: hybrid
OpenAlex ID: W7161153262