Thyroid Nodules: TI-RADS Criteria, Bethesda System, and Biopsy Decisions

An Incidental Finding in 68% of Adults

High-resolution ultrasound detects thyroid nodules in 19–68% of adults, with higher rates in women, the elderly, and those with iodine deficiency — yet the lifetime risk of dying from thyroid cancer in the United States is approximately 0.5% for women and 0.2% for men. The central clinical challenge is not identifying nodules (they are everywhere) but accurately determining which among the millions of incidentally found nodules harbor malignancy. Thyroid cancer represents about 3.8% of all new cancers in the US (American Cancer Society, 2024), but the vast majority are papillary thyroid carcinoma, which carries a 10-year survival above 98%. Over-investigation and over-treatment of low-risk nodules causes real harm — unnecessary surgery, lifelong thyroid hormone replacement, and procedural complications — while under-investigation misses the minority of cancers that genuinely threaten life.

The 95% Benign Baseline

Approximately 90–95% of all thyroid nodules are benign. Benign etiologies include colloid cysts (simple fluid-filled nodules), hyperplastic nodules in multinodular goiter, follicular adenomas, and thyroiditis-related nodules. Malignancy risk is higher in:

• History of head/neck radiation exposure in childhood (Chernobyl survivors: 2-fold increase per Gy exposure)
• Family history of thyroid cancer or familial polyposis syndrome (DICER1, RET/PTC)
• Male sex (higher proportion of detected nodules are malignant, despite lower overall prevalence)
• Nodule characteristics on ultrasound: taller-than-wide shape, irregular margins, microcalcifications, hypoechogenicity
• Age extremes (<20 or >70 years at presentation)

ACR TI-RADS: Standardized Ultrasound Risk Stratification

The American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS), published in 2017, assigns points based on five ultrasound feature categories. Points accumulate and the total determines TI-RADS level (TR1–5) and the size threshold at which FNA or follow-up is recommended.

A TR5 nodule (≥7 points) warrants FNA if ≥1 cm; TR4 (4–6 points) warrants FNA if ≥1.5 cm; TR3 warrants FNA if ≥2.5 cm. Pure cysts (TR1) need no follow-up. This system reduced unnecessary FNAs by an estimated 40% compared to prior guidelines while maintaining sensitivity for malignancy detection.

Bethesda System: Cytology to Risk Stratification

Fine needle aspiration (FNA) biopsy material is classified by the Bethesda System for Reporting Thyroid Cytopathology (BSRTC), now in its second edition (2017):

Molecular Testing: Afirma and Beyond

Bethesda III and IV cytology creates a diagnostic dilemma: malignancy risk is intermediate (10–40%), meaning many patients would undergo diagnostic surgery for what turns out to be benign disease. Molecular testing addresses this by providing additional genetic information from FNA material without reoperation.

Afirma Gene Sequencing Classifier (GSC), developed by Veracyte, analyzes 10,196 gene transcripts using RNA microarray technology. A "benign" result on Afirma GSC has a negative predictive value of 96% in Bethesda III and 91% in Bethesda IV nodules. A 2019 study in Thyroid showed 52% of Bethesda III/IV nodules classified by Afirma as benign avoided surgery. ThyroSeq v3 (University of Pittsburgh) takes a genomic sequencing approach, detecting point mutations, fusions, and copy number variants in 112 genes associated with thyroid cancer; negative predictive value ~97% for Bethesda III/IV.

• Afirma GSC: mRNA-based gene expression; rules out malignancy with high NPV
• ThyroSeq v3: DNA/RNA sequencing; rules in and rules out; identifies specific driver mutations for surgical planning
• RosettaGX Reveal: microRNA analysis; useful in limited tissue samples

This article is for informational purposes only. Consult a qualified healthcare professional.