The American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS) score as a proxy for training and proficiency in evaluating thyroid nodules

Background: The American College of Radiology Thyroid Imaging Reporting and Data System (TI-RADS) was introduced for standardised reporting of thyroid nodules on imaging, while establishing nodules that require fine needle aspiration (FNA). Reproducibility of TI-RADS depends on operator experience. This study explores how TI-RADS scoring and subsequent management vary depending on the operator’s level of training. Objectives: To evaluate the diagnostic impact of the TI-RADS score by analysing how each reader’s TI-RADS ratings translated into FNA biopsy, quantifying rates of potential over-diagnosis and under-estimation through correlation with cytology results in a single quaternary centre. Method: A radiology resident and a qualified radiologist assessed 135 patients from Inkosi Albert Luthuli Central Hospital Thyroid clinic, using the same ultrasound machine and standardised data sheet. Nodules that met biopsy criteria according to TI-RADS assessments underwent FNA. Significant discrepancies between trainee and specialist TI-RADS classifications that could alter biopsy recommendations also underwent FNA. Cytology served as the reference standard. Results: Pairwise t-tests, weighted Cohen’s kappa and ROC statistics demonstrated substantial agreement for TI-RADS categories (κ = 0.72), with slight agreement for descriptors such as echogenicity and composition (κ = 0.25 and 0.28, respectively). Diagnostic performance was adequate – sensitivity of 0.88 (resident), 0.90 (specialist) and NPV of > 98% for both. Specificity of specialist vs resident was similar (0.81 versus 0.80) with ROC areas of 0.90 and 0.89. Conclusion: The TI-RADS is a reliable tool regardless of the level of training; however, training, peer review and audits remain warranted to improve diagnostic accuracy relating to certain TI-RADS score components. Contribution: The TI-RADS is a reliable diagnostic tool, significantly relieving the healthcare sector’s financial burden from unnecessary invasive procedures.