Comparative Diagnostic Accuracy of the American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS) and the European Thyroid Imaging and Reporting Data System (EU-TIRADS) for Thyroid Nodule Malignancy: A Systematic Review

Background: The American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS) and European Thyroid Imaging Reporting and Data System (EU-TIRADS) estimate thyroid nodule malignancy risk but use different category definitions and fine-needle aspiration thresholds.This diagnostic test accuracy (DTA) systematic review compared their performance, focusing on ACR TR5 and EU-TIRADS 5.Methods: PubMed, Europe PMC, Semantic Scholar, OpenAlex, Google Scholar, and Crossref were searched through October 12, 2026.Adult studies reporting diagnostic findings for either system were assessed.All three authors participated in selection and extraction; disagreements were resolved by discussion.Risk of bias was assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool and certainty with the Grading of Recommendations Assessment, Development, and Evaluation approach for DTA studies (GRADE-DTA).Findings were synthesized narratively.The review was not prospectively registered; no public protocol was available.Results: Twenty-five reports representing 24 cohorts were included.Two studies provided verified 2×2 data for the highest-risk-category comparison (2,786 nodules; 2,742 patients).In Wang et al., ACR versus EU-TIRADS sensitivity was 86.1% versus 86.3%, and specificity was 90.3% versus 85.8%.In Shen et al., sensitivity was 88.2% versus 93.3%, and specificity was 87.5% versus 80.9%.QUADAS-2 identified important risk-of-bias concerns in two studies.GRADE-DTA certainty was moderate for ACR sensitivity and specificity and EU-TIRADS specificity, and low for EU-TIRADS sensitivity.Pooling and formal paired comparison were not possible.Conclusion: EU-TIRADS was more sensitive and ACR TI-RADS more specific in these two studies, but the limited evidence and absence of paired discordant classifications preclude claims of superiority.