276 chapter 3 TI-RADS Since 2009, several US-based risk stratification systems to identify nodules that warrant biopsy or sonographic follow-up have been proposed. Following the BI-RADS classification system that is widely used in breast imaging, the American College of Radiology (ACR) presented the TI-RADS (for Thyroid Imaging, Reporting and Data System). TI-RADS aims to (1) provide recommendations for reporting incidental thyroid nodules, (2) develop a set of standard terms (lexicon) for US reporting, and (3) propose a TI-RADS risk stratification system on the basis of the lexicon [394]. The ACR TI-RADS scores the composition, echogenicity, shape, margin, and echogenic foci of a thyroid nodule, all consisting of 0 up to 3 points. The total number of points determines whether a nodule is considered benign (TR1, 0 points), not suspicious (TR2, 2 points), mildly suspicious (TR3, 3 points), moderately suspicious (TR4, 4 to 6 points), or highly suspicious (TR5, ≥7 points) and also guides the decision to perform FNAC or follow-up: no FNAC or follow-up (TR1-2), FNAC if nodule maximum diameter (ø)≥2.5 cm and follow-up if ø≥1.5 cm (TR3), FNAC if ø≥1.5 cm and follow-up if ø≥1.0 cm (TR4) or FNAC if ø≥1 cm and follow-up if ø≥0.5 cm (TR5). In addition to ACR TI-RADS, the European Thyroid Association and the Korean Society of Thyroid Radiology/Korean Thyroid Association developed similar US risk stratification systems; the EU-TIRADS and K-TI-RADS, respectively [395, 396]. Also, the ATA and the American Association of Clinical Endocrinologists/American College of Endocrinology/Associazione Medici Endocrinologi propose US risk stratification systems [17, 397]. An international survey investigating the utilisation of all five aforementioned risk stratification systems with 875 respondents in 52 countries demonstrated that almost one third of respondents used more than one risk stratification system in their practice, potentially leading to confusion [398]. Grani et al. compared the risk stratification systems in 477 patients and found that the systems vary widely in their ability to reduce the number of unnecessary thyroid nodule FNACs (17.1 up to 53.4%) [399]. The ACR TI-RADS outperformed the others, classifying more than half of the biopsies as unnecessary with a false-negative rate of 2.2%. The remainder of this chapter focuses on the ACR TI-RADS. Over recent years, TI-RADS has become fully incorporated in the management of thyroid nodules [17]. As FNAC may be more systematically withheld for patients with a presumed benign nodule with TI-RADS 1, 2, and most 3, the patient population that is selected for additional diagnostic tests has potentially changed. Many of the studies on additional diagnostics have not incorporated TI-RADS yet, and it is currently unclear to say what effect the introduction of TI-RADS may have on the diagnostic accuracy and therapeutic yield of other tests. A different population, or reference class, with a larger proportion of malignant nodules, impacts the PPV and NPV, but also the sensitivity and specificity. Stratification of cytologically indeterminate nodules according to the risk of malignancy as determined by combining the TI-RADS and Bethesda system might be of interest, notwithstanding a limited body of evidence comprising of small retrospective cohorts. Larcher de Almeida et al. investigated
RkJQdWJsaXNoZXIy MTk4NDMw