Congress abstracts SPEDM

Background: Thyroid fine needle aspiration (FNA) cytology guided by ultrasound (US) is currently the standard diagnostic tool for evaluating suspicious nodules.EU-TIRADS is one US thyroid cancer risk assessment system developed for hierarchization of FNA requirements.The Bethesda System is currently the recommended report system for thyroid cytopathology.Currently thyroid nodules nature can mostly be inferred by applying these two classifications and therefore restrict many thyroid surgeries to only those patients with nodules presenting recognized risk of malignancy.Aims: To evaluate the predictive value, sensitivity and specificity of EU-TIRADS and Bethesda System classifications of thyroid nodules on patients who underwent thyroid surgery after US-guided FNA in our centre.Methods: Observational retrospective study of thyroid nodules presented at our Thyroid Outpatient Clinic submitted to US-guided FNA between January/2016 and December/2021 and submitted to Thyroid surgery.Results/Conclusions: We included 105 patients, with 153 nodules.Total thyroidectomy was performed in 104 patients with only one hemithyroidectomy.The overall malignancy rate was 45.1% (n=69).Considering histological results, two EU-TIRADS 2 nodules were malignant [predictive value of benignity (PVB) of 90.5%], as were twenty-three EU-TIRADS 3 nodules (PVB: 57.4%), eleven EU-TIRADS 4 [predictive value of malignancy (PVM) of 31.4%] and thirty-three EU-TIRADS 5 (PVM 75.0%).Defining EU-TIRADS 5 as a positive test for malignancy and excluding EU-TIRADS 4 cases, the sensitivity of this system was 56.90% [CI 43.23%-69.84%]and its specificity was 83.33% [CI 71.48%-91.71%].About the cytology system, of the malignant nodules on histology, twelve were classified as Bethesda I on FNA, eight were Bethesda II, twenty were either Bethesda III or IV nodules, seventeen were Bethesda V nodules (PVM 89.5%) and all the twelve Bethesda VI nodules were malignant (PVM 100%).Excluding Bethesda I nodules and defining as a positive test for malignancy both Bethesda V and VI, sensitivity was 50.88% [CI 37.29%-64.37%]and specificity was 95.45% [CI 84.53%-99.44%].In this center, both classifications had a suboptimal sensitivity but a high specificity, especially the Bethesda system, reinforcing this population surgical selective bias but also their power as an accurate clinical-decision method.