Endocrine Abstracts

Introduction and Aims: Thyroid lesions constitute a frequent consultation in the area of endocrinology. Fine-needle aspiration (FNA) cytology is a gold standard for the evaluation of thyroid lesions, while ultrasound examination according to various thyroid imaging reporting and data system (TIRADS) criteria is gaining importance in the field. The aim of our study was to compare the sensitivity and specificity of both techniques in patients attended in thyroid lesion unit of our hospital in relation to the final resection diagnosis.

Methods: We conducted a retrospective review of lesions studied in our unit since the introduction of the ACR TIRADS criteria (2018-2021). The evaluation included cases with Bethesda III to VI category since these had higher probability of having been operated. In the case of Bethesda III cytology a second FNA was conducted and if benign the case was labelled as benign and thus not operated. For the final analysis only the nodules with an ultrasound exam, FNA cytology and final result of histopathology (from thyroidectomy or core-needle biopsy in one case) were taken into consideration.

Results: Overall, 106 FNAs were reviewed from which 80 cases met the inclusion criteria. Out of 80 cases, 59 patients (74%) were women, and the average age was of 54 years. The risk of malignancy in presence of ACR TIRADS 3, 4, and 5 were 27%, 55% and 83% respectively and in case of Bethesda III, IV and V/VI category of 39%, 30%, and 92% respectively. To calculate sensitivity and specificity, we labelled TIRADS 2-3/Bethesda III as benign and TIRADS 4-5/Bethesda IV-VI as malignant. The overall sensitivity and specificity of ultrasound with ACR TIRADS was of 83% and 44%, and FNA of 78% and 36%.

Conclusions: Both FNA and TIRAD system showed similar trends in malignancy prediction. Ultrasound evaluation with TIRAD system demonstrated better concordance with final resection result with the exception of TIRADS 5/Bethesda V-VI group where cytological evaluation had better outcomes. Knowing the limitations of both techniques and the results of our and previous studies we consider advisable the use of both modalities adjunctively.