Acurácia diagnóstica das classificações de nódulos tireoidianos pela ultrassonografia: um estudo comparativo

• Ano de defesa: 2022
• Autor(a) principal: Mário Henrique Giordano Fontes
• Outros Autores: Bernardo Lopes Cançado Fonseca
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; Brasil; MEDICINA - FACULDADE DE MEDICINA; Programa de Pós-Graduação em Ciências Aplicadas à Cirurgia e à Oftalmologia; UFMG
• Palavras-chave em Português: Classificação.; Neoplasias da glândula tireoide; Glândula tireoide; Neoplasias da Glândula Tireoide; Classificações em Saúde; Nódulo da Glândula Tireoide
• Link de acesso: http://hdl.handle.net/1843/49267

Thyroid cancer has been increasing in its incidence in recent years. Ultrasonography in an important tool in determining which thyroid nodules need to have closer monitoring or should be further evaluated through Fine Needle Aspiration Procedure (FNA), mainly because it is a low-cost imaging modality, has wide availability and good accuracy. In recent years a few classifications of thyroid nodules have been created to create a standard of evaluation of thyroid nodules. Some of these are based on ecographic patterns and others by giving points according to the presence of some characteristics already stablished in the literature. This study aims to evaluate and compare the diagnostic accuracy of these three sonographic classifications: ATA, ACR TIRADS and EU TIRADS. Nodules that were submitted to FNA between Jan/2016 and Jan/2020 were included in this study. Two radiologists classified the nodules that filled the inclusion criteria in total of 767 nodules, using the three classifications without knowledge of the cytologic result. Then an analysis of interobserver agreement and a comparison of the three classifications were made against the cytologic result, The best agreement was obtained with ATA (substantial agreement) and the other classifications (ACR and EUTIRADS) had a moderate agreement, and these results were higher that similar studies previously published. The analysis comparing the outcome of the three classifications considering the sensibility, specificity, positive predictive value, and negative predictive value showed similar results between them. In the analysis that excluded cases of Bethesda III and IV of the sample obtained a sensibility, specificity, positive predictive value, and negative predictive for ATA (91,3%; 69,1%; 32,4%, 98,0% and 72,2%), ACR TIRADS (92,0%; 62,8%; 29,9%; 97,8% and 67,1%) e EU-TIRADS (91,9%; 67,0%; 32,0%; 98,0% and 70,6%), respectively. The performance of these three sonographic classifications of thyroid nodules is comparable, with slight variations in sensibility and specificity.