Specificity low for fine-needle aspiration in thyroid nodule risk stratification systems
Click Here to Manage Email Alerts
Two widely used risk stratification systems for determining whether thyroid nodules should undergo fine-needle aspiration have a specificity of less than 30%, according to study results published in Clinical Endocrinology.
“Both American College of Radiology Thyroid Imaging Reporting and Data System and European Thyroid Imaging Reporting and Data System have a high sensitivity, but low specificity rate when selecting thyroid nodules for fine-needle aspiration,” Lenara Golbert, MD, PhD, professor of medicine at the Federal University of Health Sciences of Porto Alegre in Brazil, and colleagues wrote. “In our sample presented with 5.9% prevalence of cancer, large median size of nodule and mostly composed of multinodular glands, the number of fine-needle aspirations possibly avoided by American College of Radiology Thyroid Imaging Reporting and Data System and European Thyroid Imaging Reporting and Data System were 26.7% and 17.8%, respectively.”
Researchers conducted a prospective study of thyroid nodules from adults who underwent fine-needle aspiration (FNA) from 2019 to 2021 in the endocrinology division at Santa Casa de Porto Alegre in Brazil. The number of nodules and features collected were: diameter, shape, composition, echogenicity, margins and calcifications. Nodules were classified on examination time analysis according to American College of Radiology Thyroid Imaging Reporting and Data System (ACR-TIRADS) and European Thyroid Imaging Reporting and Data System (EU-TIRADS). Within both systems, the likelihood of a nodule’s malignancy is indicated in risk class and eligibility for FNA is indicated by nodular size. The sonographic recommendation of FNA for each nodule was compared with its diagnosis to estimate sensitivity and specificity of each system.
There were 186 nodules analyzed in the study, of which 3.2% were classified as nondiagnostic, 80.1% as benign, 7% as being of undetermined significance, 3.8% as follicular neoplasm or suspicious for follicular neoplasm, 0.5% as suspicious for malignancy and 5.4% as malignant. Of 29 nodules referred for surgery with histopathological results available, 11 were malignant.
Most nodules in the study were classified as ACR-TIRADS 4 and EU-TIRADS 3. The rate of malignancy was 2.3% in ACR-TIRADS 4, 42.8% in ACR-TIRADS 5, 2% in EU-TIRADS 3, 0% in EU-TIRADS 4 and 60% in EU-TIRADS 5.
ACR-TIRADS indicated fewer nodules for FNA compared with EU-TIRADS (74.7% vs. 82.8%; P = .017). The sensitivity was high at 100% for ACR-TIRADS and 90.9% for EU-TIRADS, and specificity was low at 28.7% for ACR-TIRADS and 19.1% for EU-TIRADS. Using the ACR-TIRADS classification system would have resulted in 26.7% of FNA procedures being avoided and EU-TIRADS would have resulted in 17.8% of FNA procedures being avoided.
“The number of possibly avoided FNAs for both ACR‐TIRADS and EU‐TIRADS was lower than expected,” the researchers wrote. “The best performance of both U.S. systems was found when indicating FNA, hypothetically, only in highly suspicious nodules and/or in the presence of lymphadenopathy. Regarding clinical factors, we found a tendency of higher age and bigger nodules to be presented as benign. These data suggest that more prospective studies are needed for redefining the best criteria for FNA and the usefulness of nodule size for selection.”
Collapse
Read more about
Click Here to Manage Email Alerts