Thyroid fine-needle aspiration cytology: malignancy rate in the category of indeterminate significant atypia/indeterminate significant follicular lesion

Abstract

BACKGROUND: Fine needle aspiration cytology (FNAC) is a standard preoperative diagnostic modality for thyroid nodules. The Bethesda Thyroid Cytopathology Reporting System (TBSRTC) defines the FNAC atypia group as atypia of undetermined significance/follicular lesion of undetermined significance (AUS/FLUS). OBJECTIVES: Determine the risk of malignancy after surgical resection in patients with AUS/FLUS. DESIGN: Retrospective SETTING: Pathology department of a tertiary care center PATIENTS AND METHODS: All thyroid FNACs between 2015 and 2023 that were diagnosed as AUS/FLUS in Turkey. Patient demographics, preoperative ultrasonographic features, and follow-up data were collected. MAIN OUTCOME MEASURES: Relationship between AUS/FLUS diagnosis and final histopathological diagnosis. SAMPLE SIZE: 562 RESULTS: In total, 562 thyroid nodules were diagnosed as AUS/FLUS, and 267 (47.5%) were surgically excised. A malignant histopathological diagnosis was given in 28 cases (10.4%). Malignancy risk sensitivity of AUS/FLUS diagnosis was 75.68% (95% CI=58.80–88.23%), specificity was 55.24% (95% CI=50.91–59.52%), positive predictive value was 10.49% (95% CI=8.71–12.58%), and negative predictive value was 97.04% (95% CI=94.86–98.31%). In the ultrasonographic data, having symptomatic nodules, nodule calcification, and irregular nodule borders were all statistically significant signs of cancer in a one-variable analysis ( P <.01). The presence of a family history emerged as a statistically significant prognostic marker for malignancy ( P =.012). Although not statistically significant, the malignancy rate for nodules with nuclear atypia was 11.9%, significantly higher than the rate of 8.3% for nodules with architectural atypia only ( P =0.32). CONCLUSIONS: The diagnosis of AUS/FLUS has a high rate of predicting the risk of malignancy and should continue to be offered. In addition to cytopathological features, ultrasound data and family history should be taken into consideration when evaluating the case. LIMITATIONS: Retrospective design and no molecular studies.