Abstract：Objective: This study aimed to assess diagnostic value of ultrasound-guided fine-needle aspiration cytology(US-G FNAC) of thyroid nodules before operation. Methods: Data of 360 US-G FNAC of thyroid nodules in our hospital from 2009 to 2012 were retrospectively collected. Classification and interpretation of US-G FNAC using the Bethesda system standard. 197 nodules were operated on, the results of US-G FNAC were compared with final histology, the area under ROC curve(AUC) was used to assess diagnostic accuracy of US-G FNAC of thyroid nodules, and compared all the risks prediction of malignant nodules(quoted in the Bethesda system) and the actual risks of malignancy(final histology results). Results: ① Among 360 nodules, 197 were operated on, US-G FNAC results were reported as: 72 benign(BEN) (36.55%), 7 atypia of unknown significance(AUS)(3.55%), 65 suspicious for follicular neoplasm(SFN)(32.99%), 7 suspicious for malignancy(SFM)(3.55%), 33 papillary carcinoma(16.75%), 2 medullary carcinoma(1.02%), 2 other(1.02%) and 9 insufficient(INS)(4.57%). Histology results were reported as: 124 BEN(62.94%), 21 Follicular adenoma(10.65%), 2 follicular carcinoma(1.02%), 46 papillary carcinoma(23.35%), 2 medullary carcinoma(1.02%), 2 other(1.02%). US-G FNAC results were compared with final histology, AUC of US-G FNAC=0.903, diagnostic accuracy was high, with statistically significant difference. ②Risk prediction of malignant nodules(quoted the Bethesda system) was same as the actual risk of malignancy(final histology results). Conclusions: Before operation, diagnostic accuracy of US-G FNAC was high. The Bethesda system appropriately stratified lesions for risk of malignancy.
2014, Vol. 25 
