[Implementation of the Bethesda system for reporting the thyroid cytopathology: study on 5 729 cases from a cancer center].

Objective To analyze the data of the Bethesda system for reporting thyroid cytopathology applied in a comprehensive cancer center and to evaluate the diagnostic ability of fine needle aspiration (FNA). Methods We retrospectively reviewed the medical records of 5 729 cases applying this reporting system at Cancer Hospital, Chinese Academy of Medical Sciences. The series were from 5 011 patients including 1 174 men and 3 837 women, and their median age was 45 years (range, 7-88 years). FNA results were correlated with final histological diagnosis after surgery and the accuracy of FNA diagnosis and the malignancy rates for each of categories were also analyzed. Results Among 5 729 thyroid aspirates, aside from 456 (8.0%) cases with nondiagnostic or unsatisfactory (ND/UNS) outcomes, 1 055 (18.4%) cases were benign, 409 (7.1%) cases showed atypical of undetermined significance or follicular lesions with undetermined significance (AUS/FLUS), 80 (1.4%) cases were follicular neoplasm or suspicious for follicular neoplasm (FN/SFN), 982 (17.1%) cases were suspicious for malignancy (SUS), and 2 747 cases were malignant (47.9%). Of 5 729 cases, 3 239 had received thyroidectomies after FNA, 95.99% of them were proven histologically to be malignant, with following malignancy rates in individual FNA categories: ND/UNS 75.00%; benign 40.91%; AUS/FLUS 77.67%; FN/SFN 41.67%; SUS 96.86%; and malignant 99.96%. FNA predicted malignancy with sensitivity, specificity, accuracy, positive predictive value and negative predictive values of 98.8%, 60.5%, 97.7%, 98.9% and 59.1%, respectively. Conclusions The data of the Bethesda reporting system indicates high proportion of malignant diagnosis and high risk of malignancy at all FNA diagnostic categories. FNA offers high diagnostic accuracy and positive predictive value for the diagnosis of thyroid diseases. Key words: Fine needle aspiration; Thyroid; Bethesda system; Risk of malignancy; Accuracy