Results of retrospective classification of thyroid FNAs according to the Bethesda system: would this have improved accuracy?
Abstrakt
ObjectivesTo retrospectively reclassify the results of fine needle aspiration (FNA) cytology at our institution according to the Bethesda system for reporting thyroid cytopathology (TBSRTC), and to determine whether the introduction of the new classification system would have influenced the diagnostic accuracy compared with our existing system in which suspicious/indeterminate categories (categories III-IV) are managed according to clinical findings and cytopathologists' recommendations. MethodsFNAs performed under ultrasound guidance between 2001 and 2012, and subsequently verified by histology or repeat FNA and follow-up, were reviewed and retrospectively reclassified according to TBSRTC.
ResultsAmong a total of 1310 histologically verified FNAs, the positive predictive values (%) for malignancy and neoplasia (carcinoma+follicular adenomas) for category I-VI according to the TBSRTC were 5.7/11.3, 2.6/5.5, 14.4/34.2, 23.6/55.0, 57.0/64.5 and 92.2/94.2, respectively. Although all the categories predict malignancy closely, the difference in malignancy rate between categories III and IV was not statistically significant; however, the difference in total neoplasia rate reached statistical significance.
When patients with cytological and clinical follow-up were included, the malignancy rated dropped in category I to 2.4%, whereas, in category III, it remained at 13%. Repeat FNA instead of direct surgery spared half of the patients from surgery.
The cytopathologists' recommendations for histological verification in category III were associated with an insignificant increase in malignancy rate, but the total neoplasia rate reached 57.5%. ConclusionsAt our institution, the application of the TBSRTC did not improve the diagnostic accuracy for the detection of malignancy compared with current practice.
In selected cases with cytological results falling into category III (according to clinical data and cytopathologists' opinions), direct lobectomy seems to be justified.