A meta-analytic review of the Bethesda System for Reporting Thyroid Cytopathology: Has the rate of malignancy in indeterminate lesions been underestimated?

BACKGROUND

The Bethesda System for Reporting Thyroid Cytopathology (TBSRTC) provides a 6-tier diagnostic framework using uniform criteria in reports of thyroid aspirates. One of the major advantages of this framework is its association with defined risks of malignancy, allowing standardized management algorithms for each diagnosis. The objective of the current meta-analysis was to demonstrate the feasibility of using TBSRTC among specimens in the atypia of undetermined significance or follicular lesion of undetermined significance (AUS/FLUS) and follicular neoplasm or suspicious for neoplasm (FN/SFN) categories. The authors also evaluated both the morphologic features and the risk of malignancy in the presence of Hurthle cells.

METHODS

A literature search was performed of the PubMed, Scopus, and Web of Science databases for English-language studies published from January 2008 to December2014. Studies were considered eligible only if they evaluated the risk of malignancy for specimens in the AUS/FLUS and/or FN/SFN categories and included surgical follow-up.

RESULTS

In total, 51 articles were identified that used TBSRTC criteria and provided data for a total of 145,928 fine-needle aspiration (FNA) specimens. Of these, FNAs that had surgical follow-up were selected among the AUS/FLUS (N = 4475) and FN/SFN (N = 3202) specimens. The overall rate of malignancy was 27% for the AUS/FLUS category and 31% for the FN/SFN category.

CONCLUSIONS

The AUS category was characterized by limited reported follow-up and surgical outcome. The data demonstrated that FNAs with an AUS diagnosis had a higher risk of malignancy than the risk according to published TBSRTC criteria, whereas the percentage of malignancy in FNAs with an FN/SFN diagnosis did not differ from that according to TBSRTC. Hurthle cell lesions represent a challenging category, underlying the importance of further studies to define whether they can be diagnosed in the AUS/FLUS category rather than the FN/SFN category.