Sonographic Appearance Is Useful in Predicting the Extent of Initial Operative Therapy for Thyroid Nodules Classified as "Suspicious for Malignancy".

The "suspicious for malignancy" (SFM) category of the Bethesda system for reporting thyroid cytopathology predicts an incidence of malignancy from 60 to 75 per cent. Recommended treatment is via either lobectomy or total/near total thyroidectomy. Identification of predictors of malignancy in this category would be useful in selecting the extent of operative therapy in these patients. Published literature has suggested that suspicious ultrasound (US) findings are useful in predicting malignancy in both the "benign," "atypia of uncertain significance/follicular lesion of uncertain significance," and "suspicious for follicular neoplasm" categories. We hypothesized that US would be similarly useful in patients with nodules classified as SFM. We performed a retrospective analysis of patients with fine needle aspiration biopsy (FNAB) classified as SFM who underwent thyroidectomy from October 2007 to October 2012. Data collected included symptoms, suspicious imaging findings, FNAB results, and histopathology results-all obtained via chart review. Findings were compared between patients with/without a diagnosis of thyroid malignancy. Significance was set at P < 0.05. Of 3839 FNAB, 53 were classified as SFM. Of these, 36 were available for review. Twenty-four (66.7%) carcinomas were identified-19 papillary, 3 follicular variant of papillary, 1 follicular, and 1 medullary. No symptoms/signs reached significance as a predictor of malignancy. A suspicious US appearance, however, was significantly associated with an underlying carcinoma (P = 0.002). Patients with a Bethesda system for reporting thyroid cytopathology report of SFM and suspicious US findings are statistically more likely to harbor an underlying malignancy. This information is useful in selecting which patients are most likely to benefit from total/near total thyroidectomy as opposed to lobectomy as the initial operative strategy.