Ultrasound-Pathology Discordant Nodules on Core-Needle Biopsy: Malignancy Risk and Management Strategy.

BACKGROUND: The reported malignancy rates of thyroid nodules that appear suspicious on ultrasound (US) but yield benign fine-needle aspiration (FNA) cytology results range from 13.6% to 56.6%. However, the malignancy risk of nodules that are discordant on US and core-needle biopsy (CNB) pathological findings has not been evaluated. This study evaluated the malignancy risk associated with thyroid nodules that are suspicious on US but yield benign CNB results.

METHODS: Between July 2008 and November 2011, 1309 patients with 1313 thyroid nodules underwent US-guided CNB. Among these, 166 thyroid nodules that were suspicious on US but yielded benign CNB results were retrospectively reviewed. Benign nodules were diagnosed by surgery, FNA, and/or CNB that was repeated at least twice or CNB and lesion size stability on follow-up. The malignancy risk of US-pathology discordant nodules was calculated according to recommendations by Moon et al., and a further evaluation was conducted using four different US risk systems. Factors that differentiated benign from malignant nodules were assessed using the chi-square test or Fisher's exact test. Moreover, the histological findings of US-pathology discordant nodules were analyzed.

RESULTS: Patients with benign and malignant nodules did not differ significantly regarding clinical and demographic characteristics. Of the 166 US-pathology discordant nodules, seven (4.2%) were confirmed malignant. The malignancy risks associated with the five US risk systems ranged from 3.6% to 5.5%. Regarding suspicious US features, only the number of suspicious features was associated with malignancy (p = 0.002). Histological analysis of benign CNB specimens with suspicious US features revealed fibrosis (77.6%), thyroiditis (28.7%), calcifications (10.6%), and hemosiderin deposition (6.4%).

CONCLUSIONS: The malignancy risk among thyroid nodules with discordant US and CNB results ranged from 3.6% to 5.5%, depending on US risk systems. However, repeated biopsy may be necessary for selected thyroid nodules with more than two suspicious US features.