Malignancy Risk Is Similar for Solitary and Multiple Nodules in Hurthle Cell-Predominant Thyroid Fine Needle Aspirations
Vickie Y Jo, Jeffrey F Krane. Brigham and Women's Hospital & Harvard Medical School, Boston, MA
Background: Hurthle cell predominant (HC) thyroid fine-needle aspirations (FNAs) are problematic. FNAs diagnosed as “Suspicious for a Hurthle Cell Neoplasm” (HUR) have high sensitivity but low specificity for malignancy, especially since Hurthle cells are common in reactive and hyperplastic settings. The Bethesda System (TBS) indicates that the presence of multiple nodules may justify classifying HC FNAs as “Atypia of Undetermined Significance” (AUS) rather than HUR. We set out to address whether this approach is justified.
Design: Cytology records were searched for HC thyroid FNAs and cases were correlated with ultrasound (US) and surgical pathology reports. All FNAs were endocrinologist performed with US-guidance and were processed by ThinPrep (Hologic, Marlborough, MA) liquid-based preparation.
Results: 165 HC FNAs from 146 nodules were diagnosed as: Benign (5), AUS (40), HUR (116), Suspicious for Malignancy (3), and Malignant (1). By US, the median size of all targeted nodules was 1.9 cm. 21 nodules were solitary, while 122 nodules were in the setting of multiple (≥2) nodules. Results for 99 resections (60 benign, 39 malignant) were reviewed. Of the 60 benign nodules, US showed 49 (82%) associated with multinodular disease with median size of targeted nodules of 1.9 cm. There were 26 follicular adenomas with oncocytic features; the remainder had nodular hyperplasia. 21 cases (35%) had histologic lymphocytic thyroiditis (LT). 6 patients had simultaneous HC FNAs from different nodules; 4 were resected with nodular hyperplasia (1 also with LT). Of the 39 malignant nodules, 33 (85%) were multinodular by US with median size of targeted nodules of 2.3 cm. There were 15 follicular carcinomas (12 oncocytic/ type), 20 papillary carcinomas (14 follicular variants), and 4 poorly differentiated carcinomas. 13 cases (39%) had LT. There was no signifiant difference in malignancy rate for solitary or multiple nodules (P=0.79). Histologic presence or absence of LT (P=1.0) did not differ significantly between benign and malignant nodules nor did nodule size by US (P=0.49) or surgical resection (P=0.57).
Conclusions: Contrary to the notion that HC FNAs are more likely benign in the setting of nodular hyperplasia, we found no significant difference in malignancy rate of solitary or multiple nodules. Histologic LT and size did not differ between resected benign and malignant nodules. These findings suggest that the presence of multiple nodules does not warrant the use of the AUS category as an alternative to HUR in TBS.
Wednesday, March 21, 2012 1:00 PM
Poster Session VI # 70, Wednesday Afternoon