N. Eisa2, A. Khan3, M. Fensterwald3, S. Saleem3, M. A. Akhter3, M. J. Campbell1 1University Of California – Davis,Department Of Surgery,Sacramento, CA, USA 2University Of California – Davis,Department Of Internal Medicine,Sacramento, CA, USA 3University Of California – Davis,School Of Medicine,Sacramento, CA, USA
Introduction: Ultrasound with fine needle aspiration (FNA) is the most reliable diagnostic procedure for evaluation of thyroid nodules. Most FNA cytology is evaluated according to the Bethesda System for Reporting Thyroid Cytopathology, which uses six diagnostic categories to stratify the risk of malignancy. Bethesda category III, atypia of underdetermined significance (AUS), is a heterogeneous category with an anticipated risk of malignancy between 5 – 15%. It has been suggested AUS should be further sub-classified into categories describing the type of neoplasm of concern to allow for a better assessment of the risk of malignancy. The purpose of this study is to evaluate the association between the sub-classification of AUS thyroid nodules and risk of thyroid cancer.
Methods: We performed a review all patients with a thyroid nodule who underwent a FNA between January 2012 and October 2015 at a tertiary referral center. Patients with a cytologic diagnosis of AUS were sub-classified into the following four categories based on review of the cytology reports: 1) AUS-PTC (atypia with concern for papillary thyroid carcinoma), 2) AUS-FN (atypia with concern for a follicular neoplasm) 3) AUS – HCN (atypia with concern for a Hurthle cell neoplasm) and AUS-NOS where the nodule did not fall into the previous three categories or had mixed features. Continuous variables were evaluated using the student T-test and categorical variables were analyzed using the Fischer exact test. A multivariable logistic model was constructed to evaluate for independent associations with malignancy.
Results:
1,460 patients with 2,155 thyroid nodules underwent a FNA during our study period. 191 (8.8%) nodules had a cytologic diagnosis of AUS. Of the 163 nodules with AUS on their initial FNA, 27(17%) had AUS-PTC, 62 (38%) had AUS-FN, 57 (35%) had AUS-HCN, and 17 (10%) had AUS-NOS.
62(38%) nodules went on to have a repeat FNA and 24 (39%) came back benign. The frequency of benign cytology on the second FNA was similar between AUS subgroups [5/10 (50%) for AUS-PTC vs. 12/28 (43%) for AUS-FN vs. 5/19 (26%) for AUS-HCN vs. 2/5 (40%) for AUS-NOS, p= 0.189].
73 patients with 78 (48%) nodules went on to have a thyroidectomy. The overall malignancy rate for all nodules with AUS was 36%. The risk of malignancy differed between AUS sub-types as follows: 10/14(71%) for AUS-PTC vs. 7/25 (28%) for AUS-FN vs. 8/32 (25%) AUS-HCN vs. 3/7 (42%) AUS-NOS (p=0.033).
On univariate analysis AUS-PTC (p=0.004) was associated with a thyroid cancer on surgical pathology; however, on multivariable analysis AUS sub-type was not independently associated with risk of malignancy.
Conclusion:The risk of malignancy in nodules with AUS may be influenced by the type of atypia found by the cytopathologist, but the independent contribution of AUS subtype is unclear.