09.02 Chronic Lymphocytic Thyroiditis May Lower Accuracy of AUS/FLUS Cytopathology in Surgical Patients

M. B. Mulder1, K. Khazeni1, M. Sussman1, J. I. Lew1, J. C. Farrá1  1University Of Miami,DeWitt Daughtry Department Of Surgery: Division Of Endocrine Surgery,Miami, FL, USA

Introduction:
Chronic lymphocytic thyroiditis (CLT) is known to increase cytologic atypia on fine-needle aspiration (FNA) of thyroid nodules. The rate of malignancy (ROM) has been shown as high as 55% in surgical patients with atypia (or follicular lesion) of undetermined significance (AUS/FLUS) thyroid nodules. The effect of CLT on the ROM in AUS/FLUS thyroid nodules remains unclear. This study evaluates the effect of concomitant CLT on the malignancy rate of AUS/FLUS thyroid nodules in surgical patients.  

Methods:
A retrospective review of prospectively collected data for 1061 patients who underwent thyroidectomy for a dominant thyroid nodule from a single institution was performed. All patients had FNA of thyroid nodules classified according to the Bethesda System for Reporting Thyroid Cytopathology (BSRTC).  Patients with CLT on final pathology were further subdivided into two subgroups: CLT and benign thyroid nodule (CLT+B) and CLT and malignant thyroid nodule (CLT+M). Gene expression classifier (GEC) testing in patients with AUS/FLUS cytology were analyzed in both groups.  

Results:
Of the entire surgical series, there were 293 (28%) patients with AUS/FLUS cytopathology and a ROM of 56% (163/293) on final pathology. CLT was found in 189 patients overall (19%), with a ROM of 50% (95/189). Final pathology seen in CLT patients with malignancy was as follows: 53% follicular variant papillary thyroid cancer (PTC), 40% classic type PTC, 4% PTC tall-cell variant, 1% PTC diffuse sclerosing variant, 1% follicular carcinoma minimally invasive, and 1% medullary carcinoma.  AUS/FLUS cytopathology was reported in 73 (39%) of the 189 CLT patients, comprising 25% (73/293) of the entire AUS/FLUS group. Of these 73 patients, 56% (41/73) had benignity, whereas 44% (32/73) had malignancy on final pathology. Of the 36 CLT patients with AUS/FLUS cytopathology and GEC testing, 33 patients had suspicious results, of which 17 were found to be malignant on final pathology yielding a PPV of 52%.   

Conclusion:
The ROM in patients with AUS/FLUS thyroid nodules is lower when coexisting CLT is present. Additionally, GEC testing has limited value in the evaluation of thyroid nodules for malignancy in the setting of underlying CLT. Cytologic atypia due to a background of CLT may result in more AUS/FLUS categorizations for thyroid nodules, which may lead to overestimation of  the ROM in this patient population.