[340] Reporting Thyroid FNA before and after Implementation of Bethesda System

Kathriel J Brister, Remmi S Singh, Helen H Wang. Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, MA

Background: The Bethesda System for Reporting Thyroid Cytopathology was published in 2008, and was implemented at Beth Israel Deaconess Medical Center (BIDMC) in June 2010. Prior to this date, our diagnostic scheme was similar to the Bethesda System, except for the category of “Atypia/Follicular Lesion of Undetermined Significance” (AUS). Instead, we used the categories “Atypical (ATY),” “Suboptimal (SUB)” and “Indeterminate (IND)” to describe specimens that would later be classified as AUS. This study evaluates the impact of the Bethesda System on the rate and the positive predictive value (PPV) of the diagnostic categories at BIDMC.
Design: We performed a retrospective review of all thyroid FNAs during the time periods Jan 2006-Nov 2008 and June 2010-July 2011 (period 1 and 2, respectively) and identified 2355 and 1204 specimens, respectively. Each cytology report was categorized as shown in Table 1 below. All subsequent thyroidectomy specimens were identified (406 and 227 from period 1 and 2, respectively) and categorized as benign or malignant. PPV was determined for each cytologic category.
Results:

Table 1

Diagnostic Category	1st Period		2nd Period
	% of total	PPV in %	% of total	PPV in %
Non-Diagnostic (ND)	13	15	17	23
ND-cyst			4.8	33
SUB-benign, including cysts	21	15
SUB-Hürthle cell (HC) lesion or microfollicular lesion	2.5	17
SUB-papillary carcinoma (PC)	0.93	36
Benign	41	17	48	23
ATY/AUS	3.7	31	15	44
HC neoplasm	3.1	24	2.4	36
Follicular neoplasm	3.4	30	4.0	32
IND-PC	2.5	53
Suspicious	3.9	87	4.5	80
Positive	4.5	98	4.2	100

The most notable change between the two periods was an increase during the second period of the rate of ND (13% to 22% [including cysts]), Benign (41% to 48%), and AUS (3.7% to 15%) diagnoses. These increases were a result of the inclusion of those cases previously classified as SUB and IND-PC into these categories in the new scheme. Consequently, an increase in PPV was seen for these three categories, while the PPV for the other categories remained relatively unchanged. This is especially evident in the PPV of the AUS category (44%), which includes previous diagnoses with PPVs ranging widely from 17% (SUB-HC) to 53% (IND-PC).
Conclusions: Implementation of the Bethesda System to yield recommended PPVs is highly lab-dependent. Diagnoses with wide-ranging PPVs are now grouped into the AUS category, which has a PPV approaching 50% in our lab; this creates uncertainty regarding the appropriate management for this category. Our prior scheme provided greater diagnostic specificity to direct clinical management of these patients.
Category: Cytopathology

Monday, March 19, 2012 9:30 AM

Poster Session I Stowell-Orbison/Surgical Pathology/Autopsy Awards Poster Session # 57, Monday Morning