[388] Consensual Review Minimizes the Diagnosis of “Follicular Lesion of Undetermined Significance” and Improves Reproducibility and Cyto-Histologic Concordance.

Xin Jing, Stewart M Knoepp, Michael H Roh, Kim Hookim, Jeremiah Placido, Robertson D Davenport, Rodolfo FH Rasche, Claire W Michael. University of Michigan Health System, Ann Arbor

Background: There has been considerable variation among both pathologists and institutions in using the terminology of "atypia of undetermined significance/follicular lesion of undetermined significance" (AUS/FLUS). The Bethesda System for Reporting Thyroid cytology (BSRTC) defines the diagnostic criteria for AUS/FLUS and advises that AUS/FLUS should not be used indiscriminately. The current study was conducted to investigate if consensual review may minimize AUS/FLUS and may also improve reproducibility and cyto-histologic concordance.
Design: A SNOMED search of the electronic pathology database for the period of January 1998-June 2010 retrieved a total of 217 aspirates in which the original interpretation of AUS/FLUS was rendered and followed by hemi- or total thyroidectomies. The retrieved aspirates were retrospectively reviewed using the BSRTC as a guideline. A group of reviewers who were blinded to the corresponding histologic findings simutaneously evaluated slides of each aspirate at a multi-headed microscope and a consensus diagnosis was reached at the end of the group review. Interobserver agreement was calculated by dividing diagnosis of the majority by the total. The Cyto-histologic correlation was then performed between the consensus diagnoses and the corresponding histologic diagnoses. This is an ongoing study and a total of 50 aspirates have been reviewed.
Results:

CORRELATION BETWEEN CONSENSUAL CYTOLOGIC INTERPRETATION AND HISTOLOGIC DIAGNOSIS

		Histologic Diagnosis
Cytologis Diagnosis	Interobserver Agreement (%)	benign	follicular adenoma	follicular carcinoma	PTC	TOTAL(%)
non-diagnostic	100	2	-	-	-	2 (4)
benign	90.5	24	2	-	-	26 (52)
AUS/FLUS	71.6	5	3	2	1	11 (22)
FN/SFN	81.8	2	5	1	2	10 (20)
PTC	100	-	-	-	1	1 (2)
TOTAL	_	33	10	3	4	50 (100)

FN/SFN: positive/suspicious for follicular neoplasm; PTC: papillary thyroid carcinoma

The data generated from 50 cases demonstrates that using BSRTC as a guideline, consensual review of aspirates previously categorized as AUS/FLUS: 1) markedly reduces the diagnosis of AUS/FLUS (11 out of 50) without compromising diagnostic accuracy. 2) provides a better cyto-histologic concordance and archieves a diagnostic accuracy of 89.2%. 3) The average level of interobserver agreement of 88.8% is archieved
Conclusions: Using BSRTC as a guideline, consensual review of aspirates previously categorized as AUS/FLUS plays a substantial role in reducing unnecessary surgical intervention by 48%.
Category: Cytopathology

Wednesday, March 2, 2011 1:00 PM

Poster Session VI # 51, Wednesday Afternoon