[462] Utility of Repeat Thyroid FNA in the Management of Patients with Thyroid Nodules

RS Singh, JL Hecht, HH Wang. Beth Israel Deaconess Medical Center, Boston, MA

Background: In the new Bethesda System for reporting thyroid FNA, repeat FNA is recommended for an initial diagnosis of non-diagnostic or atypical cells/follicular lesion of undetermined significance. We therefore investigated the pattern of repeat thyroid FNA at our institution.
Design: We retrospectively reviewed all reports on thyroid FNAs and thyroidectomies submitted to the pathology department from 2006 to 2008 at our institution. All patients who had a repeated FNA of the same lesion and all who had both FNA and thyroidectomy in the same time period were identified.
Results: 2,717 FNA specimens were identified. 608 had thyroidectomies during the same time period. 291 patients underwent repeated FNA at least once for 301 thyroid lesions. Thirty had a second repeat and one had a third repeat. The main findings are listed below. (ND=non-diagnostic; Sub=suboptimal; B=benign; F=follicular lesion with or without cystic change that cannot be further categorized due to cellularity; MF=microfollicular; H=Hurthle cell; P=papillary carcinoma; Ind=indeterminate for malignancy that includes MF, H, or suggestive but not diagnostic for P; Sus=suspicious for malignancy; Pos=positive for malignancy).

Diagnostic category	Initial aspirates (%)*	Number (%) with at least one repeat	Diagnosis of first repeat (%)*	Diagnosis of second repeat (%)*	% of patients with surgery after initial diagnosis**
ND	316 (13)	132 (42)	53 (18)	4 (13)	9.2
Sub -B or F	495 (21)	81 (16)	61 (20)	12 (40)	13
Sub -MF+H+P	84 (3.5)	27 (32)	19 (6.3)	1 (3.3)	40
B	980 (41)	28 (2.9)	124 (41)	12 (40)	6.8
Atypical	89 (3.7)	21 (24)	11 (3.7)	1 (3.3)	49
Ind	216 (9.1)	11 (5.1)	20 (6.6)	0	73
Sus+Pos	205 (8.6)	1 (0.49)	13 (4.3)	0	69
Total	2385	301 (13)	301	30	23

*The difference in distribution among the initial, the first and the second repeat aspirates is statistically significant by chi-square with P < 0.01. **Surgery during the study period at our institution.


Conclusions: Patients with ND are most likely to undergo repeat aspirates followed by those with a Sub specimen but suggestive of a worrisome lesion, such as MF, H, or P, and then followed by those with an atypical diagnosis. Although 40% of repeat aspirates were benign, the repeat aspirates were less likely to be diagnostic of a lesions (whether MF, H, Sus or Pos) than the initial aspirates and at least as likely if not more likely to be ND or Sub. Patients with an initial atypical diagnosis are twice as likely to have surgery than to have a repeat (49% vs 24%).
Category: Cytopathology

Tuesday, March 23, 2010 9:30 AM

Poster Session III # 93, Tuesday Morning