Fine-Needle Aspiration of Follicular Patterned Lesions of the Thyroid: Diagnosis, Management, and Follow-Up according to Thyroid Bethesda System

Fax +41 61 306 12 34E-Mail [email protected]

Fine Needle Aspiration

Acta Cytologica 2012;56:361–369 DOI: 10.1159/000338218

Fine-Needle Aspiration of Follicular Patterned Lesions of the Thyroid: Diagnosis, Management, and Follow-Up according to Thyroid Bethesda System

Hüseyin Üstün Hesna Müzeyyen Astarcı Canan Altunkaya Sırma Yılmaz

Ahmet Barın Serap Ekici Muzaffer Çaydere

Pathology Department, Ankara Training and Research Hospital, Ankara , Turkey

and 243/381 (64%) cases diagnosed as SFON. The rate of ma-lignancy in AUS/FLUS cases with and without RFNA was 29 and 14% respectively, and it was 26% in SFON cases. Conclu-

sion: The current data shows that the malignancy rates differ between categories and AUS/FLUS cases are best managed by RFNA for selection of cases that can benefit from surgical excision. Copyright © 2012 S. Karger AG, Basel

Objective

The National Cancer Institute (NCI) State of the Sci-ence Conference on thyroid fine-needle aspiration (FNA) proposed that follicular patterned lesions can be divided into two diagnostic categories: follicular lesion of unde-termined significance/atypia of undetermined signifi-cance (AUS/FLUS) and suspicious for follicular neo-plasm/follicular neoplasm (SFON/FON). The former group can benefit from repeat FNA (RFNA) to achieve a more definitive diagnosis and the latter should undergo surgical excision for histologic characterization (adeno-ma vs. carcinoma). In this study, we report the experience from our hospital with thyroid FNA cases that can be placed into NCI-designated thyroid FNA diagnostic cat-egories for follicular patterned lesions.

Key Words

Bethesda � Fine-needle aspiration � Follicular patterned lesions � Thyroid

Abstract

Objective: The goal of this study was to report an experience with thyroid fine-needle aspiration (FNA) cases that can be placed into National Cancer Institute-designated thyroid FNA diagnostic categories for follicular patterned lesions di-vided into three diagnostic categories: follicular lesion of un-determined significance/atypia of undetermined signifi-cance (AUS/FLUS), suspicious for follicular neoplasm (SFON), and suspicious for malignancy. Study Design: The study co-hort included 4,284 cases for the period between January 2007 and July 2011 and all available follow-up data was ob-tained. All cases classified as ‘atypical’ and ‘rule out follicular neoplasm’ were included in the AUS/FLUS category (n = 3,903), whereas cases classified as ‘suspicious for’ or ‘consis-tent with follicular or Hürthle cell neoplasm’ (n = 381) were included in the SFON diagnostic category and compared with histopathologic follow-up. Results: During this period, 14,628 thyroid FNAs were reported in 12,238 patients. Re-peat FNA (RFNA) was performed in 1,366/3,903 (35%) pa-tients classified as AUS/FLUS. Histologic outcome data was available in 1,756/3,903 (45%) cases diagnosed as AUS/FLUS

Received: December 27, 2011 Accepted after revision: March 19, 2012 Published online: July 25, 2012

Correspondence to: Prof. Hüseyin Üstün Mithatpaşa Cad. No: 51/6, Kızılay TR–06420 Ankara (Turkey) Tel. +90 312 430 13 33 E-Mail olgupatoloji @ yahoo.com

© 2012 S. Karger AG, Basel 0001–5547/12/0564–0361$38.00/0

Accessible online at: www.karger.com/acy

http://dx.doi.org/10.1159%2F000338218

Üstün/Astarcı/Altunkaya/Yılmaz/Barın/Ekici/Çaydere

Acta Cytologica 2012;56:361–369362

Introduction

In many pathology practices, thyroid FNAs comprise a significant proportion of nongynecologic cytology specimens, reflecting the fact that nodules of the thyroid gland are very common. It is estimated that as many as 4–7% of the adult population has palpable thyroid en-largement, and up to 10 times this number of individuals have subclinical nodules. The majority of these thyroid nodules are benign, with only a fraction representing ma-lignant disease, which accounts for approximately 1.1% of all cancers annually. The follicular patterned lesion is the most commonly encountered type of thyroid FNA speci-men. Follicular-patterned lesions include benign thyroid nodules (adenomatous nodules, hyperplastic nodules, and follicular adenomas), follicular carcinomas, and the follicular variant of papillary thyroid carcinoma [1] .

Thyroid FNA is the most useful screening test for eval-uating a thyroid nodule and stratifying the risk of malig-nancy. FNA can effectively diagnose benign diseases. It can be used as a screening test for follicular carcinoma and as a diagnostic test for other thyroid carcinomas in-cluding papillary carcinoma, medullary carcinoma, un-differentiated carcinoma, and lymphoma [2–7] . This is due to two main reasons: only capsular and/or vascular invasion, which is only detected by histologic evaluation, can distinguish between follicular adenoma and mini-mally invasive follicular carcinoma, and the precise diag-nosis of many cases of follicular variant of papillary car-cinoma (FVPC) can be challenging even in histologic specimens due to the subtlety of diagnostic nuclear fea-tures [2, 8–10] .

Most thyroid FNA classification schemes have used various terms for the diagnosis of follicular patterned le-sions. These are: follicular lesion, atypical follicular le-sion, and follicular neoplasm. In the past, the American Thyroid Association (ATA) and the American Associa-tion of Clinical Endocrinologists (AACE) had proposed the term ‘indeterminate for malignancy’ for all thyroid FNA specimens which could be diagnosed as follicular lesion (FL), follicular neoplasm (FON), and suspicious for malignancy and recommend surgery for all patients, but Cooper et al. [13, 14] discussed the expended Bethesda classification, which eliminates the term indeterminate, so this outdated recommendation has been replaced by the revised ATA Management Guidelines [11–14] .

The NCI Thyroid FNA State of the Science Confer-ence held in 2007 proposed a six-tiered risk-based clas-sification scheme for the reporting of thyroid FNA spec-imens. In this classification system, commonly referred

to as The Bethesda System for Reporting Thyroid Cyto-pathology (TBSRTC), the indeterminate category was divided into three: atypia of undetermined significance (AUS)/follicular lesion of undetermined significance (FLUS), suspicious for follicular or Hürthle cell neo-plasm (SFON)/follicular or Hürthle cell neoplasm (FON), and suspicious for malignancy. The former group can benefit from RFNA to achieve a more definitive diagno-sis and the latter should undergo surgical excision for histologic characterization (adenoma vs. carcinoma). Because of the potential to overuse the AUS/FLUS cate-gory, it should be less than 7% of all thyroid FNA inter-pretations [15, 16] .

Materials and Methods

The study cohort included 4,284 cases from the MoH Ankara Training and Research Hospital. All thyroid biopsy reports were retrieved from the computerized pathology files for the period between January 2007 and July 2011. All available follow-ups in-cluding RFNA and surgical pathology reports were obtained. For this study, all cases classified as ‘atypical’ and ‘rule out follicular neoplasm’ were included in the AUS/FLUS category (n = 3,903), whereas cases classified as ‘suspicious for’ or ‘consistent with fol-licular or Hürthle cell neoplasm’ (n = 381) were included in the SFON diagnostic category.

Specimen Preparation All cases underwent biopsy under ultrasound guidance by a

radiologist. Three FNA passes were made in the different areas of the nodule using a 25-gauge needle attached to a 10-ml syringe. In each case, air-dried smears were made for MGG staining. No on-site evaluation was performed.

Adequacy Criteria We used adequacy criteria as defined by Geollner et al. [17] , i.e.

6–10 clusters of well-preserved follicular cells with 10–20 cells per group on two different slides.

Cytomorphologic Criteria A diagnosis of AUS/FLUS was made when, at the same time,

the cytologic or architectural atypia was more than would be ac-ceptable within the ‘benign’ category or the specimen showed mostly features of a benign thyroid nodule but also contained one or more of the following features: focal nuclear overlapping and crowding, a sparsely cellular aspirate with a predominance of mi-crofollicles, intranuclear grooves without other features of papil-lary carcinoma and nuclear pleomorphism (enlarged nuclei with hyperchromasia). The specimens revealing a monotonous popu-lation of follicular cells arranged in cohesive groups with nuclear overlapping and crowding, microfollicle formation in a back-ground of scant colloid or containing exclusively oncocytic fol-licular/Hürthle cells arranged in sheets, and cohesive groups with transgressing blood vessels were classified as suspicious for or consistent with follicular or Hürthle cell neoplasm (SFON).

FNA of Follicular Patterned Lesions of the Thyroid

Acta Cytologica 2012;56:361–369 363

Results

During the period between January 1, 2007, and July 1, 2011, 14,628 thyroid FNAs were reported at the MoH Ankara Training and Research Hospital on 12,238 pa-tients; on average, 1.2 FNAs were performed per patient. Only 3,817 patients (31%) had histologic follow-up.

These FNAs were reported as follows: 3,037 (20.8%) ‘nondiagnostic’, 8,504 (58.1%) ‘benign’, 2,348 (16%) ‘atyp-ia of undetermined significance’, 381 (2.6%) ‘suspicious for follicular/Hürthle cell neoplasm’, 120 (0.8%) ‘suspi-cious for malignancy’, and 238 (1.6%) ‘malignant’. Of 4,868 cases, 1,074 (22.1%) were ‘inadequate’, 2,736 (56.2%) were ‘negative’, 793 (16.3%) were ‘atypical and rule out FON’, 181 (3.7%) were ‘suspicious for, consistent with FON’, 40 (0.8%) were ‘suspicious for malignancy’, and 44 (0.9%) were ‘malignant’ during the 2007–2009 period ac-cording to the pre-Bethesda system; of 9,760 cases, 1,959 (20.1%) were ‘nondiagnostic’, 5,772 (59.1%) were ‘benign-benign follicular nodule’, 1,555 (15.9%) were ‘atypia of undetermined significance’, 200 (2.1%) were ‘suspicious for a follicular neoplasm’, 80 (0.82%) were ‘suspicious for malignancy’, and 194 (1.33%) were ‘malignant’ during the 2009–2011 period using TBSRTC ( fig. 1–3 ).

For this study, all cases classified as ‘atypical’ and ‘rule out follicular neoplasm’ were included in the AUS/FLUS category (n = 2,348 + n = 1,555 = 3,903; 16%), whereas cases classified as ‘suspicious for’ or consistent with fol-licular neoplasm (n = 181 + n = 200 = 381; 2.6%) were in-cluded in the SFON category.

RFNA was performed in 1,366/3,903 (35%) patients classified as AUS/FLUS. RFNA diagnoses were: benign follicular nodule (835 cases; 61%), AUS/FLUS (322 cases; 23.6%), SFON (138 cases; 10.1%), suspicious for papillary carcinoma (41 cases; 3%), papillary carcinoma (18 cases; 1.3%), and nondiagnostic (12 cases; 0.9%). The malignan-cy rate on surgical specimens in the AUS/FLUS group was 29 and 14% with and without RFNA, respectively, and 26% in cases diagnosed as SFON.

16% 21%

58%

2%2% 1%

Nondiagnosticspecimen

Benign (BFN)

Atypia of undeterminedsignificance

Suspicious for follicular / Hürthle cell neoplasm

Suspicious formalignancy

Malignant

Fig. 1. Distribution of thyroid FNA diagnoses from January 2007 to July 2011. Number of FNAs.

Inadequate

Negative-benign

Atypical and rule outfollicular neoplasm

Suspicious for follicular,consistent with FN


Malignant

56%

16% 22%

4% 1% 1%

Fig. 2. Distribution of thyroid FNA diagnoses from January 2007 to May 2009 (pre-Bethesda). Number of FNAs.

59%

16% 20%

2%2% 1%

Nondiagnosticspecimen

Benign (BFN)

Atypia ofundeterminedsignificance

Suspicious forfollicular/Hürthlecell neoplasm


Malignant

Fig. 3. Distribution of thyroid FNA diagnoses from June 2009 to July 2011 (post-Bethesda). Number of FNAs.

Co

lor v

ersi

on

avai

lab

le o

nlin

e

Co

lor v

ersi

on

avai

lab

le o

nlin

e C

olo

r ver

sio

n av

aila

ble

on

line



The case cohort consisted of 4,284 cases in 3,266 fe-males and 1,018 males; 3,903 cases could be classified as AUS/FLUS and 381 as SFON. Histologic outcome data was available in 1,756/3,903 (45%) cases diagnosed as AUS/FLUS and 243/381 (64%) cases diagnosed as SFON. Surgical excision was performed in 585 patients with and 1,171 without RFNA. One thousand two hundred fifty-three cases (71%) were diagnosed as benign and 228 (29%) as malignant on histologic examination. These diagnoses include:

AUS/FLUS Cases Three thousand nine hundred three cases were includ-

ed in the AUS/FLUS category (2,348 cases initially diag-nosed as atypical, rule out follicular neoplasm and 1,155 cases diagnosed as AUS/FLUS). Two thousand three hun-dred patients were females and 1,603 were males (average age 51 years). The size of the aspirated lesion was deter-mined by ultrasound examination and ranged from 1.2 to 5.4 cm (average size 2.6 cm). The reasons noted for the AUS/FLUS diagnoses were: few atypical cells showing in-

a b c

a b c

Fig. 4. A case diagnosed as AUS/FLUS on initial FNA due to the presence of focal microfollicle formation (MGG-stained air-dried smear) ( a ). An RFNA showing a monolayer sheet of benign appearing follicular cells and mac-rophages was diagnosed as a hyperplastic nodule (MGG-stained air-dried smear) ( b ). Surgical pathology follow-up showing an adenomatoid nodule (HE) ( c ).

Fig. 5. Case diagnosed as AUS/FLUS on initial FNA. The majority of the specimen showed features of nodular goiter; however, one group of follicular cells displayed nuclear elongation with chromatin clearing and intra-nuclear grooves (MGG-stained air-dried smear) ( a ). RFNA demonstrated a monotonous population of follicu-lar cells arranged in cohesive groups showing diagnostic nuclear features of papillary thyroid carcinoma (MGG-stained air-dried smear) ( b ). Surgical pathology follow-up showed an encapsulated follicular patterned lesion exhibiting diagnostic nuclear cytology of papillary carcinoma (HE) ( c ).

Co

lor v

ersi

on

avai

lab

le o

nlin

eC

olo

r ver

sio

n av

aila

ble

on

line



tranuclear grooves with and without nuclear pleomor-phism (854 cases; 22%), increased cellularity (1,236 cases; 32%), few microfollicles/cell groups with nuclear overlap-ping and crowding (986 cases; 25%), and oncocytic fol-licular cells with nuclear pleomorphism and sparse lym-phocytic infiltrate (827 cases; 21%) ( fig. 4–6 ).

RFNA was performed in 1,366/3,903 (35%) patients. RFNA diagnoses included: hyperplastic/adenomatoid nodule/colloid nodule (360 cases), AUS/FLUS (132 cases), SFON (58 cases), suspicious for papillary carcinoma (20 cases), papillary carcinoma (20 cases), and nondiagnostic (6 cases) ( table 1 ; fig. 7 ). No correlation was noted between the size of the lesion and the rate of RFNA. Surgical exci-sion was performed in 585 patients with and 1,171 with-out RFNA. One thousand two hundred fifty-three cases (71%) were diagnosed as benign and 228 (29%) as malig-nant on histologic examination. These diagnoses includ-ed: hyperplastic/adenomatoid nodule (1,089 cases; 62%), chronic lymphocytic thyroiditis (162 cases; 9%), follicular adenoma (277 cases; 16%), follicular variant of papillary carcinoma (130 cases; 7%), and follicular carcinoma (98 cases; 6%). Incidental papillary microcarcinoma was seen in 32 cases where the dominant biopsied nodule was be-nign ( table 2 ; fig. 8 ).

The rate of malignancy in AUS/FLUS cases with and without RFNA was 29 and 14%, respectively. Similarly, the rate of benign histologic diagnosis in cases undergo-ing RFNA was 71% as compared to 86% in cases without RFNA ( table 3 ; fig. 9 ).

SFON Cases Three hundred eighty-one patients were diagnosed as

SFON. Two hundred forty-two patients were females and 139 were males (average age 58 years). The average size of the aspirated nodules was 3.2 cm. Surgical excision was performed in 243 (64%) patients. One hundred seventy-

a b

Fig. 6. A case diagnosed as AUS/FLUS due to the presence of oncocytic follicular cells with nuclear pleomorphism. Notice the lymphocytic infiltrate indicating an ele-ment of chronic lymphocytic thyroiditis (MGG-stained air-dried smear) ( a ). Surgi-cal excision was performed without RFNA which showed an oncocytic follicular nod-ule with marked random nuclear atypia arising in chronic lymphocytic thyroiditis (HE) ( b ).

Table 1. R FNA diagnosis and histopathology follow-up of AUS/FLUS

RFNA Dx S urg Path Dx Total malig-nant, n (%) HN FA PTC/FVPC FCA

HN (n = 360) 46 112 0 0 0 AUS (n = 132) 7 48 14 6 20 (27%) SFON (n = 58) 9 29 9 11 20 (35%) SPTC (n = 20) 2 3 15 0 15 (75%) PTC (n = 9) 0 0 9 0 9 (100%) ND (n = 6) 0 3 3 0 3 (50%)

Total (n = 585) 64 195 50 17 67 (21%)

Dx = Diagnosis; Surg Path Dx = surgical pathology diagno - sis; HN = hyperplastic nodule; FA = follicular adenoma; PTC/FVPC = papillary thyroid carcinoma/follicular variant of papil-lary carcinoma; FCA = follicular carcinoma; SPTC = suspicious for papillary thyroid carcinoma; ND = nondiagnostic.

Table 2. H istopathology follow-up of AUS/FLUS and SFON cases

S urg Path Dx Total ma-lignant, n (%) HN CLT FA PTC/FVPC FCA

AUS/FLUS with RFNA 214

124

195 35 17 52 (29%)

without RFNA 875 38 82 95 81 176 (14%) SFON 92 0 87 40 24 64 (26%)

Total 1,181 162 364 170 122 292 (15%)

Surg Path Dx = Surgical pathology diagnosis; HN = hyperplas-tic nodule; CLT = chronic lymphocytic thyroiditis; FA = follicular adenoma; PTC-FVPC = papillary thyroid carcinoma/follicular variant of papillary carcinoma; FCA = follicular carcinoma.

Co

lor v

ersi

on

avai

lab

le o

nlin

e



HN FA PTC /FVPC FCA

100

80

112

0 07

48

14

69 9 911

2

15

3 3 30 0 0 0 0 0

29

46

60

40

20

0

120

HN(n = 360)

AUS(n = 132)

SFON(n = 58)

SPTC(n = 20)

PTC (n = 9) ND (n = 6)

900875

214

92124

38 0

195

82 8735

9540 17

8124

800

700

600

500

400

300

200

100

0

HN CLT FA PTC/FVPC FCA

AUS/FLUS with RFNA AUS/FLUS without RFNA SFON

Fig. 7. RFNA diagnosis and histopatholo-gy follow-up of AUS/FLUS.

Fig. 8. Histopathology follow-up of AUS/FLUS and SFON cases.

Malignant Benign

AUS withRFNA

Without RFNA SFON

100%

80%

60%

40%

20%

0%

29%

71%

14%

26%

74%

86%

RFNA No RFNA SFON

Surgicalexcision

Benign Malignant

1,500

1,000

500

585

243

533

995

17952

17664

1,171

0

Fig. 9. Rates of surgery and histologic follow-up of thyroid nod-ules diagnosed as AUS/FLUS with and without RFNA.

Fig. 10. Surgical pathology follow-up of AUS and SFON cases.

Co

lor v

ersi

on

avai

lab

le o

nlin

e C

olo

r ver

sio

n av

aila

ble

on

line

Co

lor v

ersi

on

avai

lab

le o

nlin

e

Co

lor v

ersi

on

avai

lab

le o

nlin

e



nine cases (74%) were diagnosed as benign and 64 (26%) as malignant ( table 4 ; fig. 10 ). The histologic diagnoses included hyperplastic/adenomatoid nodule (92 cases; 38%), follicular adenoma (87 cases; 36%), FVPC (40 cases; 16%), and follicular carcinoma (24 cases; 10%). Incidental papillary microcarcinoma was found in 18 cases (12 be-nign nodules and 6 follicular adenomas).

Discussion

FNA has proven to be the most effective means to as-sess thyroid nodules, having been shown to be a sensitive and specific test as well as safe for patients requiring sur-gery, compared to other diagnostic measures. Despite the high prevalence of nodules, malignancy is seen in only 5% of the nodules on surgical excision [17] . It can effec-tively diagnose most benign and malignant follicular and C-cell-derived tumors with 90–100% accuracy [2, 4, 10, 18] . However, the follicular patterned lesions of the thy-roid are often diagnosed as either atypical or indetermi-nate and the reported rate of malignancy in such cases ranges between 20 and 30% [7, 19–22] . The main reason for this is that lack of well-defined cytomorphologic cri-teria and other factors which test the limitations of pre-cise diagnosis. The diagnosis of follicular adenoma and carcinoma is based upon only demonstration of capsular and/or vascular invasion rather than nuclear features, and identification of the FVPC can be challenging in his-tologic specimens [7, 8, 23] .

Nowadays, most pathologists and clinicians agree that placing all follicular patterned thyroid lesions which do not fit into the benign cytologic category into one diag-nostic category of ‘indeterminate’ will lead to unneces-sary surgeries [24] .

The 2007 NCI Thyroid FNA Conference proposed a six-tiered system for classifying thyroid FNA specimens. This was devised based upon suggestions recorded on a dedicated website, audience response at the meeting, and a thorough literature review. According to this proposal, the indeterminate category was split into three: AUS/FLUS, SFON/FON, and suspicious for malignancy. The malignancy risk for each category was found to be differ-ent based upon literature review: 5–10% for the former and 20–30% for the latter category. It was also suggested that the use of the AUS/FLUS diagnosis should be kept to a minimum, i.e. 7% of all thyroid FNA diagnosis. The AUS/FLUS rate in our study is higher than reported in the literature. Although the general outlines of TBSRTC were published by Baloch et al. [15] in 2008, detailed def-initions for the AUS/FLUS category had been available after the publication of the Bethesda System Atlas in 2010. The difference between the rates of AUS/FLUS in our study and in the literature could be depending on this fact [15, 16] .

In one study, the number of cases that fit the criteria for an AUS/FLUS diagnosis (509 cases) ranged from 9 to 12% of all thyroid FNAs. RFNA was performed in 203 (40%) patients. Surgical excision was performed in 88/203 (43%) patients with RFNA and 185/306 (60%) patients without RFNA. The overall malignancy rate for AUS/FLUS diagnosis was 19%, and the malignancy rate was 27% in patients with RFNA as compared to 15% without RFNA [3] .

These results show that the term AUS/FLUS represents a heterogeneous diagnostic category for classifying thy-roid FNA specimens, and its use and follow-up (RFNA and surgical excision) will vary widely among institu-tions [2–4] .

The main questions are how this diagnosis be used to guide clinical management and what a pathologist should recommend; as clearly seen, RFNA is the procedure of choice. In this study, RFNA clearly allowed better selec-tion of patients for surgical excision. In patients who un-derwent repeat thyroid FNA (88 cases), the malignancy rate was 27% on surgical excision as compared to 15% in cases without RFNA (185 cases). In our study, RFNA also allowed better selection of patients for surgical excision. In patients who underwent repeat thyroid FNA (585 cas-es), the malignancy rate was 29% on surgical excision as

Table 3. Rates of surgery and histologic follow-up of thyroid nod-ules diagnosed as AUS/FLUS with and without RFNA

Benign Malignant

AUS/FLUS with RFNA 71% 29% AUS/FLUS without RFNA 86% 14% SFON 74% 26%

Table 4. Surgical pathology follow-up of AUS and SFON cases

Surgical excision Benign Malignant

RFNA 585 533 52 No RFNA 1,171 995 176 SFON 243 179 64



compared to 14% in cases without RFNA (1,171 cases). Though this difference may not appear significant, the total number of patients undergoing surgical excision af-ter RFNA is 41 versus 58% without RFNA. RFNA has been shown to be an effective means for arriving at a more definitive management-based diagnosis in thyroid nodules that were initially diagnosed as indeterminate and nondiagnostic [25–29] .

VanderLaan et al. [2] reviewed 4,691 thyroid FNAs and reported that 512 (10.9%) had a diagnosis of AUS/FLUS; cytologic or histologic outcome data was available for 331 cases (64.6%), 240 of which (72.5%) were benign and 91 (27.5%) of which were malignant. Baloch et al. [25] re-viewed a cohort of 226 patients who underwent RFNA due to initial diagnoses of indeterminate for neoplasm and nondiagnostic specimens. On RFNA the malignancy rate was 48% in indeterminate cases and 51% in nondiag-nostic cases [25] . In a study by Nayar and Ivanovic [30] , RFNA led to a diagnosis of malignancy in 12% of patients diagnosed as indeterminate on initial FNA. According to a recent report by Yassa et al. [4] , 120 nodules classified as ‘atypical cells of undetermined significance’ under-went RFNA which resulted in a definitive diagnosis in 96 (80%) cases.

In light of the results presented in this study as well as those in other studies as mentioned above, it is evident that RFNA has an important role in the management of patients with thyroid nodules diagnosed as AUS/FLUS. On the basis of these collective results, the question aris-es of why the malignant tumors, i.e. papillary carcino-mas, were missed on initial FNA [8, 21, 31] . The most conceivable answer for this is either sampling of the le-sion or the inherent morphology of the lesion [4, 21, 32] . The former is less likely in our study since all cases were biopsied by an experienced radiologist and/or endocri-nologists under ultrasound guidance. However, we be-lieve the latter, i.e. inherent morphology, is the most rea-sonable explanation. Almost all of the papillary carcino-mas found on surgical pathology follow-up (excluding incidental PTC) were FVPC. It is well known that the di-agnosis of FVPC can be challenging in some cases due to a paucity or lack of diffuse and well-defined nuclear fea-tures of papillary carcinoma [9, 20, 33–36] . Therefore, it is not unusual that the initial FNA sample contained few cells or follicular groups which show some but not all of the features of PTC leading to a diagnosis of AUS/FLUS.

In the current study, 381 cases were classified as SFON/FON. As expected with this diagnosis, a majority of pa-tients (243 patients; 64%) underwent surgical excision. The malignancy rate was 26% (64 cases) on surgical exci-

sion and the malignant diagnoses included FVPC (40 cases) and follicular carcinoma (24 cases). These findings are basically similar to those reported by other authors [3, 37–40] .

In this and other reported studies, half or more of the malignant cases diagnosed as SFON/FON were found to be FVPC at surgical excision [36, 41] . As discussed above, a subset of encapsulated FVPC shows a focal rather than diffuse distribution of nuclear features of papillary carci-noma. We believe the sampling of the areas lacking diag-nostic nuclear features is the main reason for the under-diagnosis of these cases as SFON/FON [35, 36, 42] .

However, there may be an increasing tendency among surgical pathologists to diagnose FVPC and the diagnosis may vary from benign to malignant in same cases when shown to thyroid pathology experts [23, 43] . The increase in the number of FVPC in the follow-up of SFON/FON may be related in part to the variability in the histologic diagnosis of this tumor [3] .

Conclusion

Our findings support the NCI recommendation for thyroid FNA specimens initially classified as indetermi-nate being split into three categories: AUS/FLUS, SFON/FON, and suspicious for malignancy. The current data and those reported by others show that the malignancy rates differ between both categories, and the cases diag-nosed as AUS/FLUS are best managed by RFNA for selec-tion of cases that can benefit from surgical excision.

References 1 Faquin WC: Diagnosis and reporting a fol-licular-patterned thyroid lesions by fine nee-dle aspiration. Head and Neck Pathol 2009; 3: 82–85.

2 VanderLaan PA, Marqusse E, Krane JF: Clin-ical outcome for atypia of undetermined sig-nificance in thyroid fine-needle aspirations: should repeated FNA be the preferred initial approach? Am J Clin Pathol 2011; 135: 770–775.

3 Faquin WC, Baloch ZW: Fine-needle aspira-tion of follicular patterned lesions of the thy-roid: diagnosis, management, and follow-up according to National Cancer Institute (NCI) recommendations. Diagn Cytopathol 2010; 38: 731–739.

4 Yassa L, Cibas ES, Benson CB, et al: Long-term assessment of a multidisciplinary ap-proach to thyroid nodule diagnostic evalua-tion. Cancer 2007; 111: 508–516.



5 Somma J, Schlecht NF, Fink D, et al: Thyroid fine needle aspiration cytology: follicular le-sions and the grey zone. Acta Cytol 2010; 54: 123–131.

6 Marcheusky AM, Walts AE, Bose S, et al: Ev-idence based evaluation of the risks of malig-nancy predicted by thyroid fine-needle aspi-ration biopsies. Diagn Cytopathol 2010; 38: 252–259.

7 Baloch ZW, Fleisher S, LiVolsi VA, Gupta PK: Diagnosis of ‘follicular neoplasm’: a gray zone in thyroid fine-needle aspiration cytol-ogy. Diagn Cytopathol 2002; 26: 41–44.

8 Baloch ZW, Livolsi VA: Follicular-patterned lesions of the thyroid: the bane of the pathol-ogist. Am J Clin Pathol 2002; 117: 143–150.

9 Zacks JF, de las Morenas A, Beazley RM, O’Brien MJ: Fine-needle aspiration cytology diagnosis of colloid nodule versus follicular variant of papillary carcinoma of the thy-roid. Diagn Cytopathol 1998; 18: 87–90.

10 Oertel YC, Burman K, Boyle L, et al: Inte-grating fine-needle aspiration into a daily practice involving thyroid disorders: the Washington Hospital Center approach.Diagn Cytopathol 2002; 27: 120–122.

11 Wang HH: Reporting thyroid fine-needle as-piration: literature review and a proposal. Diagn Cytopathol 2006; 34: 67–76.

12 Lewis CM, Chang KP, Pitman M, Faquin WC, Randolph GW: Thyroid fine-needleaspiration biopsy: variability in reporting. Thyroid 2009; 19: 717–723.

13 Cooper DS, Doherty GM, Haugen BR, et al: Management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid 2006; 16: 109–142.

14 Cooper DS, Doherty GM, Haugen BR, etal: Revised American Thyroid Association management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid 2009; 19: 1167–1214.

15 Baloch ZW, Cibas ES, Clark DP, et al: The National Cancer Institute Thyroid Fine Nee-dle Aspiration State of the Science Confer-ence: a summation. Cytojournal 2008; 5: 6.

16 Ali SZ, Cibas ES: The Bethesda System for Reporting Thyroid Cytopathology. New York, Springer, 2010.

17 Goellner JR, Gharib H, Grant CS, Johnson DA: Fine needle aspiration cytology of the thyroid. Acta Cytol 1987; 31: 587–590.

18 Greenblatt DY, Woltman T, Harter J, et al: Fine needle aspiration optimizes surgical management in patients with thyroid cancer. Ann Surg Oncol 2006; 13: 859–863.

19 Schlinkert RT, van Heerden JA, Goellner JR, et al: Factors that predict malignant thyroid lesions when fine-needle aspiration is ‘suspi-cious for follicular neoplasm’. Mayo Clin Proc 1997; 72: 913–916.

20 Goodell WM, Saboorian MH, Ashfaq R: Fine-needle aspiration diagnosis of the fol-licular variant of papillary carcinoma. Can-cer 1998; 84: 349–354.

21 Renshaw AA: Follicular lesions of the thy-roid. Am J Clin Pathol 2001; 115: 782–785.

22 Mijovic T, Rochon L, Gologan O, et al: Fine-needle aspiration biopsies in the manage-ment of indeterminate follicular and Hurth-le cell thyroid lesions. Otolaryngol Head Neck Surg 2009; 140: 715–719.

23 Elsheikh TM, Asa SL, Chan JK, et al: Interob-server and intraobserver variation among experts in the diagnosis of thyroid follicular lesions with borderline nuclear features of papillary carcinoma. Am J Clin Pathol 2008; 130: 736–744.

24 Baloch ZW, LiVolsi VA, Asa SL, et al: Diag-nostic terminology and morphologic criteria for cytologic diagnosis of thyroid lesions: a synopsis of the National Cancer Institute Thyroid Fine-Needle Aspiration State of the Science Conference. Diagn Cytopathol 2008; 36: 425–437.

25 Baloch Z, LiVolsi VA, Jain P, et al: Role of re-peat fine-needle aspiration biopsy (FNAB) in the management of thyroid nodules. Diagn Cytopathol 2003; 29: 203–206.

26 Renshaw AA: An estimate of risk of malig-nancy for a benign diagnosis in thyroid fine-needle aspirates. Cancer Cytopathol 2010; 118: 190–195.

27 Renshaw AA: Does a repeated benign aspi-rate change the risk of malignancy after an initial atypical thyroid fine-needle aspira-tion? Am J Clin Pathol 2010; 134: 788–792.

28 Renshaw AA: Non-diagnostic rates for thy-roid fine-needle aspiration are negatively correlated with positive for malignancy rates. Acta Cytol 2011; 55: 38–41.

29 Flanagan MB, Ohori NP, Carty SE, Hunt JL: Repeat thyroid nodule fine-needle aspira-tion in patients with initial benign cytologic results. Am J Clin Pathol 2006; 125: 698–702.

30 Nayar R, Ivanovic M: The indeterminate thyroid fine-needle aspiration: experience from an academic center using terminology similar to that proposed in the 2007 Nation-al Cancer Institute Thyroid Fine Needle As-piration State of the Science Conference. Cancer Cytopathol 2009; 117: 195–202.

31 Renshaw A: Misclassification of cytologic diagnoses in patients with follicular lesions or follicular neoplasms of the thyroid gland. Cancer 2003; 99: 318–319, author reply 319–320.

32 Renshaw AA: Accuracy of thyroid fine-nee-dle aspiration using receiver operator char-acteristic curves. Am J Clin Pathol 2001; 116: 477–482.

33 Renshaw AA: Focal features of papillary car-cinoma of the thyroid in fine-needle aspira-tion material are strongly associated with papillary carcinoma at resection. Am J Clin Pathol 2002; 118: 208–210.

34 Mesonero CE, Jugle JE, Wilbur DC, Nayar R: Fine-needle aspiration of the macrofollicu-lar and microfollicular subtypes of the fol-licular variant of papillary carcinoma of the thyroid. Cancer 1998; 84: 235–244.

35 Baloch ZW, Gupta PK, Yu GH, Sack MJ, LiVolsi VA: Follicular variant of papillary carcinoma: cytologic and histologic correla-tion. Am J Clin Pathol 1999; 111: 216–222.

36 Jain M, Khan A, Patwardhan N, Reale F, Saf-ran M: Follicular variant of papillary thyroid carcinoma: a comparative study of histo-pathologic features and cytology results in 141 patients. Endocr Pract 2001; 7: 79–84.

37 Tulecke MA, Wang HH: ThinPrep for cyto-logic evaluation of follicular thyroid lesions: correlation with histologic findings. Diagn Cytopathol 2004; 30: 7–13.

38 Clary KM, Condel JL, Liu Y, Johnson DR, Grzybicki DM, Raab SS: Interobserver vari-ability in the fine needle aspiration biopsy diagnosis of follicular lesions of the thyroid gland. Acta Cytol 2005; 49: 378–382.

39 Carling T, Udelsman R: Follicular neoplasms of the thyroid: what to recommend. Thyroid 2005; 15: 583–587.

40 Smith J, Cheifetz RE, Schneidereit N, Berean K, Thomson T: Can cytology accurately pre-dict benign follicular nodules? Am J Surg 2005; 189: 592–595, discussion 595.

41 Yang GC, Liebeskind D, Messina AV: Diag-nostic accuracy of follicular variant of papil-lary thyroid carcinoma in fine-needle aspi-rates processed by ultrafast Papanicolaou stain: histologic follow-up of 125 cases. Can-cer 2006; 108: 174–179.

42 Kurian EM, Dawlwtt M, Wang J, et al: The triage efficacy of fine needle aspiration bi-opsy for follicular variant of papillary thy-roid carcinoma using the Bethesda reporting guidelines. Diagn Cytopathol 2011, DOI: 10.1002/dc.21718.

43 Renshaw AA, Gould EW: Why there is the tendency to ‘overdiagnose’ the follicular variant of papillary thyroid carcinoma. Am J Clin Pathol 2002; 117: 19–21.

http://dx.doi.org/10.1002%2Fdc.21718

Documents

Fine-Needle Aspiration of Follicular Patterned Lesions of the Thyroid: Diagnosis, Management, and Follow-Up according to Thyroid Bethesda System