Head Shodayu

Takashima, Tomiyama,

Noriyuki

MD MD

Fumio Matsuzuka, Takahiro Kozuka,

#{149} #{149}

Nodules

Thyroid

Thyroiditis:

terms:

273.12985 273.37 273.292 Radiology

Biopsies,

#{149} Thyroid, #{149}Thyroid,

1992;

technology, 273.126, neoplasms, 273.34, 273.36,

US, 273.12985

#{149} Thyroiditis,

Tomofumi

#{149}

Associated

Assessment

Sonographic findings in 109 thyroid nodules in 104 patients with Hashimoto thyroiditis were retrospectively analyzed. Sixty patients underwent ultrasound (US)-guided fine-needle aspiration (FNA) biospy (65 nodules, 24 palpable, 27 surgically confirmed), 14 of whom also underwent palpation-guided FNA (14 nodules, all palpable, all surgically confirmed). Forty-four patients (44 nodules, 25 palpable) underwent diagnostic US followed up with surgery. US-guided FNA helped differentiate between benign (n = 11) and malignant (n = 16) diseases (sensitivity, 100%). Two papillary carcinomas were falsely diagnosed as Hashimoto thyroiditis with palpation-guided FNA. Eight malignancies were not palpable, and correct diagnosis was obtamed with US-guided FNA. Hyperechoic nodules were usually benign, and isoechoic nodules had a low frequency (13%) of malignancy. Most carcinomas were found in hypoechoic masses, and almost all lymphomas were found in markedly hypoechoic masses. Lymphoma was indistinguishable from pseudotumor in Hashimoto thyroiditis or adenomatous hyperplasia at US. US-guided FNA is helpful in making the histologic diagnosis in thyroid nodules associated with Hashimoto thyroiditis. US-guided FNA is clinically beneficial because it can help select patients who need surgery, avoiding unnecessary surgery for patients with nonneoplastic disorders. Index

MD MD

H

Nagareda,

with with

Hashimoto

US’

thyroiditis is an autodisease that predominantly affects elderly women (1-6). Patients with this disease usually present with a diffuse thyroid goiter with a firm consistency (1-3). However, these patients often develop an asymmetric swelling or a local lump, for which unnecessary surgical intervenlion may be instituted due to the inaccuracy of preoperative needle biopsy and a suspicion of carcinoma (1-6). Histopathologic changes in the thyroid gland with this disorder indude lymphocyte aggregates with germinal centers and an oxyphilic change of the epithelium (2-4). Various degrees of follicular atrophy are also seen. At sonography, the thyroid gland is diffusely enlarged and of inhomogeneous echogenicity, with a general decrease in echogenicity (7-10). Although a high frequency of coexistent nodules in the thyroid gland has been documented, to our knowledge, all reports lack pathologic confirmation (7,9). Primary thyroid lymphoma is

(US)-guided fine-nee(FNA) biopsy is widely make the histopathologic diagnosis of thyroid nodules because it is safe and allows biopsy of nonpalpable nodules and more accurate Selection of a biopsy site in palpable nodules (12,13). There are some reports of palpation-guided FNA of nodules associated with Hashimoto thyroidilis (14,15); however, to our knowledge, US-guided FNA has not been performed in this patient populalion. In this study, we evaluated the role of US-guided FNA in the evalualion of thyroid nodules associated with Hashimoto thyroidilis and the sonographic findings.

strongly

characteristics of the nodules. of the 60 patients presented

ASHIMOTO immune

associated

with

Hashimoto

thyroiditis, and patients with Hashimoto thyroiditis have a high risk of malignant lymphoma (5,11). Thyroid lymphoma is usually large at diagnosis, and patients frequently complain of obstructive symptoms (10,11). Limited surgery is usually required in these patients to obtain enough tissue for specific typing of the lymphoma. However, extensive surgery is not the treatment of choice for patients with this disorder (10). Therefore, differentiation of lymphoma from other thyroid nodules without surgery is an important clinical issue.

185:125-130

27; accepted

May

28. Address

Radiology

MD

I From the Departments of Radiology (ST., NT., T.K.) and Pathology (T.N.), Osaka University Medical School, 1-1-50 Fukushima, Fukushima-ku, Osake 553,Japan; and Department of Surgery, Kuma Hospital, Kobe,Japan (F.M.). Receivedjanuary 13, 1992; revision requested February 26; revi-

sion received April C RSNA, 1992

Neck

and

reprint

requests

Ultrasound dle aspiration used to help

MATERIALS

AND

From January 178 patients

1989 to September

with

clinically

cally proved Hashimoto went US of the thyroid University

Hospital.

onstrated

METHODS

in the

1991,

or pathologi-

thyroiditis undergland at Osaka Nodules

thyroid

were

gland

dem-

of 75 of

these 178 patients (42%) at US. Of these 75 patients, 60 subsequently underwent USguided FNA to determine the histologic

goiter patients 11 had

of various

Fifty-eight with a thyroid

consistencies.

Of the 60

who underwent US-guided FNA, two or more nodules and 49 had

one nodule. In patients with two or more nodules, aspiration was performed for every nodule that appeared different from the others. When the nodules appeared similar, only one was aspirated. Thus, 65 nodules

were

aspirated.

The aspirated to 4.8 cm

greatest nodules,

nodules

(mean, diameter.

ranged

from

0.5

1 .9 cm) in transverse Of the 65 aspirated

24 (mean,

2.4

cm)

were

palpable

and 41 (mean, 1.5 cm) were not. The diagnosis was surgically confirmed in 27 of the 65 nodules

(25 patients).

Of these,

seven

nonneoplastic lesions were surgically resected (either with open biopsy or thyroidectomy) because of a coexistent para-

to ST.

Abbreviation:

FNA

=

fine-needle

aspiration.

Echogenicity

and Histologic

Diagnosis

in 109 Lesions

in 104 Patients

with

Thyroiditis

Hashimoto

Histolo gic Diagnosis Pseudotumor No. of

Echogenicity

Lesions

Markedly hypoechoic Hypoechoic Isoechoic Hyperechoic Not interpretable Total

Cystic

Follicular

Follicular

Papillary

Anaplastic

Malignant

Lesion

Adenoma

Carcinoma

Carcinoma

Carcinoma

Lymphoma

55 (42) 22 (12) 16(10)

0 1

4 3 (2)

0

13 (4)

0

3 (3)

0

109 (71)

1

0 0

1 (1) 6 (6)

0 1 (1)

2(1)

1

1(1)

0

5 (2)

0

0

0

0

0

2 (2)

0

1

10 (10)

14 (5)

was not interpretable in three lesions (two papillary are numbers of surgically resected lesions.

Note.-Echogenicity Numbers in parentheses

Adenomatous

Hyperplasia

30 (30) 1 (1)

1 (1)

Thyroiditis

Undetermined

9 (8) 3 (2)

10 (3) 6

1 1

0

6(5)

6(3)

0

0

5 (2)

3

0

0

1 (1)

0

0

31 (31)

24 (18)

and one adenomatous

carcinomas

in

Hashimoto

25 (6)

hyperplasia)

2

of extensive

because

calcification.

tumor (n = 1), an associated carcinoma (n = 1), or because there was no regression in their size (regardless of suppressive therapy) (n = 5). thyroid

papillary

Palpation-guided formed

FNA was also per-

in 14 of the 65 nodules

gery

by

a head

and

neck

before

specialist.

surThe

remaining 51 nodules were initially aspirated under US guidance because they

were nonpalpable or located in crucial anatomic areas or because physicians considered that US could better for biopsy. US-guided FNA 22-gauge

needle

ray probe Tokyo).

lesions they

target

Medical was

not

less than represent

(15). Specimens

nodules

was performed with a 5-MHz

(Yokogawa Sampling

cystic because

help

were

with linear-ar-

in

in diameter follicles

not immediately

alyzed to determine whether adequate for diagnosis. Each stained with both Papanicolaou

a.

Systems,

performed

0.5 cm dilated

a

an-

they were section was and May-

Figure

1.

Isoechoic

mass

in follicular

defined, isoechoic mass (arrowheads) gland. A = common carotid artery, scan

obtained

during

the hypoechoic during tologic

US-guided

T

adenoma.

(a) Transverse

with

in the right lobe of the enlarged thyroid V = internal jugular vein. (b) Transverse

=

FNA

mass. Adenomatous

a halo

trachea, shows the

hyperplasia

surgery for coexistent papillary examination. T = trachea.

carcinoma

needle

US scan

shows

as a hyperechoic

was diagnosed. and

proved

a 1.7-cm,

area

This nodule

(arrow)

well-

US

within

was resected

to be follicular

adenoma

at his-

Ciemsa methods. Nodules were classified as malignant, suspicious for malignancy, cellular

atypia,

material

benign,

for cytologic

and

insufficient

interpretation

(16). were

Lesions in the first three categories considered malignant. All specimens were reviewed by one pathologist (TN.), and a specific histologic diagnosis was made. In general, repeat aspiration

was

recommended

if the

origi-

nal aspirate material was inadequate for diagnosis, cellular atypia was diagnosed, or the nodule was clinically suspected of being malignant regardless of a benign cytologic diagnosis. However, the decision to perform repeat FNA was made by the patient’s physician. When repeat aspiration was performed, the most positive diagnosis was adopted as the final cytologic

diagnosis.

126

before

character of the margin (regular or irregular, well-defined or ill-defined); and the number of lesions (single or multiple). Lesions that had less echogenicity than that of the thyroid gland but more than

malignant

that of the adjacent

surgery.

#{149} Radiology

Of

the

26 patients

lymphoma

or well-differenti-

ated follicular carcinoma Thus, the sonographic of 109 lesions of various teristics

in

104

patients

was

suspected. findings in a total histologic characwith

Hashimoto

thyroiditis were retrospectively reviewed. There were 93 women and 1 1 men, ranging in age from 22 to 85 years (mean, 53 years). Attention was also paid to the presence or absence of obstructive symptoms such

as dyspnea,

dysphagia,

muscle

tamed

either

sector

scanner

of the

with with

thyroid

gland

a real-time a 10-MHz

ob-

were

US scans,

microcalcification

hyperechoic

with

discrete

to that of the thyroid

(17).

defined

acoustic

as

without

Dense

calcifica-

as hyperechoic

areas

shadowing.

mechanical transducer

Aloka, Tokyo) or a slow mechanical arc scanner with a 7.5-MHz annular-array transducer housed in a water bath (SSD-520; Aloka). Special attention was paid to the echogenicity of the masses edly hypoechoic, hypoechoic, hyperechoic, or anechoic); absence of a cystic portion, rim surrounding the mass), (microcalcification or dense

Le-

was defined

particles

shadowing

was

RESULTS

(SSD-125;

relative

sternocleidomastoid as hypoechoic.

defined

sions that had less echogenicity than that of the thyroid gland and equal or less echogenicity than that of the muscle were classified as markedly hypoechoic. Echogenicity of the residual gland with Hashimoto thyroiditis was also evaluated. On

tion scans

were

acoustic

or hoarse-

ness. US

In addition, sonographic and pathologic findings were compared in 44 consecutive patients with Hashimoto thyroiditis who had surgically confirmed thyroid nodules (44 nodules, 25 palpable); 18 patients had nonneoplastic thyroid nodules (three had pseudotumor in Hashimoto thyroiditis and 15 had adenomatous hyperplasia), and 26 had primary thyroid lymphoma. All 44 patients were treated in Kuma Hospital between January 1985 and December 1990, and none underwent US-guided FNA

with thyroid lymphoma, 15 (58%) presented with a diffuse swelling of the thyroid gland, and 11 presented with a local lump. Surgery was performed in 18 patients with nonneoplastic masses because

gland

(mark-

isoechoic, presence or halo (transonic or calcification calcification);

Palpation-

and

US-guided

FNA

Three of the 65 nodules (5%) in which US-guided FNA was performed yielded insufficient aspirates for

diagnosis.

Repeat

aspiration

was

performed in one of these nodules and in three other nodules. At repeat aspiration, the cytologic classification October

1992

_

.

-SCM-

-

-

-. .-

-

+ :

l

a..

i,.-. a.

b.

Figure

2.

Hypoechoic

cm, hypoechoic

mass

mass

in adenomatous

without

a halo

hyperplasia.

(a) Transverse

US scan

shows

a 1.8-

in the left lobe of the enlarged thyroid gland. The nodule SCM = sternocleidomastoid muscle, V = internal jugular

has a well-defined, irregular margin. vein. (b) Transverse US scan obtained during US-guided biopsy demonstrates row) in the lesion. Adenomatous hyperplasia was diagnosed. This nodule ing surgery for coexistent parathyroid adenoma, and histologic verification

a needle was removed was obtained.

(ardur-

-

-

:-

-

-

A

, -

a.

b.

Figure

3.

0.6-cm,

tumor

respectively). The three patients with lymphoma had stage I disease, which was histologically proved with surgical biopsy in all. Of the 10 patients with papillary carcinoma, two (with 0.9-cm and 2.2-cm carcinomas, respectively) had extensive metastases to the cervical nodes; in both patients, a correct histologic diagnosis was made with palpation-guided FNA of these nodes, not of the primary tumors. At surgery, a 0.9-cm carcinoma was found to have invaded the trachea in one of these two patients. Of the 14 nodules in which palpation-guided FNA was performed, correct diagnosis was obtained in 12 (six benign and six malignant lesions); however, two papillary carcinomas were incorrectly diagnosed as Hashimoto thyroiditis as a result of a sampling error. Surgery was not performed in the remaining 35 patients with 38 nodules. Cytologic findings in these 38 nodules were as follows: insufficient material for diagnosis (n = 2), benign lesion (n = 35; 18 pseudotumor in Hashimoto thyroiditis, 10 adenoma, six adenomatous hyperplasia, and one cystic lesion), and cellular atypia (n = 1), in which a well-differentiated follicular carcinoma was suspected and findings from repeat FNA revealed follicular adenoma.

Markedly

markedly

hypoechoic

hypoechoic sonogram

in papillary

mass

without a halo and irregular margin. A

has a well-defined

(b) Transverse mass.

mass

obtained

during

US-guided

carcinoma.

in the right =

common biopsy

(a) Transverse

US scan

shows

lobe of the thyroid gland. carotid artery, T = trachea. shows

needle

(arrow)

This

in the

a

Clinical, Findings

US,

None

of the

lesions

from

insufficient

material

for

diagnosis to benign tumor (adenomatous hyperplasia) in one nodule, from atypical cells to malignant tumor

(papillary carcinoma) in one, from benign lesion (pseudotumor in Hashimoto thyroiditis) to malignant lymphoma in one, and from atypical cells (suspected

follicular

carcinoma)

to

benign lesion (follicular adenoma) in one. Of the 14 nodules in which palpalion-guided FNA was performed, two (14%) did not have sufficient aspirates for diagnosis. Repeat aspiration provided benign cytologic diagnoses in both nodules (one of pseudotumor in Hashimoto nomatous

thyroiditis hyperplasia).

and

one

of ade-

Surgical confirmation was obtained in 27 nodules; 10 were papillary carcinoma, five were non-Hodgkin lymphorna, four were follicular adenorna, Volume

185

Number

#{149}

1

four were adenomatous hyperplasia, three were pseudotumor in Hashimoto thyroiditis, and one was anaplastic carcinoma. US-guided FNA enabled differentiation between benign and malignant disease in all 27 surgically

confirmed

nodules.

How-

ever, four surgically confirmed follicular adenomas were incorrectly diagnosed with US-guided FNA, one as pseudotumor in Hashimoto thyroiditis, two as adenomatous hyperplasia, and one as H#{252}rthle cell tumor. In the remaining 23 lesions, a specific histologic diagnosis was correctly established with US-guided FNA. Thirteen of the 27 surgically confirmed nodules were not palpable. Eight of these 13 nodules were malignant; five were papillary carcinoma (0.5-1.2 cm in diameter), and three were non-Hodgkin lymphoma (1.5 cm, 2.0 cm, and 2.2 cm in diameter,

Histologic

61 patients

presented

symptoms. toms were

changed

and

with

with

Conversely, documented

these sympin 13 of the

31 patients (42%) with thyroid phoma, one of the 10 patients with papillary carcinoma, and tient

(100%)

noma. Of the mal-sized

with

benign

obstructive

anaplastic

lym(10%) the pa-

carci-

104 patients, two had a northyroid gland, and the re-

maining 102 inhomogeneous

had

an enlarged echogenicity.

gland Echo-

of

genicity of the thyroid gland was greater than that of the muscle in 92 patients and equal to or less than that of the muscle in the other 12 patients. Histologic diagnoses were not obtamed in two of the 109 nodules because of insufficient aspirates. Of the 107 nodules, was found Hashimoto

malignant lymphorna in 31, pseudotumor in thyroiditis in 25, adeno-

matous hyperplasia in 24, follicular adenoma in 14, papillary carcinoma in 10, cystic lesion in one, follicular carcinoma in one, and anaplastic carcinoma in one. Radiology

127

#{149}

US

scans

of three

of these

107

nod-

ules were not interpretable, except for findings of calcification, because of extensive calcification within the lesions; two contained papillary carcinoma, and one proved to be adenomatous hyperplasia at histologic examination. Of the 109 nodules in which sonographic findings were analyzed, 13 were hyperechoic, 16 were isoechoic, 22 were hypoechoic, and 55 were markedly hypoechoic; images of three nodules were not interpretable (Table). All hyperechoic lesions were benign. Two of the 16 isoechoic nodules (12%) (Fig 1) were malignant. Of the 22 hypoechoic lesions (Fig 2), eight (36%) were malignant, 13 were benign (59%), and one had nondiagnostic findings. Of the 55 markedly hypoechoic lesions, 31 (56%) were malignant, 23 (42%) were benign (Figs 3-?), and one had nondiagnostic findings. Pathologic study of the adenomatous

hyperplasias

creased although

with

echogenicity there was

licular

size,

markedly

to small

cytes

within

were

of underlying

were

Irregular

identified

tumor

in five

of the

(36%),

eight

nomas

(80%),

in pseudotu-

margins

were

28 of the (90%),

seen

adenomas

10 papifiary

carci-

31 malignant

11 of the

24 ade-

nomatous hyperplasias (46%), 18 of the 25 pseudotumors in Hashimoto thyroiditis (72%), the cystic lesion (100%), and the anaplastic carcinoma (100%). Ill-defined tumor margins were seen in five of the 14 follicular adeno-

128

lesions

Radiology

#{149}

carotid

were

seen

sonogram

in the

A

=

right

common

biopsy

lobe

of the

carotid

shows

needle

artery, (arrow)

artery.

-

.

.

,

of the in five

31 of

b.

adjacent

to the nodule;

(Hematoxylin-eosin

these findings

stain;

are compatible with magnification, x 5.)

original

those

of Hashimoto

the 14 follicular adenomas (36%), five of the 10 papillary carcinomas (50%), 23 of the 31 malignant lymphomas (74%), 12 of the 24 adenomatous hy-

cation

perplasias pseudotumors

the 31 malignant and the one case

(50%),

nine of the in Hashimoto

the

25 thyroidi-

anaplastic

detected

carci-

in eight

of the

14 follicular adenomas (57%), one the 10 papillary carcinomas (10%),

lymphomas (3%), 12 of the 24 adenomatous hyperplasias (50%), 16 of the 25 pseudotumors in Hashimoto thyroiditis (64%), the cystic lesion (100%), and the anaplastic carcinoma (100%). Total involvement of the thyin three (10%).

(a) Transverse border

Figure 5. Markedly hypoechoic mass in adenomatous hyperplasia. (a) Transverse US scan shows multiple, markedly hypoechoic masses in the right lobe of the enlarged thyroid gland. Neither a calcification nor a halo is seen. US-guided biopsy was not performed in this case. Malignant lymphoma was suspected, and right lobectomy was performed in another hospital. T = trachea. (b) Histologic specimen shows adenomatous nodule (arrows) consisting mainly of normal-sized to small follicles and scattered lymphocytes. A poorly developed capsule is also seen. Severe follicular atrophy and diffuse lymphocytic infiltration are seen in the area

tis (36%), and noma (100%). A halo was

mas (36%), six of the 10 papillary adenomas (60%), one of the 31 malignant

was seen lymphomas

lymphoma.

also

pathologic

14 fofficular of the

lymphomas

Single

in malignant

,

-

mor in Hashimoto thyroiditis that appeared as markedly hypoechoic areas at US; one of massive focal lymphocyte aggregates associated with frequent germinal centers (Fig 6b) and the other of local severe oxyphilic changes of the follicular epithelium with lymphocytic accumulation (Fig Th).

roid gland malignant

mass

hypoechoic mass with a well-defined was discovered at routine sonography. US scan obtained during US-guided

a.

identified. Two types changes

hypoechoic

fol-

were more predominant than follicles (Fig 5b). A poorly develcapsule and scattered lymphonodules

Markedly

a 1.5-cm, markedly gland. This tumor T = trachea. (b) Transverse in the mass. A = common

that in fol-

licles large oped

the

4.

shows thyroid

de-

revealed a variation

normal-sized

Figure

of 12

of the 24 adenomatous hyperplasias (50%), three of the 25 pseudotumors in Hashimoto thyroiditis (12%), and the follicular carcinoma (100%). Cystic change was detected in one of the 14 follicular adenomas (7%), two of the 31 malignant

of the (38%),

lymphomas

(6%),

nine

24 adenomatous hyperplasias two of the 25 pseudotumors

Hashimoto cystic lesion

Sonography

thyroiditis (100%).

depicted

(8%),

and

in the

in four

hyperplasias

licular

adenomas

papillary

calcifi-

24 adenomatous one

(7%),

carcinomas

of the

three (30%),

14 fol-

of the three

10 of

lymphomas (10%), of anaplastic carcinoma (100%). All areas of calcification in lymphoma were in the margin of the tumor. Because this study was retrospectively performed, pathologic analysis

in whole

of the nodule nodules with were

contiguous

sections

was not possible. dense calcification

surgically

resected;

Six at US

pathologic

examination confirmed US findings in all six (two adenomatous hyperplasias, three papillary carcinomas, and one anaplastic carcinoma). Psammoma bodies were also identified in one papillary carcinoma. In the nodule

with

cation dense

of the (17%),

thyroiditis.

coexistent

anaplastic

was

located papillary

carcinoma,

in the

calcifi-

areas

with

carcinoma.

October

1992

that the thyroid nodules were behind the enlarged thyroid,

had

a firm

ent

Hashimoto

suggest

consistency

Thus,

US-guided

useful

FNA

in making

carcinoma

node

cervical

we

is clini-

histologic

noses of thyroid nodules tient population. One of our 10 patients papillary

to coexist-

thyroiditis.

that

cally

due

hidden which

diag-

in this with

occult

showed

metastases

pa-

extensive

and

tracheal

invasion. The presence of occult papillary carcinoma does not influence the survival period (19,20). However, Figure

6.

Markedly

verse US scan shows the right lobe of the formed. Lymphoma

A = common lymphocytic the markedly fication,

hypoechoic

mass

in pseudotumor

in Hashimoto

thyroiditis.

16%

(a) Trans-

the

a 2.3-cm, markedly hypoechoic mass with a partly ill-defined border in enlarged thyroid gland. In this case, only palpation-guided FNA was perwas diagnosed, and right lobectomy was performed in another hospital. artery, T = trachea. (b) Histologic specimen reveals that massive local

carotid aggregates hypoechoic

accompanied area

frequently

seen

with germinal

at sonography.

centers

were responsible

(Hematoxylin-eosin

stain;

original

of these lymph

for

mass

magni-

roid and

some

and

a clinically

nonpalpable

“.

--..-

.

1

thyroid

fern eter

.

t

carcinoma

s..-

k: b.

Figure dinal

7. Markedly US scan of the

well-defined hospital.

corresponds scattered

margin.

hypoechoic right lobe

This

nodule

(b) Histologic specimen to the area of local lymphocytic infiltrates

was suspicious

in Hashimoto shows a markedly

for lymphoma

demonstrates that the severe oxyphilic changes (arrows). (Hematoxylin-eosin

and

thyroiditis. hypoechoic

(a) Longituarea with

was resected

in another

hypoechoic area seen at sonography of the epithelium associated with stain; original magnification,

x5.)

combined

three

with

of the

dense

calcification

10 papillary

consistency, in

carcinomas

dition, thyroid

(30%). Calcification and focal clusters of psammoma bodies were verified pathologically in two of those three

bility

nodules, and only dense fibrous bands were identified in one. Sonography demonstrated microcalcification in one of the 14 follicular adeno-

tion wise

(7%);

in this

surgery

was

not

performed

patient.

In our study, sensitivity of palpation in the detection of thyroid nodules was 37% . Some nodules were incorrectly diagnosed with palpationguided FNA because of sampling error. Detection of a nodule with palpalion is dependent on nodule size, Volume

185

Number

#{149}

I

of a nodule.

Christensen

Tibblin (18) noted that of clinical examination

Most were ably

of the

nodules

in their

colloid goiters, which had a soft consistency.

of our

(44%)

were

16 thyroid relatively

palpated

histologic nancies

diagnoses were obtained

guided

FNA.

in an was

a

roiditis avoid priate

with a delay therapy,

series presumAlthough

and,

(1 cm

or

US-guided in initiating influence

and

may

FNA should an approthe surgical

reduce

in our study,

with thyroid ent Hashimoto sidual thyroid

recur-

most

patients

of the

with

thyroid

Hashimalignant

gland

(5).

lymphomas

with US, and correct histologic diagnosis was established with US-guided FNA in the early stage of the lymphoma. Because major surgery is not a suitable method

of treatment

rate

therefore,

lymphoma nosis with

The low sensitivity of palpation and the sampling errors of palpationguided FNA must be due to the fact

the

lymphoma have coexistthyroiditis in the regland (11), and a small

We found three small ( < 2.5 cm in diameter)

at palpaor were

in these maligonly with US-

diam-

carci-

rate.

lymphoma

other38%.

the

papillary

percentage of patients moto thyroiditis develop

malignancies large

in diameter) and firm they were not palpable

questionably

and

the sensitivity in the detec-

of nodules occurring normal thyroid gland

seven more lion,

DISCUSSION

and location (15). In adthe character of the residual gland should affect the palpa-

that

with

the location and making a diagnosis of a nonpalpable carcinoma in Hashimoto thy-

histologic primary

As seen microcalcification

(23) reported

primary

procedure,

US demonstrated

and

diagnosis was FNA but not FNA. Mazza-

correlated

of the

rence

mas

Young

termining

mass in pseudotumor of the thyroid gland

(22),

noma and that primary tumors larger than I .5 cm in diameter recurred subsequent to initial therapy with over twice the frequency of smaller tumors. Therefore, we suggest that de-

‘-

a.

and

recurrences

..

.

thy-

a locally invasive tumor is associated with a poor prognosis (23). In another patient with nonoccult

-.

carcinoma, the correct made with US-guided with palpation-guided -

to of the

mass have greater recurrence mortality rates than those with

palpable ...-,

metastasize and

tumors can metastasize to the remote organs (19,21). Patients with thyroid carcinoma who have a regional neck

x5.)

....

tumors nodes,

for patients

depends

largely

and

the

this

disorder

with on

the

survival

stage

of the

(10), early histologic diagthis procedure followed by

minimal surgical of appropriate

biopsy therapy

and may

initiation improve

the survival rate for patients with thyroid lymphoma. Although the residual gland was Radiology

#{149} 129

affected with Hashimoto thyroiditis, hyperechoic masses were usually benign, and isoechoic masses had a low frequency of malignancy; most carcinomas were found in hypoechoic nodules. These results are compatible with those from previous reports in which echogenicity of the nodules was compared with that of the otherwise normal thyroid gland (7-9). Nodules consisting of fofficular adenoma, papillary carcinoma, adenomatous hyperplasia, pseudotumor in Hashimoto thyroidilis, and malignant lymphoma were markedly hypoechoic at US. Of these, almost all malignancies were malignant lymphoma. The only other malignancy was an occult papillary carcinoma. Follicular adenoma frequently had a round margin and a halo. At sonography, however, malignant lymphoma was indistinguishable from adenomatous hyperplasia or pseudotumor in Hashimoto thyroiditis because sonographic findings in these conditions overlapped considerably. Differentiation of lymphoma from other nonneoplastic disorders with US-guided FNA should be clinically beneficial because unnecessary surgery can be avoided for patients with nonneoplaslic conditions. Although the presence or absence of obstructive symptoms can help differentiate thyroid lymphoma from other benign conditions, final diagnosis must rely on histologic findings. In our series, unnecessary surgery was avoided in 35 of the 60 patients (58%) and appropriate surgery was instituted in 10 (17%) on the basis of results of USguided FNA. US-guided FNA was not very accurate in helping differentiate follicular adenoma from other diseases such as adenomatous hyperplasia, pseudotumor in Hashimoto thyroiditis, or Hurthle cell tumor. However, this will cause no serious problems in the treatment of patients with these nodules. Conversely, diagnostic accuracy in differentiating between malignant and benign nodules in our series was high (100%). This might be partly due

to the fact that no patient had surgically verified well-differentiated carcinoma. The prevalence of this tumor will reduce the sensitivity of FNA because histologic diagnosis of the tumor is difficult with FNA-even with large-core biopsy specimens or frozen sections (24-26). Thus, US-guided FNA is useful in making a histologic diagnosis of thyroid nodules associated with Hashimoto thyroidilis. This procedure can help select patients who need surgical

11.

intervention

15.

and

can

prevent

essary surgery in patients neoplaslic thyroid diseases.

12.

13.

14.

non-

#{149}

4.

5.

6.

7.

8.

9.

10.

P,NiCkeISJ,

Kivisaaii

gland:

US screening in midwomen with no previous thyroid Radiology 1989; 173:507-510.

disease. Kini SR, MillerJM, HamburgerJL Problems in the cytologic diagnosis of the “cold” thyroid nodule in patients with lymphocytic thyroiditis. Acts Cytol 1981; Cuarda

LA, Baskin

HJ.

Inflammatory

17.

gland. Chang

Radiology

#{149}

in

Acts Cytol TC, Hong

1985; 29:850-855. CT, Chang SL, Hsieh HC, Liaw KY, How SW. Correlation between sonography and pathology in thyroid diseases. J Formosan Med Assoc 1990; 89:777-

OH,

DHJr. Arch

Woolner

Surgical asSurg 1960; 80:720-

LB, Engel

18.

S, McCona-

hey WM. Needle biopsy of the thyroid gland and management of lymphocytic thyroiditis. Surg Cynecol Obstet 1962; 114: 636-639. Holmes HBJr, Kreutner A, O’Brien PH. Hashimoto’s thyroiditis and its relationship to other thyroid diseases. Surg Cynecol

20.

144:887-890. M, Shimaoka

K, Rao U, Tsukada K. Diagnosis of chroniclymphocytic thyroiditis (nodular presentation) by needle aspiration. Acts Cytol 1981; 25:513-522. Thomas vention

M, Tamura

21.

22.

CC, Rutledge RC. Surgical interin chronic (Hashimoto’s) thyroid-

ins. Ann Surg 1981; 193:769-776. Ott PA, Calandra DB, McCall A, Shah KH, Lawrence AM, Paloyan E. The incidence of thyroid carcinoma in patients with Hashimoto’s thyroiditis and solitary cold nodules. Surgery 1985; 98:1202-1206. Simeone

JF, Daniels

CH,

Mueller

23.

24.

PR, et al.

S, Morimoto

S, IkezoeJ,

et al.

Primary thyroid lymphoma: comparison CT and US assessment. Radiology 1989;

Beierwaltes WH. Natural history of thymid cancer. In: Ingbar SH, Braverman LE, eds. The thyroid. 5th ed. Philadelphia: Lippincott, 1986; 1319-1328. Fukunaga FH, Yatam R. Ceographic pathology of occult thyroid carcinomas. Cancer 1975; 36:1095-1099. Harach HR, Franssila KO, Wasenius V. Occult papillary carcinoma of the thyroid. Cancer 1985; 56:531-538. Nussbaum M, Bukachevsky R. Thyroid carcinoma presenting as a regional neck mass. Head Neck 1990; 12:114-117. Mazzaferri EL, Young RL Papillary thyroid carcinoma: a 10-year follow-up report of the impact of therapy in 576 patients. Am J Med 1981; 70:511-518. Lowhagen

T, WfflemsJ,

blad R, Cranberg

High-resolution real-time sonography of the thyroid. Radiology 1982; 145:431-435. Solbiati L, Volterram L, Rizzatto C, et al. The thyroid gland with low uptake lesions: evaluation by ultrasound. Radiology 1985; 155:187-191. Butch RJ, SimeoneJF, Mueller PR. Thymid and parathyroid ultrasonography. Radiol Clin North Am 1985; 23:57-71. Takashima

SB, Tibblin S. The reliability examination of the thyroid gland: a prospective study of 100 consecutive patients surgically treated for hyperparathyroidism. Ann Chir Cynaecol 1985; 74:151-154.

19.

1977;

Friedman Y, Cavigan

Christensen

of the clinical

of

cytology 25.

26.

in diagnosis

P.

Lundell

C, Sund-

Aspiration

biopsy

of thyroid

cancer.

WorldJ Surg 1981; 5:61-73. Ramacciotti CE, Pretorius HT, Chu EW, Barsky SH, Brennan MF, RobbinsJ. Diagnostic accuracy and use of aspiration biopsy in the management of thyroid nodules. Arch Intern Med 1984; 144:1169-1173. Silverman p;, West RL, Larkin EW, et al. The role of fine-needle aspiration biopsy in the rapid diagnosis and management of thyroid neoplasm. Cancer 1986; 57:1164-. 1170.

171:439-443.

130

and

lesions of the thyroid gland: cyby fine-needle aspiration. Am J Clin Pathol 1987; 87:14-22. Akerman M, TennvallJ, Bi#{246}rklundA, MArtensson H, M#{246}ller T. Sensitivity and spec-

783.

Pollock WF, Sprung pects of thyroiditis.

Obstet

Viikinkoski

Thyroid

ificity of fine-needle aspiration cytology the diagnosis of tumors of the thyroid

References

3.

S, et aL

lymphoid topathology

16.

732. Beahrs

Morimoto

25:506-512.

We thank Tetsuro Kobayashi, MD, and Shin-ichiro Takai, MD, for providing clinical data; Hideo Yoshimura for his advice on pathologic studies; Miwako Fukuda for secretarial assistance; and Anna Takashima for encouragement.

2.

BranderA, die-aged

ACknowledgments:

1.

S, IkezoeJ,

Primary thyroid lymphoma: evaluation with CT. Radiology 1988; 168:765-768. Eftekhari F, Peuchot M. Thyroid metastases: combined role of ultrasonography and fine-needle aspiration biopsy.JCU 1989; 17:657-660. L

unnec-

with

Takashima

October1992