Neuroradiolog y/Head and Neck Imaging Original Research

Bonavita et al. Thyroid Ultrasound

Neuroradiology/Head and Neck Imaging Original Research

Pattern Recognition of Benign Nodules at Ultrasound of the

Thyroid: Which Nodules Can Be Left Alone
OBJECTIVE. The purpose of this study was to evaluate morphologic features
?John A. Bonavita Jason
predictive of benign thyroid nodules.
11

Mayo1 James Babb

Genevieve Bennett12
Thaira Oweity Michael
Macari11 Joseph Yee

Bonavita JA, Mayo J, Babb J, et al.

MATERIALS AND METHODS. From a registry of the records of 1,232 fne-needle aspiration biopsies
performed jointly by the cytology and radiology departments at a single institution between 2005 and 2007, the
cases of 650 patients were identifed for whom both a pathology report and ultrasound images were available.
From the alphabetized list generated, the frst 500 nodules were reviewed. We analyzed the accuracy of individua
sonographic features and of 10 discrete recognizable morphologic patterns in the prediction of benign histologic
fndings.
RESULTS. We found that grouping of thyroid nodules into reproducible patterns of morphology, or pattern
recognition, rather than analysis of individual sonographic features, was extremely accurate in the identifcation o
benign nodules. Four specifc patterns were identifed: spongiform confguration, cyst with colloid clot, giraffe
pattern, and diffuse hyperechogenicity, which had a 100% specifcity for benignity. In our series, identifcation of
nodules with one of these four patterns could have obviated more than 60% of thyroid biopsies.
CONCLUSION. Recognition of specifc morphologic patterns is an accurate method of identifying benign
thyroid nodules that do not require cytologic evaluation. Use of this approach may substantially decrease the
number of unnecessary biopsy procedures

.Keywords: fne-needle aspiration, nodule, thyroid, ultrasound

DOI:10.2214/AJR.08.1820
Received September 12, 2008; accepted after revision October 24, 2008.
1Department

of Radiology, Langone Medical Center, New York University School of Medicine, 550 First Ave., New York, NY 10016. Address correspondence
to J. Bonavita (john.bonavita@nyumc.org).
2Department of

Pathology, Langone Medical Center, New York University School of Medicine, New York, NY.

AJ R 2009; 193:207213
0 361 8 03 X/ 0 9 /1 9 3 1 2 0 7

O
Down
loade
d
from
www.
ajronli
ne.org
by
36.79.
73.83
on
04/06/
16
from
IP
addres
s
36.79.
73.83.
Copyr
ight
ARRS
. For
perso
nal
use
only;
all
rights
reserv
ed

2007, the cases of 650 patients (436 women, All diagnostic ultrasound examinations
The fnal diagnosis was based on the
64 men; average age, 54.7 years; range, 17 and FNA biopsies were performed with an cytologic result; fnal pathologic confrmation
88 years) were identifed in which both
Acuson 300 or Antares unit (both Siemens was limited to the 20 malignant tumors
pathology reports and ultrasound images
Healthcare). All FNA biopsies were
resected. In the 20 patients with these
were available. From the alphabetized list performed by a group of four cytologists
tumors, there was no discrepancy between
generated, the frst 500 nodules were
(average experience, 5 years) under
the initial cytologic and the fnal pathologic
reviewed. This HIPAA-compliant study was ultrasound guidance by one of fve
result. The cytologic results were divided
approved by our institutional review board radiologists (average experience, 20.5
into three categories: 1, benign nodules,
with a waiver of informed consent. We
years). The biopsies were performed with including colloid nodules, hyperplastic
analyzed the accuracy of individual
25-gauge spinal needles in most instances; a nodules, and localized thyroiditis; 2,
sonographic features and of 10 discrete
27-gauge needle was used for hypervascular intermediate nodules, including follicular
recognizable morphologic patterns in the
lesions. At least two passes were made for and Hrthle cell neoplasms; and 3,
prediction of benign histologic fndings.
each nodule (average, 3.2 passes per nodule; carcinoma. Type 1 nodules were determined
range, 26 passes). All specimens
to be nodules that did not require biopsy;
types 2 and 3 were nodules requiring biopsy.
Bonavita et al.
Ultrasound Technique
were evaluated immediately by the
cytologists to confrm sample adequacy.

Data Analysis

The sensitivity, specifcity, positive

predictive value, and negative predictive
value were defned for each individual
Ultrasound Interpretation
sonographic feature in the detection of
In this retrospective study the ultrasound nonbenign masses. The Blyth-StillCasella
images of all nodules were reviewed in
procedure for construction of exact CI for a
consensus by two blinded radiologists: one binomial proportion was used to derive a
an attending radiologist with 31 years of
95% CI for the negative predictive value
ultrasound experience, the other a second- associated with each classifcation factor
year radiology resident. Each nodule was when used to identify benign masses. All
evaluated for the presence or absence of
reported p values were twosided signifcance
individual sonographic features and was
levels and were declared statistically
assigned one of 10 distinct recognizable mor signifcant at less than 0.05. SAS software
phologic patterns.
(version 9.0, SAS Institute) was used for all
statistical computations. Each p value was
derived from a Fishers exact test performed
Histologic Analysis
to determine whether the classifcation factor
was associated with benignity.

Results The individual ultrasound

features of each nodule analyzed
were size, number, texture
A CB

Fig. 1Individual
ultrasound features of
nodules. A, 85-year-old
woman with
subcentimeter papillary
carcinoma. Ultrasound
scan shows hypoechoic
nodule. B, 46-year-old
woman with papillary
carcinoma. Ultrasound
scan shows nodule with
ill-defned borders. C, 36year-old man with
papillary carcinoma.
Ultrasound scan shows

microcalcifcations (a r ro
w ), which are easily
confused with comet-tail
shadowing. Important
fnding is
hypoechogenicity of
nodule. D, 37-year-old
woman with medullary
carcinoma. Ultrasound
scan shows
macrocalcifcation. E, 37year-old woman with
papillary carcinoma.
Color Doppler
ultrasound image shows
hypervascular nodule.

DE
208 AJR:193, July 200

Thyroid Ultrasound
TABLE 1: Diagnostic Characteristics of Each Classification in Identification of Benign Masses

Down
loade
d
from
www.
ajronli
ne.org
by
36.79.
73.83
on
04/06/
16
from
IP
addres
s
36.79.
73.83.
Copyr
ight
ARRS
. For
perso
nal
use
only;
all
rights
reserv
ed

Classifiation Sensitivity (%) Specifcity (%) Positive Predictive Value (%) Negative Predictive Value (%) p Presence of
sharp border 62.5 (25/40) 61.7 (284/460) 12.4 (25/201) 95.0 (284/299) 0.0017 Absence of calcifcation 25.0 (10/40)
93.3 (429/460) 24.4 (10/41) 93.5 (429/459) 0.0005 Absence of halo 32.5 (13/40) 75.9 (349/460) 10.5 (13/124) 92.8
(349/376) 0.0731 Presence of hyperechogenicity 100.0 (40/40) 8.9 (41/460) 8.7 (40/459) 100.0 (41/41) 0.0282
Absence of hypoechogenicity 52.5 (21/40) 92.2 (424/460) 36.8 (21/57) 95.7 (424/443) < 0.0001 Absence of
isoechogenicity 35.0 (14/40) 78.7 (362/460) 12.5 (14/112) 93.3 (362/388) 0.023 Absence of hypervascularity 35.0
(14/40) 90.4 (416/460) 24.1 (14/58) 94.1 (416/442) < 0.0001 Presence of spongiform confguration 90.0 (36/40) 57.8
(266/460) 15.7 (36/230) 98.5 (266/270) < 0.0001 Absence of edge refraction 7.5 (3/40) 97.8 (450/460) 23.1 (3/13)
92.4 (450/487) 0.0625 Absence of ring vascularity 22.5 (9/40) 92.2 (424/460) 20.0 (9/45) 93.2 (424/455) 0.0042
Presence of classifcation 14 100.0 (40/40) 65.9 (303/460) 20.3 (40/197) 100.0 (303/303) < 0.000
1NoteValues in parentheses are numbers
of nodules.

(Fig. 1A), margination (Fig. 1B), presence of internal densities or

calcifcations (Figs. 1C and 1D), edge refraction, and vascularity relative to
the rest of the gland [13,25, 26] (Fig. 1E). Analysis of the presence or
absence of individual sonographic features revealed no feature with
consistently high sensitivity or specifcity for malignancy (Table 1). In our
study, sensitivity for the presence or absence of specifc features was 35
100% and specifcity, 8.997.8%. There was no correlation between
diagnosis and nodule size, which was categorized as less than 1 cm (n = 7),
12 cm (n = 288), and larger than 2 cm (n = 206) (Table 2). However,
several features were found to have a statistically signifcant negative
predictive value. These individual features, the absence of which was
common in benign disease, included calcifcation, halo, hypoechogenicity,
isoechogenicity, and ring or peripheral hypervascularity.TABLE 2: Size
Versus Diagnosis
Diagnosis Nodule Diameter (cm)< 1 12 > 2 Benign 6 265 190 Follicular 0 10
10 Malignant 1 13 6 Total 7 288 206
NoteThere was no correlation between diagnosis and nodule size

.Each nodule was assigned to one of

10 discrete morphologic groupings.

These patterns, which were based on
a previous report [23] and expanded
according to our experience, were as
follows: 1, spongiform without
hypervascularity (Fig. 2A); 2, cyst
with avascular colloid plug (Fig.
2B); 3, giraffe pattern (Fig. 2C) with
blocks of hyperechogenicity, or
white, separated by bands of
hypoechogenicity, or black; 4,
uniform hyperechogenicity (white
knight) (Fig. 2D); 5, intense
hypervascularity (red light) (Fig.
2E); 6, hypoechogenicity (Fig. 2F);
7, isoechogenicity without halo (Fig.
2G); 8, isoechogenicity with halo
(Fig. 2H); 9, ring of fire or nodules
with intense peripheral vascularity
(Fig. 2I); and 10, other (Fig. 2J), or a
mixed pattern or pattern that did not
fitthe other categories (Table 3). A
distinct pattern emerged in which it
became evident that there were
specific morphologic groupings or
patterns that were accurate predictors
of benign disease. Specifically, there
were no malignant nodules in the

303 patients (61%) with patterns 14

(Table 4). Spongiform
nonhypervascular masses were the
most common type of nodule seen,
210 of 210 being found benign at
FNA biopsy. All 53 of the cysts with
internal colloid clot, all 23 giraffe
pattern nodules, and all 17
hyperechoic nodules were benign.
The results in patterns 510 were
unpredictable, ranging from 35 of 37
isoechoic nodules without halo
biopsied being benign to only 31 of
45 hypoechoic nodules being
benign.
Discussion A thyroid nodule is a
discrete lesion,

sonographically distinct from the

surrounding thyroid parenchyma
[27]. Rather than a single disease,
nodules are manifestations of a
gamut of thyroid diseases [28].
Although some thyroid nodules may
be discovered at physical
examination, many are incidental
fndings of other imaging studies,
such as CT and MRI of the neck or
chest and carotid ultrasound
imaging. FNA of thyroid nodules has
replaced blind surgical excision as
the procedure of choice in the
diagnosis of thyroid nodules. Use of
FNA has led to a considerable
decrease in the number of surgical
excisions and to a twofold increase
in the diagnosis of carcinoma [4, 5,
29]. The relative ease of FNA
compared with surgery and the
increased frequency and refnement
of imaging studies has resulted in
what some authors have referred to
as an epidemic of thyroid nodules [3,
30]. In view of their ubiquity, it is
not feasible to biopsy every thyroid
nodule discovered with ultrasound.
Reasons for limiting thyroid biopsy,

which is relatively painless and safe,

include the small percentage of
malignant lesions, the small number
of cases of thyroid cancer in which
early diagnosis may actually have an
infuence, the economic and societal
Down
loade
d
from
www.
ajronli
ne.org
by
36.79.
73.83
on
04/06/
16
from
IP
addres
s
36.79.
73.83.
Copyr
ight
ARRS
. For
perso
nal
use
only;
all
rights
reserv
ed

AJR:193, July 2009 209

costs, the strain on radiology

resources, and the patient
uncertainty and anxiety incumbent
on a potentially malignant diagnosis.
Hence, reliable guidelines for
nodules that may not require biopsy

have become essential. Not

surprisingly in view of the
experience of other authors [31], we
concluded that no individual
sonographic feature had both high
sensitivity and high specifcity in the

Bonavita et al.
AC
BD FE
GI
H phologic patterns. A, 41-year-old man with colloid nodule. Ultrasound scan shows spongiform nodule. Similarity of
J nodule to water-flled sponge is evident. B, 52-year-old man with colloid cyst. Ultrasound scan shows cyst with colloid
Down
loade
d
from
www.
ajronli
ne.org
by
36.79.
73.83
on
04/06/
16
from
IP
addres
s
36.79.
73.83.
Copyr
ight
ARRS
. For
perso
nal
use
only;
all
rights
reserv
ed

F
i
g
.
2

M
o
r

clot. When cystic portion of nodule is subtracted, type 1 or spongiform nodules remain. C, 21-year-old woman with
Hashimotos thyroiditis. Ultrasound scan shows nodule that looks like giraffe hide, having light blocks separated by
black bands. D, 34-year-old woman with Hashimotos thyroiditis. Ultrasound scan shows white knight, or
hyperechoic, nodule. E, 61-year-old woman with follicular adenoma. Color Doppler ultrasound image shows red
light, or hypervascular, nodule. F, 29-year-old woman with papillary carcinoma. Ultrasound scan shows hypoechoic
nodule. G, 70-year-old woman with papillary carcinoma. Ultrasound scan shows isoechoic nodule without halo.
Coincidental microcalcifiations (a r ro w s ) are evident. H, 25-year-old man with nodular goiter. Ultrasound scan
shows isoechoic nodule with halo. I, 55-year-old woman with hyperplastic nodule. Color Doppler ultrasound image
shows ring of fre, or peripheral hypervascularity. J, 61-year-old man with colloid nodule. Ultrasound scan shows
nodule that fts into no other pattern.

210 AJR:193, July 2009

Thyroid Ultrasound
TABLE 3: Features of Morphologic Types of Thyroid Nodules
Down
loade
d
from
www.
ajronli
ne.org
by
36.79.
73.83
on
04/06/
16
from
IP
addres
s
36.79.
73.83.
Copyr
ight
ARRS
. For
perso
nal
use
only;
all
rights
reserv
ed

Pattern Texture Vascularity Margins Densities 1, Spongiform or puff pastry Spongiform internal cysts None or
isovascular Well-defned Present or absent comet tail 2, Cyst with colloid clot Cystic with mural clot None or isovascular Welldefned Present or absent comet tail 3, Giraffe Hyperechoic block, black bands None or isovascular Any Absent 4, Hyperechoic,
or white knight Hyperechoic None or isovascular Well-defned Absent 5, Intensely hypervascular, or red light Any Central
hypervascularity Any Present or absent 6, Hypoechoic Hypoechoic None or isovascular Any Present or absent 7, Isoechoic
without halo Isoechoic None or isovascular Any Present or absent 8, Isoechoic with halo Isoechoic None or isovascular Welldefned Present or absent 9, Ring of fre Any Peripheral hypervascularity Well-defned Present or absent 10, Other Any Any Any
Present or absent

TABLE 4: Number of Nodules With Pattern Categorized by Suggested Management and Diagnosis
(n = 500)
Pattern Benign, Watch (n = 460) Malignant, Biopsy (n = 40)Total Colloid Hashimotos Thyroiditis Hyperplasia Total Follicular
Malignant 1, Spongiform 210 196 6 8 0 0 0 2, Cyst with colloid clot 53 52 1 0 0 0 0 3, Giraffe 23 12 10 1 0 0 0 4, White
knight 17 9 8 0 0 0 0 5, Red light 37 29 5 3 15 11 4 6, Hypoechoic 31 19 8 4 14 1 13 7, Isoechoic without halo 35 26 4 5 2 0
2 8, Isoechoic with halo 37 33 1 3 4 1 3 9, Ring of fre 6 5 0 1 4 4 0 10, Other 11 10 1 0 1 0 1
NotePatterns 14 are invariably associated with benign conditions. Patterns 510 are variable
.detection of malignancy.
The persistent combination of some Like Reading et al. [23], we found

Nonetheless, many of these

previously described high-risk
features, such as calcifcation,
hypoechogenicity, poor defnition,
and hypervascularity, were found to
be absent over and over again in
nodules that did not require biopsy.

AJR:193, July 2009 211

of these common individual ultrasound that use of a pattern approach to

characteristics, or, more properly, their thyroid nodules is highly sensitive
and specifc for the presence of
absence, led us to consider a more
pattern-oriented approach, such as that benignity. Our patterns differed
advocated by Reading et al. [23] as an somewhat from those proposed
alternative to the analysis of individual previously, yet there are defnite
features. Those authors described eight similarities. Analysis of our data
revealed four patterns that were
typical appearances of commonly
invariably benign at FNA biopsy
encountered benign and malignant
(Table 5). The most common overall
nodules, allowing them to separate
more than one half of thyroid nodules pattern is a nodule with diffuse
internal linear cysts, described as
into those that could be observed
spongiform or honeycomb, our type
versus those requiring biopsy.
1 pattern. In our cases, this fnding
According to their results, the
was commonly described as a puff
following four classic patterns
pastry pattern similar to the
necessitate biopsy: 1, a hypoechoic
ultrathin layers of faky pastry in
nodule with microcalcifcations; 2,
desserts such as napoleons. This
coarse calcifcations in a hypoechoic
pattern was characteristic of colloid
nod-ule; 3, well-marginated, ovoid,
solid nodules with a thin hypoechoic nodules or goiter. The only
halo; and 4, a solid mass with refractive spongiform nodule not classically
benign was a single nodule that also
shadowing from the edges, which is
believed to occur as a result of fbrosis. was intensely hypervascular. Our
type 1 or spongiform nodule
The four classic patterns of nodules
that did not require biopsy in that series consequently is defned as avascular
were the following: 1, small (< 1 cm) or, occasionally, isovascular in
colloidflled cystic nodules; 2, a nodule relation to the rest of the gland. The
second pattern (type 2) was a cystic
with a honeycomb appearance
consisting of internal cystic spaces with nodule containing a central plug of
avascular colloid, similar to the
thin echogenic walls; 3, a large
previously described small or large
predominantly cystic nodule; and 4,
cyst patterns [23]. In our initial
diffuse multiple small hypoechoic
analysis of individual features, size
nodules with intervening echogenic
of cyst was deemed
bands, which are indicative of
insignificant.Important, however,
Hashimotos thyroiditis.
was the characterization of the plug
as avascular and puff pastry. All of
these nodules

Bonavita et al.
TABLE 5: Patterns of Nodules That Do Not Require Biopsy Versus Patterns of Reading et al. [23]
Current Study Classifcation of Reading et al. 1, Spongiform, or puff pastry 2, Honeycomb of internal cystic spaces
with thin echogenic walls 2, Cyst with colloid clot 1, Small (< 1 cm) colloid-flled cystic nodules; 3, large predominantly cystic
nodule 3, Giraffe 4, Diffuse, multiple small hypoechoic nodules with intervening echogenic bands indicative of Hashimotos
thyroiditis 4, Hyperechoic, or white knight

TABLE 6: Patterns of Nodules Requiring Biopsy Versus Patterns of Reading et al. [23]
Current Study: Indeterminate Finding, Biopsy Necessary Reading et al.: High Risk of Malignancy, Biopsy Necessary 5, Red
light, central hypervascularity 6, Hypoechoic 1, Hypoechoic nodule with microcalcifcations; 2, coarse calcifcations in a
hypoechoic nodule 7, Isoechoic without halo 4, Solid mass with refractive shadowing from the edges, believed to be due to
fbrosis 8, Isoechoic with halo 3, Well-marginated, ovoid, solid nodule with a thin hypoechoic halo 9, Ring of fre, peripheral
vascularity 10, Othe
rwere also colloid nodules. If the
We identifed other common patterns, who will decide whether biopsy should

cystic portion of the lesion is

including the type 5 red light pattern, be performed. Analysis of interobserver
subtracted visually, a type 1
variability for assigning nodules to a
or an intensely hypervascular lesion
spongiform nodule remains. The
that on Doppler images glowed like a specifc pattern will be analyzed, as will
third pattern (type 3), or giraffe
the characterizations with fnal cytologic
red stoplight. This pattern was
pattern, was characterized by
result.
commonly seen in lesions with
globular areas of hyperechogenicity
abundant cellularity, including,
surrounded by linear thin areas of
commonly, follicular neoplasms and,
hypoechogenicity, similar to the two- less commonly, hyperplastic nodules
tone blocklike coloring of a giraffe.
and carcinoma. Other nodule types
This pattern was quite characteristic
included type 9 ring-of-fre nodules
of Hashimotos thyroiditis. A
with intense peripheral vascularity and
variation of this pattern is our type 4 nodules described as other (type 10),
white knight, or hyperechoic,
which did not ft any of the classic
nodule, which was found commonly patterns. Calcifcation, although
to be a regenerative nodule of
commonly seen in nodules requiring
Hashimotos thyroiditis.
biopsy, was never seen as an isolated
Analysis of our other patterns
fnding. The likelihood of benignity of
revealed more variability in final
these nodules (type 510) ranged from
cytologic findings (Table 6). Such
60% (type 9, ring of fre) to 91% (type
nodules included both insignificantand 10, other). Because of this lack of
significantlesions with such variability predictability, we believed that these
that prediction before biopsy was not nodules should be considered for FNA
reliable. These nodules had the four
biopsy.
biopsy-recommendation patterns
The limitations of our study are
described earlier, such as isoechoic
related to the fact that most of the
nodule with a surrounding halo or
diagnoses were based on cytologic
refractive edges, which came to be
rather than histologic fndings, the
simplifiedin our series as isoechoic
retrospective nature of the study, and
nodules with or without a halo (types 7 the fact that nodule characterization
and 8). A hypoechoic nodule with or
was dependent on only two observers.
without central microcalcifcation or
The readers were blinded to the
with central macrocalcifcation in other cytologic results at the time of nodule
series [25, 26, 32], for which biopsy
characterization. The period 2005
was recommended, was the most
2007 was chosen to minimize the
worrisome pattern (type 6) in our study. potential for recall bias. To answer our
concerns with respect to these
limitations, we are preparing a study in
which we train radiologists with
varying degrees of experience in this
pattern approach. A series of
consecutive thyroid biopsies will be
chosen prospectively in the weeks
before their performance, and the
images will be shown to these readers,

Down
loade
d
from
www.
ajronli
ne.org
by
36.79.
73.83
on
04/06/
16
from
IP
addres
s
36.79.
73.83.
Copyr
ight
ARRS
. For
perso
nal
use
only;
all
rights
reserv
ed

We conclude that biopsy of a large should be performed regardless of the year follow-up. Radiology 2000; 215:801
806
number of thyroid nodules (in our
individual features or pattern of the
4. Jemal A, Murray T, Ward E, et al. Cancer
study, 61%) can be avoided when a
nodule.
statistics, 2005. CA Cancer J Clin 2005;
pattern approach to nodule
55:1030 [Erratum in CA Cancer J Clin
characterization is used. Specifc
References
2005; 55:259]
morphologic patterns are highly
1. Ezzat S, Sarti DA, Cain DR, et al. Thyroid
predictive of benignity. Specifcally, a incidentalomas: prevalence by palpation and 5. Harnsberger H. Diagnostic imaging:
head and neck. Salt Lake City, UT: Amirsys,
nodule that has a uniform
ultrasonography. Arch Intern Med 1994;
2004:2440
154:18381840
nonhypervascular spongiform
6. Kountakis SE, Skoulas IG, Maillard AA.
appearance, is a cystic lesion with a
2. Frates MC, Benson CB, Charboneau JW,
The radiologic work-up in thyroid surgery:
et
al.
Management
of
thyroid
nodules
colloid clot, has a giraffelike pattern, or
fne-needle biopsy versus scintigraphy and
detected at US: Society of Radiologists in
is diffusely hyperechoic can be
ultrasound. Ear Nose Throat J 2002; 81:151
Ultrasound
consensus
conference
statement.
observed rather than biopsied. If,
154
Radiology 2005; 237:794 800
conversely, a nodule does not
7. Rago T, Vitti P, Chiovato L, et al. Role
3. Brander AE, Viikinkoski VP, Nickels JI, of conventional ultrasonography and
correspond to one of these four
color flw-Doppler sonography in
patterns, according to our data biopsy Kivisaari LM. Importance of thyroid
abnormalities detected at US screening: a 5- predicting malignancy in cold

212 AJR:193, July 2009

Down
loade
d
from
www.
ajronli
ne.org
by
36.79.
73.83
on
04/06/
16
from
IP
addres
s
36.79.
73.83.
Copyr
ight
ARRS
. For
perso
nal
use
only;
all
rights
reserv
ed

thyroid nodules. Eur J Endocrinol 1998; 16. Frates MC, Benson CB, Doubilet PM, 23. Reading CC, Charboneau JW, Hay ID,
138:4146 8. Giuffrida D, Gharib H.
et al. Prevalence and distribution of
Sebo TJ. Sonography of thyroid nodules: a
Controversies in the mancarcinoma in patients with solitary and
classic pattern diagnostic approach.
agement of cold, hot, and occult thyroid multiple thyroid nodules on Thyroid
Ultrasound Q 2005; 21:157 165
nodules. Am J Med 1995; 99:642650 Ultrasound
24. Hegedus L. Thyroid ultrasound.
9. Papini E, Guglielmi R, Bianchini A, et
Endocrinol Metab Clin North Am 2001;
al. Risk of malignancy in nonpalpable
30:339360
sonography. J Clin Endocrinol Metab
thyroid nodules: predictive value of
25. Hoang JK, Lee WK, Lee M, Johnson D,
2006; 91: 34113417
ultrasound and color-Doppler features. J
Farrell S. US features of thyroid malignancy:
17. Cooper DS, Doherty GM, Haugen BR,
Clin Endocrinol Metab 2002; 87:1941
pearls and pitfalls. RadioGraphics 2007;
et al. Management guidelines for patients
1946
27:847860
with thyroid nodules and differentiated
10. Jun P, Chow LC, Jeffrey RB. The
thyroid cancer. Thyroid 2006; 16:109142 26. Takashima S, Fukuda H, Nomura N,
sonographic features of papillary thyroid
Kishimoto H, Kim T, Kobayashi T. Thyroid
carcinomas: pictorial essay. Ultrasound Q 2005; 18. Pacini F, Schlumberger M, Dralle H,
nodules: reevaluation with ultrasound. J Clin
Elisei R, Smit JW, Wiersinga W; European
21:3945
Ultrasound 1995; 23:179184
Thyroid Cancer Taskforce. European
11. Kim EK, Park CS, Chung WY, et al.
27. Vandermeer FQ, Wong-You-Cheong J.
consensus for the management of patients
New sonographic criteria for recommending
with differentiated thyroid carcinoma of the Thyroid nodules: when to biopsy. Appl Radiol
fne-needle aspiration biopsy of nonpalpable
follicular epithelium. Eur J Endocrinol 2006; 2007; 36:819
solid nodules of the thyroid. AJR 2002;
154:787803
28. Pacini F, Burroni L, Ciuoli C, Di
178:687691
19. British Thyroid Association and Royal Cairano G, Guarino E. Management of
12. Koike E, Noguchi S, Yamashita H, et al.
thyroid nodules: a clinicopathological,
College of Physicians. Guidelines for the
Ultrasonographic characteristics of thyroid
evidence-based approach. Eur J Nucl Med
management of thyroid cancer in adults.
nodules: prediction of malignancy. Arch
London, UK: Publication Unit of the Royal Mol Imaging 2004; 31:14431449
Surg 2001; 136: 334337
College of Physicians, 2002
29. Castro MR, Gharib H. Thyroid fne13. Chan BK, Desser TS, McDougall IR, et al.
20. Rodrigues FJ, Limbert ES, Marques AP, needle aspiration biopsy: progress, practice,
Common and uncommon sonographic features
et al.; Grupo de Estudo da Tiride. Treatment and pitfalls. Endocr Pract 2003; 9:128136
of papillary thyroid carcinoma. J Ultrasound
and follow up protocol in differentiated
30. Ross DS. Nonpalpable thyroid nodules:
Med 2003; 22:10831090
thyroid carcinomas of follicular origin [in
managing an epidemic. J Clin Endocrinol
14. Ahuja A, Chick W, King W, Metreweli Portuguese]. Acta Med Port 2005; 18:216 Metab 2002; 87: 19381940
C. Clinical signifcance of the comet-tail
21. Societ Italiana di Endocrinologia,
31. Sahin M, Sengul A, Berki Z, Tutuncu
artifact in thyroid ultrasound. J Clin
Associazione Italiana di Medicina Nucleare NB, Guvener ND. Ultrasound-guided fneUltrasound 1996; 24:129133
ed Imaging Molecolare, Associazione
needle aspiration biopsy and
15. Kovacevic DO, Skurla MS. Sonographic Italiana di Fisica Medica. Linee Guida SIE- ultrasonographic features of infracentimetric
diagnosis of thyroid nodules: correlation with AIMN-AIFM per il trattamento e follow-up nodules in patients with nodular goiter:
the results of sonographically guided fnedel carcinoma tiroideo differenziato della correlation with pathological fndings.
needle aspiration biopsy. J Clin Ultrasound tiroide. Rome, Milan, and Gazzada, Italy: Endocr Pathol 2006; 17:6774
2007; 35:6367
SIE, AIMN, and AIFM, 2004:175
32. Iannuccilli JD, Cronan JJ, Monchik
22. Van De Velde CJ, Hamming JF, Goslings JM. Risk for malignancy of thyroid
nodules as assessed by sonographic
BM, et al. Report of the consensus
development conference on the management criteria: the need for biopsy. J Ultrasound
Med 2004; 23:1455146
of differentiated thyroid cancer in The
Netherlands. Eur J Cancer Clin Oncol 1988;
24:287292
4AJR:193, July 2009 213