THYROID THYROID RADIOLOGY AND NUCLEAR MEDICINE

Volume 22, Number 9, 2012

ª Mary Ann Liebert, Inc.
DOI: 10.1089/thy.2012.0005

Risk of Malignancy in Thyroid Incidentalomas Detected

by 18F-Fluorodeoxyglucose Positron Emission Tomography:
A Systematic Review

Kerstin Kathrine Soelberg, Steen Joop Bonnema, Thomas Heiberg Brix, and Laszlo Hegedüs

Background: The expanding use of 18F-fluorodeoxyglucose positron emission tomography (18F-FDG PET) has
led to the identification of increasing numbers of patients with an incidentaloma in the thyroid gland. We aimed
to review the proportion of incidental thyroid cancers found by 18F-FDG PET or PET/computed tomography
imaging.
Methods: Studies evaluating thyroid carcinomas discovered incidentally in patients or healthy volunteers by
18
F-FDG PET were systematically searched in the PubMed database from 2000 to 2011. The main exclusion
criteria were known thyroid disease, lack of assigned diagnoses, investigation of diffuse uptake only, or in-
vestigation of patients with head and neck cancer, or cancer in the upper part of the thorax.
Results: Twenty-two studies met our criteria comprising a total of 125,754 subjects. Of these, 1994 (1.6%) had
unexpected focal hypermetabolic activity, while 999 of 48,644 individuals (2.1%) had an unexpected diffuse
hypermetabolic activity in the thyroid gland. A diagnosis was assigned in 1051 of the 1994 patients with a focal
uptake, 366 of whom (34.8%) had thyroid malignancy. Likewise, a diagnosis was assigned in 168 of 999 patients
with a diffuse uptake, 7 of whom (4.4%) had thyroid malignancy. In the eight studies reporting individual
maximum standardized uptake values (SUVmax), the mean SUVmax was 4.8 (standard deviation [SD] 3.1) and 6.9
(SD 4.7) in benign and malignant lesions, respectively ( p < 0.001).
Conclusions: Incidentally found thyroid nodules, using 18F-FDG PET, are at high risk of harboring malignancy if
uptake is focal. SUV are significantly higher in malignant than in benign nodules. The pronounced inhomo-
geneity and other shortcomings of the studies are discussed.

Introduction a much higher rate than normal tissues. However, and im-
portantly, high cellular glucose uptake is nonspecific. Thus,

T hyroid nodules are frequent in the general population.

The prevalence increases with age and degree of iodine
deficiency (1,2), and is highest if ultrasound or autopsy is used
for diagnosis. Accordingly, the prevalence is reported be-
tween 8% and 65% (3). Incidental thyroid nodules are a
common finding due to the increasing use of neck imaging
with ultrasound, computed tomography (CT), magnetic res-
onance imaging, and positron emission tomography (PET). In
recent years thyroid incidentalomas, defined as newly iden-
tified focal thyroid lesions encountered during imaging with
18
F-fluorodeoxyglucose PET (18F-FDG PET), have been given
much attention (4–10).
18
F-FDG PET, with or without CT, is a noninvasive tech-
nique, widely used as a diagnostic tool and for the preoper-
ative staging of various malignancies (11,12). 18F-FDG PET
exploits the fact that many malignancies metabolize glucose at
tissues also harboring inflammatory cells have increased
glucose metabolism (13), which constitutes a differential di-
agnostic challenge when 18F-FDG PET is employed in search
of malignancy.
The increasing use of 18F-FDG PET or PET/CT has led
to the identification of increasing numbers of patients with
an incidentaloma in the thyroid gland (14–16). The prev-
alence of malignancy in such lesions has varied between
0% (5) and 63.6% (17). The risk seems dependent on
whether the 18F-FDG uptake is diffuse [prevalence of ma-
lignancy from 0% (5) to 13% (18)] or focal [prevalence from
10% (19) to 63.6% (17)]. As a potential way of discrimi-
nating between benign and malignant tissues, several
studies have investigated maximum standardized uptake
values (SUVmax) as a semiquantitative indicator of 18F-FDG
uptake (17,20).

Department of Endocrinology and Metabolism, Odense University Hospital, Odense, Denmark.

918
18
F-FDG PET AND THYROID INCIDENTALOMAS
FIG. 1. Flowchart showing the data extraction process.
Our aim was to review studies dealing with the risk of
malignancy in incidental findings in the thyroid gland found
with 18F-FDG PET or PET/CT imaging. The focus being on
the relative importance of whether the uptake is focal or dif-
fuse, on the potential role of determining SUV, and on
methodological differences and shortcomings.
Methods
Literature search
English language articles published from January 2000 to
June 2011, as a source for descriptions of thyroid in-
cidentalomas, were identified via a search in the PubMed
database (Fig. 1). The following search strategy was used: #1,
carcinoma OR cancer OR malignancy OR gland OR lesion OR
disease staging; #2, positron emission tomography OR focal
uptake OR fluorodeoxyglucose OR SUV OR diffuse uptake;
#3, incidentally OR incidental OR incidentaloma OR in-
cidentalomas OR unexpected; #4 thyroid. The categories were
combined as follows: #1 AND #2 AND #4 OR #1 AND #3
AND #4. Our search was restricted to humans and resulted in
a total of 602 abstracts extracted for further evaluation.
Selection of literature sources
The 602 abstracts were evaluated using predefined inclu-
sion and exclusion criteria. All of the following inclusion cri-
teria needed to be met: incidental finding in the thyroid gland;
incidentalomas being defined by the authors of the study as a
focal as well as a diffuse thyroid uptake, or solely as a focal
uptake, were included; articles were included regardless of
their definition of limits for anatomical area of uptake, and the
presence of a nodule was not a requirement; the participants
in the studies were investigated with 18F-FDG PET or 18F-FDG
PET/CT; investigated individuals were either healthy volun-
teers or individuals undergoing staging, restaging, or assess-
ment of treatment response for nonthyroidal malignancies;
there needed to be clinical follow-up or confirmation of
thyroid disease by fine-needle-aspiration (FNA) or surgery.
919
Studies were excluded if they included participants with a
prior history of thyroid disease; exclusively investigated dif-
fuse uptake in the thyroid gland; were case reports; evaluated
< 10 patients; only investigated patients with head and neck
cancer, or cancer in the upper part of the thorax.
Data extraction
Any study identified as relevant by reviewing its abstract
was retrieved and analyzed as a full text. This yielded a total
of 22 studies published as peer-reviewed articles (4–
10,14,15,17–29). The reference lists of all these studies were
scrutinized for additional relevant studies. Data extracted
from the studies included first author, country of origin, year
of publication, number of patients examined, imaging tech-
niques, study period, number of patients with focal or diffuse
thyroid uptake, patients without an assigned (this terminol-
ogy is used when also patients without cytological or histo-
logical confirmation are included) thyroid diagnosis or lost to
follow-up, patients with confirmed thyroid pathology, pa-
tients without pathology but follow-up, SUVmax for focal
uptake, and cytological or histological diagnosis (Table 1).
Data synthesis
Based on the results of the 22 studies, the prevalence of
malignancy in thyroid incidentalomas detected with 18F-FDG
PET, with or without CT, was calculated. The total number of
patients with an 18F-FDG PET uptake was reached by adding
the patients from all the publications. To find the prevalence
for unexpected focal as well as diffuse uptake in the thyroid
gland, the ratio of patients with positive thyroid findings to
the total number of patients was calculated. This was done
separately for focal and diffuse uptake.
All findings were categorized into benign, malignant, and
indeterminate, as evaluated in the studies by cyto- or histo-
pathology, other imaging such as ultrasound, thyroid 99mTc
scan, laboratory evaluation, or with follow-up clinical exam-
inations. The malignant findings were further classified ac-
cording to type of malignancy: papillary thyroid cancer (PTC)
and follicular variant of PTC, follicular thyroid cancer (FTC),
medullary cancer, Hürthle cell carcinoma, unspecified thy-
roid cancer, lymphoma, and metastatic disease from a non-
thyroid malignancy. The prevalence of malignancy in patients
with a focal uptake or a diffuse uptake was calculated sepa-
rately by the ratio of assigned cases with malignancy and the
total number of patients with follow-up in each group. A
subanalysis was performed in eight studies that reported
SUVmax for confirmed focal lesions, and thereby allowed
calculation of the overall mean SUVmax for benign and
malignant focal lesions.
In case of missing information an attempt was made
to obtain this by contacting the authors of the articles in
question.
Results
Pertinent data from the 22 studies are presented in Table 1.
Twenty studies were retrospective studies and 2 were pro-
spective cohort studies (9,28). The number of nodules per
patient was not evaluated. Therefore, one patient with several
focal nodules in a study was only registered once; for exam-
ple, Choi et al. (6) reported two foci of thyroid malignancy in
 Table 1. Characteristics of 22 Studies Evaluating Prevalence of Malignancy in Incidental Findings
in the Thyroid Gland Using Positron Emission Tomography

Focal Diffuse Patients Benigna,b Malignanta,b

Subjects Indication Imaging uptake uptake with (n) [mean (n) [mean
Study Country/year (n) for imaging technique (n1/n2/n3) (n4/n2/n3) pathologya (n) SUVmax – SD] SUVmax – SD]

Cohen et al. (4) United States/2001 4525 Disease work-upc PET 71/14/7 31/01/0 14 6 [3.4 – 0.2d] 7 [6.9 – 1.5d]
Kang et al. (8) Korea/2003 1330 Disease work-up & healthy PET 21/15/4 8/8/0 15 10 [6.5 – 3.8] 4 [16.5 – 4.7]
volunteer cancer screening
Hsieh et al. (19) Taiwan/2003 477 Disease work-up & healthy PET 12/10/1 NA 8 9 [2.9 – 0.7] 1 [3.2]
volunteer cancer screening
Ishimory et al. (24) United States/2005 1912 Disease work-up PET/CT 29/12/6 NA 11 6 [NA] 6 [NA]
Chen et al. (25) Taiwan/2005 4803 Healthy volunteer cancer PET or PET/CT 60/50/7 NA 50 43 [2.6 – 1.0d] 7 [6.7 – 3.7d]
screening
Kim et al. (5) Korea/2005 4136 Disease work-up PET 45/40/16 45/34/0 32 15e [6.1 – 7.0] 16 [5.1 – 4.3]
Choi et al. (6) Korea/2006 1763 Disease work-up & healthy PET/CT 70/45/17 NA 30 27 [6.7 – 5.5] 17 [10.7 – 7.8]
volunteer cancer screening
Chu et al. (7) United States/2006 6241 Disease work-up PET 76/14/4 NA 14 9 [NA] 4 [NA]
Even-Sapir et al. (23) Israel/2006 2360f Disease work-up PET/CT 59g/41/13 NA 30 28 [NA] 13 [NA]
Are et al. (18) United States/2007 8800 Disease work-up PET or PET/CT 101/42/22 162/15/2 42 20 [NA] 22 [NA]
King et al. (21) United States/2007 15,711 Disease work-up PET or PET/CT 22/22/7 NA 21 15 [NA] 7 [NA]
Bogsrud et al. (20) United States/2007 7347 Disease work-up PET/CT 79/48/15 NA 42 31 [7.9 – 9.7] 15 [7.3 – 3.6]
Kwak et al. (26) Korea/2008 14,434 Disease work-up & healthy PET 88h/85/40 NA 85 45 [6.0 – 5.1] 40 [7.6 – 8.1]
volunteer cancer screening

920
Chen et al. (17) United States/2009 2594 Not reported PET/CT 53/11/7 46/21/0 11 4 [2.9 – 1.4] 7 [4.0 – 1.4]
Bae et al. (27) Korea/2009 3379 Disease work-up & healthy PET/CT 133/68/21 152/31/2 49 45 [NA] 21 [NA]
volunteer cancer screening
Eloy et al. (14) United States/2009 630 Disease work-up PET or PET/CT 30/18/5 NA 18 13 [2.9 – 1.6] 5 [3.4 – 2.6]
Kang et al. (29) Korea/2009 12,840 Disease work-up & healthy PET/CT 612/148/55 539/42/2 148 93 [3.5 – 2.9] 55 [5.9 – 5.4]
volunteer cancer screening
Zhai et al. (28) China/2010 3580 Not reported PET/CT 115/103/48 NA 96 55i [3.8 – 1.7] 48 [6.7 – 3.1]
Ohba et al. (9) Japan/2010 1503 Healthy volunteer cancer PET 20/20/11 NA 20 8 [3.4 – 0.5] 11 [5.4 – 3.4]
screening
Nishimori et al. (15) Canada/2011 4726 Disease work-up PET or PET/CT 103/50/9 57/NA/NA 38 33 [NA] 9 [8.1 – 6.8]
Kim et al. (22) Korea/2010 11,623 Disease work-up PET/CT 159/159/37 NA 140 122 [3.5 – 1.6] 37 [4.5 – 2.1]
Pagano et al. (10) Italy/2011 11,040 Disease work-up & fever of PET/CT 36/36/14 16/16/1 36 22 [NA] 14 [7.4 – 2.9]
unknown origin

n1, subjects with a focal uptake; n2, subjects with a follow-up; n3, subjects with cancer; n4, subjects with a diffuse uptake.
a
Only patients with focal uptake.
b
Some patients (from ‘‘n2’’ in the column ‘‘Focal uptake’’) were not assigned a final diagnosis due to indeterminate FNA.
c
Staging, restaging, or assessing treatment responses in various malignancies.
d
SUVmean analyzed instead of SUVmax.
e
Eight had no SUV and are therefore not included.
f
One patient had known thyroid cancer.
g
All lesions unexpected primary tumors.
h
Included only thyroid incidentalomas that underwent sonographically guided FNA.
i
Seven subjects were not included in the calculation.
CT, computed tomography; PET, positron emission tomography; NA, not analyzed; FNA, fine-needle aspiration; SUVmax, maximum standardized uptake value; SD, standard deviation.
18
F-FDG PET AND THYROID INCIDENTALOMAS
one patient. Eleven studies were carried out as a part of dis-
ease follow-up, which included staging, restaging, or asses-
sing treatment responses in various malignancies
(4,5,7,14,15,18,20–24). Six studies were based on cancer
screening in healthy volunteers and patients with suspected
or known cancer (6,8,19,26,27,29). Two studies included only
healthy volunteers (9,25). One study was carried out in pa-
tients with suspected or known cancer and patients with fever
of unknown origin (10). Two studies provided no information
on selection of subjects for 18F-FDG PET/CT (17,28). Appli-
cation of CT varied among studies (see Table 1).
Of the 125,754 individuals in total, who had an 18F-FDG
PET with or without CT scan performed, 1994 (1.6%) had a
focal incidentaloma in the thyroid gland. Fourteen studies did
not evaluate diffuse uptake (6,7,9,14,15,19–26,28). Nine stud-
ies, one (15) of which did not provide analysis of the data,
described both focal and diffuse uptake. In the remaining
eight studies, diffuse uptake was seen in 999 of 48,644 (2.1%)
individuals (4,5,8,10,17,18,27,29).
There were 1051 out of the 1994 (52.7%) patients with a
focal uptake who had a follow-up. Of them, 1025 had a di-
agnosis assigned and 26 had an indeterminate lesion. A di-
agnosis was assigned in 168 of the 999 (16.8%) patients with
diffuse uptake and none had an indeterminate lesion. There
were 1774 patients with an 18F-FDG uptake in the thyroid
gland who did not have an evaluation. This was either be-
921
FIG. 2. Patient attrition
flowchart. ‘‘Verified’’ means
that the diagnosis (benign or
malignant) is based on either
cytological or histological
specimens. aClassified as
indeterminate by fine-needle
aspiration; bBogsrud et al. (20)
had two malignant patients
with no cytology.
cause they refused evaluation, or were lost to follow-up, or
due to the advanced nature of the primary malignancy hin-
dering further examination.
Focal uptake
The prevalence of focal uptake, evaluated in all 22 studies,
varied from 0.1% (21) to 4.8% (14,29) (mean 2.0%). Among
individuals with an assigned diagnosis, 366 of 1051 with
follow-up (34.8%) were malignant, 659 (62.7%) were benign,
and 26 (2.5%) were indeterminate. In 90.1% (924 of 1025) of
patients with a benign or malignant diagnosis it was estab-
lished by cytology or histology, whereas only 29.3% (271 of
924) were confirmed by operation (see Fig. 2). Two studies,
Chen et al. (17) and Ishimory et al. (24), only provided the total
number of patients with FNA and operation. Therefore, the
number of patients operated in these two studies is unknown.
Accepting FNA as a minimum requirement for diagnosis, the
positive predictive value (PPV) of focal uptake by 18F-FDG
representing malignancy would be no higher than 39.4% (364
of 924). PTC and follicular variant of PTC were the most
prevalent thyroid cancer type (297 out of 366, corresponding
to 81.1%). Fifteen (4.1%) patients had follicular cancer, 5
(1.4%) had medullary cancer, 3 (0.8%) had Hürthle cell car-
cinoma, 31 (8.5%) had unspecified thyroid cancer, 5 (1.4%)
patients had lymphoma, and 10 (2.7%) had metastatic disease
922
from other primary malignancies. One study did not disclose
the subtype of cancer (23). Based on the above, in an inci-
dentally found thyroid lesion with focal uptake, as evaluated
in these studies, the prevalence of thyroid cancer varied from
10.0% (19) to 63.6% (17) (mean 36.0%).
In the 11 studies (5,6,8,9,19,22,25–29) conducted in Asia
the prevalence of focal uptake was 2.2% (1335 of 59,868) and
rate of thyroid cancer was 34.6% (257 of 743 with follow-up).
In the nine studies (4,7,14,15,17,18,20,21,24) conducted in
North America the prevalence of focal uptake was 1.1% (564
of 52,486) and the rate of cancer was 35.5% (82 of 231 with
follow-up). The remaining two studies (10,23) were per-
formed in Europe. In these, the prevalence of focal uptake
was 0.7% (95 of 13,400) and the rate of cancer was 35.0% (27
of 77 with follow-up). Thus, in Asia, the prevalence of focal
uptake was more than twice that in the studies from North
America and Europe while the risk of cancer was similar.
Diffuse uptake
The prevalence of diffuse uptake, evaluated in the eight
studies (4,5,8,10,17,18,27,29), varied from 0.1% (10) to 4.5%
(27) (mean 1.9%). In 77 of the 168 individuals (45.8%) with an
assigned diagnosis this was established by cytology or his-
tology. Only 14.3% (11 of 77) were confirmed by operation
(see Fig. 2). Of the 168 patients, 7 (4.2%) had cancer (3 PTC, 2
FTC, and 2 unknown), 78 (46.4%) had chronic thyroiditis, and
83 (49.4%) had unspecified benign disorders. The prevalence
of cancer in these eight studies varied from 0.0% (4,5,8,17) to
13.3% (18), with a mean of 3.9%.
Four out of eight studies (5,8,27,29) were performed in
Asia. The prevalence of diffuse uptake was 3.4% (744 of
21,685) and rate of cancer was 3.5% (4 of 115 with follow-up).
Three out of eight studies (4,17,18) were carried out in
America. Here, the prevalence of diffuse uptake was 1.5% (239
of 15,919) and the rate of cancer was 5.4% (2 of 37 with follow-
up). The remaining study (10) was performed in Italy. The
prevalence of diffuse uptake was 0.1% (16 of 11,040) and the
rate of cancer was 6.3% (1 of 16).
SUVmax for focal uptake
Eighteen of the 22 studies reported SUVs (4–6,8–
10,14,15,17–20,22,25–29). Cohen et al. (4) and Chen et al. (25)
reported individual SUVmean rather than SUVmax, and the
latter reported SUVmean only for cancer patients. Bae et al. (27)
and Are et al. (18) reported SUVmax as a combination of diffuse
and focal uptake values. Six studies (6,17,22,26,28,29) did not
list the individual SUVmax, but a mean SUVmax for all patients.
Thus, only the remaining eight studies (5,8–10,14,15,19,20)
that listed individual SUVmax were included in the SUVmax
analysis. In the study by Bogsrud et al. (20), two of the lesions
suspected of malignancy were not included in the calculation
of SUVmax since they were considered as extreme outliers.
Nishimori et al. (15) and Pagano et al. (10) only provided an
individual SUVmax for the malignant lesions.
The procedure for using 18F-FDG PET varied considerably
between studies. The period of fasting before radiotracer
administration varied from a minimum of 4 hours (14,17,
23,24,26) to at least 8 hours (5,8,25). In only 10 of 18 studies
(10,14,17,18,22,23,26–29) information was available regarding
the allowed maximum fasting serum glucose level. This level
varied from 120 to 200 mg/dL. The time from the radiotracer
SOELBERG ET AL.
administration to the PET imaging also varied between indi-
viduals (19,25,28) and between the studies, ranging from 40
minutes (4) to 90 minutes (20). Four studies (7,9,21,23) did not
provide information on this variable.
A total of 80 assigned benign and 78 assigned malignant
lesions were identified in the eight eligible studies. The mean
SUVmax was 4.8 (3.1 standard deviation [SD]) for the 80 be-
nign lesions and 6.9 (4.7 SD) for the 78 malignant lesions
( p < 0.001 between groups).
Discussion
Until not too long ago, the use of 18F-FDG PET—with or
without CT, in the management of thyroid disease—was
limited primarily to the postoperative staging of remnant
disease in patients with known differentiated thyroid carci-
noma (30). But the expanding use of PET in clinical practice
stresses the need to clarify the clinical significance of inci-
dental findings in the thyroid gland. Thus, the aim of this
review was to determine the proportion of incidental thyroid
cancers found by 18F-FDG PET—with or without CT of the
thyroid gland—in individuals with no suspicion of thyroid
disease, and to estimate the ability of PET—with or without
CT—to determine whether the 18F-FDG uptake can differen-
tiate between malignant and benign tumors. However, before
trying to amalgamate the data, a number of essential factors
need to be critically assessed.
Dissimilarities among studies
Comparing the studies is not straight forward, since there is
pronounced heterogeneity regarding a number of crucial
variables, such as inclusion and exclusion criteria, selection
bias, study size, study origin and thereby differences in
background risk of thyroid cancer, genetic and environmen-
tal/nutritional differences, and health status—whether
known preexisting nonthyroidal malignancy or normal
healthy volunteers were studied. On top of this, our inter-
pretation is clouded by methodological differences related to
performing the 18F-FDG PET—with or without the CT—and
estimating the SUV, including observer-variation in relation
to whether uptake is focal or diffuse. Generally, these issues
have been given little attention although they most certainly
influence the recorded data to a degree that is not clarified.
Thus, while using a clear set of inclusion and exclusion criteria
for this review, in order to obtain a more homogeneous group
of studies, it is clear from the following that considerable in-
homogeneity remains. Finally, as evident from Table 1, some
studies lack crucial information, which—although authors
were contacted—could not be retrieved. The studies were
performed in different parts of the world where the incidence
rate of thyroid cancer varies, as exemplified when comparing
Asia and America. Thus, the incidence of thyroid cancer
during 2004–2008 per 100,000 persons was 10.9 in Asia and 6.3
in America (31). Interestingly, the studies showed that the
prevalence of focal uptake in Asia was twice as high as in
the studies from America, while the rate of cancer was almost
the same. Additionally, the studies vary regarding the health
status of the subjects. Two studies included only healthy
volunteers (9,25) whereas the other studies reviewed patients
with known malignancy. Based on the above, it is clear that
any conclusions drawn from these very inhomogeneous
studies can only be very crude ones.
18
F-FDG PET AND THYROID INCIDENTALOMAS
Focal uptake
On average 34.8% of individuals with focal uptake were
found to have thyroid malignancy with PTC and follicular
variant of PTC being the most prevalent. Of these thyroid
malignancies, 95.9% were new primary malignancies, while
the remaining were metastases from other cancers. In 15
studies (6,10,14,15,17,18,20–25,27–29) PET/CT was used.
Choi et al. (6) found that 16 of 18 (88.9%) of the malignant
thyroid lesions had low attenuation on CT. All focal thyroid
lesions with a diffuse increase in surrounding thyroid uptake,
or very low attenuation on CT, were benign. Focally increased
18
F-FDG uptakes in the thyroid gland without a correspond-
ing discernible focal anatomic lesion on CT also indicated a
benign lesion with 100% certainty (6). Therefore, uptake of
18
F-FDG PET in incidental findings in combination with CT
should be regarded as a more precise method than 18F-FDG
PET alone, for differentiating between malignant and benign
findings.
The ultimate goal of the evaluation of uptake in the thyroid,
using 18F-FDG, is to differentiate between benign and ma-
lignant lesions. Ideally, a gold standard (e.g., surgical re-
moval) should be available to assess sensitivity and specificity
of 18F-FDG uptake for the diagnosis of thyroid disorders.
Only 29.3% (271 of 924) of focal uptake lesions were con-
firmed by operation. Thus, only a minority had a definitive
diagnosis, in part probably explained by 20 of 22 studies being
retrospective. Accepting FNA as a minimum requirement for
diagnosis, the PPV of focal uptake by 18F-FDG representing
malignancy would be no higher than 39.4%. One cannot ex-
clude that surgical confirmation was most likely obtained in
those patients with the highest likelihood of malignancy and
therefore the malignancy risk in focal uptake is overestimated.
Most importantly, it is unclear whether there is a true varia-
tion as for rate of thyroid malignancy [10.0%–63.6% (17,19)],
or if this is related to methodological variation.
Diffuse uptake
Fourteen out of 22 studies did not evaluate diffuse uptake at
all. Seven studies (15,19,21–24,26) gave no reason for only in-
vestigating focal uptake, while seven studies (6,7,9,14,20,25,28)
did not investigate diffuse uptake because previous reports
indicated that the majority represented benign disease, such as
chronic thyroiditis or Graves’ disease.
Based on the overall data from eight studies reporting on
diffuse 18F-FDG uptake, the prevalence of thyroid malignancy
is 4.2%. Kim et al. (5) found that 34 of 45 patients with diffuse
thyroid 18F-FDG uptake had chronic thyroiditis. In support,
Chen et al. (17) reported the same diagnosis in 21 of 21 pa-
tients. Karantanis et al. (32) investigated 133 patients with
diffuse thyroid uptake and found that 63 (47.4%) of them had
the clinical diagnosis of hypothyroidism or autoimmune
thyroiditis. None had thyroid cancer. Hence, diffuse 18F-FDG
uptake rarely indicates thyroid cancer, but cannot exclude it.
It is important to note that in the eight studies that evaluated
diffuse uptake, only 16.8% (168 of 999) of patients had a final
diagnosis, of whom only 45.8% (77 of 168) had a cytological or
histological confirmation. It follows that sensitivity and
specificity are affected with considerable uncertainty, and risk
of malignancy potentially overestimated. The data cannot
clarify whether patients with diffuse uptake on 18F-FDG PET
need FNA or surgery.
923
SUV evaluation
The 18 studies (4–6,8–10,14,15,17–20,22,25–29) that evalu-
ated SUVs compared their findings with the results of earlier
studies. 18F-FDG is a radioactively marked glucose analogue
that has a half-life of 110 minutes. The 18 studies varied
considerably regarding the method for calculation of the SUV.
This concerned length of fasting period before the exami-
nation, level of fasting serum glucose, volume of injected
18
F-FDG, and the time from the radiotracer administration to
the PET imaging. It follows that there is not only an individual
patient or technique variation but also a procedure variation
and therefore a direct comparison of SUV between studies
may be heavily biased.
Nine studies (6,8,14,17,20,22,26,28,29) have reported mean
SUVmax and investigated whether they could be used to dif-
ferentiate between malignant and benign lesions. Five of these
studies (6,8,22,28,29) found a significant difference. In the
present review, we collected the SUVmax from eight studies
and calculated the mean SUVmax, as being 4.8 – 3.1 for benign
lesions versus 6.9 – 4.7 for malignant lesions. This suggests a
potential utility of SUVmax in differentiating malignant lesions
from benign lesions, although there is pronounced overlap.
Using a high SUVmax as the sole indicator of cancer may be
misleading, since both benign Hürthle cell adenomas and
follicular adenomas have higher SUVmax compared with
other benign conditions (18,20). In patients with known thy-
roid nodules, Mitchell et al. (33) investigated 31 patients with
48 thyroid lesions who underwent FNA and 18F-FDG PET/
CT before surgical resection. A total of 15 of 48 (31%) lesions
were malignant. In that study the mean SUVmax was signifi-
cantly higher in malignant lesions than in benign nodules (6.5
vs. 2.4).
The high rate of malignancy in thyroid PET incidentalomas
has been explained in two ways (34). First, 18F-FDG uptake
often reveals malignant tumors because neoplastic cells exhibit
increased rates of glycolysis and glucose consumption with
accelerated production of glucose transport proteins (35). Sec-
ond, most subjects submitted to 18F-FDG PET studies are pa-
tients who are having cancer staging, restaging, or assessment
of treatment responses of various malignancies, in which the
pretest probability for secondary tumors may be higher (34). Yet
another explanation might be the increasing use of advanced
technology in the diagnosis of a variety of illnesses, whereby
otherwise unrecognized microcarcinomas are discovered.
A large-scale retrospective study from the United States
looked at the incidence and mortality of thyroid cancer over a
30-year period (36). This review found that, although the inci-
dence of thyroid cancer had more than doubled during this
period, mortality from the disease had remained stable. Further,
Ito et al. (37) performed a study over an 8-year period during
which 732 patients had an ultrasound-guided FNA diagnosis of
papillary carcinoma, 162 of whom chose observation. During
the follow-up period, 70% of the tumors in the observational
group remained the same size or decreased in size.
Patients suspected of having a new thyroid cancer, based on
increased focal uptake of 18F-FDG, are facing a new diagnosis, a
new set of concerns, and diagnostic as well as therapeutic
choices. Since it is unclear whether thyroid cancers detected in
this way, compared with any other way, have a different
prognosis it is pertinent to investigate whether the detection
mode affects prognosis, quality of life, and mortality.
924
Conclusion
We found that, despite the inhomogeneity in the studies, the
risk of cancer seems markedly higher when patients have a focal
uptake compared with a diffuse uptake on 18F-FDG PET. The
overall risk is probably overestimated since pathological con-
firmation was obtained in patients with the highest probability
of malignancy. The potential value of SUV determination, al-
though higher in malignant than in benign lesions, needs more
study with standardized methodology. It remains to be clarified
whether malignancy detected by 18F-FDG PET alters prognosis,
mortality, and most importantly quality of life of the patients.
Disclosure Statement
None of the authors have received any financial or other
type of compensation related to the subject of this article.
There are no competing financial interests.
References
1. Hegedus L 2004 Clinical practice. The thyroid nodule. N
Engl J Med 351:1764–1771.
2. Hegedus L, Bonnema SJ, Bennedbaek FN 2003 Management
of simple nodular goiter: current status and future per-
spectives. Endocr Rev 24:102–132.
3. Dean DS, Gharib H 2008 Epidemiology of thyroid nodules.
Best Pract Res Clin Endocrinol Metab 22:901–911.
4. Cohen MS, Arslan N, Dehdashti F, Doherty GM, Lairmore
TC, Brunt LM, Moley JF 2001 Risk of malignancy in thyroid
incidentalomas identified by fluorodeoxyglucose-positron
emission tomography. Surgery 130:941–946.
5. Kim TY, Kim WB, Ryu JS, Gong G, Hong SJ, Shong YK 2005
18F-fluorodeoxyglucose uptake in thyroid from positron
emission tomogram (PET) for evaluation in cancer patients:
high prevalence of malignancy in thyroid PET incidentaloma.
Laryngoscope 115:1074–1078.
6. Choi JY, Lee KS, Kim HJ, Shim YM, Kwon OJ, Park K, Baek
CH, Chung JH, Lee KH, Kim BT 2006 Focal thyroid lesions
incidentally identified by integrated 18F-FDG PET/CT:
clinical significance and improved characterization. J Nucl
Med 47:609–615.
7. Chu QD, Connor MS, Lilien DL, Johnson LW, Turnage RH,
Li BD 2006 Positron emission tomography (PET) positive
thyroid incidentaloma: the risk of malignancy observed in a
tertiary referral center. Am Surg 72:272–275.
8. Kang KW, Kim SK, Kang HS, Lee ES, Sim JS, Lee IG, Jeong
SY, Kim SW 2003 Prevalence and risk of cancer of focal
thyroid incidentaloma identified by 18F-fluorodeoxyglucose
positron emission tomography for metastasis evaluation and
cancer screening in healthy subjects. J Clin Endocrinol Metab
88:4100–4104.
9. Ohba K, Nishizawa S, Matsushita A, Inubushi M, Nagayama
K, Iwaki H, Matsunaga H, Suzuki S, Sasaki S, Oki Y, Okada
H, Nakamura H 2010 High incidence of thyroid cancer in
focal thyroid incidentaloma detected by 18F-fluorodeox-
yglucose [corrected] positron emission tomography in rela-
tively young healthy subjects: results of 3-year follow-up.
Endocr J 57:395–401.
10. Pagano L, Sama MT, Morani F, Prodam F, Rudoni M, Bol-
dorini R, Valente G, Marzullo P, Baldelli R, Appetecchia M,
Isidoro C, Aimaretti G 2011 Thyroid incidentaloma identi-
fied by (18)F-fluorodeoxyglucose positron emission tomog-
raphy with CT (FDG-PET/CT): clinical and pathological
relevance. Clin Endocrinol (Oxf) 75:528–534.
SOELBERG ET AL.
11. Kayani I, Groves AM 2006 18F-fluorodeoxyglucose PET/CT
in cancer imaging. Clin Med 6:240–244.
12. Palmedo H, Bucerius J, Joe A, Strunk H, Hortling N, Meyka
S, Roedel R, Wolff M, Wardelmann E, Biersack HJ, Jaeger U
2006 Integrated PET/CT in differentiated thyroid cancer:
diagnostic accuracy and impact on patient management. J
Nucl Med 47:616–624.
13. Zhuang H, Alavi A 2002 18-fluorodeoxyglucose positron
emission tomographic imaging in the detection and moni-
toring of infection and inflammation. Semin Nucl Med
32:47–59.
14. Eloy JA, Brett EM, Fatterpekar GM, Kostakoglu L, Som PM,
Desai SC, Genden EM 2009 The significance and manage-
ment of incidental [18F]fluorodeoxyglucose-positron-
emission tomography uptake in the thyroid gland in
patients with cancer. AJNR Am J Neuroradiol 30:1431–1434.
15. Nishimori H, Tabah R, Hickeson M, How J 2011 Incidental
thyroid ‘‘PETomas’’: clinical significance and novel descrip-
tion of the self-resolving variant of focal FDG-PET thyroid
uptake. Can J Surg 54:83–88.
16. Salvatori M, Melis L, Castaldi P, Maussier ML, Rufini V,
Perotti G, Rubello D 2007 Clinical significance of focal and
diffuse thyroid diseases identified by (18)F-fluorodeox-
yglucose positron emission tomography. Biomed Pharmaco-
ther 61:488–493.
17. Chen W, Parsons M, Torigian DA, Zhuang H, Alavi A 2009
Evaluation of thyroid FDG uptake incidentally identified on
FDG-PET/CT imaging. Nucl Med Commun 30:240–244.
18. Are C, Hsu JF, Schoder H, Shah JP, Larson SM, Shaha AR
2007 FDG-PET detected thyroid incidentalomas: need for
further investigation? Ann Surg Oncol 14:239–247.
19. Hsieh H, Lin S, Yang B, Chu Y, Chang C, Liu R 2003 The
clinical relevance of thyroid incidentalomas detected by 18F-
fluorodeoxyglucose positron emission tomography. Ann
Nucl Med Sci 16:53–58.
20. Bogsrud TV, Karantanis D, Nathan MA, Mullan BP, Wise-
man GA, Collins DA, Kasperbauer JL, Strome SE, Reading
CC, Hay ID, Lowe VJ 2007 The value of quantifying 18F-
FDG uptake in thyroid nodules found incidentally on
whole-body PET-CT. Nucl Med Commun 28:373–381.
21. King DL, Stack BC Jr, Spring PM, Walker R, Bodenner DL 2007
Incidence of thyroid carcinoma in fluorodeoxyglucose positron
emission tomography-positive thyroid incidentalomas. Oto-
laryngol Head Neck Surg 137:400–404.
22. Kim BH, Na MA, Kim IJ, Kim SJ, Kim YK 2010 Risk strati-
fication and prediction of cancer of focal thyroid fluor-
odeoxyglucose uptake during cancer evaluation. Ann Nucl
Med 24:721–728.
23. Even-Sapir E, Lerman H, Gutman M, Lievshitz G, Zuriel L,
Polliack A, Inbar M, Metser U 2006 The presentation of
malignant tumours and pre-malignant lesions incidentally
found on PET-CT. Eur J Nucl Med Mol Imaging 33:541–
552.
24. Ishimori T, Patel PV, Wahl RL 2005 Detection of unexpected
additional primary malignancies with PET/CT. J Nucl Med
46:752–757.
25. Chen YK, Ding HJ, Chen KT, Chen YL, Liao AC, Shen YY,
Su CT, Kao CH 2005 Prevalence and risk of cancer of focal
thyroid incidentaloma identified by 18F-fluorodeoxyglucose
positron emission tomography for cancer screening in
healthy subjects. Anticancer Res 25:1421–1426.
26. Kwak JY, Kim EK, Yun M, Cho A, Kim MJ, Son EJ, Oh KK
2008 Thyroid incidentalomas identified by 18F-FDG PET:
sonographic correlation. AJR Am J Roentgenol 191:598–603.
18
F-FDG PET AND THYROID INCIDENTALOMAS
27. Bae JS, Chae BJ, Park WC, Kim JS, Kim SH, Jung SS, Song BJ
2009 Incidental thyroid lesions detected by FDG-PET/CT:
prevalence and risk of thyroid cancer. World J Surg Oncol 7:63.
28. Zhai G, Zhang M, Xu H, Zhu C, Li B 2010 The role of 18F-
fluorodeoxyglucose positron emission tomography/com-
puted tomography whole body imaging in the evaluation of
focal thyroid incidentaloma. J Endocrinol Invest 33:151–155.
29. Kang BJ, JH O, Baik JH, Jung SL, Park YH, Chung SK 2009
Incidental thyroid uptake on F-18 FDG PET/CT: correlation
with ultrasonography and pathology. Ann Nucl Med 23:
729–737.
30. Cooper DS, Doherty GM, Haugen BR, Kloos RT, Lee SL,
Mandel SJ, Mazzaferri EL, McIver B, Pacini F, Schlumberger
M, Sherman SI, Steward DL, Tuttle RM 2009 Revised
American Thyroid Association management guidelines for
patients with thyroid nodules and differentiated thyroid
cancer. Thyroid 19:1167–1214.
31. National Cancer Institute. Available at http://seer.cancer
.gov/csr/1975_2008/sections.html (accessed on December
14, 2011).
32. Karantanis D, Bogsrud TV, Wiseman GA, Mullan BP, Sub-
ramaniam RM, Nathan MA, Peller PJ, Bahn RS, Lowe VJ
2007 Clinical significance of diffusely increased 18F-FDG
uptake in the thyroid gland. J Nucl Med 48:896–901.
33. Mitchell JC, Grant F, Evenson AR, Parker JA, Hasselgren PO,
Parangi S 2005 Preoperative evaluation of thyroid nodules
with 18FDG-PET/CT. Surgery 138:1166–1174.
925
34. Van den BA, Maes A, De PT, Mortelmans L, Drijkoningen M,
Van DB, Delaere P, Bouillon R 2002 Clinical relevance of
thyroid fluorodeoxyglucose-whole body positron emission
tomography incidentaloma. J Clin Endocrinol Metab 87:
1517–1520.
35. Ciampi R, Vivaldi A, Romei C, Del GA, Salvadori P, Cosci B,
Pinchera A, Elisei R 2008 Expression analysis of facilitative
glucose transporters (GLUTs) in human thyroid carcinoma cell
lines and primary tumors. Mol Cell Endocrinol 291:57–62.
36. Davies L, Welch HG 2006 Increasing incidence of thyroid
cancer in the United States, 1973–2002. JAMA 295:2164–
2167.
37. Ito Y, Uruno T, Nakano K, Takamura Y, Miya A, Kobayashi
K, Yokozawa T, Matsuzuka F, Kuma S, Kuma K, Miyauchi
A 2003 An observation trial without surgical treatment in
patients with papillary microcarcinoma of the thyroid.
Thyroid 13:381–387.
Address correspondence to:
Laszlo Hegedüs, M.D., D.M.Sc.
Department of Endocrinology and Metabolism
Odense University Hospital
Dk-5000 Odense C
Denmark
E-mail: laszlo.hegedus@ouh.regionsyddanmark.dk