Original article

Comparison of whole-body 18F-FDG PET, 99mTc-MIBI SPET,

and post-therapeutic 131I-Na scintigraphy in the detection
of metastatic thyroid cancer
Masahiro Iwata1, Kanji Kasagi2, Takashi Misaki3, Keiichi Matsumoto4, Yasuhiro Iida5, Takayoshi Ishimori5,
Yuji Nakamoto4, Tatsuya Higashi5, Tsuneo Saga5, Junji Konishi5
1 Department of Radiology, Hikone Municipal Hospital, Hikone, Shiga, Japan
2 Department of Nuclear Medicine, Takamatsu Red Cross Hospital, Takamatsu, Japan
3 Radioisotope Center, Tenri Hospital, Tenri, Japan
4 Institute of Biomedical Research and Innovation, Kobe, Japan
5 Department of Nuclear Medicine and Diagnostic Imaging, Kyoto Graduate School of Medicine, Kyoto, Japan

Received: 28 July 2003 / Accepted: 29 October 2003 / Published online: 10 December 2003
Springer-Verlag 2003

Abstract. The usefulness of fluorine-18 fluorodeoxyglu-
cose (FDG) positron emission tomography (PET) in dif-
ferentiated thyroid cancer (DTC) has been demonstrated
by many investigators, but in only a small number of
studies have FDG-PET images been compared with
those obtained using other non-iodine tumour-seeking
radiopharmaceuticals. In most of the studies, planar im-
aging was performed for comparison using thallium-201
chloride or technetium-99m 2-methoxyisobutylisonitrile
(99mTc-MIBI). Furthermore, FDG-PET studies were not
always performed in the hypothyroid state with in-
creased levels of thyroid stimulating hormone (TSH),
which are known to increase FDG uptake by DTC. The
aim of this study was to compare the ability of FDG-PET
to detect metastatic DTC with that of 99mTc-MIBI whole-
body single-photon emission tomography (SPET) and
post-therapeutic iodine-131 scintigraphy, evaluated un-
der TSH stimulation. Nineteen patients (8 men, 11 wom-
en; age range, 3872 years, mean 60 years; 17 thyroidec-
tomised and 2 inoperable patients following 131I ablation
of the remaining thyroid tissue; 16 papillary and 3 follic-
ular carcinomas) with metastatic DTC underwent FDG-
PET whole-body scan (WBS) and 99mTc-MIBI SPET
WBS at an interval of less than 1 week, followed by 131I
therapy. The SPET images were reconstructed using the
maximum likelihood expectation maximisation (ML-
EM) method. All patients were hypothyroid at the time
of each scan. 131I WBS was performed 35 days after
oral administration of the therapeutic dose. A total of 32
lesions [10 lymph node (LN), 15 lung, 6 bone, 1 muscle]
Masahiro Iwata ()
Department of Radiology, Hikone Municipal Hospital,
1882 Hassakacho, 522-8539 Hikone, Shiga, Japan
e-mail: miwata@municipal-hp.hikone.shiga.jp
Tel.: +81-749-226050, Fax: +81-749-260754
were diagnosed as metastases, as confirmed by histopa-
thology and/or other imaging modalities (X-ray, US, CT,
MRI, bone, 201Tl and 131I scans). FDG-PET, 99mTc-MIBI
SPET and post-therapeutic 131I scintigraphy respectively
revealed a total of 26 (81.3%), 20 (62.5%) and 22
(68.8%) lesions. These techniques respectively demon-
strated nine (90.0%), eight (80.0%) and six (60.0%) LN
metastases, and eleven (73.3%), seven (46.7%) and ten
(66.7%) lung metastases. They each demonstrated five
of the six bone metastases (83.3%). FDG-PET and
99mTc-MIBI SPET were positive in 17 (78.3%) and 14
(63.6%) of the 22 131I-positive lesions, respectively, and
also in nine (90.0%) and six (60.0%) of the ten 131I-nega-
tive lesions, respectively. Three of the five 131I-positive
and FDG-PET-negative lesions were miliary type lung
metastases with a maximal nodular diameter of less than
10 mm. Comparison of FDG-PET with 99mTc-MIBI
SPET revealed concordant results in 24 lesions, and dis-
cordant results in eight lesions (seven with positive
FDG-PET alone and one with positive 99mTc-MIBI
SPET alone). In conclusion: (a) even using whole-body
SPET, FDG PET is superior to 99mTc-MIBI in terms of
ability to detect metastases of DTC; (b) the higher sensi-
tivity of FDG-PET compared with the previous studies
could partly be due to increased serum TSH.
Keywords: Thyroid cancer FDG PET 99mTc-MIBI
SPET
Eur J Nucl Med Mol Imaging (2004) 31:491498
DOI 10.1007/s00259-003-1403-y

European Journal of Nuclear Medicine and Molecular Imaging Vol. 31, No. 4, April 2004
492
Introduction
Differentiated thyroid cancer (DTC) is a disease with a
relatively good prognosis. However, the outcome of pa-
tients with recurrent or metastatic DTC depends on the
size and extent of tumour when detected, indicating that
early discovery of the tumour lesion is important [1]. Flu-
orine-18 fluorodeoxyglucose positron emission tomogra-
phy (FDG-PET) has been useful especially for the evalu-
ation of patients with negative radioactive iodine scintig-
raphy (RIS) and elevated thyroglobulin (Tg) levels [2, 3,
4, 5, 6, 7, 8, 9, 10, 11, 12]. Thus, FDG-PET and RIS play
a complementary role in the detection of metastatic DTC.
As regards comparison with images obtained using
other non-iodine radiopharmaceuticals, FDG-PET is
equal to or more sensitive than thallium-201 or techne-
tium-99m 2-methoxyisobutylisonitrile (99mTc-MIBI)
scanning [2, 3, 4, 5, 12]. However, in most such studies,
including all of the aforementioned, FDG-PET was com-
pared with 201Tl or 99mTc-MIBI planar imaging, not with
single-photon emission tomography (SPET).
Recent studies have revealed that FDG uptake by me-
tastases is enhanced under TSH stimulation [13, 14], How-
ever, most of the studies on DTC have been performed un-
der TSH suppression [7, 8, 10] or under various conditions
showing widely ranging TSH values [3, 4, 6, 9, 12].
In the present study, the ability of the whole-body
FDG-PET to detect metastases in the hypothyroid state
with elevated serum TSH levels was compared to that of
99mTc-MIBI SPET or post-therapeutic 131I scans in pa-
tients with sizeable metastases detected by other imaging
modalities (X-ray, US, CT, MRI and bone scintigraphy)
and with elevated levels of serum Tg.
Materials and methods
Patients. We studied 19 patients (8 men, 11 women; age range,
3872 years, meanSD, 60.010.0 years; 17 thyroidectomised and the femur to the base of the cerebellum. Attenuation correction
2 inoperable patients; 16 papillary and 3 follicular carcinomas)
with metastatic DTC prospectively. FDG PET and subsequent
99mTc-MIBI planar imaging/SPET were performed at an interval
of 16 days, followed by 131I therapy 15 days later, between
November 1999 and July 2001. FDG-PET and 99mTc-MIBI studies
were not performed on the same day because of the possibility that
FDG might interfere with SPET acquisition.
All patients had been given replacement doses of thyroxine
(T4). Five weeks before 131I treatment, T4 supplement was replaced
by triiodothyronine (T3), which was continued for 2 weeks. FDG-
PET and 99mTc-MIBI planar imaging/SPET were performed 1019
and 1620 days after stopping T3 replacement therapy, respective-
ly, when serum TSH levels were considered to be elevated.
Two inoperable patients with advanced thyroid cancer (patients
4 and 17 in Table 1) had previously undergone ablation therapy
(3 GBq) of the normal thyroid tissue.
Diagnosis. Metastasis from thyroid cancer was defined based on
the appearances on X-ray, neck US, CT, MRI and bone, 201Tl and
131I scans, histological findings, increased serum Tg levels and
European Journal of Nuclear Medicine and Molecular Imaging Vol. 31, No. 4, April 2004
evolution of disease during the subsequent follow-up. Absence of
non-thyroidal tumours was confirmed by negative in vitro test re-
sults for other tumour markers and/or by radiographic examina-
tions when indicated.
TSH and Tg measurements. Serum TSH and Tg concentrations
were measured by immunoradiometric assay (TSH: Riagnost
hTSH, CIS Diagnostic, Chiba, Japan; normal range 0.33.9 mU/l;
thyroglobulin: Thyroglobulin-IRMA, Daiichi Radioisotope Labo-
ratories, Tokyo, Japan; normal range <45 ng/ml) 38 h before the
administration of the therapeutic dose of Na131I. Anti-Tg antibod-
ies were determined in all patients and were negative except in
one patient (patient 2 in Table 1).
99mTc-MIBI imaging. 99mTc-MIBI planar and SPET images were
obtained 15 min and 2 h after intravenous injection of 600 MBq
99mTc-MIBI respectively. All planar and SPET images were obtained
with a dual-headed SPET system (RC-2500 IV, Hitachi) equipped
with a low-energy, high-resolution collimator. Planar images
(256256 matrix, 106 counts) were acquired in the anterior and pos-
terior projections and peaked at 140 keV with a symmetrical 15%
window. Whole-body SPET was performed with a 128128 matrix
for 32 projections (232; 5.625/step) and an imaging time of 12 s per
projection. Transverse and coronal sections were reconstructed using
the maximum likelihood-expectation maximisation (ML-EM) meth-
od with a Butterworth filter with a cut-off frequency of 2.15 cy-
cles cm1 and obtained from the femur to the base of the cerebellum.
The iteration number of ML-EM was 30. Slice thickness was 5 mm.
Neither attenuation correction nor scatter correction was performed.
FDG-PET imaging. FDG-PET was performed 211 days prior to
131I treatment using a high-resolution, whole-body PET scanner
with an 18-ring detector arrangement (Advance; General Electric
Medical Systems, Milwaukee, WI). The system permitted the si-
multaneous acquisition of 35 axial images with inter-slice spacing
of 4.25 mm. Axial resolution was 4.2 mm full-width at half-maxi-
mum intensity, allowing multidirectional reconstruction of the im-
ages without loss of resolution. The field of view and pixel size of
the reconstructed images were 256 and 2 mm, respectively.
Approximately 370 MBq of FDG was injected intravenously after
a fasting period of 6 h. All patients were relaxed prior to and after
injection of FDG. Whole-body PET images were acquired from
was performed in all patients except nos. 15.
131I imaging. Post-therapy 131I whole-body scanning (WBS) was
performed using a large field of view gamma camera (Bodyscans,
Siemens, Hoffman Estate, IL) fitted with a high-energy parallel-
hole collimator. The photo peak was 364 keV with a symmetrical
20% window. The images were acquired 35 days after oral ad-
ministration of the therapeutic dose (3.36.7 GBq; mean dose
SD, 5.030.95 GBq).
Evaluation. All images were visually interpreted by three experi-
enced nuclear medicine physicians on a lesion basis for detection
of metastases in lymph node (LN), lung, bone and other regions.
Only one patient (no. 19) had a metastasis in a region other than
LN, lung and bone, namely in the muscle of the leg. Consensus
was reached concerning the visualisation or non-visualisation of
lesions. The number of lesions in each site was defined as one
even if multiple lesions were observed. As for 99mTc-MIBI, a find-
ing was defined as positive if the lesion was visualised either on
planar images or on SPET.
 Table 1. Clinical backgrounds of 19 patients with differentiated thyroid cancer in relation to FDG-PET, 99mTc-MIBI and 131I results

Patient no. Age/sex Histology Tga (TSH) Tgb (TSH) Metastatic site Maximal FDG-PET MIBI (P) MIBI (S) Planar (n) 131I

on T4 off T4 diameter (cm) vs SPET (n)

1 68/F PAC 19,668 (0.05) 80,424 (>49) Lung 3.5 + + + ND +

Bone (sternum) 3.0 + + + ND +
2 67/M PAC 2,719 (0.12)c 21,834 (>49)c Neck LN 2.5 + + + P(9)<S(11) +
3 62/F PAC 348 (NA) 617 (>49) Neck and mediastinal LN 1.8 + ND +
4 55/M PAC 514 (NA) 811 (21.7) Neck LN 2.8 + + + ND +
Lung 2.0 + + + ND +
5 68/M FAC 6,681 (0.18) 19,511 (>49) Lung 3.0 + + + ND +
Bone (cervical spine) 2.2 + + ND +
6 43/F PAC 891 (0.58) 7,985 (>49) Neck and mediastinal LN 1.5 + + P(0)<S(1)
Lung 1.2 + ND
7 65/M PAC 1,136 (1.3) 1,842 (>49) Neck LN 2.4 + + + ND +
Lung 1.3 + ND +
8 39/F PAC 90 (0.12) 703 (>49) Neck LN 1.1 + + + ND
Lung (miliary) 0.8 + + P(0)<S(1)
9 64/M PAC 6.1 (<0.03) 71 (>49) Lung 1.2 + ND +
10 61/F FAC 3,190 (0.79) 63,431 (>49) Lung (miliary) 1.0 ND +
Bone (sternum, rib, sacrum) 3.0 + + + ND +
11 57/F PAC 103 (0.05) 939 (>49) Lung 1.8 + + + P(3)<S(5) +
12 71/M PAC 100 (0.05) 633 (>49) LN (neck, supraclavicular) 2.4 + + + P(1)<S(2)
Lung (miliary) 1.0 ND
Bone (rib) d + ND
13 71/F PAC 2,438 (0.16) 30,599 (>49) Neck LN 2.0 + + + P(3)<S(5) +
14 38/M PAC 175 (0.21) 1,158 (>49) Neck LN 1.5 ND +
Lung (miliary) 0.9 ND +

European Journal of Nuclear Medicine and Molecular Imaging Vol. 31, No. 4, April 2004
15 72/F PAC 334 (<0.03) 10,197 (>49) Neck LN 1.7 + + + ND
Lung 2.2 + + + P(4)<S(8)
16 55/F PAC 155 (<0.03) 2,701 (>49) Lung (miliary) 1.0 + ND
17 60/F PAC 190 (1.4) 398 (>49) Lung (miliary) 0.6 ND +
18 70/F FAC 3,751 (0.03) 10,879 (43.3) Bone (sacroiliac joint) 3.5 + + + ND +
19 53/M PAC 508 (<0.03) 7,045 (>49) Lung 1.1 + + + ND +
Bone (rib) 1.3 + + + ND +
Other (muscle of the leg) d + ND +

PAC, Papillary adenocarcinoma; FAC, follicular adenocarcinoma; MIBI (P), 99mTc-MIBI planar imaging; MIBI (S), 99mTc-MIBI SPET imaging; NA, not available; ND, no differ-
ence; n, number of detectable lesions at each site (LN, lung, bone, others)
a Serum Tg levels under TSH suppression (Tg: ng/ml, TSH: U/ml)
b Serum Tg levels under TSH stimulation
c TgAb positive
d The size could not be assessed because only scintigraphic findings including bone, 131I, 99mTc-MIBI and FDG were available
493
494

Table 2. Comparison of 131I, 99mTc-MIBI

FDG-PET and 99mTc-MIBI A SPET
SPET in metastatic DTC FDG PET Positive Negative Total
Positive 19 (8 LN, 7 lung, 4 bone) 7 (1 LN, 4 lung, 1 bone, 1 muscle) 26
Negative 1 (1 bone) 5 (1 LN, 4 lung) 6
Total 20 12 32
P<0.05
B 131I

FDG PET Positive Negative Total

Positive 17 (5 LN, 7 lung, 4 bone, 1 muscle) 9 (4 LN, 4 lung, 1 bone) 26
Negative 5 (1 LN, 3 lung, 1 bone) 1 (1 lung) 6
Total 22 10 32
NS
C 131I

99mTc-MIBI SPET Positive Negative Total

Positive 14 (4 LN, 5 lung, 5 bone) 6 (4 LN, 2 lung) 20
The total number of lesions for Negative 8 (2 LN, 5 lung, 1 muscle) 4 (3 lung, 1 bone) 12
LN, lung, bone and muscle were Total 22 10 32
10, 15, 6 and 1, respectively NS
NS, No significant association

Fig. 1AD. A 43-year-old

woman with cervical lymph
nodes (LNs) and lung metastas-
es of poorly differentiated pap-
illary thyroid carcinoma (pa-
tient 6). A Neck US and chest
CT show cervical lymphaden-
opathy and multiple small lung
metastases, respectively. LN,
Lymph node; C, common ca-
rotid artery. B 131I scan shows
no definite abnormality. C
99mTc-MIBI planar imaging

also shows no significant accu-

mulation, but the SPET image
demonstrates LN metastasis
(arrow). D The cervical LNs
and lung metastases are detect-
ed by FDG-PET scan

Statistical analysis. Chi-square test was used to determine whether
there was a significant association between any two of the scan re-
sults using 131I, FDG and 99mTc-MIBI. The size of FDG-positive
and FDG-negative lesions was compared using Students t test. A
P value of <0.05 was considered statistically significant.
Results
A total of 32 lesions (10 LN, 15 lung, 6 bone, 1 muscle)
were diagnosed as metastases, as confirmed by histopathol-
ogy and/or by other imaging modalities, including
X-ray, US, CT, MRI, bone scan, 201Tl scan and whole-body
radioactive iodine scan (RIS) in all patients. Thirty-one of
the 32 lesions (96.9%) were detected by at least one of the
three modalities FDG-PET, 99mTc-MIBI and RIS (Table 1).
Small lung metastases were not visualised by any of the
three modalities in one patient (no. 12 in Table 1). Two false
positive lesions were detected by FDG-PET. One was a be-
nign colon polyp diagnosed by biopsy (patient 4). The other
lesion was attributed a radiological diagnosis of retroperito-

European Journal of Nuclear Medicine and Molecular Imaging Vol. 31, No. 4, April 2004
 495
Table 3. Comparison of tumour size on US, CT or MRI between FDG-positive lesions and FDG-negative lesions

FDG positive (n=24) FDG negative (n=6) Statistical significance

Tumour size (cm) 2.000.82 1.200.57 P<0.05

RIS positive (n=16) RIS negative (n=8) RIS positive (n=5) RIS negative (n=1)
Tumour size (cm) 2.260.81*,** 1.490.58* 1.240.63** 1.0 *,**: P<0.05

RIS, Radioactive iodine scan

Two cases with bone and muscular metastases (patients 12 and 19) were excluded because the size could not be assessed accurately by
CT or MRI

Fig. 2AD. A 38-year-old man

with cervical LNs and lung me-
tastases of well-differentiated
thyroid follicular carcinoma
(patient 14). A Neck US and
chest CT show mild cervical
lymphadenopathy and multiple
lung metastases (miliary type),
respectively. LN, Lymph node;
J, internal jugular vein; C,
common carotid artery. B 131I
scan shows cervical LNs and
lung metastases. C 99mTc-MIBI
planar and SPET images show
no significant accumulation at
either site. D FDG-PET scan is
also negative

neal tumour, but histological diagnosis was lacking (patient
5). Since this tumour showed no 131I uptake and no signifi-
cant change in size during the follow-up period of 5 years,
we did not include it in the present study. No false-positive
lesions were identified by either 99mTc-MIBI or RIS.
FDG-PET, 99mTc-MIBI SPET and post-therapeutic
131I scintigraphy respectively revealed a total of 26
(81.3%), 20 (62.5%) and 22 (68.8%) lesions. These tech-
niques respectively demonstrated nine (90.0%), eight
(80.0%) and six (60.0%) LN metastases, and eleven
(73.3%), seven (46.7%) and ten (66.7%) lung metastas-
es. They each demonstrated five of the six bone metas-
tases (83.3%). (Tables 1, 2). FDG-PET and 99mTc-MIBI
SPET were positive in 17 (78.3%, 12 patients) and 14
(63.6%, ten patients) of 22 RIS-positive lesions (14 pa-
tients), respectively, and also in nine (90.0%, five pa-
tients, Fig. 1) and six (60.0%, three patients) of ten RIS-
negative lesions (five patients), respectively (Table 2).
When the size of positive metastases was compared be-
tween FDG-positive and -negative groups, the maximal
diameter of FDG-positive lesions was significantly
greater than that of FDG-negative lesions (P<0.05, Ta-
ble 3). The FDG-positive and RIS-positive lesions were
significantly larger than the FDG-negative and RIS-posi-
tive lesions as well as the FDG-positive and RIS-nega-
tive lesions (P<0.05, Table 3). Only 18 lesions (56.3%)
showed concordant results in respect of FDG-PET and
131I uptake, and, similarly, only 18 showed concordant
results regarding 99mTc-MIBI and 131I uptake. Thus,
there was no significant association between 131I uptake
and FDG or 99mTc-MIBI uptake. Three of the five 131I-
positive and FDG-PET-negative lesions (four patients)
were miliary type lung metastases with a maximal nodu-
lar diameter of less than 10 mm (Fig. 2). Of six patients

European Journal of Nuclear Medicine and Molecular Imaging Vol. 31, No. 4, April 2004
496
Fig. 3AD. A 68-year-old man
with lung (non-miliary type)
and bone (cervical spine) me-
tastases of well-differentiated
follicular thyroid carcinoma
(patient 5). A The metastases
are demonstrated on MRI and
spine CT images. B 131I scan
shows lung and bone metastas-
es. C 99mTc-MIBI scan detects
lung and bone metastases even
by planar imaging. D Lung me-
tastases, but no bone metastas-
es, were detected by FDG-PET
scan. A retroperitoneal tumour
is visualised as a false positive
result (arrow)

with miliary lung metastases, four had negative FDG-
PET scans and five, negative 99mTc-MIBI scans.
Comparison of FDG-PET with 99mTc-MIBI SPET is
also demonstrated in Table 2. FDG uptake was concor-
dant with 99mTc-MIBI uptake in 24 lesions (75.0%).
There was a significant association between FDG up-
take and 99mTc-MIBI uptake (P<0.05). FDG-PET results
correlated better to 99mTc-MIBI than to RIS results.
There were seven lesions with positive FDG-PET and
negative 99mTc-MIBI, in marked contrast to only one le-
sion with negative FDG and positive 99mTc-MIBI
(Fig. 3). The size of FDG-positive and 99mTc-MIBI-pos-
itive metastases, with a maximal diameter of
2.190.82 cm (meanSD; n=19), was significantly
greater than that of FDG-positive and 99mTc-MIBI-nega-
tive metastases (1.300.30 cm; n=5) and that of FDG-
negative and 99mTc-MIBI-negative metastases (1.00
0.32 cm; n=5). Thus, small metastases seem not to be
visualised on 99mTc-MIBI images.
There were 20 99mTc-MIBI-positive lesions (20/32;
62.5%). Among them, two lesions (LN metastasis in pa-
tient 6 and lung metastasis in patient 8) were detected by
SPET alone. The remaining 18 lesions were positive on
both 99mTc-MIBI planar and SPET images. The number
of metastases detectable by SPET was greater than that
detectable by planar imaging in five metastatic sites
from five patients (LN: patients 2, 12 and 13; lung: pa-
tients 11 and 15). These five patients had a total of 11
small metastases that were missed on planar images.
Thus, the use of SPET increased the detectability of
metastatic DTC in seven patients (nos. 2, 6, 8, 11, 12, 13
and 15).
Discussion
Various nuclear medicine imaging techniques using ra-
dioactive iodine, 201Tl, 99mTc-MIBI, 99mTc-tetrofosmin

European Journal of Nuclear Medicine and Molecular Imaging Vol. 31, No. 4, April 2004

and FDG-PET have been employed in patients with re-
current and/or metastatic DTC, the reported sensitivity
being 4262%, 4594%, 5088%, 8689% and 5078%
respectively [15].
To our knowledge, there have been only four studies
in which the ability of FDG-PET to detect metastatic
and/or recurrent DTC was compared with that of 99mTc-
MIBI scintigraphy [3, 4, 5, 12]. Grnwald et al. [4] and
Dietlein et al. [5] performed 99mTc-MIBI SPET studies
(head and neck, and neck and chest, respectively) in ad-
dition to planar imaging. However, whole-body SPET, as
presented in this study, has never been employed. Ac-
cording to the aforementioned studies, FDG-PET has a
higher sensitivity than [3, 4] or a similar sensitivity to
99mTc-MIBI [5]. Fridrich et al. [12] reported that 18F-
FDG rectilinear scanning and 99mTc-MIBI provided simi-
lar results. The lower sensitivity (50%, 11/22 for both
FDG and 99mTc-MIBI) reported by Dietlein et al. [5] as
compared with the present study (81.3% and 62.5%, re-
spectively) could be explained by patient selection, since
their patients showed lower levels of serum Tg, indicat-
ing that they had less advanced disease compared with
our patients. On the other hand, the sensitivity of FDG-
PET and 99mTc-MIBI was 75% and 53%, respectively,
according to the multicentre study reported by Grnwald
et al. [3], which are similar to the results of the present
study.
99mTc-MIBI is known to be a diagnostic tool for pa-
tients with recurrent and/or metastatic DTC [16, 17, 18,
19]. Miyamoto et al. [18] reported that 99mTc-MIBI has
an advantage over 201Tl because it detects more macro-
nodular lesions owing to its better image quality. With
regard to the relation between FDG and 99mTc-MIBI up-
take, the results were concordant in 75% of the lesions in
this study. Such a high percentage is similar to that re-
ported by Dietlein et al. [5] (81.8%) and Grnwald et al.
[4] (76.9%). 99mTc-MIBI uptake is known to be related
to the metabolic demand, which influences mainly the
mitochondrial potential, and can be expected to be in the
same range as FDG uptake, which reflects glucose me-
tabolism in differentiated thyroid carcinoma [4].
Comparison of 99mTc-MIBI planar and SPET images,
obtained 15 min and 2 h following injection, respective-
ly, revealed that the use of whole-body SPET helped to
localise lesions and increased the detectability in several
cases in spite of the evidence that delayed scanning is
less sensitive than early scanning [18]. Nevertheless, in
the present study, FDG-PET was more sensitive for de-
tection of tumours than 99mTc-MIBI SPET. Seven of the
eight discrepant results were FDG-positive and 99mTc-
MIBI-negative. We consider that FDG is superior to
99mTc-MIBI in terms of ability to detect metastases of
DTC, even if the comparison is made between PET and
whole-body SPET, although the possibility that perfor-
mance of earlier SPET scans might have increased the
sensitivity to some extent cannot be ruled out. The ob-
served lower sensitivity of 99mTc-MIBI is probably
European Journal of Nuclear Medicine and Molecular Imaging Vol. 31, No. 4, April 2004
497
caused by the inferior spatial resolution of SPET imag-
ing (1012 mm, compared with 46 mm for PET) [3],
since the small metastases were missed by 99mTc-MIBI
imaging. However, it should be borne in mind that both
FDG-PET and 99mTc-MIBI failed to detect miliary lung
metastases, indicating that spiral CT is necessary in such
cases.
Previous studies have revealed the role of FDG-PET
in patients with negative radioactive iodine scan (RIS)
and elevated Tg, with reported sensitivities of 6494%
(sensitivity in individual studies: 64.3% [2], 85% [3],
90.5% [6], 75% [7], 63.7% [8], 69.4% [9], 94% [10],
82% [11] and 8294% [15]). On the other hand, the sen-
sitivity of FDG in detecting RIS-positive lesions has
been reported to be as low as 1665% (sensitivity in in-
dividual studies: 39.4% [2], 65.0% [3], 45.5% [4], 16.4%
[6] and 47.6%[11]) according to previous studies. It is
generally accepted that glucose metabolism is increased
in poorly differentiated carcinomas with enhanced ex-
pression of glucose transporter genes (GLUT), in partic-
ular GLUT1 associated with a loss of radioiodine uptake
[20, 21]. We agree that FDG-PET is more useful for thy-
roid cancer patients with RIS-negative and elevated Tg
levels than for those with positive RIS, since identifica-
tion of the focus of disease activity can appropriately di-
rect additional therapy such as surgical excision and ex-
ternal beam radiation. The sensitivity of FDG for the de-
tection of RIS-positive lesions (78.3%) in this study was
higher than that reported in previous studies [2, 3, 4, 6,
11]. Such a high sensitivity can be explained by the en-
rolment of patients with extensive disease. Wang et al.
[22] reported that the survival rate was lower in patients
with distant metastases which concentrate both 131I and
FDG than in those with positive RIS but negative FDG
scan. In the present study, the size of metastases with 131I
and FDG uptake was significantly greater than that of
metastases with 131I-positive but FDG-negative metas-
tases. Large tumour masses are known to be radioresis-
tant, and surgery is occasionally indicated, even if the tu-
mour concentrates radioiodine. In the case of a large
functioning metastasis in the bone, which cannot be
readily excised surgically, external radiation may be ben-
eficial to relieve pain [23]. Uptake of radioiodine in an
FDG-avid lesion implies that the malignant cells contain
both differentiated and undifferentiated phenotypes.
Thus, we would propose that the utility of FDG scan is
not limited to RIS-negative cases.
Mueller et al. [17] reported that 99mTc-MIBI uptake in
thyroid carcinoma is independent of TSH stimulation.
Although an inter-individual comparison has revealed
that FDG uptake is also independent of TSH stimulation
[3], Moog et al. [13] recently presented an intra-individ-
ual comparison between FDG-PET and serum TSH lev-
els, in which FDG uptake by recurrent and metastatic
DTC was increased under TSH stimulation during with-
drawal of thyroid hormone replacement therapy. Petrich
et al. [14] reported that the administration of recombi-
498
nant human thyrotropin (rhTSH) resulted in increased tu-
mour to background ratios and standardised uptake val-
ues, which were significantly different from those ob-
tained under TSH suppression. In the present study, FDG
PET studies were performed under TSH stimulation in
all cases. Although intra-individual comparison was not
made, the higher sensitivity of FDG PET (81.3%) com-
pared with previous studies (5078%) [15] could partly
be due to increased serum TSH.
In conclusion, even when whole-body SPET was
used, FDG PET was superior to 99mTc-MIBI in terms of
ability to detect metastases of DTC, and we consider that
FDG-PET is a valuable diagnostic tool for the detection
not only of 131I-negative lesions but also of 131I-positive
lesions of metastatic DTC. For patients who are sched-
uled to undergo 131I therapy for recurrent and/or meta-
static DTC, we would recommend that FDG-PET be per-
formed just before 131I therapy under TSH stimulation.
Acknowledgements. We sincerely thank Miss Yuko Ono for excel-
lent technical and secretarial assistance.
References
1. Schlumberger MJ. Diagnostic follow-up of well-differentiated
thyroid carcinoma: historical perspective and current status.
J Endocrinol Invest 1999; 22 (11 suppl):37.
2. Shiga T, Tsukamoto E, Nakada K, Morita K, Kato T, Mabuchi
M, Yoshinaga K, Katoh C, Kuge Y, Tamaki N. Comparison
of18F-FDG, 131I-Na, and 201Tl in diagnosis of recurrent or
metastatic thyroid carcinoma. J Nucl Med 2001; 42:414419.
3. Grnwald F, Klicke T, Feine U, Lietzenmayer R, Scheidhauer
K, Dietlein M, Schober O, Lerch H, Brandt-Mainz K, Burchert
W, Hiltermann G, Cremerius U, Biersack HJ. Fluorine-18 flu-
orodeoxyglucose positron emission tomography in the thyroid
cancer: results of multicentre study. Eur J Nucl Med 1999;
26:15471552.
4. Grnwald F, Menzel C, Bender H, Palmedo H, Willkomm P,
Ruhlmann J, Franckson T, Biersack HJ. Comparison of18FDG-
PET with 131iodine and 99mTc-sestamibi scintigraphy in differ-
entiated thyroid cancer. Thyroid 1997; 7:327335.
5. Dietlein M, Scheidhauser K, Voth E, Theissen P, Schicha H.
Follow-up of differentiated thyroid cancer: what is the value
of FDG and sestamibi in the diagnostic algorithm? Nu-
klearmedizin 1998; 37:611.
6. Feine U, Lietzenmayer R, Hanke JP, Held J, Whrle H, Mller-
Schauenburg W. Fluorine-18-FDG and iodine-131-iodine up-
take in thyroid cancer. J Nucl Med 1996; 37:14681472.
7. Helal BO, Merlet P, Toubert ME, Franc B, Schvartz C,
Gauthier-Koelesnikov H, Prigent A, Syrota A. Clinical impact
of18F-FDG PET in thyroid carcinoma patients with elevated
thyroglobulin levels and negative 131I scanning results after
therapy. J Nucl Med 2001; 42:14641469.
8. Alnasfisi NS, Driedger AA, Coates G, Moote DJ, Raphael SJ.
FDG PET of recurrent or metastatic131I-negative papillary thy-
roid carcinoma. J Nucl Med 2000; 41:10101015.
European Journal of Nuclear Medicine and Molecular Imaging Vol. 31, No. 4, April 2004
9. Schlter B, Bohuslavizki KH, Beyer W, Plotkin M, Buchert R,
Clausen M. Impact of FDG PET on patients with differentiat-
ed thyroid cancer who present with elevated thyroglobulin and
negative131I scan. J Nucl Med 2001; 42:7176.
10. Chung JK, So Y, Lee JS, Choi CW, Lim SM, Lee DS, Hong
SW, Youn YK, Lee MC, Cho BY. Value of FDG PET in papil-
lary thyroid carcinoma with negative131I whole-body scan. J
Nucl Med 1999; 40:986992.
11. Dietlein M, Scheidhauser K, Voth E, Theissen P, Schicha H. Flu-
orine-18 fluorodeoxyglucose positoron emission tomography
and iodine-131 whole-body scintigraphy in the follow-up of dif-
ferentiated thyroid cancer. Eur J Nucl Med 1997; 24:13421348.
12. Fridrich L, Messa C, Landoni C, Lucignani G, Moncayo R,
Kendler D, Riccabona G, Fazio F. Whole-body scintigraphy
with99mTc-MIBI, 18F-FDG and 131I in patients with metastatic
thyroid carcinoma. Nucl Med Commun 1997; 18:39.
13. Moog F, Linke R, Manthey N, Tiling R, Knesewitsch P, Tatsch
K, Hahn K. Influence of thyroid-stimulating hormone levels
on uptake of FDG in recurrent and metastatic differentiated
thyroid carcinoma. J Nucl Med 2000; 41:19891995.
14. Petrich T, Brner AR, Otto D, Hofmann M, Knapp WH. Influ-
ence of rhTSH on [18F]fluorodeoxyglucose uptake by differen-
tiated thyroid carcinoma. Eur J Nucl Med Mol Imaging 2002;
29:641647.
15. Wong CO, Dworkin HJ. Role of FDG PET in metastatic thy-
roid cancer. J Nucl Med 1999; 40:993994.
16. Dadparvar S, Chevres A, Tulchinsky M, Krishna-Badrinath L,
Khan AS, Slizofski WJ. Clinical utility of technetium-99m
methoxyisobutylisonitrile imaging in differentiated thyroid
carcinoma: comparison with thallium-201 and iodine-131 Na
scintigraphy, and serum thyroglobulin quantitation. Eur J Nucl
Med 1995; 22:13301338.
17. Mueller SP, Piotrowski B, Guth-Tougelides B, Reiners C.
Technetium-99m MIBI and Tl-201 uptake in thyroid carcino-
ma [abstract]. J Nucl Med 1988; 29 (Suppl):854.
18. Miyamoto S, Kasagi K, Misaki T, Alam MS, Konishi J. Evalu-
ation of technetium-99m-MIBI scintigraphy in metastatic dif-
ferentiated thyroid carcinoma. J Nucl Med 1997; 38:352356.
19. Alam MS, Kasagi K, Misaki T, Miyamoto S, Iwata M, Iida Y,
Konishi J. Diagnostic value of technetium-99m methoxyisobutyl
isonitrile (99mTc-MIBI) scintigraphy in detecting thyroid cancer
metastases: a critical evaluation. Thyroid 1998; 8:10911110.
20. Lazar V, Bidart JM, Caillou B, Mah C, Lacroix L, Filetti S,
Schlumberger M. Expression of the Na+/I symporter gene in
human thyroid tumors: a comparison study with other thyroid-
specific genes. J Clin Endocrinol Metab 1999; 84:32283234.
21. Grnwald F, Biersack HJ. FDG PET in thyroid cancer: thyrox-
ine or not?. J Nucl Med 2000; 41:19961998.
22. Wang W, Larson SM, Fazzari M, Tickoo SK, Kolbert K,
Sgouros G, Yeung H, Macapinlac H, Rosai J, Robbins RJ.
Prognostic value of [18F]fluorodeoxyglucose positron emis-
sion tomographic scanning in patients with thyroid cancer.
J Clin Endocrinol Metab 2000; 85:11071113.
23. Schlumberger M, Challeton C, Vathaire FD, Travagli JP,
Gardet P, Lumbroso JD, Francese C, Fontaine F, Ricard M,
Parmentier C. Radioactive iodine treatment and external radio-
therapy for lung and bone metastases from thyroid carcinoma.
J Nucl Med 1996; 37:598605.