Article

The Usual Ultrasonographic Features of Thyroid Cancer Are Less Frequent in Small Tumors That Develop After a Long Latent Period After the Chernobyl Radiation Release Accident

Department of Thyroid Disease Research, Belarusian Medical Academy for Postgraduate Education, Minsk, Republic of Belarus.
Thyroid: official journal of the American Thyroid Association (Impact Factor: 4.49). 06/2009; 19(7):725-34. DOI: 10.1089/thy.2008.0238
Source: PubMed
ABSTRACT
The Chernobyl accident resulted in an unprecedented number of radiation-induced thyroid cancers in young individuals as detected by national and international screening programs. The vast majority of thyroid malignancies were papillary carcinomas that, despite being similar by histopathology, displayed large variability in clinical course. The correlations between ultrasound (US) and clinicopathological features in young patients with radiation-induced thyroid cancer, however, have not been well studied. Because of the importance of US for deciding which subjects should have fine-needle aspiration biopsy, we assessed the US features of papillary thyroid carcinoma in patients exposed to Chernobyl fallouts.
We performed a retrospective multivariate logistic regression analysis of US features, clinicopathological data, and the latency period between radiation exposure and the diagnosis of cancer in 94 patients who were 10.6-34.3 years old (16.5 +/- 6.2, mean +/- standard deviation) at the time of diagnosis and 0.1-18.0 years old (5.6 +/- 4.2) at the time of the Chernobyl accident.
Nodules greater than 10 mm were associated with the higher frequency of irregular margins (p = 0.001), longer period of latency (p = 0.016), and bilateral lymph node involvement (p = 0.025). Irregular tumor margins correlated with the shorter period of latency (p = 0.009) and unilateral nodal disease (p = 0.010). Hypoechoic nodules were observed more frequently in female patients (p = 0.012), in the absence of halo (p = 0.003) or calcifications (p = 0.005). Hypoechogenicity also correlated with the shorter latency (p = 0.015) and younger age of patients (p = 0.048).
Irregular nodule margins, a usual sign of malignancy, are less useful in detecting thyroid cancers in radiation-exposed patients with tumors less than 10 mm. Thyroid cancers that are detected after longer latent periods display less of the US features characteristic of a malignant process, while benign US features are observed more frequently. Therefore, we recommend fine-needle aspiration biopsy to ensure early diagnosis of thyroid cancer for patients with a history of radiation exposure, even if their nodules are less than 10 mm.

Full-text

Available from: Yuri Demidchik
The Usual Ultrasonographic Features of Thyroid Cancer
Are Less Frequent in Small Tumors That Develop
After a Long Latent Period After the Chernobyl
Radiation Release Accident
Valentina M. Drozd,
1
Maxim L. Lushchik,
1
Olga N. Polyanskaya,
1
Michail V. Fridman,
1
Yuri E. Demidchik,
1
Andrej P. Lyshchik,
2
Johannes Biko,
3
Christoph Reiners,
3
Yoshisada Shibata,
4
Vladimir A. Saenko,
5
and Shunichi Yamashita
4,5
Background: The Chernobyl accident resulted in an unprecedented number of radiation-induced thyroid cancers
in young individuals as dete cted by national and international screening programs. The vast majority of thyroid
malignancies were papillary carcinomas that, despite being similar by histopathology, displayed large vari-
ability in clinical course. The correlations between ultrasound (US) and clinicopathological features in young
patients with rad iation-induced thyroid ca ncer, however, have not been well studied. Because of the importance
of US for deciding which subjects should have fine-needle aspiration biopsy, we assessed the US features of
papillary thyroid carcinoma in patients exposed to Chernobyl fallouts.
Design: We performed a retrospective multivariate logistic regression analysis of US features, clinicopathological
data, and the latency period between radiation exposure and the diagnosis of cancer in 94 patients who were
10.6–34.3 years old (16.5 6.2, mean standard deviation) at the time of diagnosis and 0.1–18.0 years old
(5.6 4.2) at the time of the Chernobyl accident.
Results: Nodules greater than 10 mm were associated with the higher frequency of irregular margins ( p ¼ 0.001),
longer period of latency ( p ¼ 0.016), and bilateral lymph node involveme nt ( p ¼ 0.025). Irregular tumor margins
correlated with the shorter period of latency ( p ¼ 0.009) and unilateral nodal disease ( p ¼ 0.010). Hypoechoic
nodules were observed more frequently in female patients ( p ¼ 0.012), in the absence of halo ( p ¼ 0.003) or
calcifications ( p ¼ 0.005). Hypoechogenicity also correlated with the shorter latency ( p ¼ 0.015) and younger age
of patients ( p ¼ 0.048).
Conclusions: Irregular nodule margins, a usual sign of malignancy, are less useful in detecting thyroid cancers in
radiation-exposed patients with tumors less than 10 mm. Thyroid cancers that are detected after longer latent
periods display less of the US features characteristic of a malignant process, while benign US features are
observed more frequently. Therefore, we recommend fine-need le aspiration biopsy to ensure early diagnosis of
thyroid cancer for patients with a history of radiation exposure, even if their nodules are less than 10 mm.
Introduction
C
hildhood thyroid cancer is rare, but early diagnosis is
important in pediatric practice. Real-time ultrasound
(US) examination is the most useful means of screening for
thyroid cancer. US is widely available, has a low cost, requires
short examination time, is not painful, and does not expose
the patient to radiation (1). When US is combined with US-
guided fine-needle aspiration biopsy (FNAB) this can provide
early diagnosis of thyroid disease in high-risk individuals
exposed to radiation (2–5).
Among the two million children exposed to Chernobyl
fallout in Belarus, over 2000 cases of thyroid cancer have
been diagnosed in young patients during the 20-year period
1
Department of Thyroid Disease Research, Belarusian Medical Academy for Postgraduate Education, Minsk, Republic of Belarus.
2
Department of Radiology and Radiological Sciences, Vanderbilt University Medical Center, Nashville, Tennessee.
3
Clinic and Polyclinic of Nuclear Medicine, University of Wuerzburg, Wuerzburg, Germany.
Departments of
4
Molecular Medicine and
5
International Health and Radiation Research, Nagasaki University Graduate School of Bio-
medical Sciences, Nagasaki, Japan.
THYROID
Volume 19, Number 7, 2009
ª Mary Ann Liebert, Inc.
DOI: 10.1089=thy.2008.0238
725
Page 1
(1986–2006) after the accident (6). The mortality rate has been
extremely low. This is probably due, at least in part, to US
screening. Had the disaster occurred a decade before US was
available, the mortality would probably have been substan-
tially higher (7).
In adults the US features associated with malignancy are
hypoechogenicity, mostly nonhomogenous structure, calcifi-
cations, and irregular (microlobulated and blurred) margins
(1,3,8–18) as summarized in Table 1. In the pediatric popula-
tion the potential usefulness of these features is less well
studied. Most of the data relating to the US features of thyroid
malignancy are based on studies of sporadic cases. Less is
known about the US features of radiation-associated thyroid
cancer, particularly as it relates to the interval between radi-
ation exposure and US.
In this study we retrospectively analyzed young patients
with thyroid cancer who had been exposed to the accidental
release of radiation from the Chernobyl reactor in 1986. We
found that the US features of malignant nodules in this group
were more characteristic of what has been described for thy-
roid cancer with increasing tumor size and less characteristic
of thyroid cancer with increasing duration between radiation
exposure and surgical diagnosis.
Materials and Methods
Patients
The study group included 94 patients (47 males and 47
females, sex ratio 1:1) from Belarus with papillary thyroid
carcinoma for whom pathological findings, US imaging, and
demographic data were available at the database of Belar-
usian Medical Academy for Postgraduate Education. Seven
additional cases in the database distributed evenly over the
period of data acquisition (1990–2003) were not included be-
cause of excessively missing clinicopathological information.
The patients’ age in the study ranged 0.1–18.0 years with
the mean standard deviation (SD) of 5.6 4.2 at the time of
the Chernobyl accident, and 10.6–34.3 years with the
mean SD of 16.5 6.2 years at diagnosis. The latent period,
defined as the interval between radiation exposure (April 26,
1986) and the first thyroid surgery, ranged from 45.6 to 207.6
months with the mean SD of 128.4 49.2 months.
US-guided FNAB followed by cytological analysis was
routinely done for all thyroid nodules measuring 5 mm or
greater on US. In all cases in this study the preoperative cy-
tological diagnosis was papillary thyroid carcinoma or sus-
pected thyroid carcinoma. All the patients were surgically
treated at the Thyroid Cancer Center, Minsk. The cytological
diagnosis of papillary thyroid carcinoma was confirmed by
pathological examination after surgery.
Clinicopathological data were retrieved from medical re-
cords. The study protocol was approved by the Belarusian
Medical Academy for Postgraduate Education and Nagasaki
University Graduate School of Biomedical Sciences institu-
tional review boards.
US examination
Thyroid nodules were initially detected during annual
screenings in most radiocontaminated regions of Belarus.
Table 1. Ultrasound Features Suggestive of a Malignant or Benign Thyroid Nodule
Malignant (references) Benign (references)
Number of nodules
Solitary Frates et al. (8)
Multiple Frates et al. (8)
Structure
Homogeneous
Heterogeneous Hegedus (1), Drozd et al. (3)
Echogenicity
Hypoechoic Hegedus (1), Drozd et al. (3),
Cooper et al. (9), Shimura et al. (10), Tae et al. (11)
Wienke et al. (12)
Isoechoic Drozd et al. (3) Hegedus (1)
Mixed
Margins
Regular Drozd et al. (3) Hegedus (1), Shimura et al. (10),
Tae et al. (11)
Irregular Hegedus (1), Drozd et al. (3), Cooper et al. (9),
Shimura et al. (10), Tae et al. (11), Papini et al. (13),
Koike et al. (14), Rago et al. (15), Chan et al. (16)
Halo
Absent Hegedus (1), Peccin et al. (17)
Present Drozd et al. (3) Hegedus (1)
Calcifications
Absent Tae et al. (11)
Present Hegedus (1), Cooper et al. (9) Hegedus (1), Shimura et al. (10),
Koike et al. (14)
Lymphadenopathy
Absent Hegedus (1)
Present Hegedus (1), Drozd et al. (3), Pellegriti et al. (18)
Hypoechoic
Isoechoic Drozd et al. (3)
726 DROZD ET AL.
Page 2
Table 2. Clinical, Ultrasonographic, and Pathological Parameters by Nodule Size
Group
Parameters Nodule size 10 mm n ¼ 25 Nodule size >10 mm n ¼ 69 p-Value
a
Clinical characteristics
Sex
Male 11 (44.0)
b
36 (52.2)
Female 14 (56.0) 33 (47.8) 0.64
Age at the time of accident (years) 5.0 3.9
c
5.7 4.9 0.88
Age at the time of diagnosis (years) 14.8 5.9 17.11 6.6 0.14
Latent period (months) 117.9 46.6 133.6 48.7 0.089
Ultrasonographic characteristics
Nodule volume (mL) 0.2 0.1 1.6 2.5 <0.001
Number of nodules
Solitary 21 (84.0) 59 (85.5)
Multiple 4 (16.0) 10 (14.5) 1.00
Structure
Homogeneous 14 (58.3) 26 (37.7)
Heterogeneous 10 (41.7) 43 (62.3) 0.097
Unknown 1 0
Echogenicity
Hypoechoic 18 (75.0) 36 (52.9)
Isoechoic 4 (16.7) 18 (26.5)
Mixed 2 (8.3) 14 (20.6) 0.18
Unknown 1 1
Margins
Regular 10 (43.5) 10 (14.7)
Irregular 13 (56.5) 58 (85.3) 0.009
Unknown 2 1
Halo
Absent 22 (88.0) 56 (85.5)
Present 3 (12.0) 10 (14.5) 1.00
Unknown 0 3
Calcifications
Absent 25 (100.0) 59 (85.5)
Present 0 (0.0) 10 (14.9) 0.058
Lymphadenopathy
Absent 15 (60.0) 39 (56.5)
Present 10 (40.0) 30 (43.5) 0.82
Hypoechoic 9 (36.0) 21 (30.4)
Isoechoic 1 (4.0) 8 (11.6) 0.61
Unknown 0 1
Pathological characteristics
Tumor size (mm)
d
7.8 2.6 13.2 5.3 <0.001
T stage
T1 24 (96.0) 63 (91.3)
T2 0 (0.0) 4 (5.8)
T3 1 (4.0) 2 (2.9) 0.64
N stage
N0 13 (52.0) 20 (29.4)
N1a 10 (40.0) 18 (26.5)
N1b 2 (8.0) 30 (44.1) 0.003
Unknown 0 1
Histopathological variant
Solid 3 (12.0) 6 (8.9)
Papillary and follicular 22 (88.0) 61 (91.1) 0.99
Unknown 0 2
Invasiveness
Intrathyroidal 21 (84.0) 60 (86.9)
Extrathyroidal 4 (16.0) 9 (13.1) 0.74
a
Based on Wilcoxon rank-sum or Kruskal-Wallis test for continuous variables and Fisher’s exact test for categorical variables; patients
without available information were excluded.
b
Number of patients with percentage within parentheses.
c
Mean standard deviation.
d
Largest measurement.
ULTRASOUND FEATURES OF CHERNOBYL THYROID CANCER 727
Page 3
Individuals with US thyroid abnormalities underwent health
examination, including thyroid ultrasonography, at the hos-
pital. US examination was done with a Toshiba SSA 240A
(7.5 MHz sector probe) ultrasonic real-time scanner, with pa-
tient in the supine position with the neck slightly hyper-
extended.
For the purpose of this study, archived US images were
reanalyzed by experienced radiologists (V.M.D., M.L.L., and
A.P.L.) using identical criteria. Sonograms were interpreted in
terms of nodule size (largest dimension), number of nodules
(solitary and multiple), echogenicity (hypoechoic, isoechoic, or
mixed), echostructure (homogeneous or heterogeneous), mar-
gins (regular or irregular), presence or absence of a halo (rim),
calcifications, and regional lymphadenopathy. A consensus es-
timate was reached for every parameter in all cases under
study.
Thyroid and nodule volumes (mL) were estimated, using
the height (H), width (W), and length (L) of the thyroid lobes
or of a nodule measured in the transverse and longitudinal
scans (cm) by the formula for ellipsoid volume: V ¼ H W
L =6 (19).
Histopathology
All cases were pathologically diagnosed as papillary car-
cinomas. Solid variant of papillary thyroid carcinoma (less
mature histotype) was observed in nine (9.6%) cases. Typical
papillary morphology and follicular variant of papillary
thyroid carcinoma (more differentiated histotypes) were
documented in 85 (90.4%) cases. Tumor size varied from 5 to
30 mm with the mean SD of 11.7 3.9 mm. The presence or
absence of capsular and extrathyroidal invasion was re-
corded. Tumor staging was according to the International
Union Against Cancer (UICC) TNM Classification of Malig-
nant Tumors (20).
Statistical analysis
All patients were divided into subgroups according to
tumor size on ultrasonography (10 mm and >10 mm), type
of margins (regular or irregular), and echogenicity (hy-
poechoic, isoechoic, or mixed). Subgroups thus defined were
compared with respect to clinical, ultrasonographic, or path-
ological characteristics by Fisher exact test for categorical
data, and by Wilcoxon rank-sum test or Kruskal-Wallis test
for quantitative measurements. Association of these charac-
teristics with tumor size, type of margins, or echogenicity was
evaluated by logistic regression analysis, where the most
appropriate model was selected by Akaike information crite-
rion starting from the full model. The FREQ, NPAR1WAY,
and LOGISTIC procedures in the SAS system (21) were used
for the calculations. A p-value less than 0.05 was regarded as
indicating statistical significance.
Results
Clinical, ultrasonic, and pathological characteristics
with respect to nodule size by US
The tumor size on ultrasonography was 10 mm in 25
patients and >10 mm in 69 patients (Table 2). No significant
difference was observed between the two groups in age at the
time of accident, age at diagnosis, or sex ratio. Although not
significant, the latent period tended to be longer in the pa-
tients with larger nodules.
Irregular margins were detected more frequently in the
patients with larger nodules ( p ¼ 0.009). There was a very
strong trend toward the frequency of calcifications being
higher in the patients with larger nodules ( p ¼ 0.058).
No significant differences were observed in other US charac-
teristics between the two subgroups. By definition, tumors
smaller than 1 cm are classified as pT1. However, 91% of
the larger tumors in this series also were classified as pT1
(with diameter smaller than 2 cm). There were no pT4 tu-
mors; the low prevalence of advanced stage tumors is most
probably due to extensive screenings. Lymph node involve-
ment (on pathology) significantly correlated with the
larger ultrasonic nodule size ( p ¼ 0.003). No significant dif-
ference was observed between the two subgroups regard-
ing the frequency of histopathological variants or tumor
invasiveness.
Three factors—latent period, margins, and N stage—were
found by logistic regression analysis to be associated with the
frequency of larger tumor size (Table 3). The frequency of
larger tumors was significantly increased with the latency
(
p ¼ 0.016), was significantly higher among the nodules
with irregular margins ( p ¼ 0.001), and was significantly
higher in the patients with bilateral lymph node involvement
( p ¼ 0.025).
Clinical, ultrasonic, and pathological characteristics
with respect to nodule margins
Tumor margins were regular or irregular in 20 and 72 pa-
tients, respectively (Table 4). The age at diagnosis was sig-
nificantly lower ( p ¼ 0.006) and the latency period was
significantly shorter ( p ¼ 0.021) in the patients with irregular
margins of cancer nodules.
A heterogeneous US structure was observed more fre-
quently in nodules with irregular margins ( p ¼ 0.022). US
signs of lymph node involvement as well as N1a tumor stage
were significantly more frequent in the patients with nodules
with irregular margins ( p ¼ 0.022 and p ¼ 0.025, respectively).
The frequency of isoechoic lymph nodes, not considered a
suspicious sign, also tended to correlate with the irregular
margins of thyroid nodules, although the significance was
marginal ( p ¼ 0.064). An isoechoic pattern and the presence of
Table 3. Parameters Associated with the Frequency
of Large Nodule Size (>10 mm), Logistic
Regression Analysis
Parameters Comparison Odds ratio 95% CI p-Value
a
Number of
nodules
Singular vs.
multiple
0.24 0.05–1.12 0.071
Latent period By 1-month
increment
1.02 1.00–1.03 0.016
Margins Irregular vs.
regular
12.75 2.99–69.3 0.001
N stage
N1a vs. N0 0.51 0.13–1.92 0.331
N1b vs. N0 7.08 1.48–48.7 0.025
a
Based on likelihood ratio test.
CI, confidence interval.
728 DROZD ET AL.
Page 4
halo were more frequent in the nodules with regular margins
( p ¼ 0.051 and p ¼ 0.029, respectively).
Three factors, that is, latent period, nodule size, and N stage,
were found by logistic regression analysis to be associated
with the frequency of irregular margins (Table 5). Irregular
margins were decreased in frequency with longer latency
( p ¼ 0.009), were significantly frequent in the larger nodules
(>10 mm) as compared to smaller ones (10 mm; p ¼ 0.042),
and were observed significantly more frequently in the pa-
tients with unilateral lymph node involvement ( p ¼ 0.010).
Table 4. Clinical, Ultrasonographic, and Pathological Parameters by Tumor Margin Type
Group
Parameters Regular margins n ¼ 20 Irregular margins n ¼ 72 p-Value
a
Clinical characteristics
Sex
Male 11 (55.0)
b
36 (50.0)
Female 9 (45.0) 36 (50.0) 0.80
Age at the time of accident (years) 7.2 6.1
c
5.1 4.2 0.27
Age at the time of diagnosis (years) 20.2 7.5 15.3 5.9 0.006
Latent period (months) 155.9 34.3 122.7 49.9 0.021
Ultrasonographic characteristics
Nodule size (mm)
d
12.6 5.3 15.4 6.7 0.052
Nodule volume (mL) 0.7 0.8 1.4 2.5 0.019
Number of nodules
Solitary 17 (85.0) 60 (83.3)
Multiple 3 (15.0) 12 (16.7) 1.00
Structure
Homogeneous 13 (65.0) 25 (34.7)
Heterogeneous 7 (35.0) 47 (65.3) 0.022
Echogenicity
Hypoechoic 9 (45.0) 44 (61.1)
Isoechoic 9 (45.0) 13 (18.1)
Mixed 2 (10.0) 15 (20.8) 0.051
Halo
Absent 14 (70.0) 65 (90.3)
Present 6 (30.0) 7 (9.7) 0.029
Calcifications
Absent 18 (90.0) 64 (88.9)
Present 2 (10.0) 8 (11.1) 1.00
Lymphadenopathy
Absent 16 (80.0) 37 (51.4)
Present 4 (20.0) 35 (48.6) 0.022
Hypoechoic 4 (20.0) 26 (36.1)
Isoechoic 0 (0.0) 8 (11.1) 0.064
Unknown 0 1
Pathological characteristics
Tumor size (mm)
d
10.3 3.5 12.4 5.6 0.15
T stage
T1 20 (100.0) 64 (89.0)
T2 0 (0.0) 4 (5.5)
T3 0 (0.0) 4 (5.5) 0.50
N stage
N0 12 (60.0) 20 (28.2)
N1a 2 (10.0) 24 (33.8)
N1b 6 (30.0) 27 (38.0) 0.025
Unknown 0 1
Histopathological variant
Solid 1 (5.0) 9 (12.5)
Papillary and follicular 19 (95.0) 63 (87.5) 0.70
Invasiveness
Intrathyroidal 18 (90.0) 60 (83.3)
Extrathyroidal 2 (10.0) 12 (16.7) 0.72
a
Based on Wilcoxon rank-sum or Kruskal-Wallis test for continuous variables and Fisher’s exact test for categorical variables; patients
without available information were excluded.
b
Number of patients with percentage within parentheses.
c
Mean standard deviation.
d
Largest measurement.
ULTRASOUND FEATURES OF CHERNOBYL THYROID CANCER 729
Page 5
Clinical, ultrasonic, and pathological characteristics
with respect to nodule echogenicity
Tumors were hypoechoic in 54 patients, isoechoic in 22, and
mixed in 17 (Table 6). The age at diagnosis was significantly
higher ( p ¼ 0.048) and latent period was significantly longer
( p ¼ 0.015) in the patients with isoechoic tumors (usually no
sign of malignancy) than in those with hypoechoic or mixed-
type tumors. The frequency of isoechoic tumors was signifi-
cantly higher in male patients ( p ¼ 0.003).
Heterogeneous US structure was significantly more fre-
quent in mixed-type tumors ( p ¼ 0.021). Irregular margins
were more frequent in hypoechoic or mixed-type tumors
( p ¼ 0.051). The presence of halo was significantly more fre-
quent in isoechoic tumors ( p < 0.001), while calcifications
occurred more frequently in the mixed-type tumors
( p < 0.001). Tumor size on pathology was significantly larger
in the mixed-type tumors ( p ¼ 0.023).
Three factors—sex, halo, and calcification—were found by
logistic regression analysis to be associated with the frequency
of hypoechoic tumors (Table 7). It was significantly higher in
female patients ( p ¼ 0.012), in the nodules without halo
( p ¼ 0.003) or calcifications ( p ¼ 0.005).
Discussion
Easy detection of thyroid nodules on US, besides of obvious
benefits, brings about a variety of associated problems. These
are the differential diagnosis, follow-up of a great number of
small nodules occurring in 3–50% of population, and estab-
lishment of evidence-based criteria for FNAB in different age
groups (1,2,7,22). Regarding the latter, the Consensus of So-
ciety of Radiologists in US and the American Thyroid Asso-
ciation recommend FNAB for the nodules measuring 1 cm or
greater because of ‘uncertainty as to whether diagnosis of
smaller cancers improves life expectancy as well as a concern
that assessment of small nodules would lead to an excessive
number of biopsies’ (9,22). Nodules not exceeding 1 cm in
diameter are recommended for follow-up or evaluation in
case of suspicious US findings, history of head and neck ir-
radiation, or documented thyroid cancer in consanguine rel-
ative(s). Recent guidelines of the American Association of
Clinical Endocrinologists and Associazione Medici En-
docrinologi also recommend evaluation of thyroid nodules
smaller than 1 cm in radiation-exposed patients (23). With
regard to the young individuals who were internally irradi-
ated by iodine isotopes, substantiated recommendations for
management of small thyroid nodules need further study, in
our opinion.
In the present work we specifically focused on a group of
young patients with radiation-induced papillary thyroid
carcinoma. Nodules in all cases had been initially detected
due to the implementation of post-Chernobyl US screening
programs in Belarus; all nodules exceeding 5 mm were sub-
jected to FNAB and cytological examination. Unlike in other
studies, here we not addressed the differential cancer diag-
nosis but attempted to determine clinicopathological and ul-
trasonic criteria with regard to tumor size, ultrasonographic
margin type, and nodule echogenicity, that is, with the most
important US parameters taken into account for clinical de-
cision making.
The results of analyses demonstrated an association of
tumor size (under vs. over 10 mm) with tumor margin type
and bilateral lymph node involvement. With increasing
nodule size, irregular tumor margins (a usual suspicious sign)
and bilateral lymph node involvement (an indicator of tumor
aggressiveness) became more frequent. Similar to our find-
ings, Pellegriti et al. reported a progressive accumulation of
signs of tumor aggressiveness (multifocality, bilaterality, ex-
trathyroidal invasion, and nodal disease) with increasing
nodule size at presentation (18). Also, in patients with cancers
larger than 1 cm, a close relationship between tumor size and
aggressiveness has been demonstrated (24–26). Apparently,
smaller tumors display more benign characteristics on US, but
these should not be misleading as nodule size per se is not a
predictor of malignancy (8,13,27) and may not necessarily
correlate with an indolent clinical course or lower risk of
recurrence (10,18).
Type of margins in our series correlated not only with
tumor size, but also with other US features, in particular
lymph node involvement and duration of latency. The fre-
quency of irregular margins (a usual suspicious sign) was
increasing in parallel with those of heterogeneous structure,
hypoechoic or mixed echogenicity (all worrisome), and
lymphadenopathy (tumor aggressiveness). Patients with ir-
regular tumor margins were younger at diagnosis and had
shorter latent period and more aggressive tumors. Of note,
from the molecular studies of Chernobyl thyroid cancers, it is
known that mentioned characteristics correlate with the type
of genetic abnormality underlying malignancy. RET=
PTC3
rearrangements were particularly prevalent in the aggressive
tumors developing after the shorter latency; these tumors
frequently had solid morphological structure (28,29). In our
series, solid morphology was more frequently observed in
the tumors with irregular margins (as well as in the hy-
poechoic ones), but significance of difference was not reached
perhaps because of the small number of such cases. It would
Table 5. Factors Associated with the Frequency of Irregular Type of Margins, Logistic Regression Analysis
Factor Comparison Odds ratio 95% CI p-Value
a
Age at accident By 1-year increment 0.95 0.84–1.08 0.434
Latent period By 1-month increment 0.98 0.96–0.99 0.009
Nodule size on ultrasound Larger than 10 mm vs. smaller 1.17 1.02–1.37 0.042
N stage
N1a vs. N0 10.88 2.09–89.9 0.010
N1b vs. N0 0.072
a
Based on the likelihood ratio test.
CI, confidence interval.
730 DROZD ET AL.
Page 6
Table 6. Clinical, Ultrasonographic, and Pathological Parameters by Tumor Echogenicity
Group
Parameters Hypoechoic n ¼ 54 Isoechoic n ¼ 22 Mixed n ¼ 17 p-Value
a
Clinical characteristics
Sex
Male 22 (40.7)
b
18 (81.8) 7 (41.2)
Female 32 (59.3) 4 (18.2) 10 (58.8) 0.003
Age at the time of accident (years) 5.2 4.1
c
6.7 6.2 5.1 4.2 0.87
Age at the time of diagnosis (years) 14.9 5.8 19.4 8.1 16.8 5.5 0.048
Latent period (months) 117.7 47 151.9 45 139.9 47 0.015
Ultrasonographic characteristics
Nodule size (mm)
d
14.7 7.1 14.7 4.8 16.6 6.7 0.13
Nodule volume, mL 1.2 1.90 1.0 0.87 1.9 4.1 0.32
Number of nodules
Solitary 48 (88.9) 17 (77.3) 13 (76.5)
Multiple 6 (11.1) 5 (22.7) 4 (23.5) 0.38
Structure
Homogeneous 25 (47.2) 11 (50.0) 2 (11.8)
Heterogeneous 28 (52.8) 11 (50.0) 15 (88.2) 0.021
Unknown 1 0 0
Margins
Regular 9 (17.0) 9 (40.9) 2 (11.8)
Irregular 44 (83.0) 13 (59.1) 15 (88.2) 0.051
Unknown 1 0 0
Halo
Absent 53 (98.2) 12 (54.5) 15 (88.2)
Present 1 (1.8) 10 (45.5) 2 (11.8) <0.001
Calcifications
Absent 53 (98.2) 18 (81.8) 12 (70.6)
Present 1 (1.8) 4 (18.2) 5 (29.4) <0.001
Lymphadenopathy
Absent 27 (50.0) 15 (68.2) 11 (64.7)
Present 27 (50.0) 7 (31.8) 6 (35.3) 0.29
Hypoechoic 20 (37.0) 6 (27.3) 1 (4.5)
Isoechoic 6 (11.1) 1 (4.5) 1 (5.9) 0.76
Unknown 1 0 4
Pathological characteristics
Tumor size (mm)
d
11.0 5.6 11.7 4.1 14.4 5.4 0.023
T stage
T1 50 (92.6) 22 (100.0) 13 (76.5)
T2 3 (5.6) 0 (0.0) 1 (5.9)
T3 1 (1.8) 0 (0.0) 3 (17.6) 0.063
N stage
N0 23 (43.4) 6 (27.3) 4 (23.5)
N1a 17 (32.1) 5 (22.7) 4 (23.5)
N1b 13 (24.5) 11(50.0) 9 (53.0) 0.14
Unknown 1 0 0
Histopathological variant
Solid 7 (13.0) 1 (4.5) 2 (11.8)
Papillary and follicular 47 (87.0) 21 (95.5) 15 (88.2) 0.86
Invasiveness
Intrathyroidal 46 (85.2) 19 (86.4) 14 (82.4)
Extrathyroidal 8 (14.8) 3 (13.6) 2 (17.6) 0.92
Unknown 0 0 1
a
Based on Wilcoxon rank-sum or Kruskal-Wallis test for continuous variables and Fisher’s exact test for categorical variables; patients
without available information were excluded.
b
Number of patients with percentage within parentheses.
c
Mean standard deviation.
d
Largest measurement.
ULTRASOUND FEATURES OF CHERNOBYL THYROID CANCER 731
Page 7
be interesting to analyze a correlation between US, clini-
copathological, and molecular features, first of all RET=PTC3
and RET=PTC1 rearrangements, in a specially designed
study.
The irregular nodule margins on US are common in thyroid
malignancy (1,3,9–11,13–16,30,31). An analysis done by
Papini et al. demonstrated that independent risk factors of
cancer at US of nonpalpable thyroid nodules were irregular or
blurred nodule margins, intranodular vascular pattern, and
microcalcifications (13). Another study has identified nodule
margins and shape, internal echo level, but not ‘strong ech-
oes’ (calcifications) as important characteristics in differenti-
ating papillary thyroid carcinoma from benign neoplasm (10).
Our findings are in agreement with these studies only for the
tumors exceeding 10 mm in diameter. The importance of ir-
regular margins as criterion of malignancy was significantly
weaker in smaller tumors. Technically, it is not surprising that
irregular margins are found less frequently in very small
thyroid nodules by US because of limited resolution of the
latter.
The associations between cancer nodule size, type of mar-
gin, and latent period are also worth noting. The longer latent
period was characteristic to the nodules with regular margins,
which may reflect the difficulties of early diagnosis of slowly
growing tumors. Radiation-induced thyroid carcinomas de-
veloping after the shorter period of latency are often in ad-
vanced stage, are aggressive at presentation, and display
elevated risk of recurrence despite the small size (5,32–34). It is
therefore necessary to take into account that the nodules de-
tected after the longer latency may display less unusual sus-
picious signs on US, yet they may well be malignant.
Regarding tumor echogenicity, hypoechoic nodules are
more likely to be malignant in contrast to those lacking such
characteristics (10,11,15–17), especially if nodule size exceeds
15 mm (35). In our series hypoechogenicity was associated
with female sex, and the absence of both halo and calcifica-
tions, while isoechogenicity with male sex (this observation is
difficult to explain). Note also that nodule hypoechogenicity
was more frequently observed in younger patients and after
the shorter period of latency which together are supportive of
the overall suspicious nature of this US feature.
From the practical point of view, the clinical decision to
biopsy a small thyroid nodule in children or young patients is
not simple and has individual and overall healthcare cost
implications (36). If we used nodule diameter of 10 mm as cut-
off for FNAB, we would have missed 25 (26.6%) cases of
thyroid cancer in our young patients. Papini et al. obtained a
very similar result in a prospective study of 402 nodules (13).
If FNAB had been restricted to the nodules exceeding 10 mm
in size, cytological evaluation would have been omitted in
38.7% of the actually diagnosed malignancies. They proposed
a cut-off diameter of 8 mm for nonpalpable nodules to be
subjected to cytological evaluation, while follow-up of the
smaller lesions was recommended. The findings presented
here support the results of prior studies that showed that
nodule size greater or smaller than 10 mm was not an effective
arbitrary criterion for FNAB (8,13,37,38) as well as that
probability of malignancy in thyroid nodules measuring less
than 1 cm is not lower than in larger lesions (39). From this
point of view, our results also support the necessity of eval-
uation of small thyroid nodules in radiation-exposed patients
recommended by the American Thyroid Association (9) and
the American Association of Clinical Endocrinologists and
Associazione Medici Endocrinologi (23).
In addition to the nodule size issues, if we biopsied only
nodules with irregular margins, 20 (21.7%) cases would have
been missed. If we biopsied only usual suspicious hypoechoic
nodules, we would have missed 22 isoechoic tumors and 17
nodules with mixed echogenicity (41.9% cases combined).
Further, as shown in our previous study, the relative pro-
portion of cancer nodules among all thyroid nodules detected
in young individuals affected by the Chernobyl fallouts de-
creases with time after exposure due to the increasing prev-
alence of nodular goiter, nonverified small solid nodules,
and cysts (3). In some sense this may predispose to the intu-
itive expectation of a benign rather than cancerous nodule
to be developing after the longer latency. In our opinion,
physicians need to be aware of that the situation is more
complex.
This s tudy demonstrates that important US features co n-
sidered usual for thyroid cance r are less frequent in smaller
tumors and that thyroid cancers developing with a greate r
latency after radiation are less likel y to have worrisome US
features. We therefore propose that in young individuals
with a history of internal rad iation exposure FNA B should
be rec ommended even for small thyroid nodules (i.e.,
whenever technically accessible) with any margin type and
echogenicity.
Acknowledgment
This work was supported in part by Nagasaki University
Global COE program.
Disclosure Statement
The authors declare that no competing financial interests
exist.
References
1. Hegedus L 2004 Clinical practice. The thyroid nodule. N Engl
J Med 351:1764–1771.
2. Reiners C, Wegscheider K, Schicha H, Theissen P, Vaupel R,
Wrbitzky R, Schumm-Draeger PM 2004 Prevalence of thy-
roid disorders in the working population of Germany: ul-
trasonography screening in 96,278 unselected employees.
Thyroid 14:926–932.
3. Drozd V, Polyanskaya O, Ostapenko V, Demidchik Y, Biko I,
Reiners C 2002 Systematic ultrasound screening as a
Table 7. Factors Associating with Nodule
Hypoechogenicity, Logistic Regression Analysis
Factor Comparison Odds ratio 95% CI p-Value
a
Sex Male vs.
female
0.26 0.08–0.72 0.012
Halo Absent vs.
present
26.37 4.24–522.49 0.003
Calcifications Absent vs.
present
25.42 3.64–525.31 0.005
a
Based on the likelihood ratio test.
CI, confidence interval.
732 DROZD ET AL.
Page 8
significant tool for early detection of thyroid carcinoma in
Belarus. J Pediatr Endocrinol Metab 15:979–984.
4. Imaizumi M, Usa T, Tominaga T, Akahoshi M, Ashizawa K,
Ichimaru S, Nakashima E, Ishii R, Ejima E, Hida A, Soda M,
Maeda R, Nagataki S, Eguchi K 2005 Long-term prognosis of
thyroid nodule cases compared with nodule-free controls in
atomic bomb survivors. J Clin Endocrinol Metab 90:5009–
5014.
5. Bucci A, Shore-Freedman E, Gierlowski T, Mihailescu D,
Ron E, Schneider AB 2001 Behavior of small thyroid cancers
found by screening radiation-exposed individuals. J Clin
Endocrinol Metab 86:3711–3716.
6. Shevchuk VE, Gurachevsky GV (eds) 2006 20 Years after the
Chernobyl Catastrophe. Publishing House Belarus, Minsk,
Belarus, p 250.
7. Baskin HJ 2004 Thyroid ultrasound—just do it. Thyroid
14:91–92.
8. Frates MC, Benson CB, Doubilet PM, Kunreuther E,
Contreras M, Cibas ES, Orcutt J, Moore FD Jr., Larsen PR,
Marqusee E, Alexander EK 2006 Prevalence and distribution
of carcinoma in patients with solitary and multiple thyroid
nodules on sonography. J Clin Endocrinol Metab 91:3411–
3417.
9. Cooper DS, Doherty GM, Haugen BR, Kloos RT, Lee SL,
Mandel SJ, Mazzaferri EL, McIver B, Sherman SI, Tuttle RM
2006 Management guidelines for patients with thyroid nod-
ules and differentiated thyroid cancer. Thyroid 16:109–142.
10. Shimura H, Haraguchi K, Hiejima Y, Fukunari N, Fujimoto
Y, Katagiri M, Koyanagi N, Kurita T, Miyakawa M,
Miyamoto Y, Suzuki N, Suzuki S, Kanbe M, Kato Y,
Murakami T, Tohno E, Tsunoda-Shimizu H, Yamada K,
Ueno E, Kobayashi K, Kobayashi T, Yokozawa T, Kitaoka M
2005 Distinct diagnostic criteria for ultrasonographic exam-
ination of papillary thyroid carcinoma: a multicenter study.
Thyroid 15:251–258.
11. Tae HJ, Lim DJ, Baek KH, Park WC, Lee YS, Choi JE, Lee JM,
Kang MI, Cha BY, Son HY, Lee KW, Kang SK 2007 Diag-
nostic value of ultrasonography to distinguish between be-
nign and malignant lesions in the management of thyroid
nodules. Thyroid 17:461–466.
12. Wienke JR, Chong WK, Fielding JR, Zou KH, Mittelstaedt
CA 2003 Sonographic features of benign thyroid nodules:
interobserver reliability and overlap with malignancy.
J Ultrasound Med 22:1027–1031.
13. Papini E, Guglielmi R, Bianchini A, Crescenzi A, Taccogna S,
Nardi F, Panunzi C, Rinaldi R, Toscano V, Pacella CM 2002
Risk of malignancy in nonpalpable thyroid nodules: pre-
dictive value of ultrasound and color-Doppler features. J Clin
Endocrinol Metab 87:1941–1946.
14. Koike E, Noguchi S, Yamashita H, Murakami T, Ohshima A,
Kawamoto H, Yamashita H 2001 Ultrasonographic charac-
teristics of thyroid nodules: prediction of malignancy. Arch
Surg 136:334–337.
15. Rago T, Vitti P, Chiovato L, Mazzeo S, De Liperi A, Miccoli
P, Viacava P, Bogazzi F, Martino E, Pinchera A 1998 Role of
conventional ultrasonography and color flow-doppler so-
nography in predicting malignancy in ‘cold’ thyroid nod-
ules. Eur J Endocrinol 138:41–46.
16. Chan BK, Desser TS, McDougall IR, Weigel RJ, Jeffrey RB Jr.
2003 Common and uncommon sonographic features of
papillary thyroid carcinoma. J Ultrasound Med 22:1083–
1090.
17. Peccin S, de Castsro JA, Furlanetto TW, Furtado AP, Brasil
BA, Czepielewski MA 2002 Ultrasonography: is it useful in
the diagnosis of cancer in thyroid nodules? J Endocrinol
Invest 25:39–43.
18. Pellegriti G, Scollo C, Lumera G, Regalbuto C, Vigneri R,
Belfiore A 2004 Clinical behavior and outcome of papillary
thyroid cancers smaller than 1.5 cm in diameter: study of 299
cases. J Clin Endocrinol Metab 89:3713–3720.
19. Brunn J, Block U, Ruf G, Bos I, Kunze WP, Scriba PC 1981
Volumetrie der Schilddru
¨
senlappen mittels Real-Time-
Sonographie. Dtsch Med Wochenschr 106:1338–1340.
20. Sobin LH, Wittekind CH (eds) 2002 International Union
Against Cancer. TNM Classification of Malignant Tumors,
sixth edition. John Wiley & Sons, Inc., New York, p 239.
21. SAS Institute 1999 SAS=STAT User’s Guide Version 8. SAS
Institute, Cary, NC.
22. Frates MC, Benson CB, Charboneau JW, Cibas ES, Clark OH,
Coleman BG, Cronan JJ, Doubilet PM, Evans DB, Goellner
JR, Hay ID, Hertzberg BS, Intenzo CM, Jeffrey RB, Langer JE,
Larsen PR, Mandel SJ, Middleton WD, Reading CC,
Sherman SI, Tessler FN 2005 Management of thyroid nod-
ules detected at US: Society of Radiologists in Ultrasound
consensus conference statement. Radiology 237:794–800.
23. AACE=AME Task Force on Thyroid Nodules 2006 American
Association of Clinical Endocrinologists and Associazione
Medici Endocrinologi medical guidelines for clinical practice
for the diagnosis and management of thyroid nodules. En-
docr Pract 12:63–102.
24. Hay ID, Grant CS, van Heerden JA, Goellner JR, Ebersold JR,
Bergstralh EJ 1992 Papillary thyroid microcarcinoma: a
study of 535 cases observed in a 50-year period. Surgery
112:1139–1146; discussion 1146–1147.
25. Mazzaferri EL, Massoll N 2002 Management of papillary
and follicular (differentiated) thyroid cancer: new paradigms
using recombinant human thyrotropin. Endocr Relat Cancer
9:227–247.
26. Cappelli C, Castellano M, Braga M, Gandossi E, Pirola I, De
Martino E, Agosti B, Rosei EA 2007 Aggressiveness and
outcome of papillary thyroid carcinoma (PTC) versus mi-
crocarcinoma (PMC): a mono-institutional experience. J Surg
Oncol 95:555–560.
27. Alexander EK, Heering JP, Benson CB, Frates MC, Doubilet
PM, Cibas ES, Marqusee E 2002 Assessment of non-
diagnostic ultrasound-guided fine needle aspirations of
thyroid nodules. J Clin Endocrinol Metab 87:4924–4927.
28. Nikiforov YE, Rowland JM, Bove KE, Monforte-Munoz H,
Fagin JA 1997 Distinct pattern of ret oncogene rearrange-
ments in morphological variants of radiation-induced and
sporadic thyroid papillary carcinomas in children. Cancer
Res 57:1690–1694.
29. Rabes HM, Demidchik EP, Sidorow JD, Lengfelder E,
Beimfohr C, Hoelzel D, Klugbauer S 2000 Pattern of
radiation-induced RET and NTRK1 rearrangements in 191
post-Chernobyl papillary thyroid carcinomas: biological,
phenotypic, and clinical implications. Clin Cancer Res 6:
1093–1103.
30. Takashima S, Fukuda H, Nomura N, Kishimoto H, Kim T,
Kobayashi T 1995 Thyroid nodules: re-evaluation with ul-
trasound. J Clin Ultrasound 23:179–184.
31. Kim EK, Park CS, Chung WY, Oh KK, Kim DI, Lee JT, Yoo
HS 2002 New sonographic criteria for recommending fine-
needle aspiration biopsy of nonpalpable solid nodules of the
thyroid. AJR Am J Roentgenol 178:687–691.
32. Williams ED, Becker D, Dimidchik EP, Nagataki S, Pinchera
A, Tronko ND 1996 Effects on the thyroid in populations
exposed to radiation as a result of the Chernobyl accident.
ULTRASOUND FEATURES OF CHERNOBYL THYROID CANCER 733
Page 9
In: Demir M (ed) One Decade after Chernobyl: Summing up
the Consequences of the Accident. International Atomic
Energy Authority, Vienna, pp 207–230.
33. Farahati J, Demidchik EP, Biko J, Reiners C 2000 Inverse
association between age at the time of radiation exposure
and extent of disease in cases of radiation-induced child-
hood thyroid carcinoma in Belarus. Cancer 88:1470–1476.
34. Demidchik YE, Demidchik EP, Reiners C, Biko J, Mine M,
Saenko VA, Yamashita S 2006 Comprehensive clinical as-
sessment of 740 cases of surgically treated thyroid cancer in
children of Belarus. Ann Surg 243:525–532.
35. Lyshchik A, Drozd V, Demidchik Y, Reiners C 2005 Diag-
nosis of thyroid cancer in children: value of gray-scale and
power doppler US. Radiology 235:604–613.
36. Rosen JE, Stone MD 2006 Contemporary diagnostic ap-
proach to the thyroid nodule. J Surg Oncol 94:649–661.
37. Davies TF 2006 Is consensus a good thing in the manage-
ment of thyroid nodules? Thyroid 16:205.
38. Burman KD 2006 Micropapillary thyroid cancer: should we
aspirate all nodules regardless of size? J Clin Endocrinol
Metab 91: 2043–2046.
39. Berker D, Aydin Y, Ustun I, Gul K, Tutuncu Y, Isik S,
Delibasi T, Guler S 2008 The value of fine-needle aspiration
biopsy in subcentimeter thyroid nodules. Thyroid 18:603–
608.
Address correspondence to:
Vladimir A. Saenko, Ph.D.
Department of International Health and Radiation Research
Nagasaki University Graduate School of Biomedical Sciences
1-12-4 Sakamoto
Nagasaki 852-8523
Japan
E-mail: saenko@net.nagasaki-u.ac.jp
734 DROZD ET AL.
Page 10
Show more