ARTICLE IN PRESS

Ultrasound in Med. & Biol., Vol. 00, No. 00, pp. 110, 2019
Copyright © 2019 World Federation for Ultrasound in Medicine & Biology. All rights reserved.
Printed in the USA. All rights reserved.
0301-5629/$ - see front matter

https://doi.org/10.1016/j.ultrasmedbio.2019.03.004

Original Contribution

MULTI-PARAMETRIC ULTRASOUND EVALUATION OF PEDIATRIC THYROID

DYSHORMONOGENESIS

TAGEDPIBRAHIM ADALETLI,* ZUHAL BAYRAMOGLU,* EMINE CALISKAN,y RAVZA YILMAZ,*

ZEYNEP NUR AKYOL SARI,* FIRDEVS BAS,z ASLI DERYA KARDELEN,z
SUKRAN POYRAZOGLU,z and FEYZA DARENDELILERzAGEDNTE
* Istanbul Medical Faculty, Radiology Department, Istanbul University, Istanbul, Turkey; y Radiology Department, Seyhan State
Hospital, Adana, Turkey; and z Istanbul Medical Faculty, Pediatric Endocrinology Department, Istanbul University, Istanbul, Turkey
(Received 8 November 2018; revised 1 February 2019; in final from 5 March 2019)

Abstract—The aim of this study was to assess the diagnostic contribution of gray-scale ultrasonography, color
Doppler, superb microvascular imaging and shear wave elastography in thyroid dyshormonogenesis (TD). From
October 2017 to February 2018, the prospective study included 31 patients (13.6 y; 1114 y) diagnosed with TD
based on thyroid scintigraphy and perchlorate discharge tests and 40 healthy pediatric volunteers (12.8 y; 1016 y).
Median resistive indices (RIs), peak systolic and end-diastolic velocities, vascularity indices (VIs) via superb micro-
vascular imaging and shear wave elastography parameters were evaluated. Median VI values were significantly
higher and median RI values were significantly lower in the study group than the control group. No significant
difference was found between shear wave elastography parameters of the TD and control group. VI was significantly
correlated with median total thyroid gland volumes (p = 0.002, r = 0.28), medication dosage (p = 0.03, r = 0.48) and
2-h radioactive iodine uptake values (p = 0.008, r = 0.57). VI is a clinically significant and novel parameter useful for
diagnosing TD. (E-mail: [email protected]) © 2019 World Federation for Ultrasound in Medicine & Biology.
All rights reserved.
Key Words: Dyshormonogenesis, Pediatric, Shear wave elastography, Superb microvascular imaging, Thyroid,
Vascularity index.

INTRODUCTION Diagnosis of TD depends on biochemical analysis,

thyroid ultrasonography, radioisotope thyroid scanning
Congenital hypothyroidism (CH) is the most common
and genetic analysis. TD is responsible for elevated
endocrine disorder, with an incidence of 1/3500, in new-
plasma thyroglobulin and thyroid-stimulating hormone
borns (Medeiros-Neto et al. 2002). Permanent CH is
(TSH) levels, which is supported by identifying nodules
identified as thyroid hormone deficiency requiring life-
in some patients and a eutopic thyroid gland on ultraso-
long replacement treatment starting at birth. The major-
nographic examination. Prolonged hyperthyrotropinemia
ity of patients with permanent CH are diagnosed with
stimulates the thyroid gland, resulting in a goiter and
thyroid dysgenesis, and only 15% of the patients with
hypervascularity. Because of the increased nodule preva-
permanent CH are considered to have thyroid dyshormo-
lence in TD (Chang et al. 2011), a further sonomorpho-
nogenesis (TD). In some patients, thyroidal ontogenetic
logic examination would be performed. Color Doppler
defects may be associated with TD. Possible causes of
(CD) evaluation of TD has revealed increased vascular-
TD, such as autoimmune and environmental factors,
ity, mimicking the thyroid inferno pattern (Voss et al.
have yet to be confirmed. A higher prevalence of TD in
2013). However, a quantitative algorithm by ultrasonog-
girls and in some ethnic groups, in association with
raphy-based applications using microvascular imaging
extra-thyroidal congenital anomalies, suggests a genetic
or ultrasoud elastography to reveal the presence or sever-
element in the pathogenesis of TD (Bikker et al. 1994).
ity of TD has not yet been exhibited.
Superb microvascular imaging (SMI) is a novel
microvascular imaging technique that can clearly and
Address correspondence to: Zuhal Bayramoglu, MD, Department
of Radiology, Istanbul Faculty of Medicine, Fatih 34390, Istanbul, quantitatively display microcirculation by exhibiting a
Turkey. E-mail: [email protected] vascularity index (VI), which is defined as the ratio of

1
 ARTICLE IN PRESS
2 Ultrasound in Medicine & Biology Volume 00, Number 00, 2019

color pixels to the total pixels within the selected region (Aplio 500 Platinum, Canon, Tokyo, Japan). All exami-
of interest (ROI). Because of the hyperthyrotropinemia- nations were performed with the patient in a supine
related hypervascularity, diagnosis of TD based on VI position with the neck hyperextended. We calculated
via SMI would be more efficient than isolated morpho- the thyroid volumes with an automated formula
logic findings on gray-scale ultrasound (US) imaging. In (length £ width £ depth £ 0.52). We categorized the
addition, because of increased nodule prevalence in TD, thyroid gland echogenicity as hypoechoic, isoechoic or
examination of biomechanical properties of the paren- hyperechoic relative to the cricoid cartilage. We evalu-
chyma would provide additional diagnostic contribution. ated the parenchyma for the presence of nodules, as
Here, we aimed to investigate further gray-scale imaging well as the visibility of tiny echogenic reticulations
findings, diagnostic validity and feasibility of VI and (Fig. 1a, 1b).
shear wave elasticity parameters of thyroid glands of All patients and volunteers underwent CD and SMI
patients with TD in comparison with those of the normal (Fig. 2a, 2b) examinations. Spectral wave analysis on
pediatric population. CD US evaluation was performed in each patient from
the intra-thyroidal vessels of the upper, middle and infe-
MATERIALS AND METHODS rior zones of each lobe displayed on transverse images.
Pulse repetition frequency (PRF), Doppler angle and
Patients frame rate values were set between 850900 Hz, 30˚
This prospective study was conducted between 60˚ and 20 frames per s, respectively. Median peak sys-
October 2017 and February 2018, after the local ethics tolic velocities (PSVs), end-diastolic velocities (EDVs)
committee approved the study and informed consents and resistive indices (RIs) were obtained by averaging
were obtained from the parents of the participants. This six measurements for each participant.
comparative study employed 31 consecutive pediatric
patients already diagnosed with TD and under follow up SMI
by the pediatric endocrinology department and 40 Superb microvascular images provided by a
healthy pediatric volunteers. We searched the TD 14-MHz linear-array transducer (Aplio 500 Platinum,
patients from the hospital records who had positive new- Canon) were assessed on transverse images. All study
born screening tests regarding elevated TSH (>20 mIU/ and control patients underwent SMI using a PRF set as
mL) and decreased free thyroxine (fT4) levels, were 150180 Hz and a frame rate of 50 frames per s. After
diagnosed with CH and whose disease was confirmed by the SMI images were frozen, the margins of each thyroid
the presence of eutopic thyroid parenchyma on US, fol- lobe were manually outlined to exhibit the ROI. A quan-
lowed by radioactive iodine uptake (RAIU) and perchlo- titative evaluation of VI was achieved after gray-scale
rate discharge tests. An RAIU scan that revealed uptake pixels were eliminated automatically by an algorithm.
ratios higher than 10% and a perchlorate discharge test The percentage exhibiting color pixels to the total num-
that revealed a 10%50% iodide discharge indicated a ber of pixels within the ROI on transverse sections repre-
partial organification defect (Park et al. 2018). We sented VI (Fig. 3a, 3b). The median VI of each lobe was
excluded those patients who had not underwent scintig- calculated by averaging VI values obtained from upper,
raphy and had a dysgenetic thyroid gland, such as ecto- middle and inferior zones at the transverse section. The
pia, hypoplasia or aplasia. We recorded the age and median VI values for each patient were calculated by
gender of the participants, number of family members averaging the mean VI values of each lobe. Since nod-
diagnosed with TD, actual TSH levels, 1-h and 2-h val- ules can be developed in patients with TD, we calculated
ues of RAIU and dosage of medication. Previously per- VI of the parenchyma outside of nodules.
formed RAIU and perchlorate discharge test results were
scanned from the hospital records. The patients under- SWE
went gray-scale US, CD, SMI and shear wave elastogra- SWE images provided by 14-MHz linear-array
phy (SWE) examinations within the week in which the transducer (Aplio 500 Platinum, Canon) were assessed
actual thyroid functions were obtained. Healthy volun- on longitudinal images. Three measurements with a
teers were selected among the patients who had thyroid round-shaped ROI 3 mm in diameter were obtained
gland with normal ultrasonographic appearance and thy- from each lobe in the longitudinal plane from two differ-
roid functions within the normal range. ent sections (Fig. 4a, 4b). All measurements were
recorded both as m/s and kPa. The mean stiffness value
Gray scale and Doppler US evaluation for each lobe was calculated by averaging the six meas-
Ultrasonographic assessments were performed by urements per lobe. During elastography measurements,
a pediatric radiologist who had more than 7 y of US ROIs were placed on parenchyma with the stiffest part
experience using a 14-MHz linear-array transducer of the gland with TD.
 ARTICLE IN PRESS
Pediatric Thyroid Dyshormonogenesis US
I. ADALETLI et al. 3

Fig. 1. (a) Gray scale image of an 8-y-old, male patient. The thyroid gland is diffusely enlarged and hypoechoic with
tiny, horizontal echogenic reticulations. (b) Transverse and longitudinal sections of a cervical ectopic thymus in a
5-y-old, female patient with punctate echogenic foci. Echogenicities of thymic tissue and the dyshormonogenetic gland
are similar and hypoechoic to the cricoid cartilage. D2 = anteroposterior diameter of ectopic cervical thymus; D3 =cra-
niocaudal diameter of the cervical ectopic thymus.

Fig. 2. Superb microvascular images colored (a) and monochromatic scale (b) of dyshormonogenetic thyroid lobes.
Superb microvascular images allows to calculate vascularity index based on the clear visualization of the
microcirculation.
 ARTICLE IN PRESS
4 Ultrasound in Medicine & Biology Volume 00, Number 00, 2019

Fig. 3. VI is calculated in a normal thyroid lobe (a) and dyshormonogenetic thyroid lobe (b) by encircling the margins of
the lobes. VI is significantly higher in patients with TD. ROI = region of interest; SMI = superb microvascular imaging;
TD = thyroid dyshormonogenesis; VI = vascularity index.

Statistical analysis 1114]). The control group included 40 healthy patients

Statistical analysis was performed using SPSS (version (22 boys, 18 girls; median age: 12.8 y [IQR: 1016]).
21.0, SPSS, Chicago, IL, USA). Categorical variables were The differences between the median ages were not statis-
expressed as frequencies. Quantitative parameters were tically significant. Five siblings of the 31 patients with
expressed as median and range. The distribution of the data TD (16%) were also diagnosed with TD and receiving
was analyzed by the Kolmogorov-Smirnov test. Mann- thyroid hormone replacement. Thyroid glands were
Whitney U test and Kruskal-Wallis test were used for com- found to be eutopic in all patients. A descriptive analysis
parison of independent samples concerning medians of vol- of quantitative variables is provided in Table 1. Total
ume (cm3), PSV (cm/s), EDV (cm/s), RI, VI (%), SWE thyroid volumes were significantly higher in patients
(kPa) and shear wave velocity (SWV; m/s). Spearman’s with TD (median: 8.1 cm3, IQR: 3.3010.6 cm3) com-
correlation analysis was performed to investigate the rela- pared with control patients (median: 4.5 cm3, IQR:1
tionship of the VI with Doppler parameters, SWE parame- 7.4 cm3) (p < 0.001). Solid nodules were found in seven
ters, scintigraphy results and dosage of medication. of 31 patients (22.6%), of which three patients had mul-
The diagnostic utility of the CD, VI and SWE param- tiple nodules (0.09%). Thyroid gland echogenicities
eters were assessed using receiver operating characteristic were hypoechoic relative to the cricoid cartilage in seven
curves. The best cutoff for VI, elasticity and velocity values patients (22.6%), isoechoic in 12 patients (38.7%) and
representing the diagnosis of TD were calculated. Sensitiv- hyperechoic in 12 patients (38.7%). Tiny echogenic
ity, specificity, positive predictive value (PPV), negative reticulations were found in nine patients (29%), whose
predictive value (NPV), area under the curve (AUC) and thyroid echogenicities were either isoechoic or hypoe-
diagnostic accuracy were calculated. choic to the cricoid cartilage.
When we compared the median values of CD
parameters, there were significant differences among RI
RESULTS
values but neither among EDV nor PSV (Table 1).
The study included 31 patients with TD (20 boys, Median RI values were significantly lower in the TD
11 girls, median age: 13.6 y [interquartile range (IQR): group compared with the control group (0.51 [IQR:
 ARTICLE IN PRESS
Pediatric Thyroid Dyshormonogenesis US
I. ADALETLI et al. 5

Fig. 4. Shear wave elastography images of a patient with TD (a) and a normal patient (b). Shear wave velocity values were
similar in normal children and patients with TD. Ave = average; TD = thyroid dyshormonogenesis; SD = standard deviation.

0.480.53] vs. 0.55 [IQR: 0.490.63], p < 0.01). The difference of median elasticity and SWV val-
Median VI values were significantly higher in the study ues were not statistically significant between the TD
group (11%, IQR: 6.546%) compared with the control group (9.75 kPa, IQR: 9.110.85 kPa; vs. 1.82 m/s,
group (7.42%; IQR: 310%) (p < 0.008). IQR: 1.751.92 m/s, respectively) compared with the
 ARTICLE IN PRESS
6 Ultrasound in Medicine & Biology Volume 00, Number 00, 2019

Table 1. Descriptive statistics regarding demographic, biochemical, radiologic and scintigraphic parameters of the normal patients
and patients with dyshormonogenesis
Demographic and quantitative data Thyroid dyshormonogenesis Mann-Whitney U test Normal
N = 31 (M: 20, F: 11) N = 40 (M: 22, F: 18)
Median IQR p Median IQR

Age (y) 13.6 1114 0.05 12.8 1016

Total Thyroid Volume (cm3) 8.1 3.3010.6 0.001 4.5 17.4
VI (%) 11 6.546 0.008 7.42 310
EDV (cm/s) 8.4 612 0.05 9.8 3.814.7
PSV (cm/s) 16 12.527 0.98 15.1 1021
SWE (kPa) 9.75 9.110.85 0.2 9.5 8.510.3
SWV (m/s) 1.82 1.751.92 0.1 1.78 1.691.86
RI 0.51 0.480.53 0.01 0.55 0.490.63
TSH (mIU/L) 7.5 0.1523 0.01 4.1 3.24.5
L T4 Dosage (mcg/kg/d) 1.15 0.851.89
Perchlorate Discharge (%) 58.9 4167
EDV = end-diastolic velocity; F = female; IQR = interquartile range; L T4 = levothyroxine; M = male; PSV = peak systolic velocity; RI = resistive
index; SWE = shear wave elastography; SWV = shear wave velocity; TSH = thyroid stimulating hormone; VI = vascularity index.
Bold values indicate statistically significant results.

control group (9.5 kPa, IQR: 8.510.3 kPa; 1.78 m/s, When the cutoff VI value was set at 10%, the sensi-
IQR: 1.691.86 m/s, respectively) (p > 0.1). tivity, specificity, PPV, NPV and diagnostic accuracy
VI via SMI was positively correlated with median for differentiating TD from normal were found to be
total thyroid gland volumes (p < 0.002, r = 0.28), EDV 70%, 63%, 68%, 65% and 71% respectively (p = 0.008)
(p < 0.001, r = 0.34) and PSV (p < 0.001, r = 0.37) (Table 3, Fig. 5). When the cutoff RI value was set at
(Table 2). However, VI was negatively correlated with 0.53, the sensitivity, specificity, PPV, NPV and diagnos-
RI (p < 0.018, r = 0.22). VI exhibited moderate posi- tic accuracy for differentiating TD from normal were
tive correlation with medication dosage (p = 0.008, found to be 28%, 44%, 34%, 42% and 36%, respectively
r = 0.49) and also RAIU (2-h uptake; p = 0.008, r = 0.57). (p = 0.001).
No significant association was found between the medi-
cation dosage and RAIU with CD parameters, such as
DISCUSSION
PSV, EDV and RI (p > 0.05).
SWE (p = 0.02; r = 0.41) and SWV (p = 0.011; TD, as a less common pediatric health problem,
r = 0.44) values were positively correlated with the age constitutes 10%15% of the CH etiologies that affect
of the participants. the neurologic development, growth curve and cognitive
The 2-h RAIU values were moderately correlated capacity in early childhood (Voss et al. 2013). Thyroid
with medication dosage (p < 0.03; r = 0.48). Since TSH US is the initial imaging modality in patients with a thy-
and fT4 levels were normalized by medication, there roid function disorder. A qualitative description of
was no association between actual TSH values and VI in markedly heterogeneous and enlarged thyroid glands
patients that were under treatment. with hypo- or isoechogenicity, as well as pronounced
hypervascularity mimicking a thyroid inferno pattern
Table 2. Spearman correlation analysis among color Doppler seen in Graves’ disease, have been reported for TD
and SMI parameters with RAIU and medication in study group (Voss et al. 2013). However, well-described gray-scale
Parameters p rho US and quantitative Doppler characteristics have yet to
be determined for TD. We categorized the echogenicity
VI vs. Volume 0.002 0.28
VI vs. EDV 0.001 0.34
(hypoechoic, isoechoic and hyperechoic) of the thyroid
VI vs. PSV 0.001 0.37 gland compared with the cricoid cartilage to increase the
VI vs. RI 0.018 0.22 sensitivity of the mild echogenicity changes. We identi-
2-h RAIU vs. Dosage 0.03 0.48 fied horizontally oriented tiny echogenic reticulations
VI vs. Dosage 0.008 0.49
VI vs. 1-h RAIU 0.012 0.54 within the either hypo- or isoechoic gland, resembling
VI vs. 2-h RAIU 0.008 0.57 thymus echostructure. At first glance; significant hypoe-
SWE vs. Age 0.02 0.41 chogenicity of the gland having smooth and sharp mar-
SWV vs. Age 0.011 0.44
gins may rule out thyroiditis presenting blunt contours
EDV = end-diastolic velocity; PSV = peak systolic velocity; and may consider an ectopic cervical thymus with thy-
RAIU = radioactive iodine uptake; RI = resistive index; SWE = shear
wave elastography; SWV = shear wave velocity; VI = vascularity index. roid agenesis. Although the echogenicities would be
Bold values show statistically significant results. similar with the thymus, the thymic tissue would be
 ARTICLE IN PRESS
Pediatric Thyroid Dyshormonogenesis US
I. ADALETLI et al. 7

Table 3. Diagnostic performances of SMI, CD and SWE parameters, including VI, elasticity, velocity, RI, PSV, EDV, based on cut-
off values in differentiating TD from normal
Parameter Cutoff value Sensitivity Specificity PPV NPV Diagnostic accuracy (%) AUC (%) p

VI (%) 10 70 63 68 65 71 (6079) 68 ** 0.008

PSV (cm/s) 15.5 54 44 56 47 52 (3964) 50 0.88
EDV (cm/s) 8 58 55 60 50 56 (4468) 52 0.44
RI 0.53 28 44 34 42 36 (2545) 32 0.001
SWV (m/s) 1.76 67 54 65 52 62 (5170) 60 0.21
SWE (kPa) 10 41 70 62 64 61 (5271) 63 0.08
AUC = area under the curve; CD = color Doppler; CI = Confidence interval; EDV = end-diastolic velocity; NPV = negative predictive value;
PPV = positive predictive value; PSV = peak systolic velocity; RI = resistive index; SMI = superb microvascular imaging; SWE = shear wave elastog-
raphy; SWV = shear wave velocity; TD = thyroid dyshormonogenesis; VI = vascularity index.
Bold values indicate statistically significant results.

hypovascularized and echogenic structures in the diffusely hypoechoic thyroid gland would include TD,
absence of acoustic shadowing would be punctate in Graves’ disease and, in some cases, Hashimoto thyroid-
shape within the thymic tissue (Erol et al. 2017; Kim itis as differential diagnoses. However, diffuse autoim-
et al. 2012). Significant hypervascularity along with a mune thyroid disorders would be heterogeneous in

Fig. 5. Representative ROC curves between patients with TD and healthy control volunteers investigating vascularity
index (VI) by SMI; EDV, PSV and RI by CD; shear wave elasticity (SWE) and shear wave velocity (SWV). CD = color
Doppler; EDV = end-diastolic velocity; PSV = peak systolic velocity; RI = resistive index; ROC = receiver operating
characteristic; SMI = superb microvascular imaging; SWE = shear wave elastography; SWV = shear wave velocity;
TD = thyroid dyshormonogenesis; VI = vascularity index.
 ARTICLE IN PRESS
8 Ultrasound in Medicine & Biology Volume 00, Number 00, 2019

contrast to a homogeneously enlarged and hypoechoic cutoff value of 10.58% has revealed better diagnostic
gland in TD. It would be relatively simple to differenti- accuracy (70%) for diagnosing Hashimoto thyroiditis
ate autoimmune thyroiditis from CH via biochemical than CD parameters, including PSV, EDV and RI. This
analysis by evaluating TSH, fT4 and autoantibody levels is the first study investigating the role of SMI and SWE
and via clinicodemographic evaluation as well as US in TD. We depicted significant diagnostic utility of VI in
examination. Along with increased thyroid function lev- differentiating TD from normal in correlation with medi-
els, VI could help differentiate permanent from transient cation dosage and RAIU ratios. Quantitative evaluation
hypothyroidism, reduce the need of repetition of scintig- of thyroid vascularity by SMI and VI also demonstrated
raphy and refer the patients to the final diagnostic tests, the superiority of the SMI compared with other Doppler
such as genetic analysis instead of scintigraphy. parameters in terms of diagnosing TD, estimating medi-
Sodium iodide symporter uptake of 123I is associ- cation status and also the scintigrafic results. In the pres-
ated with TSH stimulation, as well as the organification ent study, unlike VI, commonly used CD parameters,
process. The normal level of 123I uptake is about 12% § such as EDV, PSV and RI, were not correlated with
6% at 120 min for a mean thyroid volume of about RAIU or medication dosage. Velocities obtained with
12 mL and mean TSH level of about 1.2 mU/L for a spectral analysis depend on multiple factors, such as the
physiologic stimulation (Clerc 2014). Thyroid scintigra- operator experience, vessel diameters and the insonation
phy with 99mTc-pertechnetate or 123I are suggested as angle and also take longer to perform. SMI images can
the standard screening modalities in the evaluation of be obtained in a few seconds, are less operator dependent
thyroid dysgenesis or TD, as well as ectopic thyroid. In and the VI is easy to calculate. Therefore, by demon-
selected patients, the 99mTc pertechnetate and perchlo- strating tiny and slow flow vessels within the gland with
rate discharge test is combined to identify possible higher resolution and by suppressing motion artifacts
defects during the organification process. Well-known (Yongfeng et al. 2016), SMI would be superior to other
limitations of thyroid scintigraphy can be put in order as vascular imaging modalities in diagnosing TD, as well
radiation exposure to the radiosensitive tissue, expense as estimating RAIU ratios.
and availability at large medical centers (L
eger et al. Viscoelastic tissue properties of thyroid gland in
2014). Although 99mTc-pertechnetate is selected based patients with CH have been investigated in a recent study
on its relatively lower thyroid and total body radiation (Sarıca et al. 2016). They found higher strain index ratios,
dose (approximately 0.04 mSv compared with 0.35 with a cutoff value of 0.69, in CH compared with control
mSv) (Becker et al. 1999), the final result of these patients. Since TD consists of heterogeneous genetic
examinations is radiation exposure to the children. The mutations, several degrees of thyroid hormone synthesis
correlative qualitative results of the US evaluation with defects and lack of inflammatory and fibrotic processes,
scintigraphy have been reported in several studies as seen in the chronic autoimmune thyroiditis, lack of sig-
(Chang et al. 2011; Supakul et al. 2012), especially in nificant difference in median SWE values among patients
thyroid dysgenesis. Furthermore, in selected patients, with TD and the normal population was expected. We
US evaluation provides valuable data beyond the scintig- found SWV values just about 1.8 m/s in both control and
raphy by demonstrating thyroid tissue that could not be study group. SWE values greater than 11 kPa and SWV
visualized on scintigraphy (Tamam et al. 2009). In addi- values greater than 1.95 m/s have been revealed as helpful
tion to this, we propose vascularity quantification via for diagnosing Hashimoto thyroiditis (Kandemirli et al.
SMI as a novel Doppler modality with considerable clin- 2018). When mean strain index cutoff value is accepted
ical significance, similar to scintigraphy. Also, we found as 0.31, the greater ratio has a sensitivity of 92% to diag-
a highly significant positive correlation in the VI and nose Hashimoto thyroiditis (Öztürk et al. 2017). The opti-
RAIU values for determining TD, which has not been mal cutoff SWV value via acoustic radiation force
revealed for biochemical parameters such as TSH and impulse quantification for differentiating Hashimoto thy-
thyroglobulin. SMI outperforms scintigraphy in patients roiditis has been reported as 1.41 m/s with 80% of the
with eutopic thyroid gland on US without significant AUC (Yucel et al. 2017). In another study, a cutoff value
RAIU (Perry et al. 2006). Both RAIU and vascularity of 1.86 m/s for SWV to distinguish Hashimoto thyroiditis
are stimulated with hyperthyrotropinemia in TD patients. by virtual tissue touch quantification provided a sensitiv-
The presented moderate positive correlation of volume ity of 82.5% (Lin et al. 2018). The duration of hypothy-
and VI with RAIU could consider that detailed evalua- roidism and status of medication would affect the SWE
tion of thyroid gland with SMI would be more than a results in TD as well. There have not been comparable
complementary modality to the scintigraphy. data regarding SWE evaluation of thyroid gland in
VI via SMI has been published as a novel Doppler patients with several types of hypothyroidism.
parameter in diagnosing and grading Hashimoto thyroid- The prevalence of thyroid nodules, which is 1.5%
itis in children (Bayramoglu et al. 2018). VI with a among adolescents, increases in patients with CH due to
 ARTICLE IN PRESS
Pediatric Thyroid Dyshormonogenesis US
I. ADALETLI et al. 9

hyperthyrotropinemia (Niedziela 2006) and might result per lobe, as well as the median VI for each patient. We
in dysplasia followed by neoplasia. In the present study, did not obtain urinary iodine levels in these patients
we found nodules in seven of the 31 TD patients (22%), because these tests cannot exclude iodine deficiency.
which is considerably higher than in previous studies. Although further long-lasting studies employing larger
Along with the increased incidence, more than one nod- sample sizes are required to address these limitations,
ule was found in three patients. Based on the knowledge our findings suggest that SMI potentially anticipates
that the presence of a nodule in a euthyroid patient youn- RAIU values and medication dosage by detecting
ger than 10-y-old is more suggestive for malignancy the microvascular network in patients with hypothyroid-
(Clerc et al. 2008) and hyperfunctional nodules might ism. The repetitive scintigraphic evaluations in non-
confound the RAIU values, follow-up US examinations cooperative children would be replaced with VI via
would be crucial in the dyshormonogenetic thyroid SMI.
gland. Dyshormonogenetic goiter, as well as the insuffi- In conclusion, this is the first study indicating the
cient dietary iodine, is a major risk factor for papillary clinical utility of this promising novel vascular imaging
and follicular thyroid carcinoma development depending modality in diagnosis and follow up of TD. Scintigraphic
on the duration of high levels of thyrotropin in the evaluations can be allocated to the patients with discor-
plasma (Chertok et al. 2012; Şıklar et al. 2012). This dant ultrasonographic results and used to demonstrate
explains why patients with TD who have been on thyroid ectopic tissue.
hormone therapy should be examined regularly and the
diagnosis of TD via novel US applications could take
REFERENCES
place in future approaches.
Autosomal recessive inherited mutations have been Adibi A, Haghighi M, Hosseini SR, Hashemipour M, Amini M, Hovse-
reported in the pathogenesis of TD, causing an increased pian S. Thyroid abnormalities among first-degree relatives of chil-
dren with congenital hypothyroidism: An ultrasound survey. Horm
prevalence (15-fold higher) in patients with familial dis- Res Paediatr 2008;70:100–104.
ease, which would be more than expected by chance Bayramoglu Z, Kandemirli SG, Caliskan E, Yilmaz R, Kardelen AD,
alone (Park and Chatterjee 2005). Developmental thy- Poyrazoglu S, Bas F, Adaletli I, Darendeliler F. Superb microvas-
cular imaging in assessment of Hashimoto’s thyroiditis in the pedi-
roid abnormalities were found among 3.5% of parents of atric population. Clin Radiol 2018;73:1059.e9–1059.e15.
the patients with CH and 10.5% of their siblings (Adibi Becker D, Charkes ND, Dworkin H, Hurley J, McDougall IR, Price D,
et al. 2008). Based on our results, five siblings of the 31 Royal H, Sarkar S. Procedure guideline for thyroid scintigraphy:
1.0. Society of Nuclear Medicine. J Nucl Med 1996;37:1264–1266.
patients (16%) were also diagnosed with TD and receiv- Bikker H, den Hartog MT, Baas F, Gons MH, Vulsma T, de Vijlder JJ.
ing thyroid hormone replacement. For this reason, evalu- A 20-box pair duplication in the human thyroid peroxidase gene
ating patients with biochemical analysis along with VI results in a total iodine organification defect and congenital hypo-
thyroidism. J Clin Endocrinol Metab 1994;79:245–252.
via SMI would be a radiation-free and cost-effective Chang YW, Lee DH, Hong YH, Hong HS, Choi DL, Seo DY. Congeni-
diagnostic screening method in patients with suspected tal hypothyroidism: Analysis of discordant US and scintigraphic
TD, which could also be extended to their siblings. Also, findings. Radiology 2011;258:872–879.
Chertok Shacham E, Ishay A, Irit E, Pohlenz J, Tenenbaum-Rakover Y.
an experiment of stopping levothyroxine when the Minimally invasive follicular thyroid carcinoma developed in dys-
patient is about 3-y-old is often undertaken for a few hormonogenetic multinodular goiter due to thyroid peroxidase gene
days to distinguish permanent from transient hypothy- mutation. Thyroid 2012;22:542–546.
Clerc J. Imaging the thyroid in children. Best Pract Res Clin Endocrinol
roidism (Supakul et al. 2012). Prompt evaluation of the Metab 2014;28:203–220.
parenchyma with SMI might reduce the need for treat- Clerc J, Monpeyssen H, Chevalier A, Amegassi F, Rodrigue D, Leger
ment interruption in some patients. Although the differ- FA, Richard B. Scintigraphic imaging of paediatric thyroid dys-
function. Horm Res Paediatr 2008;70:1–3.
ent causes of CH would not alter the disease Erol OB, Şahin D, Bayramoglu Z, Yılmaz R, Akpınar YE, Ünal OF, €
management, being aware of dyshormonogenesis would Yekeler E. Ectopic intrathyroidal thymus in children: Prevalence,
suggest treatment for life that is clear and cautious for imaging findings and evolution. Turk J Pediatr 2017;59:387–394.
Kandemirli SG, Bayramoglu Z, Caliskan E, Sari ZNA, Adaletli I.
parents. Quantitative assessment of thyroid gland elasticity with shear-wave
Our study has several limitations. First, we included elastography in pediatric patients with Hashimoto’s thyroiditis.
a limited number of patients, although we included age J Med Ultrasond (2001) 2018;45:417–423.
Kim HG, Kim MJ, Lee MJ. Sonographic appearance of intrathyroid
and gender-matched control patients. Second, we could ectopic thymus in children. J Clin Ultrasound 2012;40:266–271.
not obtain genetic analyses from all of the patients con- L
eger J, Olivieri A, Donaldson M, Torresani T, Krude H, van Vliet G,
firming the defective stage of the hormone synthesis. Polak M, Butler. Congenital Hypothyroidism Consensus Confer-
ence Group. European Society for Paediatric Endocrinology con-
Third, the ultrasonographic examination was carried out sensus guidelines on screening, diagnosis, and management of
with the knowledge of the final diagnosis of TD by a congenital hypothyroidism. J Clin Endocrinol Metab 2014;99:
radiologist, and the interval from the scintigraphy to US 363–384.
Lin ZM, Wang Y, Liu CM, Yan CX, Huang PT. Role of virtual touch
examination remained variable. To reduce the operator- tissue quantification in Hashimoto’s thyroiditis. Ultrasound Med
dependent variability, we provided three measurements Biol 2018;44:1164–1169.
 ARTICLE IN PRESS
10 Ultrasound in Medicine & Biology Volume 00, Number 00, 2019

Medeiros-Neto G, Knobel M, DeGroot LJ. Genetic disorders of the Şıklar Z, Berberoglu M, Yagmurlu A, Hacıhamdioglu B, Savaş Erdeve
thyroid hormone system. Philadelphia: Lippincott Williams & €
S, Fit€oz S, Kır M, Oçal G. Synchronous occurrence of papillary car-
Wilkins; 2002. p. 375–402. cinoma in the thyroid gland and thyroglossal duct in an adolescent
Niedziela M. Pathogenesis, diagnosis and management of thyroid with congenital hypothyroidism. J Clin Res Pediatr Endocrinol
nodules in children. Endocr Relat Cancer 2006;13:427–453. 2012;4:30.
€
Oztürk M, Yildirim R. The usefulness of strain wave elastography in Supakul N, Delaney LR, Siddiqui AR, Jennings SG, Eugster EA, Kar-
the diagnosis and grading of Hashimoto’s thyroiditis in children. mazyn B. Ultrasound for primary imaging of congenital hypothy-
La radiologia medica 2017;122:960–966. roidism. Am J Roentgenol 2012;199:360–366.
Park AY, Seo BK, Woo OH, Jung KS, Cho KR, Park EK, Cha SH, Cha Tamam M, Adalet I, Bakir B, Türkmen C, Darendeliler F, Baş F, Sanli
J. The utility of ultrasound superb microvascular imaging for evalu- Y, Kuyumcu S. Diagnostic spectrum of congenital hypothyroidism
ation of breast tumour vascularity: Comparison with colour and in Turkish children. Pediatr Int 2009;51:464–468.
power Doppler imaging regarding diagnostic performance. Clin Voss E, Stierkorb E, Pohlenz J, Hermanns P, Staatz G, Rohrer T, Ham-
Radiol 2018;73:304–311. mersen G. Ultrasound findings in congenital hypothyroidism due to
Park SM, Chatterjee VK. Genetics of congenital hypothyroidism. dyshormonogenesis. Ultraschall in Med 2013;34. doi: 10.1055/s-
J Med Genet 2005;42:379–386. 0033-1354925. WS_SL19_08.
Perry RJ, Maroo S, Maclennan AC, Jones JH, Donaldson MD. Com- Yongfeng Z, Ping Z, Wengang L, Yang S, Shuangming T. Application
bined ultrasound and isotope scanning is more informative in the of a novel microvascular imaging technique in breast lesion evalua-
diagnosis of congenital hypothyroidism than single scanning. Arch tion. Ultrasound Med Biol 2016;42:2097–2105.
Dis Child 2006;91:972–976. Yucel S, Ceyhan Bilgici M, Kara C, Can Yilmaz G, Aydin HM, Elmali
Sarıca MA, Dalkıran T, Menzilcioglu MS, Duymuş M, Yurttutan N. M, Tomak L, Saglam D. Acoustic radiation force impulse quantifi-
Ultrasound elastography in pediatric congenital hypothyroid cation in the evaluation of thyroid elasticity in pediatric patients
patients. Pediatr Endocrinol Rev 2016;14:48–53. with Hashimoto thyroiditis. J Ultrasound Med 2017;37:1143–1149.