DOI:10.2214/AJR.07.3137
AJR 2008; 190:W255-W260
© American Roentgen Ray Society
Hypervascular Thyroid Nodules on Time-Resolved MR Angiography at 3 T: Radiologic–Pathologic Correlation
Derek G. Lohan1,
Anderanik Tomasian1,
Roya Saleh1,
Mayil Krishnam1 and
J. Paul Finn1
1 All authors: Department of Radiological Sciences, David Geffen School of
Medicine at UCLA, Peter V. Ueberroth Bldg., Ste. 3371, 10945 Le Conte Ave.,
Los Angeles, CA 90095-7206.
Received September 10, 2007;
accepted after revision October 12, 2007.
Address correspondence to D. G. Lohan
(derek.lohan{at}gmail.com).
WEB
This is a Web exclusive article.
Abstract
OBJECTIVE. Detection of a thyroid nodule, either incidental or as a
result of related symptomatology, is an extremely common event, often inducing
considerable uncertainty regarding the requirement for and best means of
further investigation. Whereas tissue sampling represents the sole means of
true characterization of these lesions, a number of imaging characteristics
have been suggested as potential indicators of the presence of malignancy. The
potential value of time-resolved MR angiography, whereby a minimal dose of IV
contrast agent is dynamically depicted during the first pass of the bolus
through the various compartments of circulation, has recently been realized,
particularly so with regard to supraaortic angiography. However, it is not
uncommon during such temporal imaging to identify focal hyperenhancing thyroid
nodules, the significance of which has not previously been described in the
literature. We describe the frequency of occurrence and potential significance
of this finding, using pathologic correlation where available.
CONCLUSION. The prevalence of malignancy in incidentally detected
focal hyperenhancing thyroid parenchymal nodules during time-resolved MR
angiography is significant, representing 8.3% (1/12) of patients for whom
cytologic correlation was available. Further investigation is certainly
warranted when encountering such a lesion in clinical practice, particularly
because it appears as though time-resolved MR angiography is of no value in
the pathologic discrimination of such incidentally identified lesions.
Keywords: enhancement • hypervascular • MR angiography • thyroid • time-resolved
Introduction
Detection of a thyroid nodule, whether by palpation or imaging, is an
extremely common occurrence because thyroid nodules are present in 4–7%
of the asymp tomatic population
[1] and 50% of persons at
autopsy [2]. Characterization
of such nodules often presents a challenge, involving a combination of
clinical, radiologic, biochemical, and often cytologic or histologic tools.
Certain characteristics have been associated with increased likelihood of
malignancy, such as hypervascular central flow on color Doppler sonography
[3].
Recent reports suggest that contrast-enhanced MR angiography at 3 T shows
particular promise for evaluation of the carot id circulation
[4,
5]. Time-resolved MR angio
graphy, in which multiple time points are sampled rapidly during the first
pass of contrast-enhanced imaging, provides functional information that
complements high-spatial-resolution MR angiography
[6].
We have frequently noted focal nodular hyperenhancement within the thyroid
gland during the first pass of contrast-enhanced imaging relative to remote,
normal thyroid tissue. To the best of our knowledge, the pathologic
significance of such a finding has not been previously reported. The purpose
of this study was to document the frequency of occurrence of such
hypervascular nodules within a single radiology department in a population of
patients referred for contrast-enhanced MR angiography of the carotids and,
where possible, to provide pathologic correlation with the MRI findings.
Materials and Methods
After institutional review board approval of this HIPAA-compliant study,
retrospective review of all head and neck MR angiographic examinations at a
single center on a 3-T MR system (Magnetom Trio, Siemens Medical Solutions)
was performed. All examinations were clinically indicated for sus pected
supraaortic arterial disease. The MR scan ner was equipped with 32 independent
receiver channels and a rapid three-axis gradient system, providing a peak
gradient amplitude of 45 mT/m and maximum slew rate of 200 mT/m/ms. For the
time-resolved acquisition, 18 of the 32 available channels were interfaced to
individual coil elements for parallel acquisition and optimal signal reception
over a 500-mm field of view.
Time-Resolved MR Angiography Technique
Time-resolved MR angiography was performed as a complementary technique
before 3D high-spatial-resolution contrast-enhanced MR angio g-raphy in all
cases and as a separate contrast injection. Patients were positioned on the MR
table in a supine orientation and moved head first into the magnet bore. A
20-gauge IV cannula was sited in an antecubital vein and connected to an
electronic power injector (MR Spectris, Medrad). After acquisition of
multiplanar localizers, time-resolved MR angiography was performed using a
standard dose of 4 mL of gadolinium-based contrast agent (gadopentetate
dimeglumine, [Mag nevist, Bayer HealthCare]) injected at a rate of 2.0 mL/s
and flushed with a 20-mL saline bolus at the same rate.
Time-resolved MR angiography was performed using an ultrafast 3D gradient
recalled-echo (GRE) sequence integrated with a time-resolved echo-shared
angiographic technique (TREAT). An asymmetric k-space sampling scheme (partial
Fourier 6/8) was applied in all three dimensions to optimize both the echo and
acquisition times for each 3D data set. The following sequence parameters were
used: TR/TE, 2.57/1.06; flip angle, 24°; bandwidth, 750 Hz/pixel; field of
view, 500 x 132 mm; matrix, 512 x 410 using 24 partitions with a
thickness of 4 mm (interpolated to 2.56 mm); and voxel size, 1.2 x 1.0
x 4.0 mm. Parallel imaging, using a generalized auto calibrating
partially parallel acquisition (GRAPPA) algorithm, was applied with an
acceleration factor of 3, with 30 reference k-space lines for calibration in
the left-to-right phase-encoding direction.
Time-resolved MR angiography was performed in the coronal plane using a 3D
imaging slab that extended from below the aortic arch to the cranial vertex
and laterally to both subclavian arteries and incorporating the aortic arch,
carotid system, and vertebral arterial system. Twelve sequential measurements,
commencing 5 seconds after the start of the contrast injection and updated
every 1.8 seconds, were obtained during respiratory suspension at
end-inspiration, ensuring data acquisition from before the systemic arterial
pass of the contrast bolus to the jugular venous phase.
Magnitude subtraction of the first (mask) image set from all subsequent
image sets was performed online, as was coronal maximum-intensity-projection
(MIP) reconstruction of all phases of enhancement. All examinations were
formally assessed by a single fellowship-trained diagnostic cardiovascular
imaging radiologist with the purpose of thyroid parenchymal evaluation for the
presence of solitary or multiple foci of hypervascular enhancement.
Lesion Evaluation
Thyroid lesions thus detected were then blindly evaluated by two separate
radiologists and classified as homogeneous nodule hyper enhancement, 1; peri
pheral nodular hyper enhancement with central hypo enhancement, 2; or
heterogeneous nodular hyper enhancement, 3. Furthermore, dynamic flow curves
were generated for each thyroid nodule, determining the maximal pixel signal
intensity within each lesion during the 12 sequential measurements obtained
compared with that of adjacent normal thyroid parenchyma on the same temporal
image.
Results
During a 30-month period, 624 MR angiography neck examinations
incorporating the time-resolved technique described were performed on a single
3-T MRI system, yielding solitary or multiple areas of hypervascular thyroid
parenchymal enhancement on time-resolved MR angiography in 49 patients (37
women and 12 men; mean age, 64.5 years; age range, 36–90 years), for a
prevalence of 7.8%. Cytologic characterization of these lesions was performed
in 12 patients, one of whom subsequently underwent total thyroidectomy, with
histologic confirmation of locally metastatic thyroid papillary carcinoma.
The mean age in this group of 12 patients with radiologic–cytologic
correlation was 56.9 years (10 women and two men; age range, 36–80
years). The presence of parenchymal nodules had been known before
time-resolved MR angiography in 25% of the patients (three of 12). In the
remaining nine patients, focal thyroid hyperenhancement on time-resolved MR
angiography prompted fine-needle aspiration biopsy (FNAB). Each of the three
patients in whom nodules were known to be present underwent subsequent FNAB to
confirm the benignity of these lesions. When more than one nodule was present,
FNAB of each nodule was performed to exclude the presence of more than one
pathologic process. Of note, no history of thyroid hormone supplementation was
obtained from any patient, either at the time of time-resolved MR angiography
or in the preceding 12 months. Individual patient demographic details and
information relating to lesion characteristics on complementary imaging
techniques are provided in Table
1.
Time-Resolved MR Angiography
The conspicuity of these lesions at time-resolved MR angiography resulted
from early hyperenhancement compared with the adjacent normal thyroid
parenchyma, which tended to occur in a diffusely homogeneous manner. Solitary
lesions were detected in eight patients, with two or more nodules in the
remaining four patients. These nodules ranged in size from 4.9 to 59.7 mm,
with a mean lesional diameter of 25.2 mm. Figures
1A and
1B illustrates one such example
of focal thyroid hypervascularity (patient 3 in
Table 1), with parenchymal
enhancement curves compared with those of uninvolved thyroid tissue.
View larger version (64K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A —47-year-old woman with isthmic hypervascular thyroid nodule
(bold region of interest in A). Single arterial phase image from
dynamic time-resolved MR angiography reveals conspicuity of this enhancing
lesion.
|
|
View larger version (10K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B —47-year-old woman with isthmic hypervascular thyroid nodule
(bold region of interest in A). Graph shows dynamic enhancement curves
for hypervascular nodule and normal thyroid parenchyma. Note more rapid
time-to-peak (image number 7) and overall higher signal intensity for
hypervascular nodule compared with thyroid parenchyma.
|
|
Blinded classification of the 12 lesions detected at time-resolved MR
angiography by the independent observers is shown in
Table 1. There was perfect
interobserver agreement in 11 of 12 patients (91.7%): four lesions showing
homogeneous hyper en hancement; two, peripheral hyper enhance ment with
central hypoenhancement; and five, heterogeneous hyperenhancement. Dis
agreement existed with regard to a single nodule that was subsequently
determined to represent a benign colloid nodule. Observer 1 designated this
lesion as displaying hetero geneous hyperenhancement, whereas observer 2
determined this nodule to have relative central hypoenhancement.
Analysis of the peak pixel signal intensity within the thyroid nodule
relative to adjacent normal thyroid parenchyma (signal intensity ratio) is
also shown in Table 1. Of note,
benign follicular lesions (n = 2, average signal intensity ratio =
1.41) had the lowest ratio of relative enhancement, whereas multinodular
goiter had the highest ratio (n = 1, signal intensity ratio = 2.66).
Benign colloid nodules, Hashimoto's thyroiditis, papillary carcinoma, and
hyperplastic nodules revealed similar degrees of nodular enhancement.
Correlative Findings on Sonography, CT, and Nuclear Scintigraphy
Sonography was performed in all 12 patients at the time of FNAB. As shown
in Table 1, a broad range of
echo patterns was observed, including hypoechoic (n = 5), isoechoic
(n = 1), hyperechoic (n = 2), and mixed hyperechoic and
hypoechoic (n = 4) lesional characteristics.
CT correlation was available in five of these patients in the form of both
unenhanced and contrast-enhanced imaging in three patients and
contrast-enhanced acquisitions alone in the remaining two individuals. In the
three patients in whom dual-phase imaging was performed (including two benign
colloid nodules and one hyperplastic nodule), these lesions appeared as
hypoattenuating foci relative to adjacent thyroid tissue on unenhanced
imaging, with mixed hyper- and isoattenuation on arterial phase
contrast-enhanced imaging. Both the remaining two patients (including one with
papillary carcinoma and one with benign colloid nodule) underwent
contrast-enhanced CT only, with the nodules in these patients appearing in
isoattenuation to adjacent thyroid, likely a manifestation of contrast
equalization during the blood-pool phase of imaging (Figs.
2A,
2B,
2C,
2D,
2E,
2F,
2G,
2H,
3A and
3B).
View larger version (52K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A —45-year-old woman with incidentally noted left thyroid
hypervascular nodule, subsequently diagnosed as papillary carcinoma.
Sequential frames from time-resolved MR angiography reveal presence of
relatively more rapid and more intense nodular enhancement
(arrows).
|
|
View larger version (57K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —45-year-old woman with incidentally noted left thyroid
hypervascular nodule, subsequently diagnosed as papillary carcinoma.
Sequential frames from time-resolved MR angiography reveal presence of
relatively more rapid and more intense nodular enhancement
(arrows).
|
|
View larger version (57K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2C —45-year-old woman with incidentally noted left thyroid
hypervascular nodule, subsequently diagnosed as papillary carcinoma.
Sequential frames from time-resolved MR angiography reveal presence of
relatively more rapid and more intense nodular enhancement
(arrows).
|
|
View larger version (56K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2D —45-year-old woman with incidentally noted left thyroid
hypervascular nodule, subsequently diagnosed as papillary carcinoma.
Sequential frames from time-resolved MR angiography reveal presence of
relatively more rapid and more intense nodular enhancement
(arrows).
|
|
View larger version (54K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2E —45-year-old woman with incidentally noted left thyroid
hypervascular nodule, subsequently diagnosed as papillary carcinoma.
Sequential frames from time-resolved MR angiography reveal presence of
relatively more rapid and more intense nodular enhancement
(arrows).
|
|
View larger version (52K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2F —45-year-old woman with incidentally noted left thyroid
hypervascular nodule, subsequently diagnosed as papillary carcinoma.
Sequential frames from time-resolved MR angiography reveal presence of
relatively more rapid and more intense nodular enhancement
(arrows).
|
|
View larger version (87K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2G —45-year-old woman with incidentally noted left thyroid
hypervascular nodule, subsequently diagnosed as papillary carcinoma. Duplex
Doppler sonogram (G) confirms presence of nodular vascularity. Note
isoattenuation of this lesion compared with adjacent thyroid parenchyma on
blood-pool phase CT (H), with markers (red) showing extent of
thyroid nodule.
|
|
View larger version (61K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2H —45-year-old woman with incidentally noted left thyroid
hypervascular nodule, subsequently diagnosed as papillary carcinoma. Duplex
Doppler sonogram (G) confirms presence of nodular vascularity. Note
isoattenuation of this lesion compared with adjacent thyroid parenchyma on
blood-pool phase CT (H), with markers (red) showing extent of
thyroid nodule.
|
|
View larger version (65K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A —38-year-old man with hyperplastic thyroid nodule
(arrows). Note presence of peripheral hyperenhancement on arterial
phase image (A), with loss of lesion definition on subsequent image
(B) acquired only 3.6 seconds later. Loss of definition is due to
relatively delayed thyroid parenchymal enhancement in B.
|
|
View larger version (65K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B —38-year-old man with hyperplastic thyroid nodule
(arrows). Note presence of peripheral hyperenhancement on arterial
phase image (A), with loss of lesion definition on subsequent image
(B) acquired only 3.6 seconds later. Loss of definition is due to
relatively delayed thyroid parenchymal enhancement in B.
|
|
In three patients, 123I scintigraphy was performed (one with
Hashimoto's thyroiditis, one with a hyperplastic nodule, and one with a benign
colloid nodule) showing normal radiotracer uptake in all cases.
Discussion
The results of our study suggest that visualization of nodular
hypervascular foci within the thyroid parenchyma during time-resolved MR
angiography is common, occurring in 7.8% of our patient population, and that
there is a finite probability of malignancy (occurring in one of 49 [2%]
patients with enhancing nodules and one of 12 [8.3%] patients in whom
cytologic analysis was performed). However, given the limited number of cases
with cytologic comparison, it is conceivable that the true prevalence of
malignancy in this setting may be higher or lower than suggested. Nonetheless,
to our knowledge, our results represent the only such information available to
date, suggesting that the identification of a hyper vascular thyroid nodule on
time-resolved MR angio graphy should warrant further evaluation with FNAB if
necessary.
Incidental detection of thyroid ab normalities on sono graphy or
cross-sectional imaging is not rare. This poses a challenge for the
interpreting radiologist for whom an optimal management algorithm must
consider the specificity of a technique in detecting a malignant process and
the prevalence of this condition in the asymptomatic population.
Unfortunately, although a number of studies have revealed trends that may
suggest the presence of malignancy, none has been sufficiently specific to
preclude the requirement for FNAB.
Marked intrinsic vascularity exceeding that of adjacent thyroid parenchyma
has been shown to be the most common pattern in thyroid malignancy, occurring
in 69–74% of such lesions
[7]. However, this pattern is
nonspecific, with more than 50% of in trinsically hypervascular nodules being
benign [2]. Other features
suggestive of malignancy were microcalcifications, marked lesional
hypoechogenicity, local invasion, lesions taller than they are wide, and
abnormal lymph node echogenicity, whereas a surrounding uniform halo and
nodular avascularity are reassuring for benignancy
[8]. However, none of these
appearances are pathognomonic for either the benign or malignant processes.
Shetty et al. [9] reviewed 230
patients with incidentally detected thyroid abnormalities on CT, concluding
this technique had no role in the differentiation of benign from malignant
processes. Notably, these authors found a 3.9% prevalence of malignancy and
7.4% prevalence of potential malignancy in this population group.
In addition, 123I-scintigraphy has an established role in the
evaluation of potential thyroid malignancy, with a cold nodule being
characteristic of malignancy, although with insufficient specificity (<
10%) to establish this diagnosis
[10]. Similarly, although the
utility of conventional MRI in the evaluation of the presence and extent of
regional invasion in known thyroid malignancy is established
[11], this technique has not
as yet been determined as being of sufficient discriminatory value to
influence the diagnostic algorithm
[12]. We suggest that
time-resolved MR angiography is of similarly limited utility in the
differentiation of hypervascular thyroid lesions. When such lesions are
incidentally detected, neither the pattern of enhancement, regarding which we
observed a high level of interobserver agreement, nor the ratio of nodule
enhancement relative to adjacent thyroid parenchyma are of sufficient accuracy
to enable suggestion of a likely pathologic diagnosis.
In conclusion, incidental detection of a thyroid nodule during head and
neck imaging continues to represent a challenge with re gard to derivation of
an optimal management algorithm. High-resolution sonography, CT, radionuclide
scintigraphy, and MRI represent valuable techniques in the identification of
the presence and extent of such a lesion, in addition to allowing surveillance
over time for those patients in whom FNAB is not performed. However, these
techniques have insufficient predictive values to eliminate the requirement
for FNAB in the majority of cases. As a result, when such a thyroid lesion is
detected, the next step in the diagnostic algorithm depends heavily on
knowledge of the prevalence of malignancy when such a nodule is encountered on
the image technique in question. This report details the initial experience of
a single center with hyper vascular thyroid nodules when detected on
time-resolved MR angiography and suggests that the prevalence of malignancy in
the presence of these lesions is significant (8.3%, one of 12 patients) and
that further investi gation is warranted when such a nodule is encountered.
Time-resolved MR angiography is, however, of no value in the pathologic
discrimination of such incidentally identified lesions.
References
- Mazzaferri EL. Management of a solitary thyroid nodule.
N Engl J Med 1993;328
: 553-559[Free Full Text]
- Frates MC, Benson CB, Doubilet PM, Cibas ES, Marqusee E. Can color
Doppler sonography aid in the prediction of malignancy of thyroid nodules?
J Ultrasound Med 2003;22
: 127-131[Abstract/Free Full Text]
- Pacella CM, Guglielmi R, Fabbrini R, et al. Papillary carcinoma in
small hypoechoic thyroid nodules: predictive value of echo color Doppler
evaluation—preliminary results. J Exp Clin Cancer
Res 1998; 17:127
-128[Medline]
- Sodickson DK, McKenzie CA, Li W, Wolff S, Manning WJ, Edelman RR.
Contrast-enhanced 3D MR angiography with simultaneous acquisition of spatial
harmonics: a pilot study. Radiology 2000;217
: 284-289[Abstract/Free Full Text]
- Nael K, Ruehm SG, Michaely HJ, et al. High spatial-resolution
CE-MRA of the carotid circulation with parallel imaging: comparison of image
quality between 2 different acceleration factors at 3.0 Tesla.
Invest Radiol 2006;41
: 391-399[CrossRef][Medline]
- Nael K, Michaely HJ, Villablanca P, Salamon N, Laub G, Finn JP.
Time-resolved contrast enhanced magnetic resonance angiography of the head and
neck at 3.0 tesla: initial results. Invest Radiol2006; 41:116
-124[CrossRef][Medline]
- Chan BK, Desser TS, McDougall IR, Weigel RJ, Jeffrey RB. Common and
uncommon sonographic features of papillary thyroid carcinoma. J
Ultrasound Med 2003; 22:1083
-1090[Abstract/Free Full Text]
- Hoang JK, Lee WK, Lee M, Johnson D, Farrell S. US features of
thyroid malignancy: pearls and pitfalls. RadioGraphics2007; 27:847
-860[Abstract/Free Full Text]
- Shetty SK, Maher MM, Hahn PF, Halpern EF, Aquino SL. Significance
of incidental thyroid lesions detected on CT: correlation among CT,
sonography, and pathology. AJR 2006;187
: 1349-1356[Abstract/Free Full Text]
- Meller J, Becker W. The continuing importance of thyroid
scintigraphy in the era of high-resolution ultrasound. Eur J Nucl
Med Mol Imaging 2002;29
[suppl 2]:S425
-S438[Medline]
- Takasima S, Takayama F, Wang J, Kobayashi S, Kadoya M. Using MR
imaging to predict invasion of the recurrent laryngeal nerve by thyroid
carcinoma. AJR 2003;180
: 837-842[Abstract/Free Full Text]
- Mihailovic J, Stefanovic L, Prvulovic M. Magnetic resonance imaging
in diagnostic algorithm of solitary cold thyroid nodules. J
BUON 2006; 11:341
-346[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
