Download Clinical Investigative Study Detectability of Neural Tracts and Nuclei

Clinical Investigative Study
Detectability of Neural Tracts and Nuclei in the Brainstem
Utilizing 3DAC-PROPELLER
Taro Nishikawa, MD, PhD, Kouichirou Okamoto, MD, PhD, Hitoshi Matsuzawa, MD, PhD, Makoto Terumitsu, DDS,
PhD, Tsutomu Nakada, MD, PhD, Yukihiko Fujii, MD, PhD
From the Department of Neurosurgery, Brain Research Institute, University of Niigata, Niigata, Japan (TN, KO, YF); Center for Integrated Human Brain Science, Brain Research
Institute, University of Niigata, Niigata, Japan (HM, MT, TN, YF).
ABSTRACT
Despite clinical importance of identifying exact anatomical location of neural tracts and
nuclei in the brainstem, no neuroimaging studies have validated the detectability of these
structures. The aim of this study was to assess the detectability of the structures using
three-dimensional anisotropy contrast-periodically rotated overlapping parallel lines with
enhanced reconstruction (3DAC-PROPELLER) imaging. Forty healthy volunteers (21 males,
19 females; 19-53 years, average 23.4 years) participated in this study. 3DAC-PROPELLER
axial images were obtained with a 3T-MR system at four levels of the brainstem: the lower
midbrain, upper and lower pons, and medulla oblongata. Three experts independently
judged whether five tracts (corticospinal tract, medial lemniscus, medial longitudinal
fasciculus, central tegmental and spinothalamic tracts) and 10 nuclei (oculomotor and
trochlear nuclei, spinal trigeminal, abducens, facial, vestibular, hypoglossal, prepositus,
and solitary nuclei, locus ceruleus, superior and inferior olives) on each side could be
identified. In total, 240 assessments were made. The five tracts and eight nuclei were
identified in all the corresponding assessments, whereas the locus ceruleus and superior
olive could not be identified in 3 (1.3%) and 16 (6.7%) assessments, respectively. 3DACPROPELLER seems extremely valuable imaging method for mapping out surgical strategies
for brainstem lesions.
Introduction
Even a tiny lesion in the brainstem is likely to cause significant and critical neurological deterioration because neural tracts
and nuclei as well as reticular formation important in maintaining fundamental brain functions such as consciousness, respiratory regulation, and motor functions are densely packed
within this small part of the brain. Although surgery for brainstem lesions is challenging, advancements in technologies such
as neuroimaging, navigation systems, and intraoperative functional monitoring have made it is possible to be safely perform
brainstem surgery without worsening of neurological symptoms
and signs.1–3 When diagnosing and treating patients with lesions in the brainstem, especially with surgical intervention, it
is quite important to precisely map the anatomical relationships
between the lesions and neural structures such as neural tracts
and nuclei. However, no studies have validated a neuroimaging
method that can detect a number of unspecific neural tracts and
nuclei, including thin tracts and small nuclei, in the brainstem.
Magnetic resonance imaging (MRI) is clearly the best
method for imaging brain structures in clinical settings. Among
various MRI contrast imaging modalities, three-dimensional
anisotropy contrast-periodically rotated overlapping parallel
lines with enhanced reconstruction (3DAC-PROPELLER) on
238
Copyright
Keywords: 3DAC-PROPELLER, MRI,
neural tract, nucleus, brainstem.
Acceptance: Received June 11, 2012,
and in revised form March 4, 2013. Accepted for publication March 11, 2013.
Correspondence: Yukihiko Fujii, MD,
PhD, Department of Neurosurgery, Brain
Research Institute, University of Niigata,
1-757 Asahimachi-dori, Niigata 9518585, Japan. E-mail: [email protected].
ac.jp.
J Neuroimaging 2014;24:238-244.
DOI: 10.1111/jon.12027
a high-field system seems a suitable method for demonstrating
the neural structures, including neural tracts and nuclei, in all
areas of the human brain, especially in the brainstem.4, 5 In this
study we aimed to verify the detectability of neural tracts and
nuclei in the brainstem using 3DAC-PROPELLER imaging on
a 3 Tesla MRI system.
Subjects and Methods
Forty healthy volunteers (21 males and 19 females; 19-53 years,
median 23 years) participated in this study. Informed consent
was obtained from all participants according to the human research guidelines of the Internal Review Board of the University
of Niigata.
A General Electric (Waukesha, WI, USA) Signa 3.0T system
with an eight-channel phased-array head coil was used to perform all the imaging studies. Axial diffusion-weighted images
(DWI) with the PROPELLER method were obtained using the
following parameter settings: FOV 22 cm × 22 cm; matrix
256 × 256; slice thickness 5 mm; intersection gap 2.5 mm;
echo train length 12; TR 4,000 milliseconds; TE 78.7 milliseconds; NEX 5. The b-value was 1,100 seconds/mm2 for each
axis, with the three combinations of diffusion gradient vectors
◦ 2013 by the American Society of Neuroimaging
C
㸫㸫
Table 1. A List of Assessed Neural Tracts in the Brainstem
1. Corticospinal tract
2. Medial lemniscus
3. Medial longitudinal fasciculus
4. Central tegmental tract
5. Spinothalamic tract
3DAC-PROPELLER, three-dimensional anisotropy contrast-periodically rotated
overlapping parallel lines with enhanced reconstruction.
Table 2. A List of Assessed Nuclei in the Brainstem
a. Oculomotor nucleus/trochlear nucleus
b. Spinal trigeminal nucleus
c. Abducens nucleus
d. Facial nucleus
e. Vestibular nucleus
f. Hypoglossal nucleus/prepositus nucleus
g. Solitary nucleus
h. Locus ceruleus
i. Superior olive
j. Inferior olive
3DAC-PROPELLER, three-dimensional anisotropy contrast-periodically rotated
overlapping parallel lines with enhanced reconstruction.
as follows: (1, 0, 0), (0, 1, 0), (0, 0, 1), where the (x, y, z) direction
corresponds to (right to left, anterior to posterior, superior to
inferior) of the brain in the supine position. Considering the
specific absorption rate, the number of slices was limited to 4.
From each participant, DWI images were obtained at four levels of the brainstem: lower midbrain, upper pons, lower pons,
and medulla oblongata. The total scanning time necessary to
obtain images of four slices was approximately 17 minutes.
PROPELLER is an ingenious method of motion correction
found to be useful for eliminating artifacts on rapid-DWI images
by acquiring data in a series of rotating strips (blades) in kspace.6 In 3DAC image processing, three primary colors—red,
green, and blue—are respectively assigned to the gray scale of
the three anisotropic DWI PROPELLER images: x-, y-, and zaxes. These three primary color images are then combined pixel
by pixel to form a single-color image in full-color spectrum. The
final images are displayed negatively to obtain a one-to-one
correlation between each of the three colors (red, green, and
blue) and their respective axes (x, y, and z).7 For details of 3DAC
processing and further mathematical formulations, readers are
referred to references 4, 19, and 20.
Evaluation of Neural Tracts and Nuclei in the Brainstem
3DAC imaging is a type of fiber orientation-weighted imaging.
On axial 3DAC-PROPELLER images, neural fibers running
in the right-left direction (x-axis) appear as red, whereas fibers
running in the anterior-posterior direction (y-axis) and in the direction of the body axis (ie, up-down, z-axis) appear as green and
blue, respectively. Nuclei appear as light colors, nearly white.
Because of their clinical importance, five neural tracts
(Table 1) and 10 cranial nerve nuclei (Table 2) were selected
and evaluated on each side of the brainstem. In several anatomical textbooks and atlases,8–11 three experts (two neurosurgeons
Fig 1. Representative axial 3DAC-PROPELLER image at the level of the lower midbrain. The dense blue region between the
frontopontine tract (12) located anteromedially and the occipitaltemporopontine tract (13) located posterolaterally in the cerebral peduncle is the corticospinal tract (1). The medial lemniscus (2) is located in the tegmentum on the boundary with the cerebral peduncle
and is depicted as a thin blue region adjacent to the substantia nigra (l). The oculomotor/trochlear nuclei (a) can be observed as a
pale and small region medially located on the front edge of the central gray matter (k). The small blue region lateral to the central gray
matter (k) is the mesencephalic tract and nucleus of the trigeminal
nerve (6). The medial longitudinal fasciculus (3) is displayed as a
blue region anterolateral to the oculomotor/trochlear nuclei (a). The
central tegmental tract (4), albeit a main ascending tract of reticular
formation, is observed as a blue area lateral to the medial longitudinal
fasciculus (3). The spinothalamic tract (5) is identified as a blue area
just posterior to the medial lemniscus (2) in the lateral margin of the
tegmentum. The decussation of superior cerebellar peduncles (11) is
also observed as a red structure in the center. In the interpeducular
cistern between the cerebral peduncles the oculomotor nerves (III)
are seen.
and a neuroradiologist) independently identified the five neural tracts, first as blue regions clearly demonstrated on 3DACPROPELLER images; they marked these regions on the images
with arrows. Next, they identified the nuclei as light-colored,
nearly white areas in the vicinity of the neural tracts, which
served as anatomical indices for identifying nuclei. On the basis of the aforementioned anatomical textbooks and atlases,8–11
diagrams showing the four levels of brainstem structures were
also created (Figures 1–4). Detailed mapping of the neural structures on 3DAC-PROPELLER images of each level of the brainstem were described in the figure legends.
Nishikawa et al: Detecting Brainstem Neural Tracts and Nuclei with 3DAC-PROPELLER
㸫㸫
239
Fig 2. Representative axial 3DAC-PROPELLER image at the level
of the upper pons. The corticospinal tract (1) is observed as multilayered dark blue regions among the transverse pontine fibers displayed
in red. The medial lemniscus (2) is shown as a transverse blue region
in the middle of the upper pons. The medial longitudinal fasciculus
(3) is seen as a small blue area on the paramedian floor of the fourth
ventricle. On the lateral side of the fasciculus, the central tegmental
tract (4) is observed as a round dark blue area. In the margin of the
pontine tegmentum, the spinothalamic tract (5) is depicted as a blue
region located adjacent and posterolateral to the medial lemniscus
(2). The locus ceruleus (h) can be observed as a small pale area
anterolateral and adjacent to the central gray matter (k). The raphe
nucleus (n) is shown as a light green longitudinal region on the midline. The superior cerebellar peduncle (8) appears as a light blue
area lateral to the fourth ventricle and central tegmental tract (4). The
middle cerebellar peduncle (9) is observed as a green large region in
the lateral margin of the pontine base. The mesencephalic trigeminal
tract and nucleus (6) are depicted as a small dark blue area posterior
to the locus ceruleus (h) on the boundary between the central gray
matter (k) and superior cerebellar peduncle (8).
The neural tracts and nuclei on a 3DAC-PROPELLER image were classified into one of three groups: (1) excellent, structure could be easily identified without any references; (2) good,
identification required reference to anatomical structures on
the other side or the adjacent images; and (3) poor, identification was difficult or impossible. Tracts and nuclei classified
as excellent and good were regarded as detectable and those
as poor as undetectable. In total, 240 assessments of each neural tract or nucleus were made (ie, 40 subjects × 2 sides ×
3 experts).
These assessments based on 3DAC-PROPELLER images
also estimated the reproducibility and accuracy of the experts’
findings for each structure (neural tract or nucleus), defined as
240
Fig 3. Representative axial 3DAC-PROPELLER image at the level
of the lower pons. The corticospinal tract (1) is depicted as a large
dark blue region between two red compartments (ie, the superficial
layer and the deep layer of the transverse pontine fibers in the middle
of the ventral part of the pons). The medial lemniscus (2) is observed
as a blue transverse structure behind the deep layer of the transverse
pontine fibers. The medial longitudinal fasciculi (3) are depicted in the
form of inverted V on the midline of the floor of the fourth ventricle.
The central tegmental tract (4) is shown a blue region in the reticular
formation located behind the medial lemniscus (2). The spinothalamic tract (5) is displayed as a light blue region lateral to the medial
lemniscus (2). The abducens nucleus (c) is shown as a pale oval region anterolateral to the medial longitudinal fasciculus (3). The spinal
trigeminal nucleus nerve (b) is shown as a pale region adjacent to
the spinal trigeminal tract (7) depicted in blue. The vestibular nucleus
(e) is observed as pale region lateral to the lateral wall of the fourth
ventricle, which is in contact with the inner surface of the inferior cerebellar peduncle (10). The superior olive (i) is displayed as a tiny pale
area anterolateral to the central tegmental tract and behind the medial
lemniscus (2). The facial nucleus (d) is a pale round area posterolateral to the superior olive (i). The raphe nucleus (n) is observed as
a green region on the midline. The cochlear nucleus (m) composed
of anterior and posterior parts (ie, ventral and dorsal nuclei) is depicted as a pale region lateral to the inferior cerebellar peduncle (10).
The abducens nerve (VI) emerges from the anterior aspect of the
pons and is depicted in a greenish color. The facial nerves (VII) and
vestibulocochlear nerve (VIII) run through the brainstem and appear
from the lateral side of the pons and are depicted in red.
the concordance percentage among assessments (detectable or
undetectable) of the three experts and the percentage of truepositives and true-negatives, respectively. In this study, truepositives, true-negatives, false-positives, and false-negatives corresponded to detectable assessment of the structure of interest;
detectable assessment of the structure closest to the structure
of interest, which was the closest neural tract if the structure
of interest was a neural tract or the closest nucleus if it was a
Journal of Neuroimaging Vol 24 No 3 May/June 2014
㸫㸫
shown in Figures 1–4, respectively. Numbers, letters, and abbreviations representing anatomical structures in the figures are
listed in Tables 1 and 2.
Detectability of Neural Tracts and Nuclei
The five neural tracts, displayed as blue structures, were all
judged as detectable (ie, excellent or good) on all four slices in
all assessments (Table 3). In all five tracts, the reproducibility
and accuracy were also 100%, ie, none of the five neural tracts
were difficult to identify (ie, poor) and none were misinterpreted
as another neural tract (no false-positive).
Among the 10 nuclei, displayed as light color areas, eight
were judged as detectable in all assessments (Table 4). The remaining two, the locus ceruleus and superior olive, were judged
as difficult to identify in 3 (1.3%) and 16 (6.7%) assessments, respectively. For the majority of the nuclei, reproducibility and
accuracy were 100% (Table 4).
Discussion
Fig 4. Representative axial 3DAC-PROPELLER image at the level
of the medulla oblongata. The corticospinal tract (1) is shown as
a dark blue area in the anterior margin of the medulla oblongata.
The inferior olive (j) is displayed as a green oval region posterior to
the tract. Between both the inferior olives (j), the medial lemniscus
(2) is shown as a paramedian and longitudinal blue structure, behind which the medial longitudinal fasciculus (3) is also shown as
a paramedian blue region adjacent to the floor of fourth ventricle.
The central tegmental tract (4) is displayed as a dark blue and round
structure abutting the dorsal side of the inferior olive (j), behind which
the spinothalamic tract (5) is depicted as a dark blue area lateral to
the central tegmental tract (4) and on the margin of the retro-olivary
groove. Posterior to the groove, the spinal trigeminal tract (7) and
nucleus (b) are shown as blue and pale areas, respectively. The
pale small region observed lateral to the medial longitudinal fasciculus (3) is the hypoglossal nucleus/prepositus nucleus (f), behind
which the dorsal longitudinal fasciculus (14) is observed as a small
blue region beneath the fourth ventricle floor. The vestibular nuclei
(e) composed of medial vestibular nucleus and inferior vestibular nucleus are observed as pale regions beneath the floor of the fourth
ventricle and medial to the inferior cerebellar peduncle (10). On the
ventral side of the nuclei, the solitary nucleus (g) around the solitary tract is displayed as a pale small area. The tectospinal tract
(15), which connects the quadrigeminal body and the spinal cord, is
shown as the blue region between the medial lemniscus (2) and the
medial longitudinal fasciculus (3). On the lateral margin of the inferior cerebellar peduncle (10), the cochlear nucleus (m) is depicted
as a light-colored region. Outside the medulla, the glossopharyngeal
nerve (IX) and vestibulocochlear nerve (VIII) are shown as red streak
structures.
nucleus; misinterpretation of the structure of interest as its closest structure; and undetectable assessment of the structure of
interest, respectively.
Results
Representative axial 3DAC-PROPELLER images of the lower
midbrain, upper pons, lower pons, and medulla oblongata are
This study utilizing 3DAC-PROPELLER images4, 5 revealed
that the five neural tracts assessed were detectable at all four
levels of the brainstem in all subjects and the 10 nuclei assessed
were detectable in most subjects, although the locus ceruleus
and superior olive were undetectable in a few subjects (Table 4).
Hence, 3DAC-PROPELLER imaging is capable of detecting
a significant number of the nuclei and neural tracts in the
brainstem.
3DAC-PROPELLER
On conventional MRI images, the majority of neural tracts and
nuclei in the brainstem are difficult to detect, although the locus
ceruleus can be imaged with a specialized high-resolution T1TSE MR sequence as a small area of elevated signal intensity
adjacent to the fourth ventricle.12 In various pathological conditions or diseases, the degenerated neural tracts or nuclei can
be depicted as areas of high signal intensity on conventional
T2WI images.13–16 However, the depiction of these anatomical
structures in the brainstem is limited to specific pathological
conditions.
The contrast based on water diffusion in MRI imaging may
be capable of depicting fine structures in the brainstem. The
anisotropy of the diffusion of water molecules in human brains,
which is noninvasively measurable using MRI, arises almost
exclusively from neural fibers, principally axons. There are
many applications for neuroimaging methods based on diffusion anisotropy, known as diffusion tensor imaging (DTI),
including 3DAC imaging, fiber tracking (ie, tractography), and
fractional anisotropy imaging.17, 18 The main drawback of DTIs
other than 3DAC imaging is the need to use algorithms to
calculate eigenvalues and to determine eigenvectors based on
these numerical estimations, which lead to potential degradation of image quality and difficulty in detecting fine neural tracts
and nuclei in the brainstem.7 The quality of the reconstructed
images—based on estimated parameter values such as eigenvalues of the apparent diffusion tensor—has been disappointing.19
This can be understood intuitively by considering the difference between T1 images (ie, images reconstructed on the basis
Nishikawa et al: Detecting Brainstem Neural Tracts and Nuclei with 3DAC-PROPELLER
㸫㸫
241
Table 3. Detectability of the Five Neural Tracts Assessed in the Brainstem
Detection, n (%)
Yes
Neural Tracts (Label in Figures)
Lower midbrain
Corticospinal tract (1)
Medial lemniscus (2)
Medial longitudinal fasciculus (3)
Central tegmental tract (4)
Spinothalamic tract (5)
Upper pons
Corticospinal tract (1)
Medial lemniscus (2)
Medial longitudinal fasciculus (3)
Central tegmental tract (4)
Spinothalamic tract (5)
Lower pons
Corticospinal tract (1)
Medial lemniscus (2)
Medial longitudinal fasciculus (3)
Central tegmental tract (4)
Spinothalamic tract (5)
Medulla oblongata
Corticospinal tract (1)
Medial lemniscus (2)
Medial longitudinal fasciculus (3)
Central tegmental tract (4)
Spinothalamic tract (5)
No
Excellent
Good
Poor
Overall
Detectability,
n (%)
240 (100)
240 (100)
240 (100)
237 (98.7)
240 (100)
0 (.0)
0 (.0)
0 (.0)
3 (1.3)
0 (.0)
0 (.0)
0 (.0)
0 (.0)
0 (.0)
0 (.0)
240 (100)
240 (100)
240 (100)
240 (100)
240 (100)
100
100
100
100
100
100
100
100
100
100
240 (100)
240 (100)
240 (100)
237 (98.7)
240 (100)
0 (.0)
0 (.0)
0 (.0)
3 (1.3)
0 (.0)
0 (.0)
0 (.0)
0 (.0)
0 (.0)
0 (.0)
240 (100)
240 (100)
240 (100)
240 (100)
240 (100)
100
100
100
100
100
100
100
100
100
100
240 (100)
240 (100)
240 (100)
224 (93.3)
221 (87.5)
0 (.0)
0 (.0)
0 (.0)
16 (6.7)
19 (8.1)
0 (.0)
0 (.0)
0 (.0)
0 (.0)
0 (.0)
240 (100)
240 (100)
240 (100)
240 (100)
240 (100)
100
100
100
100
100
100
100
100
100
100
240 (100)
240 (100)
240 (100)
240 (100)
240 (100)
0 (.0)
0 (.0)
0 (.0)
0 (.0)
0 (.0)
0 (.0)
0 (.0)
0 (.0)
0 (.0)
0 (.0)
240 (100)
240 (100)
240 (100)
240 (100)
240 (100)
100
100
100
100
100
100
100
100
100
100
Reproducibility, %
Accuracy, %
Table 4. Detectability of the 10 Nuclei Assessed in the Brainstem
Detection, n (%)
Excellent
Good
Poor
Overall
Detectability,
n (%)
233 (97.1)
7 (2.9)
0 (.0)
240 (100)
Yes
Nuclei (Label in Figures)
Lower midbrain
Oculomotor
Nucleus/trochlear nucleus (a)
Upper pons
Locus ceruleus (h)
Lower pons
Spinal trigeminal nucleus (b)
Abducens nucleus (c)
Facial nucleus (d)
Vestibular nucleus (e)
Superior olive (i)
Medulla oblongata
Spinal trigeminal nucleus (b)
Vestibular nucleus (e)
Solitary nucleus (g)
Hypoglossal nucleus/prepositus nucleus (f)
Inferior olive (j)
No
Reproducibility, %
Accuracy, %
100
NA
225 (93.7)
12 (5.0)
3 (1.3)
237 (98.7)
100
NA
224 (93.3)
240 (100)
227 (94.6)
240 (100)
211 (87.9)
16 (6.7)
0 (.0)
13 (5.4)
0 (.0)
13 (5.4)
0 (.0)
0 (.0)
0 (.0)
0 (.0)
16 (6.7)
240 (100)
240 (100)
240 (100)
240 (100)
224 (93.3)
100
100
100
100
98.8
100
100
96.7
100
96.7
224 (93.3)
240 (100)
224 (93.3)
240 (100)
240 (100)
16 (6.7)
0 (.0)
16 (6.7)
0 (.0)
0 (.0)
0 (.0)
0 (.0)
0 (.0)
0 (.0)
0 (.0)
240 (100)
240 (100)
240 (100)
240 (100)
240 (100)
100
100
100
100
100
100
100
100
100
100
NA = not available (no nucleus to compare with the nucleus in this study).
of estimated T1 values) and T1-weighted images (ie, images
obtained using a spin-echo sequence without numerical estimations). However, 3DAC is an algorithm introduced to avoid
the difficulties of performing full tensor analysis.7, 19, 20 In the
3DAC algorithm, the mathematical process is performed in im-
242
age form, effectively preventing unwanted degradation of image quality. Consequently, 3DAC images have exquisitely high
anatomical resolution sensitive to physiological perturbations,
resulting in the potential capability of identifying thin neural
tracts and nuclei.7, 19, 20
Journal of Neuroimaging Vol 24 No 3 May/June 2014
㸫㸫
When trying to detect microstructures in the brainstem,
higher signal-to-noise ratio (S/N) obtained using high-field MRI
systems is undoubtedly advantageous, but the higher susceptibility effect remains a significant trade-off. The high susceptibility effect is likely to produce a susceptibility artifact resulting in
degradation of image quality, especially on echo-planar imaging (EPI), which is usually used to obtain DTI. EPI is inherently
subject to susceptibility artifacts, particularly at the skull base
or brainstem region. Accordingly, several methods have been
proposed to reduce or correct image distortion caused by the
susceptibility effect.17, 18, 21, 22
Among them, the PROPELLER sequence is a suitable
method to effectively eliminate these undesired artifacts. Because PROPELLER is a spin-echo sequence, a motion probing
gradient can be readily implemented for DWI and can eliminate the susceptibility artifacts inherent in DWI.23 Therefore,
PROPELLER is especially effective for application to DWI
with high-field systems. Hence, a combination of 3DAC and
PROPELLER with a 3 Tesla MRI system seems an optimal
imaging method in the clinical setting for depicting anatomical
microstructures such as neural tracts and nuclei in the whole
brainstem.4
Detectability of Neural Tracts and Nuclei in the Brainstem
In this study utilizing 3DAC-PROPELLER images, the five
neural tracts were clearly visualized as blue regions because of
strong diffusion anisotropy in the direction of the body axis
(z-axis). The 10 nuclei were well depicted as light-colored structures because of both weak diffusion anisotropy and the existence of neural tracts clearly identified in their vicinity, which
serve as anatomical indices for identifying nuclei. The oculomotor nucleus/trochlear nucleus, spinal trigeminal nucleus, abducens nucleus, facial nucleus, vestibular nucleus, hypoglossal
nucleus/prepositus nucleus, solitary nucleus, and inferior olive
could be identified in all subjects. In several subjects, however,
the locus ceruleus and superior olive were difficult to detect because they are smaller than the other nuclei assessed and their
signal contrast decreases because of the partial volume effect
in thick slices. Indeed, thinner slices seem preferable for identifying smaller neural structures because they reduce the partial
volume effect. That leads, however, to decreased signal-to-noise
ratio, resulting in reduced signal contrast or longer examination
time to allow for more NEX to avoid loss of contrast. Furthermore, the locus ceruleus and superior olive are poorly demarcated nuclei intermingled with numerous scattered neurons in
the reticular formation, leading to the difficulty in identifying
them as clusters of neurons (ie, nuclei).24, 25
Reproducibility and Accuracy
Reproducibility and accuracy were 100% for the majority
of neural structures of the brainstem assessed in this study—
especially for all the five neural tracts, even the medial longitudinal fasciculus and the central tegmental tract (Tables 3 and 4).
These two are small tracts running very close together in a similar direction. Their high reproducibility and accuracy seems
to be plausibly explained by the fact that 3DAC-PROPELLER
images can clearly demonstrate a distinction between the two
tracts: the former appears as oval, blue structures running in the
Fig 5. Axial 3DAC-PROPELLER images of four levels of the brainstem in three subjects demonstrate the reproducibility of assessments of the neural tracts and nuclei.
paramedian region and close to the forth ventricle and the latter
as round, blue structures running laterally in the deep portion
of the brainstem (Figures 1–5).
Clinical Application
Based on normal anatomical structures of the brainstem, several
safe entry zones, where incision and retraction of the brainstem
could be safely made without significant neurological deficits
due to the absence of important neural tracts and nuclei, have
been proposed.26, 27 However, it may be difficult to apply these
safe entry zones to patients with large intra-axial mass lesions in
the brainstem because of the displacement of neural tracts and
nuclei. For these patients, it seems feasible to identify neural
tracts and nuclei directly and precisely and to determine real
safe entry zones using 3DAC-PROPELLER preoperatively.
Conclusion
Using 3DAC-PROPELLER images, five neural tracts in the
brainstem were identified in all subjects and 10 nuclei were
identified in most subjects. According to our results, 3DACPROPELLER imaging is helpful for identifying both neural
tracts and nuclei in the brainstem, which seems to be an extremely valuable imaging method for mapping out surgical
strategies for brainstem lesions.
References
1. Bertalanffy H, Gilsbach JM, Eggert HR, et al. Microsurgery of
deep-seated cavernous angiomas: report of 26 cases. Acta Neurochir
(Wien) 1991;108:91-99.
Nishikawa et al: Detecting Brainstem Neural Tracts and Nuclei with 3DAC-PROPELLER
㸫㸫
243
2. Wang C, Zhang J, Liu A, et al. Surgical management of medullary
hemangioblastoma: report of 47 cases. Surg Neurol 2001;56:218226.
3. Wang C, Zhang J, Liu A, et al. Surgical treatment of primary midbrain gliomas. Surg Neurol 2000;53:41-51.
4. Nakada T, Matsuzawa H, Fujii Y, et al. Three-dimensional
anisotropy contrast periodically rotated overlapping parallel lines
with enhanced reconstruction (3DAC PROPELLER) on a 3.0T
system: a new modality for routine clinical neuroimaging. J Neuroimaging 2006;16:206-211.
5. Terajima K, Matsuzawa H, Shimohata T, et al. Tract-by-tract morphometric and diffusivity analyses in vivo of spinocerebellar degeneration. J Neuroimaging 2009;19:220-226.
6. Pipe JG. Motion correction with PROPELLER MRI: application
to head motion and free breathing cardia imaging. Magn Reson Med
1999;103:963-969.
7. Nakada T, Nakayama N, Fujii Y, et al. Clinical application of magnetic resonance axonography. J Neurosurg 1999;90:791-795.
8. Martin JH. Atlas II: myelin-stained sections through the central
nervous system. In: Martin JH, ed. Neuroanatomy, Text and Atlas.
New York: McGraw-Hill; 1989:425-488.
9. Paxinos G, Huang XF. Atlas of the Human Brainstem. Waltham, MA:
Academic Press; 1995.
10. Haines DE. Neuroanatomy. 7th ed. Philadelphia, PA: Lippincott
Williams & Wilkins; 2007.
11. Naidich TP, Duvernoy HM, Delman BN, et al. Duvernoy’s Atlas of
the Human Brain Stem and Cerebellum. Berlin, Germany: Springer;
2008.
12. Keren NI, Lozar CT, Harris KC, et al. In vivo mapping of the
human locus coeruleus. NeuroImage 2009;47:1261-1267.
13. Kuhn MJ, Johnson KA, Davis KR. Wallerian degeneration: evaluation with MR imaging. Radiology 1988;168:199-202.
14. Kitajima M, Korogi Y, Shimomura O, et al. Hypertrophic olivary degeneration: MR imaging and pathologic findings. Radiology
1994;192:539-543.
244
15. Okamoto K, Furusawa T, Ishikawa K, et al. Focal T2 hyperintensity
in the dorsal brain stem in patients with vestibular schwanoma.
AJNR Am J Neuroradiol 2006;27:1307-1311.
16. Yagishita A, Oda M. Progressive supranuclear palsy: MRI and
pathological findings. Neuroradiology 1996;38(Supple 1):S60S66.
17. Finsterbusch J. High-resolution diffusion tensor imaging with inner
field-of-view EPI. J Magn Reson Imaging 2009;29:987-993.
18. Nagae-Poetscher LM, Jiang H, Wakana S, et al. High-resolution
diffusion tensor imaging of the brain stem at 3 T. AJNR Am J Neuroradiol 2004;25:1325-1330.
19. Nakada T, Matsuzawa H. Three-dimensional anisotropy magnetic resonance imaging of the rat nervous system. Neurosci Res
1995;22:389-398.
20. Nakada T, Matsuzawa H, Kwee IL. Magnetic resonance axonography of the rat spinal cord. NeuroReport 1994;5:2053-2056.
21. Mangin JF, Poupon C, Clark C, et al. Distortion correction and
robust tensor estimation for MR diffusion imaging. Med Image Anal
2002;6:191-198.
22. Heidemann RM, Anwander A, Feiweier T, et al. k-space and qspace: combining ultra-high spatial and angular resolution in diffusion imaging using ZOOPPA at 7T. NeuroImage 2012;967-978.
23. Pipe JG. Motion correction with PROPELLER MRI: application
to head motion and free breathing cardiac imaging. Magn Reson Med
1999;103:963-969.
24. Brodal A. Neurological Anatomy in Relation to Clinical Medicine. 3rd
ed. New York: Oxford University Press; 1981:394-447.
25. Wang D. Reticular formation and spinal cord injury. Spinal Cord
2009;47:204-212.
26. Cantore G, Missori P, Santoro A. Cavernous angiomas of the brain
stem: intra-axial anatomical pitfalls and surgical strategies. Surg
Neurol 1999;52:84-93.
27. Kyoshima K, Kobayashi S, Gibo H, et al. A study of safe entry
zones via the floor of the fourth ventricle for brain-stem lesions:
report of three cases. J Neurosurg 1993;78:987-993.
Journal of Neuroimaging Vol 24 No 3 May/June 2014
㸫㸫