-
Neurosurg Focus / Volume 32 / May 2012
Neurosurg Focus 32 (5):E6, 2012
1
Vascular lesions of the spine are rare and difficult to diagnose
by means of noninvasive imaging modalities.2,6,7,26 Spinal dural
arteriovenous fis-tulas are the most common type of SVM; patients
with SDAVFs usually present with progressive myelopathy and
weakness,21,25 and the lesions require prompt diag-nosis and
treatment.3 Spinal AVMs are less common than SDAVF, and the
classification of these lesions has been a
subject of considerable debate.17,29 Aneurysms also occur in the
spinal arterial circulation and are commonly as-sociated with
fistulas. Spinal aneurysm rupture may lead to acute neurological
deficits. Over the last 2 decades, imaging techniques have evolved
to better visualize these heterogeneous lesions that may occur in
the spinal vas-culature.
Multiple case series8,23,28 and meta-analysis27 have underscored
the utility and benefits of microsurgical treatment for spinal
vascular lesions. Precise localiza-tion and detailed knowledge of
the vascular architecture is essential for preoperative planning.
Multiple imaging modalities have been used for the diagnosis of
vascular malformation of the spine including CTA, MRI, and
cath-eter-based DSA. Although traditional DSA is the most sensitive
methodology for detecting SVMs, it is limited
Rotational angiography for diagnosis and surgical planning in
the management of spinal vascular lesions
*AlexAnder e. ropper, M.d., ning lin, M.d., BrAdley A. gross,
M.d., HekMAt k. ZArZour, M.d., rutH tHiex, M.d., pH.d., JoHn H.
CHi, M.d., M.p.H., rose du, M.d., pH.d., And kAi u. FreriCHs,
M.d.Department of Neurosurgery, Brigham and Women’s Hospital,
Harvard Medical School, Boston, Massachusetts
Object. The management of spinal vascular malformations has
undergone significant evolution with the advent of advanced
endovascular and angiographic technology. Three-dimensional
rotational spinal angiography is an ad-vanced tool that allows the
surgeon to gain a better appreciation of the anatomy of these
spinal vascular lesions and their relation to surrounding
structures. This article describes the use of rotational
angiography and 3D reconstruc-tions in the diagnosis and management
of spinal vascular malformations.
Methods. The authors present representative cases involving
surgical treatment planning for spinal vascular mal-formations with
focus on the utility and technique of rotational spinal
angiography. They report the use of rotational spinal angiography
for a heterogeneous collection of vascular pathological
conditions.
Results. Eight patients underwent rotational spinal angiography
in addition to digital subtraction angiography (DSA) for the
diagnosis and characterization of various spinal vascular lesions.
Postprocessed images were used to characterize the lesion in
relation to surrounding bone and to enhance the surgeon’s ability
to precisely localize and obliterate the abnormality. The
reconstructions provided superior anatomical detail compared with
traditional DSA. No associated complications from the rotational
angiography were noted, and there was no statistically significant
difference in the amount of radiation exposure to patients
undergoing rotational angiography relative to traditional
angiography.
Conclusions. The use of rotational spinal angiography provides a
rapid and powerful diagnostic tool, superior to conventional DSA in
the diagnosis and preoperative planning of a variety of spinal
vascular pathology. A more detailed understanding of the anatomy of
such lesions provided by this technique may improve the safety of
the surgical
approach.(http://thejns.org/doi/abs/10.3171/2012.1.FOCUS11254)
key Words • spinal angiography •
spinal artery aneurysm •
spinal dural arteriovenous fistula •
3D rotational angiography
1
Abbreviations used in this paper: AP = anteroposterior; ASA =
anterior spinal artery; AVM = arteriovenous malformation; CTA = CT
angiography; DAP = dose-area product; DSA = digital subtrac-tion
angiography; MIP = maximum intensity projection; PSA = posterior
spinal artery; RA = rotational angiography; SAH = sub-arachnoid
hemorrhage; SDAVF = spinal dural arteriovenous fistula; SVM =
spinal vascular malformation.
* Drs. Ropper and Lin contributed equally to this work.
Unauthenticated | Downloaded 06/04/21 05:05 AM UTC
-
A. E. Ropper et al.
2 Neurosurg Focus / Volume 32 / May 2012
by its planar nature, and anatomical correlation with the
surrounding soft tissue and bony anatomy can be diffi-cult. The
combination of high-resolution vascular images from selective
transarterial DSA with the tomographic images of a cross-sectional
modality such as CT or MRI provides greater anatomical detail for
localization of vas-cular pathology and for preoperative planning
of either endovascular or surgical treatments. The use of 3D RA has
become routine in the management of intracranial aneurysms.10,11
The utility of this technique in the diagno-sis and treatment
planning of SVMs is not nearly as well established.14,19,24
We report a heterogeneous collection of cases of SVMs involving
patients who underwent 3D RA with de-tailed postprocessing of the
rotational dataset, including standard and optimized tomographic
maximum inten-sity projections (MIPs) and 3D volume rendering prior
to operative obliteration or radiosurgery. Postprocessed images
from the 3D RA provided superior diagnostic information compared
with traditional DSA in operative planning for SVMs by precise
localization of the lesion in relation to the surrounding bony
anatomy and should be considered part of the standard angiographic
work-up for these lesions.
MethodsThe study population consisted of all patients who
underwent diagnostic spinal angiography at the Brigham and
Women’s Hospital between 2007 and 2010. Angio-grams performed for
the purpose of targeted intervention (for example, preoperative
tumor embolization) or post-operative evaluations were excluded
from the study. All medical records were reviewed to extract
demographic information, clinical symptoms, imaging findings,
an-giography parameters, and information about hospital courses.
The study was approved by the Brigham and Women’s Hospital and
Partners Healthcare Institutional Review Board.
All angiographic procedures were performed on the General
Electric Innova 3131 biplane fluoroscopy system (GE Healthcare).
Patients received intravenous conscious sedation and local
anesthesia prior to the procedure. All angiograms were performed
via 5-Fr transfemoral sheath access. A 4-Fr Berenstein II catheter
was used to access the internal iliac, vertebral, and carotid
arteries as well as the thyro- and costocervical trunks, and a 5-Fr
Mickelson catheter was used to access the thoracolumbar segmental
and middle sacral arteries. The rotational angiogram was configured
to carry out a 200° spin from the AP x-ray projector with one of 3
preset rotation speeds (40°/second, 20°/second, or 10°/second),
depending on the amount of soft tissue and bony information desired
for the recon-structed images. Contrast medium (Ultravist 240) was
injected at a rate of 1–2 ml/second for a total of 10–20 ml during
each rotational angiogram depending on the duration of the
rotation. Postprocessing analyses were completed on an AW
workstation (Innova 3D software, GE Healthcare) to reconstruct 3D
vascular models and produce adjustable maximum intensity projection
(MIP) images of various thicknesses. Images for postprocessing
were available in near real-time (with a delay of only 60–90
seconds) following acquisition, and postprocessing could therefore
be carried out with the catheter remain-ing in the selected vessel
to ensure that the quality of the acquired dataset was adequate or
if further manipulation was required.
Radiation exposure was evaluated with cumula-tive air kerma (in
Gy) and dose-area product (DAP, in Gy⋅cm2), both of which were
obtained directly from the angiography station. The cumulative air
kerma (or cumu-lative dose) was measured 15 cm below the isocenter
of the fluoroscopy tubing, and the DAP was calculated as the
integral of air kerma across the x-ray beam emission. These
radiation exposure parameters were routinely re-corded for all
neuroangiography procedures.
Differences in demographic and clinical character-istics for
patients who received 3D RA and traditional spinal angiography were
examined using chi-square and 2-tailed t-tests for categorical and
continuous variables, respectively. Statistical significance was
defined as a probability of Type I error less than 0.05. All
statistical analyses were performed using SAS version 9.2 (SAS
In-stitute, Inc.) and Excel 2007 (Microsoft Corporation).
ResultsBetween 2007 and 2010, 37 patients underwent diag-
nostic spinal angiography at the Brigham and Women’s Hospital;
of these, 3D RA was performed in 8 patients. Demographic and
clinical characteristics are summa-rized in Table 1. The average
age of patients who under-went 3D RA was 46.5 years, and 5 of these
8 patients
TABLE 1: Demographic and clinical information for patients who
underwent spinal angiography*
Variable2D DSA
(29 patients)3D RA
(8 patients) p Value
age at procedure (yrs) 0.63 mean 52.5 ± 15.6 46.5 ± 17.4 range
25–79 20–65sex 0.71 male 16 5 female 13 3diagnosis 0.001 negative
for SVM 23 0 SDAVF 5 3 spinal AVM 1 3 aneurysm 0 2radiation dosage
(Gy) 0.61 mean 2.96 ± 1.94 3.44 ± 2.45 range 0.72–6.43 0.55–6.75DAP
(Gy⋅cm2) 0.66 mean 372.7 ± 267.8 409.5 ± 318.6 range 65.9–808.0
60.9–932.0
* Values represent numbers of patients unless otherwise
indicated. Mean values are presented with SDs.
Unauthenticated | Downloaded 06/04/21 05:05 AM UTC
-
Neurosurg Focus / Volume 32 / May 2012
Rotational spinal angiography
3
were male. Although the mean cumulative dose and DAP were higher
for patients who were examined with 3D RA than those examined with
2D traditional DSA, the differ-ence was not statistically
significant (p = 0.61).
Table 2 summarizes the clinical and specific imaging findings in
the 8 patients who underwent 3D RA. Cases 1 and 3, who had
intradural aneurysms (one of which was associated with an SDAVF),
and Case 8, who had a cer-vical intramedullary AVM, presented with
spontaneous intradural hemorrhage and acute onset of neurological
deficits, whereas the patients with Type I SDAVFs and a pial AVM
presented with progressive neurological defi-cits. Case 6 presented
with radiographic recanalization of an intramedullary AVM that was
previously embolized. Initial MRI and/or CTA demonstrated stigmata
of vascu-lar pathology, and all patients underwent spinal
angiogra-phy prior to operative or radiosurgical management. No
intra- or periprocedural complications were encountered in those
undergoing 3D RA.
Illustrative CasesCase 1
This 52-year-old woman presented with sudden onset of severe
headache, right-sided neck spasm, upper-extrem-ity paresthesias,
and episodic vertigo after chiropractic manipulation of her neck.
Physical examination revealed minimal left biceps weakness, and a
noncontrast head CT scan demonstrated evidence of hydrocephalus
with dilated temporal horns and SAH in the basal cisterns (Fig.
1A). Magnetic resonance imaging revealed SAH in the anterior
cervical canal, extending down to the level of C-7 (Fig. 1B).
Standard cerebral and cervical spinal angiograms demonstrated an
arteriovenous fistula supplied by the ASA and musculoskeletal
branches of the left vertebral artery, draining superiorly into the
anterior median spinal vein
(Fig. 1C). The fistula was associated with a small aneu-rysm.
Incidentally, the ASA had an aberrant supply from the costocervical
trunk. Three-dimensional RA was per-formed from the right
costocervical trunk, which showed the dural AVF to be located on
the left anterior surface of the cervical spinal cord associated
with a 3-mm ASA aneurysm located between the levels of C-4 and C-5
(Fig. 1D–F, rotational angiogram slow spin of 20°/second). The
ability to freely tumble the 3D model allowed analysis of the
anatomy in any projection (Fig. 1G, rotational angio-gram fast spin
of 40°/second). The patient underwent C3–6 laminectomies for
resection of the fistula and clipping of the aneurysm. The patient
recovered well from surgery, and her initial mild proximal left
upper extremity weak-ness had resolved at 6-month follow-up.
Case 2This 58-year-old man presented with progressively
worsening paresthesias and weakness in his legs over 6 weeks. He
also reported occasional urinary retention and constipation.
Neurological examination demonstrated bi-lateral ankle clonus but
no strength or sensory deficits. An MRI study of the spine revealed
T2 prolongation and gadolinium enhancement from T-9 to the conus
medul-laris, consistent with chronic venous congestion, and
mul-tiple flow voids in thoracic and lumbar spine (Fig. 2A and B).
Spinal DSA demonstrated an SDAVF fed by the radic-ulomedullary
branches of the left T-12, bilateral L-1, and left L-2 segmental
arteries and draining superiorly into a perimedullary vein as well
as a paraspinal vein (Fig. 2C and D). Three-dimensional RA was
performed from the left L-1 and left T-12 segmental arteries and
showed that the fistula was directly behind the L-1 vertebral body,
anterior to the spinal cord, and just medial to the left L-1
pedicle (Fig. 2E–G, slow spin of 20°/second; Fig. 2H, fast spin of
40°/second). The arteriovenous transition could be
TABLE 2: Demographic and clinical summary for patients who
underwent 3D rotational spinal angiography*
Case No.
Age (yrs), Sex Symptoms
Noninvasive Imaging Diagnosis (levels) Arterial Supply
Management
FU (mos)
mRS at FU
1 52, F headache, neck pain, paresthesias, lt arm weakness
MRI, CTA SDAVF (C4–5), ASA aneurysm
ASA, lt VA 3D RA, surgical obliteration
15 1
2 58, M paresthesias MRI SDAVF (T12–L2) lt T-12, bilat L-1,
& lt L-2 segmental artery
3D RA, surgical obliteration
2 0
3 42, M back pain, lt leg weakness & numb- ness
MRI PSA aneurysm (T-11)
lt L-1 segmental artery 3D RA, surgical obliteration
3 1
4 61, M urinary retention, myelopathy MRI pial AVM (T-6) rt T-6
segmental artery 3D RA, surgical obliteration
11 1
5 65, F myelopathy, urinary retention MRI SDAVF (T-10) rt T-10
segmental artery 3D RA, surgical obliteration
2 1
6 20, M paraplegia, radiographic residual spinal AVM after
embolization
MRI residual intramedul- lary AVM (T-8)
lt T-9 segmental artery 3D RA, radiosur- gery
6 5
7 53, F paresthesias MRI SDAVF (L-1) rt L-1 segmental artery 3D
RA, surgical obliteration
1 0
8 21, M hemiplegia, paresthesias MRI intramedullary AVM
(C2–3)
ASA 3D RA, radiosur- gery
5 0
*
FU = follow-up; mRS = modified Rankin Scale score; VA = vertebral artery.
Unauthenticated | Downloaded 06/04/21 05:05 AM UTC
-
A. E. Ropper et al.
4 Neurosurg Focus / Volume 32 / May 2012
traced through the thin-slice (2-mm) tomographic MIP images from
coronal reconstructions of the 3D RA (Fig. 3A–H, slow spin of
20°/second). The reconstructed im-ages enhanced the view of the
fistulous point in multiple projections. The patient underwent
T12–L2 laminecto-mies and resection of the SDAVF with clipping of
the feeding artery (Fig. 2I). He recovered well from the op-eration
with no neurological deficits and resolution of the preoperative
paresthesias.
Case 3This 42-year-old man developed acute onset of low
back pain and left leg numbness and weakness, which persisted
for 1 day. His left leg weakness rapidly wors-ened and he was
unable to ambulate. Physical examina-tion revealed a plegic left
leg, distal right leg weakness, a left T-11 sensory level, and
absent rectal tone. Magnetic resonance imaging showed an acute
intradural hem-orrhage in the thoracic and lumbar spine without
clear evidence of a vascular abnormality (Fig. 4A). Spinal
an-giogram with 3D RA demonstrated a fusiform aneurysm of the left
posterior spinal artery at level of T-11, fed by the left L-1
segmental artery (Fig. 4B). Tomographic MIP reconstructions from
the 3D RA allowed tracing of the
feeding artery from the anterior canal to the posterior ca-nal,
demonstrating that the aneurysm was fed by the pos-terior spinal
artery (Fig. 4C–F, slow spin of 20°/second). He underwent urgent
T10–L1 decompressive laminecto-mies and obliteration of the left
posterior spinal artery aneurysm. The patient made a complete
recovery after surgery, was able to ambulate without difficulty,
and had normal bowel and bladder functions.
Case 4This 61-year-old man presented to another institution
with urinary retention and increased tone in both legs that
limited his gait. He underwent L3–5 laminectomies for presumed
lumbar stenosis, but his symptoms did not improve following
surgery. At presentation to our institu-tion, 10 days after his
laminectomies, he had preserved strength in his lower extremities
but bilateral ankle clo-nus and increased tone in both legs (to a
greater extent on the right). There was numbness over the dorsal
and ventral aspects of both feet, but it was more pronounced on the
right. An MRI study showed increased T2 signal in the lower
thoracic cord suggestive of venous conges-tion, and a formal
angiogram was performed. The right T-6 segmental artery injection
demonstrated arteriove-
Fig. 1. Case 1. A 52-year-old woman with headache and neck pain.
She had a C4–5 SDAVF and an ASA aneurysm at that lev-el, which was
successfully treated with surgical clipping. A: Noncontrast head CT
scan obtained at admission, demonstrating acute SAH in the basal
cisterns and evidence of hydrocephalus with dilated temporal horns.
Subarachnoid blood is also present in both sylvian fissures. B:
Sagittal T1-weighted MR image demonstrating SAH in the anterior
cervical canal (arrowhead). C: A DS angiogram, AP view, showing a
fistula at C4–5 supplied by the ASA, which has an aberrant supply
via the costocervical trunk. The arrowhead indicates the ASA; the
single arrow, a median spinal vein; the double arrows, a lateral
perimedullary vein; and the asterisk, the aneurysm. D–F:
Tomographic MIP axial (D), coronal (E), and sagittal (F)
reconstructions of a 3D rotational angiogram showing the presence
of the SDAVF and ASA aneurysm in relation to the vertebral bodies
and laminae. The location of the aneurysm in the anterior portion
of the spinal canal is elucidated by the sagittal reconstruction.
The arrowhead indicates the ASA; the single arrow, a median spinal
vein; and the asterisk, the aneurysm. G: Magnified 3D
reconstruction of the SDAVF and the ASA aneurysm. The asterisk
indicates the aneurysm; the arrowhead, the ASA; the single arrow, a
median spinal vein; the double arrows, a lateral perimedullary
vein.
Unauthenticated | Downloaded 06/04/21 05:05 AM UTC
-
Neurosurg Focus / Volume 32 / May 2012
Rotational spinal angiography
5
nous shunting with early opacification of a large, caudally
draining perimedullary vein (Fig. 5A). Analysis of the 3D RA
suggested that this lesion represented an AVM with a plexiform
nidus located on the right posterior surface of the spinal cord,
displacing the cord anteriorly (Fig. 5B–E, fast spin of
40°/second). He underwent T5–6 laminecto-mies and
ligation-resection of the AVM. Postoperatively, the patient’s
sensory symptoms improved.
DiscussionSpinal vascular malformations are rare in the gen-
eral population, and the majority (60%–80%) are in the form of
SDAVFs.13 They are usually supplied by dural branches of radicular
arteries and drain into medullary veins at the dural sleeve of a
nerve root and ultimately into the coronal venous plexus of the
spinal cord.27 Symp-toms are thought to arise secondary to vascular
steal or venous congestion and may include myelopathy, weak-ness,
sensory disturbance, gait disturbance, and bowel or urinary
problems. These lesions are classically difficult to diagnose and
may be confused with radiculopathy from degenerative disc disease,
neuromuscular disease, or de-myelinating neuropathy. The median
time from onset of
Fig. 2. Case 2. A 58-year-old man with paresthesias secondary to
a T12–L2 SDAVF. A and B: Sagittal (A) and axial (B) T2-weighted MR
images of the thoracolumbar spine demonstrating increased T2 signal
intensity within the parenchyma, cord expansion, and large flow
voids, suggestive of an SDAVF. C and D: Oblique AP views of the
DSA, obtained with selective left L-1 segmental artery injection
(C) and left L-2 segmental artery injection (D) demonstrating an
SDAVF. The asterisks indicate the fistula point; the arrowheads, an
epidural venous pouch; the single arrows, a draining perimedullary
vein; the double arrows, a draining paraspinal vein. E–G:
Tomographic MIP axial (E), coronal (F), and sagittal (G)
reconstructions of a 3D rotational angiogram showing the presence
of the SDAVF. The fistula point is visible just medial to the left
pedicle and is indicated by an asterisk in each image. The single
arrows indicate a draining perimedullary vein; the double arrows, a
draining paraspinal vein; the arrowheads, an epidural venous pouch.
H: A 3D reconstruction based on the original angiogram
demonstrating multiple feeding arteries supplying the SDAVF. The
double arrowheads indicate the left L-1 segmental artery; the
asterisk, the fistula point; the arrowhead, an epidural venous
pouch; the single arrow, a draining perimedullary vein. I:
Intraoperative photograph of the SDAVF, showing the clipping of a
feeding artery in the predicted location of the fistulous point (to
the left of center). Multiple dilated perimedullary veins are
visible.
Unauthenticated | Downloaded 06/04/21 05:05 AM UTC
-
A. E. Ropper et al.
6 Neurosurg Focus / Volume 32 / May 2012
Fig. 3. Case 2. A–H: Sequential (anterior to posterior) thin-cut
coronal reconstructions of the 3D angiogram (the asterisk
in-dicates the fistula point). This technique allows for detailed
anatomical views of the SDAVF in relation to the surrounding
pedicles, laminae, and disc spaces, essential for preoperative
planning.
Fig. 4. Case 3. A 42-year-old man with acute left leg numbness
and weakness. Imaging demonstrated a PSA aneurysm that was
subsequently clipped. A: Sagittal T2-weighted MR image showing
evidence of subarachnoid blood (arrowheads) surrounding the
thoracolumbar spinal cord. B: An AP-view DS angiogram obtained with
selective left L-1 segmental artery injection demonstrating an
aneurysm (asterisk) rostral to the injected level. C–E: Tomographic
MIP axial (C), coronal (D), and sagittal (E) reconstructions of a
3D rotational angiogram showing the presence of a PSA aneurysm
(asterisk) inferior to the left pedicle of T-11. These images
helped to demonstrate that the aneurysm was in the posterior
portion of the spinal canal and its precise level relative to the
vertebrae. F: Magnified 3D reconstruction of the PSA aneurysm at
the T-11 level, fed by the left L-1 segmental artery
inferiorly.
Unauthenticated | Downloaded 06/04/21 05:05 AM UTC
-
Neurosurg Focus / Volume 32 / May 2012
Rotational spinal angiography
7
symptoms to diagnosis of an SDAVF is between 12 and 44 months.13
Spinal artery aneurysms are even less common than fistulas and may
occur either in association with an SDAVF22 or in isolation. A
recent review reported only 26 cases of ruptured isolated spinal
artery aneurysms in the literature.15 Seven of those 26 aneurysms
were located on the ASA, 5 on the artery of Adamkiewicz, 4 on the
PSA, and the remaining aneurysms were fed by segmental or
radiculomedullary branches.
Despite the improvements in technology discussed below, neither
cross-sectional imaging modality—MRI or CT—currently approaches the
necessary degree of sensitivity and spatial resolution of standard
DSA to rule out spinal vascular lesions. Therefore, DSA remains the
“gold standard” in the diagnosis of these often-elusive spinal
vascular lesions. Magnetic resonance imaging has emerged as the
standard initial diagnostic screening tool for the workup of SVMs.
Characteristic MRI findings of SDAVF include centrally located T2
hyperintensity with peripheral sparing12 and tortuous “flow voids”
on both T1- and T2-weighted images.7 Contrast enhanced MRA has been
used to visualize SVMs in multiple studies with promising results.
Binkert and colleagues4 reviewed MRA and DSA studies performed in
12 consecutive pa-tients with suspected SVMs and found that MRA
correct-ly identified the categories of 9 vascular lesions (6 AVMs,
3 SDAVFs). Mull et al.20 studied how accurately MRA could localize
SVMs compared with DSA and reported that MRA-derived spinal levels
agreed with DSA in 14
of 19 SDAVF cases. Ali et al.1 used the newer technology of
dynamic multiphase time-resolved MRA in 11 patients with suspected
SVMs. The authors correctly diagnosed 6 vascular lesions and were
able to localize within 1 verte-bral level in 5 of the 6 cases. In
general, however, the im-aging quality of MR-based modalities is
easily affected by motion degradation secondary to respirations,
espe-cially in the thoracolumbar region, and the spatial
resolu-tion does not yet approach the sensitivity or anatomical
detail provided by standard DSA.19
Three-dimensional CTA has also been used to di-agnose spinal
vascular lesions and has the advantage of excellent visualization
of bony anatomy. Lai et al.18 evalu-ated 8 patients with suspected
SDAVF via multidetector CTA and DSA and reported good correlation
between the 2 modalities in all 8 cases. Differentiation of
arterial from venous phases on “dynamic” multidetector CTA has
proven to be quite challenging and tends to degrade imag-ing
quality.19
The utility of 3D RA has been described before, spe-cifically by
those evaluating its role in the endovascular treatment of spinal
vascular lesions. Prestigiacomo et al.24 reviewed their experience
with 17 3D spinal angiograms in 14 patients, who had undergone
angiography for the diagnosis and treatment of a variety of SVMs.
Surgical planning for hemangioblastoma resection after
emboli-zation has also benefitted from the use of 3D rotational
angiograms, as described by Kern et al.16 One signifi-cant
technical difference in our approach is the ability
Fig. 5. Case 4. A 61-year-old man with a history of myelopathy
and urinary retention. Previous lumbar decompression for
radiographic stenosis did not relieve his symptoms. MR imaging of
the spine was nondiagnostic, but an angiogram revealed a pial AVM.
Surgery confirmed this diagnosis and he underwent thoracic
laminectomies and surgical obliteration of the AVM. A:
Pre-operative DS angiogram, AP view of the right T-6 segmental
artery injection showing the malformation. B–D: Tomographic MIP
axial (B), coronal (C), and sagittal (D) reconstructions of a 3D
rotational angiogram demonstrating the nidus of the pial AVM and
its relation to the vertebral canal. The asterisk indicates the
plexiform nidus; the arrow, a perimedullary draining vein. E: A 3D
reconstruction of the pial AVM (asterisk) with a T-6 segmental
artery feeder (arrowhead) and draining perimedullary vein
(arrow).
Unauthenticated | Downloaded 06/04/21 05:05 AM UTC
-
A. E. Ropper et al.
8 Neurosurg Focus / Volume 32 / May 2012
to choose varying rotational speeds depending on the amount of
soft tissue and bony details desired from the angiogram. Fast spin
(40°/second, used for all 3D render-ing models) allows crisp
vascular imaging with relatively less soft tissue information. Slow
spin (20°/second, used for Case 2 as described above) maximizes
bony detail in addition to providing good delineation of
angioarchitec-ture. The quality of rotational spinal angiography
can be influenced by the stability of the catheter position and the
degree of cooperation from the patient if he or she is not in a
state of general anesthesia. High-flow AVMs may not allow a
sufficient volume of contrast agent to be injected during the
rotation flow.19 In addition, image quality can be limited due to
respiratory artifacts.24 Rotational spinal angiography does require
marginally more contrast me-dium (10–20 ml) than standard 2D DSA,
but does not ex-pose the patient to significantly more radiation
(Table 1). Radiation exposure in our series was comparable to that
reported in the spinal angiography literature.9,24 Further-more,
the mean dose area product (DAP) (409.5 Gy⋅cm2) delivered to
patients receiving 3D RA was comparable to the mean DAP (413
Gy⋅cm2) delivered to patients during cranial aneurysm embolization
procedures.5
The cases presented in this series demonstrate the specific
benefits of 3D RA as an adjunct to traditional DSA—in particular
the utility and superior anatomical detail provided by the
adjustable tomographic MIP im-ages, as well as the volume-rendered
3D reconstructions. Rapid, near real-time postprocessing of the
image dataset allowed us to create a 3D model of the lesion and
tumble it in any desired projection. Features of the malformation
can be easily correlated with bony and even soft tissue anatomy.
The anatomical detail obtained from these re-constructions in
conjunction with the standard DSA find-ings can be used to better
differentiate an SDAVF from an AVM, as in Case 4 in the current
report. Differentia-tion of an aneurysm associated with the
anterior versus posterior spinal artery, as in Case 3, is
absolutely critical when considering treatment options such as
sacrifice of an artery. Ligation of the PSA is usually well
tolerated, whereas sacrifice of a major ASA contribution can have
devastating effects. Tracing a vascular malformation on 3D
reconstructions and the corresponding tomographic MIP
reconstructions provides precise localization of the fistulous
connection as demonstrated in the case of Case 2. Placed in the
context of the soft tissue and bony anato-my, this high-resolution
vascular dataset supplies unprec-edented detail while being
visually intuitive and therefore easily applicable clinically.
Information provided by this advanced imaging tool is likely to
improve the planning of endovascular as well as surgical approaches
required for lesion obliteration.
ConclusionsIn summary, 3D RA is an advanced imaging tool
pro-
ducing extraordinary anatomical detail in the character-ization
and delineation of spinal vascular pathology and should become a
standard part of the invasive workup of these lesions. Routine use
of this tool may improve our understanding and management of SVMs,
both for open surgical and endovascular approaches.
Disclosure
The authors report no conflict of interest concerning the
mate-rials or methods used in this study or the findings specified
in this paper.
Author contributions to the study and manuscript preparation
include the following. Conception and design: Frerichs, Ropper,
Lin, Thiex. Acquisition of data: Frerichs, Ropper, Lin, Zarzour,
Du. Analysis and interpretation of data: Frerichs, Ropper, Lin,
Gross, Zarzour. Drafting the article: Frerichs, Ropper, Lin, Gross.
Critically revising the article: all authors. Reviewed submitted
version of manuscript: all authors. Statistical analysis: Lin.
Study supervision: Frerichs.
References
1. Ali S, Cashen TA, Carroll TJ, McComb E, Muzaffar M, Shaibani
A, et al: Time-resolved spinal MR angiography: ini-tial clinical
experience in the evaluation of spinal arteriove-nous shunts.
AJNR Am J Neuroradiol 28:1806–1810, 2007
2. Aminoff MJ, Logue V: Clinical features of spinal vascular
malformations. Brain 97:197–210, 1974
3. Behrens S, Thron A: Long-term follow-up and outcome in
pa-tients treated for spinal dural arteriovenous fistula. J Neurol
246:181–185, 1999
4. Binkert CA, Kollias SS, Valavanis A: Spinal cord vascular
disease: characterization with fast three-dimensional
con-trast-enhanced MR angiography. AJNR Am J Neuroradiol
20:1785–1793, 1999
5. D’Ercole L, Mantovani L, Thyrion FZ, Bocchiola M, Azza-retti
A, Di Maria F, et al: A study on maximum skin dose in cerebral
embolization procedures. AJNR Am J Neuroradiol 28:503–507, 2007
6. Eddleman CS, Jeong H, Cashen TA, Walker M, Bendok BR, Batjer
HH, et al: Advanced noninvasive imaging of spinal vascular
malformations. Neurosurg Focus 26(1):E9, 2009
7. Gilbertson JR, Miller GM, Goldman MS, Marsh WR: Spinal dural
arteriovenous fistulas: MR and myelographic findings.
AJNR Am J Neuroradiol 16:2049–2057, 1995
8. Hanel RA, Nakaji P, Spetzler RF: Use of
microscope-integrat-ed near-infrared indocyanine green
videoangiography in the surgical treatment of spinal dural
arteriovenous fistulae. Neu-rosurgery 66:978–985, 2010
9. Heinemann MK, Brassel F, Herzog T, Dresler C, Becker H, Borst
HG: The role of spinal angiography in operations on the thoracic
aorta: myth or reality? Ann Thorac Surg 65:346–351, 1998
10. Hirai T, Korogi Y, Suginohara K, Ono K, Nishi T, Uemura S,
et al: Clinical usefulness of unsubtracted 3D digital angiogra-phy
compared with rotational digital angiography in the pre-treatment
evaluation of intracranial aneurysms. AJNR Am J Neuroradiol
24:1067–1074, 2003
11. Hochmuth A, Spetzger U, Schumacher M: Comparison of
three-dimensional rotational angiography with digital sub-traction
angiography in the assessment of ruptured cerebral aneurysms.
AJNR Am J Neuroradiol 23:1199–1205, 2002
12. Hurst RW, Grossman RI: Peripheral spinal cord hypointen-sity
on T2-weighted MR images: a reliable imaging sign of venous
hypertensive myelopathy. AJNR Am J Neuroradiol 21:781–786, 2000
13. Jellema K, Tijssen CC, van Gijn J: Spinal dural
arteriovenous fistulas: a congestive myelopathy that initially
mimics a pe-ripheral nerve disorder. Brain 129:3150–3164, 2006
14. Jiang L, Huang CG, Liu P, Yan B, Chen JX, Chen HR, et al:
3-Dimensional rotational angiography for the treatment of spinal
cord vascular malformations. Surg Neurol 69:369–374, 2008
15. Karakama J, Nakagawa K, Maehara T, Ohno K: Subarachnoid
hemorrhage caused by a ruptured anterior spinal artery aneu-rysm.
Neurol Med Chir (Tokyo) 50:1015–1019, 2010
Unauthenticated | Downloaded 06/04/21 05:05 AM UTC
-
Neurosurg Focus / Volume 32 / May 2012
Rotational spinal angiography
9
16. Kern M, Naeini R, Lehmann TN, Benndorf G: Imaging of a
thoracic spinal nerve haemangioblastoma by three-dimensional
digital angiography. J Clin Neurosci 13:929–932, 2006
17. Kim LJ, Spetzler RF: Classification and surgical manage-ment
of spinal arteriovenous lesions: arteriovenous fistulae and
arteriovenous malformations. Neurosurgery 59 (5 Suppl 3):S195–S201,
2006
18. Lai PH, Pan HB, Yang CF, Yeh LR, Hsu SS, Lee KW, et al:
Multi-detector row computed tomography angiography in di-agnosing
spinal dural arteriovenous fistula: initial experience. Stroke
36:1562–1564, 2005
19. Matsubara N, Miyachi S, Izumi T, Ohshima T, Tsurumi A,
Hososhima O, et al: Usefulness of three-dimensional digital
subtraction angiography in endovascular treatment of a spinal dural
arteriovenous fistula. Report of 2 cases. J Neurosurg Spine
8:462–467, 2008
20. Mull M, Nijenhuis RJ, Backes WH, Krings T, Wilmink JT, Thron
A: Value and limitations of contrast-enhanced MR an-giography in
spinal arteriovenous malformations and dural arteriovenous
fistulas. AJNR Am J Neuroradiol 28:1249–1258, 2007
21. Narvid J, Hetts SW, Larsen D, Neuhaus J, Singh TP, McSwain
H, et al: Spinal dural arteriovenous fistulae: clinical features
and long-term results. Neurosurgery 62:159–167, 2008
22. Ohmori Y, Hamada J, Morioka M, Yoshida A: Spinal aneu-rysm
arising from the feeding pedicle of a thoracic perimed-ullary
arteriovenous fistula: case report. Surg Neurol 64: 468–470,
2005
23. Oldfield EH, Di Chiro G, Quindlen EA, Rieth KG, Doppman JL:
Successful treatment of a group of spinal cord arteriove-nous
malformations by interruption of dural fistula. J Neuro-surg
59:1019–1030, 1983
24. Prestigiacomo CJ, Niimi Y, Setton A, Berenstein A:
Three-dimensional rotational spinal angiography in the evaluation
and treatment of vascular malformations. AJNR Am J Neu-roradiol
24:1429–1435, 2003
25. Saladino A, Atkinson JL, Rabinstein AA, Piepgras DG, Marsh
WR, Krauss WE, et al: Surgical treatment of spinal dural
arteriovenous fistulae: a consecutive series of 154 pa-tients.
Neurosurgery 67:1350–1358, 2010
26. Sharma AK, Westesson PL: Preoperative evaluation of spinal
vascular malformation by MR angiography: how reliable is the
technique: case report and review of literature.
Clin Neu-rol Neurosurg 110:521–524, 2008
27. Steinmetz MP, Chow MM, Krishnaney AA, Andrews-Hinders D,
Benzel EC, Masaryk TJ, et al: Outcome after the treatment of spinal
dural arteriovenous fistulae: a contemporary single-institution
series and meta-analysis. Neurosurgery 55:77–88, 2004
28. Symon L, Kuyama H, Kendall B: Dural arteriovenous
malfor-mations of the spine. Clinical features and surgical results
in 55 cases. J Neurosurg 60:238–247, 1984
29. Zozulya YP, Slin’ko EI, Al-Qashqish II: Spinal arteriovenous
malformations: new classification and surgical treatment. Neurosurg
Focus 20(5):E7, 2006
Manuscript submitted September 17, 2011.Accepted January 27,
2012.Please include this information when citing this paper:
DOI:
10.3171/2012.1.FOCUS11254. Address correspondence to: Kai U.
Frerichs, M.D., Department
of Neurosurgery, Brigham and Women’s Hospital, 75 Francis
Street, Boston, Massachusetts 02115. email:
[email protected].
Unauthenticated | Downloaded 06/04/21 05:05 AM UTC