Chronic Cerebrospinal
Venous Insufficiency
A new paradigm and therapy for multiple sclerosis.
BY SALVATORE J.A. SCLAFANI, MD
COVER STORY
Figure 1. Preprocedural MR venogram shows a venous
plethora of collateral veins throughout the neck. Arrows
point to perceived stenoses of the distal lower cervical
portion of the IJV.
neurological disorder of young adults. It is quite possible
that some of the protean manifestations of MS,
including fatigue and lethargy, headaches, and cognitive
dysfunction, may actually represent symptoms of CCSVI
itself.8
CCSVI is more insidious in its onset than acute
venous insufficiency. In fact, the association of CCSVI
with MS has been largely ignored despite Charcot’s original
description of the relationship of the cerebral veins
and inflammatory lesions that are the hallmark of MS.9
Zamboni proposes that CCSVI has a role in the
pathogenesis of MS. He suggests that resistance to cerebrospinal
venous outflow causes vicarious redistribution
through small collateral veins that cannot handle
high flow.10 He also suggests that tight endothelial junctions
widen to allow diapedesis of red blood cells,
T cells, and other immune cells into the brain, resulting
in inflammation and hemosiderosis that is reminiscent
of what is seen with venous insufficiency of the lower
extremities. This is supported by iron deposition as seen
on susceptibility-weighted magnetic resonance imaging
(SW-MRI), which reveals that the inflammatory MS
plaques always surround a central venous structure.
MRI shows that the central vein and surrounding
plaque have abnormal quantities of iron. Pathologically,
the basement membranes of these deep veins are thickened,
and hemosiderin deposits are present in the wall
of and adjacent to the deep cortical veins. T cells and
macrophages violating the blood-brain barrier provide a
working explanation for the autoimmune cascade that
result in demyelination and the neurological manifestations
associated with MS.
DIAGNOSIS
Ultrasound
One could argue that the diagnosis of MS is sufficient
to justify catheter venography to identify venous abnormalities
worthy of angioplasty. However, Zamboni used
ultrasound imaging to noninvasively screen patients who
might have CCSVI, and this algorithm persists as the route
of detection. His protocol includes transcranial and
extracranial Doppler to detect deranged hemodynamics
and B-mode ultrasound to detect stenoses and changes in
cross-sectional diameters in the supine and the upright
positions. He states that two of five characteristics lead to
a diagnosis of CCSVI. The five characteristics are (1) reflux
within the IJVs or vertebral veins, (2) reflux within any of
the deep cerebral veins, (3) no flow in the IJV on activation
of the thoracic pump upon inspiration, (4) failure of
the IJV to increase in diameter in the supine position
compared to the erect position, and (5) any B-mode
abnormality such as septum, stenosis, abnormal valve, etc.
MR Venography and Computed Tomographic
Venography
Others have used cross-sectional venography to
evaluate venous stenosis (Figure 1). The majority of
sites use MR venography, but occasionally, computed
tomographic venography is also used. To evaluate the
dural sinuses and the veins of the neck, two-dimensional
and two-dimensional contrast-enhanced imaging
is used. These cross-sectional studies show a variety
of findings that include venous narrowing and collateral
vessels throughout the neck. Occasionally, narrowing
or occlusions of the dural sinuses are noted,
but for the majority of times, findings are restricted to
the neck. However, there is poor correlation between
the anatomical findings on MR venography and subsequent
catheter venography. Many areas of narrowing
on MR venography are not constant and are not
reproduced during catheter-based studies.
Hemodynamics of Cerebral Venous Drainage Explain
False-Positive Findings on MR Venography
To explain this enigma, one must understand the
hemodynamics of cerebral venous outflow. The brain
has two methods of venous drainage: blood drains
42 I ENDOVASCULAR TODAY I JULY 2010
COVER STORY
Figure 2. Bilateral IJV venography shows opacification of the
vertebral plexus and drainage through the vertebral veins
(arrows).Both IJVs had focal stenoses at the confluens of the
IJV with the subclavian vein.Angioplasty improved but did
not eliminate the stenoses.
JULY 2010 I ENDOVASCULAR TODAY I 43
COVER STORY
In this article, Dr. Salvatore Sclafani presents an introduction
to chronic cerebrospinal venous insufficiency (CCSVI)
and the current understanding of its association with multiple
sclerosis (MS). Much of the initial evidence supporting
this possible relationship has been reported by Dr. Paolo
Zamboni and colleagues. Using duplex ultrasonography and
transcranial Doppler studies, they have documented the frequent
association of abnormal venous hemodynamics with
MS. In one study of 109 MS patients and 177 age- and gender-
matched controls, subjects underwent a blinded transcranial
and extracranial color Doppler sonographic assessment
(TCCS-ECD) of five parameters related to venous outflow
hemodynamics. These five criteria are detailed by Dr.
Sclafani in his review. In controls, only 2.7% of the measurements
were abnormal, whereas in MS patients, 47% of measurements
were anomalous.1
In a study comparing duplex ultrasound with contrast
venography, 40% to 70% of MS patients had evidence of
flow disturbances and/or venous stenosis by TCCS-ECD. Of
these patients, 86% and 91% had obstructive disease of the
azygos or internal jugular veins, respectively, as assessed by
traditional catheter venography.2
Some of the symptoms of MS mimic those observed in
patients with superior vena cava syndrome. Relief of superior
vena cava obstruction with venous angioplasty and stent
placement, if required, provides swift and dramatic resolution
of the symptoms of impaired cognition and fatigue.3 Thus, it
is not surprising that patients with CCSVI associated with MS
also report rapid relief of these nonlocalizing symptoms.
It is well-recognized, however, that many symptoms of MS
fluctuate and are largely subjective. It is possible that in the
initial nonrandomized patient series reported to date, the
improvement in symptoms could reflect a strong placebo
effect. Nonetheless, the biological plausibility linking cerebral
venous congestion to inflammation that is the hallmark of
MS requires serious consideration. Whether the relief of the
venous obstruction will have an impact on the course of the
neurological disease remains to be seen.
Although the initial observations relating CCSVI and MS
are interesting and potentially paradigm-shifting, they now
need rigorous testing. As Dr. Sclafani correctly points out,
there are life-threatening adverse effects that may complicate
endovascular management of CCSVI. A randomized clinical
trial is needed to assess the risks and benefits of endovascular
treatment of this condition. There are many physicians and
others who have endovascular skills who are promoting and
developing centers for treating these patients without regard
for the lack of scientific data to support therapy. Patients
with this disease have frequently suffered for long periods of
time, often without great relief of symptoms and are often
desperate for any alternative that may offer hope.
We remain very concerned about the possibility of misleading
these individuals or exposing them to additional risk, outside
of scientific efforts to get a better understanding of this potentially
exciting therapy. Given the concerns of the neurology
community, it would be unfortunate if the attempts to advance
this field suffer the consequences of premature promotion of a
procedure that could mislead patients, payors, and regulators.
Accordingly, we propose a global initiative to meticulously
document the prevalence of venous anomalies in MS, by
comparison to age- and gender-matched healthy individuals,
as well as those with neurological disease not due to MS. In
part, recent grants from the National MS Society awarded to
seven investigative groups to study CCSVI will help initiate
this effort in the United States and Canada. These observations
may provide a basis for a clinical trial in MS to assess
the long-term safety and efficacy of endovascular procedures
in restoring normal venous hemodynamics, in relieving the
nonlocalizing symptoms secondary to venous obstruction,
and in slowing or halting the inflammatory and demyelinating
processes. In parallel, the development of animal models
will advance our understanding of how CCSVI may influence
or even initiate the pathophysiology of MS.
Michael D. Dake, MD, is Professor, Department of
Cardiothoracic Surgery, Stanford University School of Medicine
in Stanford, California. He has disclosed that he receives
grant/research funding from Cook Medical. Dr. Dake may be
reached at mddake@stanford.edu.
Barry T. Katzen, MD, is Founder and Medical Director of Baptist
Cardiac and Vascular Institute and Clinical Professor of Radiology
at the University of South Florida College of Medicine in Florida.
He has disclosed that he is a member of the scientific advisory
board for W. L. Gore & Associates. Dr. Katzen may be reached at
barryk@baptisthealth.net.
1. Zamboni P, Menegatti E, Galeotti R, et al. The value of cerebral Doppler venous hemodynamics
in the assessment of multiple sclerosis. J Neurol Sci. 2009;282:21-27.
2. Zamboni P, Galeotti R, Menegatti E, et al. Chronic cerebrospinal venous insufficiency
in patients with multiple sclerosis. J Neurol Neurosurg Psychiatry. 2009;80:392-399.
3. Kee ST, Kinoshita L, Razavi MK, et al. Superior vena cava syndrome treatment with catheterdirected
thrombolysis and endovascular stent placement. Radiology. 1998;206:187-193.
COMMENTARY
The Relationship Between CCSVI and MS
BY MICHAEL D. DAKE, MD, AND BARRY T. KATZEN, MD
44 I ENDOVASCULAR TODAY I JULY 2010
COVER STORY
anteriorly through the internal jugular
system in the supine position and posteriorly
through the vertebral system when
erect. In the normal, upright patient, the
jugular vein collapses (narrows) because
there is not enough blood flow through
it to maintain distension. In the supine
position, the normal IJVs distend
because the supine position favors jugular
flow. The same issues apply when
there is increased resistance to jugular
flow. The alternate vertebral venous outflow
system shunts blood away from the
jugular veins. Because pressure is normally
low and only marginally rises with
obstruction, distension of the obstructed
system does not occur.
As a result, many of the narrowings seen in CCSVI
are caused by compression of a collapsed system by
external forces rather than due to stenoses. This may
lead to unnecessary angioplasty. The common areas of
questionably physiological stenosis seen on MR venography
are located at the skull base, adjacent to the
carotid bulb, or where strap muscles exert compression.
VENOGRAPHY AND VENOGRAPHIC
OBSERVATIONS
Venography remains the gold standard for evaluating
the anatomy of the veins draining cerebrospinal blood
flow. It should be emphasized that a reliable assessment
of the azygos system can only be done by using catheter
venography.
Technique
The venographic evaluation is begun by placing a
headhunter catheter in the left femoral vein with the
purpose of excluding May-Thurner syndrome. The
catheter is subsequently placed in the left ascending
lumbar vein to assess the lumbar veins for hypoplasia
and other abnormalities. The left renal vein is then
catheterized to look for abnormalities of the renal vein
tributaries. The purpose of these three studies is to look
for causes of increased blood flow into lumbar veins
that might be compromised by azygos stenosis.
The catheter is then placed in succession into the
AZV and both IJVs. The catheter is positioned in the
AZV at the junction with the hemiazygos vein. Contrast
venography is done twice: first at 3 mL/s for a total volume
of 10 mL to look for reflux, followed by a second,
fuller injection at 8 to 10 mL/s for a total volume of 20
to 30 mL to delineate all the anatomy. The AZV and its
tributaries are imaged to include the chest and
abdomen. Some physicians measure pressures, but I
have not found this to be helpful. Any stenosis is treated,
as will be described later.
The catheter is then withdrawn from the AZV and
advanced sequentially into each IJV. Catheterization of
the IJV may be challenging because funneled narrowing
of stenotic valve leaflets occurs near the origin of the
vessel. Occasionally, an incomplete duplication is present
posterior to the main ostium. This may make
catheterization confusing and difficult. Two contrast
injections are performed: one with a slow injection of
3 mL/s for a total volume of 10 mL and one with a fuller
injection of 8 to 10 mL/s for a total volume of 20 mL.
Film rates of 3 to 6 frames per second are necessary to
get sufficient detail of the valves and to detect ostial
narrowing that may become obscured as contrast
enters the brachiocephalic veins and overlaps the confluens
where stenosis is often located. Any stenoses or
other outflow obstructions are treated at this time.
Diluted contrast abnormalities (50:50 mixture of saline)
is helpful in the IJV evaluation because valve abnormalities
and some webs may be obscured by very dense
contrast media.
Venographic Findings
First, there are numerous collateral veins when outflow
obstructions are present (Figure 2). These veins may be
wildly abnormal and include hypoplasias and early divisions
that reconnect to a larger conduit. The vertebral
veins may be enlarged and can be confusing in their
appearance. The pathology of this disease is a truncal
malformation of the veins that is probably genetically
determined; it is not an inflammatory or postphlebitic
stenosis. Much of the resistance to blood flow is related
to abnormal valve development. Fused, reversed, thick-
Figure 3. IVUS in resting state at the level of the stenosis of the central cervical
IJV (A) shows narrowing of the IJV, which is collapsed around the catheter.
The thoracic pump was activated, and flow into the IJV improved (B). IVUS
shows distension; therefore, angioplasty was not performed.
A B
ened, and other abnormally located and developed
valves cause resistance to flow. Atresias, hypoplasias,
duplications, webs, septums, and kinks also occur. Most
of these abnormalities are located centrally near the confluens.
Challenges occur when more peripheral narrowings
are present, which may be physiological.
INTRAVASCULAR ULTRASOUND
Diagnosis by venography can also be subtle. I have
found that intravascular ultrasound (IVUS) is very helpful
in identifying some of these abnormalities, as well as
in differentiating the narrowed veins caused by inadequate
volume from the narrowed veins resulting from
stenosis (Figure 3). IVUS enables a real-time assessment
of the distensibility of collapsed veins. Simple maneuvers,
such as slow sustained inspiration by activating the thoracic
pump, allow improved distension of the vein and
confirms that the narrowing is not fixed. Further, IVUS
allows detection of improper or incomplete valve movement.
Finally, incomplete duplications of the jugular vein
may not be detected without IVUS.
TREATMENT OPTIONS
Treatment of these abnormalities is still in development,
and the ideal methodologies for treatment have
not yet been established. Essentially, only one team has
published an outcomes study.1 Results were encouraging
but showed limitations. Angioplasty with high-pressure
balloons of diameters 4 mm greater than nominal diameters
in 2- to 4-cm lengths is performed with venographic
control. Inflations to maximum pressures for 30 to 60
seconds were used several times. Some of these obstructions
are very resistant, and Cutting balloons (Boston
Scientific Corporation, Natick, MA) are used with
increasing frequency. Dr. Sinan Tariq, the leader of the
Kuwaiti national trial, has been using valvulotomy
devices with some success (personal communication,
April 2010). Stenting is performed by some investigators
for resistant narrowings. However, no reports have been
published about their outcomes. I have not used stents
in any cases yet.
AFTERCARE AND FOLLOW-UP
The procedure is performed under local anesthesia
in an ambulatory setting. Most patients are kept in
the hospital for 1 or 2 hours and then discharged.
Most physicians treat patients with clopidogrel or
short-term anticoagulation with heparins, enoxaparin,
or fondaparinux. Clinical and imaging follow-up varies
among investigators. Assessment tools are predominantly
clinical and include an expanded disability status
score (EDSS), which is a neurological assessment of
EVToday.com
EVToday.com
EVToday.com
EVToday.com
EVToday.com
EVToday.com
EVToday.com
EVToday.com
EVToday.com
EVToday.com
EVToday.com
EVToday.com
EVToday.com
EVToday.com
EVToday.com
visit www.evtoday.com for the current issue
and complete archives
46 I ENDOVASCULAR TODAY I JULY 2010
COVER STORY
eight areas of the central nervous system, along with
certain measures of disability and restriction in daily
life. These scores are added up to give a rating on the
EDSS, which ranges from 0 (normal) to 10 (death due to
MS). From step 4 onward, the ability to walk becomes
the key factor in determining the EDSS score.
OUTCOMES
It must be emphasized that only one team has published
any clinical results, and although promising,
they were not overwhelming. Zamboni’s group
described an open-label experience of patients with
MS who were allowed to stay on disease-modifying
drugs for their MS. The results were encouraging, with
statistically significant improvements in cognition and
motor function and reduced exacerbation rates, and
MRI confirmed diminished new brain lesion development.
The patients who have shown the most positive
results are those in the relapsing-remitting phase of
the disease. Patients with primary progressive MS, for
whom there is no proven treatment, had the least positive
effects.
However, the dilatations are not always durable,
with approximately half of the patients developing
restenosis between 8 to 14 months. It is interesting
that all patients who suffered from an exacerbation of
symptoms had a restenosis and that no patients who
had durable angioplasty experienced restenosis.
Overall, the procedure is well-tolerated, and patients
do not require sedation. The complications reported
in Zamboni’s trial were minimal. I have had one early
thrombosis that did not respond to thrombolytics and
one case of atrial fibrillation that I thought might have
been a response to treatment that modified autonomic
neural transmission, but resolved within 12 hours.
Those interventionists who have used stents have not
yet reported outcomes in the literature.
Dr. Zamboni cautions against stents because they
are not designed for placement at the confluens of the
jugular vein with the subclavian vein where the jugular
vein widens. Improved flow is shown to significantly
increase the diameter of these veins. He worried about
migration in his article, and indeed, one of the early
patients treated with stenting by another interventionist
is reported in the lay press to have required open
heart surgery for stent retrieval.
CAVEATS
CCSVI has not been well-accepted by the neurological
community. Many leaders strongly oppose this
treatment on the grounds that no randomized
prospective trials have taken place, and they describe