Flow Diversion for Treatment of Partially Thrombosed Aneurysms: A Multicenter Cohort

Original Article
Flow Diversion for Treatment of Partially Thrombosed Aneurysms: A Multicenter Cohort
Paul M. Foreman1,5, Mohamed M. Salem2, Christoph J. Griessenauer1,6, Adam A. Dmytriw7, Carmen Parra-Farinas8,
Patrick Nicholson7, Nicola Limbucci9, Anna Luisa Kühn3, Ajit S. Puri4, Leonardo Renieri9, Sergio Nappini9,
Kimberly P. Kicielinski2, Alejandro Bugarini1, Vitor Mendes Pereira7, Thomas R. Marotta8, Clemens M. Schirmer1,
Christopher S. Ogilvy2, Ajith J. Thomas2
BACKGROUND: Partially thrombosed intracranial aneurysms (PTIA) represent a unique subset of intracranial
aneurysms with an ill-defined natural history, posing
challenges to standard management strategies. This study
aims to assess the efficacy of flow diversion in the treatment of this pathology.
-
METHODS: A retrospective review of patients with flowdiverted PTIA at 6 cerebrovascular centers was performed.
Clinical and radiographic data were collected from the
medical records, with the primary outcome of aneurysmal
occlusion and secondary outcomes of clinical status and
complications.
-
RESULTS: Fifty patients with 51 PTIA treated with flow
diversion were included. Median age was 56.5 years.
Thirty-three (64.7%) aneurysms were saccular and 16
(31.4%) were fusiform/dolichoectatic. The most common
location was the internal carotid artery (54.9%) followed by
the vertebral and basilar arteries (17.7% and 17.7%,
respectively). Last imaging follow-up was performed at a
median of 25.1 (interquartile range, 12.8e43) months.
Complete occlusion at last radiographic follow-up was
achieved in 37 (77.1%) aneurysms. Pretreatment aneurysm
thrombosis of >50% was associated with a significantly
lower rate of complete aneurysm occlusion (58.8 vs. 87.1%,
P [ 0.026) with a trend toward better functional outcome
(modified Rankin scale <2) at last follow-up in patients
with <50% pretreatment aneurysm thrombosis (96.8 vs. 82.4;
P [ 0.08). Ischemic complications occurred in 5 (9.8%)
patients, producing symptoms in 4 (7.8%) and resultant
mortality in 2 (4.2%) patients.
CONCLUSIONS: Flow diversion treatment of PTIA has
adequate efficacy along with a reasonable safety profile.
Aneurysms harboring large amounts of pretreatment
thrombus were associated with lower rates of complete
occlusion.
-
-
Key words
Aneurysm
- Complication
- Flow diverter
- Occlusion
- Pipeline
- Thrombosed aneurysm
- Thrombus
-
Abbreviations and Acronyms
CN: Cranial nerve
CT: Computed tomography
CTA: Computed tomography angiography
DSA: Digital subtraction angiography
IQR: Interquartile range
MRA: Magnetic resonance angiography
MRI: Magnetic resonance imaging
mRS: Modified Rankin scale
PTIA: Partially thrombosed intracranial aneurysm
e164
www.SCIENCEDIRECT.com
INTRODUCTION
P
artially thrombosed intracranial aneurysms (PTIA) are a
rare and diverse subset of intracranial aneurysm with a
distinct, but ill-defined, natural history. These aneurysms
are often large or giant in size, and although they can present with
aneurysmal subarachnoid hemorrhage, a more insidious presentation due to mass effect is more common.1 Their size coupled
with the presence of organized thrombus makes surgical
From the 1Department of Neurosurgery and Neuroscience Institute, Geisinger Health System,
Danville, Pennsylvania; 2Division of Neurosurgery, and 3Division of Neuroradiology,
Department of Radiology, Beth Israel Deaconess Medical Center, Harvard Teaching Hospital,
Boston; 4Division of Neuroimaging and Intervention, Department of Radiology, University of
Massachusetts Medical Center, Worcester, Massachusetts; 5Orlando Health, Neuroscience
and Rehabilitation Institute, Orlando, Florida, USA; 6Research Institute of Neurointervention,
Paracelsus Medical University, Salzburg, Austria; 7Department of Medical Imaging, Toronto
Western Hospital, University Health Network, Toronto, Ontario, Canada; 8Department of
Interventional Neuroradiology, St. Michael’s Hospital, Toronto, Ontario, Canada; and
9
Department of Interventional Neuroradiology, University of Florence, Florence, Italy
To whom correspondence should be addressed: Paul M. Foreman, M.D.
[E-mail: pforeman@geisinger.edu]
Citation: World Neurosurg. (2020) 135:e164-e173.
https://doi.org/10.1016/j.wneu.2019.11.084
Journal homepage: www.journals.elsevier.com/world-neurosurgery
Available online: www.sciencedirect.com
1878-8750/$ - see front matter ª 2019 Published by Elsevier Inc.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.11.084
ORIGINAL ARTICLE
PAUL M. FOREMAN ET AL.
obliteration challenging and can require advanced techniques
complicated by mortality rates of 6%2 and unfavorable outcomes
in 18%.3 Standard endovascular embolization with detachable
coils has also yielded unsatisfactory long-term results due to
continued aneurysm growth and recanalization rates as high as
78%.4-6
Food and Drug Administration approval of the Pipeline
Embolization Device in 2011 added flow diverting stents to the
interventionalist’s armamentarium. Placement of flow diverting
stents within the parent artery obviates the need to enter or
manipulate the PTIA, potentially reducing complication rates. The
durability of aneurysm occlusion after parent artery reconstruction
also has the potential to overcome shortcomings of traditional
endovascular techniques while avoiding parent artery occlusion.
Despite these conceivable advantages, there is concern that
evolving thrombosis can lead to destabilization of the aneurysm
wall and predispose to rupture7,8; however, the effect of preexisting thrombus on this risk is not known. In addition, the
impact of the chemical milieu of pre-existing intra-aneurysmal
thrombus on occlusion rates is also unknown.
We present an international multicenter cohort of PTIA treated
with flow diverting stents in an effort to further define the safety
and efficacy of the procedure for these aneurysms.
FLOW DIVERSION FOR THROMBOSED ANEURYSMS
persistent filling of aneurysm; stable, no change in aneurysm
filling; larger, increased filling of aneurysm. Secondary outcomes
included clinical status at last follow-up and thromboembolic and
hemorrhagic neurologic complication rates.
The study was approved at participating centers by their
respective internal review boards. Patient consent was not required
for this retrospective chart review project.
Statistical Analysis
Continuous variables are presented as mean standard deviation
or median (interquartile range [IQR]), as appropriate according to
the data distribution, whereas categorical variables are reported as
proportions. Aneurysms were categorized into 2 groups according
to their pretreatment thrombosis status (50% vs. <50%
thrombosis). In each group, categorical variables were compared
using the c2 test and continuous variables were compared using
the Mann-Whitney U test. A multivariable logistic regression
model was constructed to identify predictors of incomplete occlusion based on potential predictors on univariate screening.
Statistical significance was set to a P value of less than 0.05. All
statistical analyses were performed using the STATA 15 software
(StataCorp, College Station, Texas, USA).
RESULTS
METHODS
A retrospective review of consecutive patients with PTIA treated
with endovascular placement of a flow diverting stent at 5 international cerebrovascular centers from 2009 to 2018 was performed. The inclusion criterion was imaging evidence of a PTIA
that was treated with endovascular placement of a flow diverting
stent. The diagnosis of a PTIA was made when the diameter of an
aneurysm of cross-sectional computed tomography (CT) or magnetic resonance imaging (MRI) was greater than the diameter of
the perfused aneurysm on computed tomography angiography
(CTA), magnetic resonance angiography (MRA), or digital subtraction angiography (DSA). In addition, aneurysms were grouped
into 2 categories (50% or <50%) based on the percentage of
aneurysm thrombosis on pretreatment axial MRI/MRA or CTA. A
dichotomous category was used to improve reliability across
centers and imaging modalities and allow for an adequate number
of aneurysms in each group for statistical comparison. Both
ruptured and unruptured aneurysms were included. All aneurysm
morphologies and locations were eligible for inclusion.
Data collected included patient age, sex, past medical history,
history of subarachnoid hemorrhage from index aneurysm, cranial
nerve (CN) dysfunction caused by index aneurysm, functional
status at presentation as measured by the modified Rankin scale
(mRS), aneurysm characteristics, antiplatelet regimen, procedural
details, complications, radiographic and clinical outcome, and
final disposition (i.e., home, skilled nursing facility). Data were
collected through review of the electronic medical record at each
center. Data collectors were not blinded to the outcome
characteristics.
The primary outcome was aneurysm occlusion at last follow-up
as determined by contrast filling of index aneurysm on CTA, MRA,
or DSA. Occlusion status was defined as follows: complete occlusion, no filling of aneurysm; incomplete occlusion, reduced but
WORLD NEUROSURGERY 135: e164-e173, MARCH 2020
Patient Characteristics
Fifty patients with 51 PTIA treated with placement of flow diverting
devices were included. Median age was 56.5 (IQR, 46e67) years
and 60% of patients were female. Seventeen (34%) patients were
current tobacco smokers and 7 (14%) were past smokers. The
most frequent comorbidities included hypertension (56%),
hyperlipidemia (28%), and diabetes (22%). Pretreatment mRS was
0 in 44% and 1e2 in 42% (Table 1).
Aneurysm Characteristics
Of the 51 PTIA, 33 (64.7%) were saccular in morphology, 16
(31.4%) were fusiform/dolichoectatic, and the remaining 2 (3.9%)
were complex, multilobulated aneurysms unable to be classified as
saccular or fusiform/dolichoectatic. Twenty-eight (54.9%) aneurysms affected the internal carotid artery, 9 (17.7%) the vertebral
artery, 9 (17.7%) the basilar artery, 2 (3.9%) the middle cerebral
artery, 2 (3.9%) the anterior cerebral artery, and 1 (2%) the posterior inferior cerebellar artery. The majority of aneurysms were
side-wall (80.4%), with 10 bifurcation aneurysms in the cohort
(19.6%). Eighteen (35.3%) aneurysms were more than 50%
thrombosed before treatment. Forty-two (82.4%) aneurysms were
unruptured at presentation and 26 (51%) produced CN dysfunction. Seven (13.7%) aneurysms had been previously treated with
endovascular methods. The median size of the maximum
dimension of contrast filling of the aneurysm on MRA, CTA, or
DSA was 13 (IQR, 8.5e17) mm. The median size of the maximum
dimension on an axial MRI or CT, including the thrombosed
portion of the aneurysm, was 17 (IQR, 12e24.8) mm (Table 1).
Procedural Details
The Pipeline Embolization Device was used to treat 40 (78.4%)
aneurysms, a SILK device in 4 (7.8%), an LVIS device in 3 (5.9%),
an eCLIP device in 2 (3.9%), a combination of a SILK device and a
www.journals.elsevier.com/world-neurosurgery
e165
ORIGINAL ARTICLE
PAUL M. FOREMAN ET AL.
FLOW DIVERSION FOR THROMBOSED ANEURYSMS
Table 1. Continued
Table 1. Baseline Descriptive Statistics
n [ 50
Treated Aneurysms Baseline Characteristics
Male
20 (40%)
Location
Female
30 (60%)
ICA
28 (54.9%)
56.5 (46e67)
VA
9 (17.7%)
BA
9 (17.7%)
39 (78%)
MCA
2 (3.9%)
Asian
5 (10%)
ACA
2 (3.9%)
African American
2 (4%)
Number of Patients
Bifurcation
Sex
Age (years), median (IQR)
Race
Caucasian
PICA
Hispanic
2 (4%)
Pretreatment thrombosis 50%
Unknown
2 (4%)
Presentation
Multiple intracranial aneurysms
Family history of aneurysms
18 (36%)
2 (4%)
Smoking history
42 (82.4%)
6 (11.8%)
Ruptured (>2 weeks)
3 (5.9%)
26 (51%)
3 (5.9%)
26 (52%)
Past smokers
7 (14%)
Ischemia due to aneurysm
Current smokers
17 (34%)
Previous treatment
Endovascular treatment
22 (44%)
1e2
21 (42%)
Repeat flow diversion
8 (16%)
Coiling
13e15
2 (4%)
Comorbidities
Hypertension
28 (56%)
Hyperlipidemia
14 (28%)
Diabetes
11 (22%)
Coronary artery disease
6 (12%)
Prior aneurysmal subarachnoid hemorrhage
8 (16%)
Prior stroke (ischemic or hemorrhagic)
8 (16%)
Congestive heart failure
2 (4%)
Peripheral vascular disease
4 (8%)
Treated Aneurysms Baseline Characteristics
n [ 51
Morphology
Saccular
33 (64.7%)
Fusiform/dolichoectatic
16 (31.4%)
Other (multilobulated)
2 (3.9%)
Sidewall versus bifurcation aneurysms
Sidewall
41 (80.4%)
Continues
e166
www.SCIENCEDIRECT.com
6 (11.8%)
1 (2%)
Filling size (greatest dimension of contrast filling) (mm)*
13 (8.5
e17)
Axial size (greatest diameter on axial MRI or CT including
thrombosed portion) (mm)*
17 (12
e24.8)
48 (96%)
3e5
7 (13.7%)
Retreatment
0
Initial GCS
1 (2%)
18 (35.3%)
Ruptured (acute)
Never smokers
3e5
10 (19.6%)
Unruptured
Cranial nerve dysfunction
Pretreatment mRS
n [ 51
ACA, anterior cerebral artery; BA, basilar artery; CT, computed tomography; GCS, Glasgow
coma scale; ICA, internal carotid artery; IQR, interquartile range; MCA, middle cerebral
artery; MRI, magnetic resonance imaging; mRS, modified Rankin scale; PICA, posterior
inferior cerebellar artery; VA, vertebral artery.
*Measurements: median (IQR).
Leo Plus stent in 1 (2%), and a DERIVO device in 1 (2%). Thirtynine (76.5%) procedures used a single flow diverter and 11 (21.6%)
used 2 flow diverters. Pretreatment antiplatelet regimens included
aspirin and clopidogrel in 48 (94.1%) cases and aspirin and ticagrelor in 3 (5.9%) cases (Table 2). Adjunctive coiling was
performed in 16 (31.4%) cases, with complete neck coverage
obtained in 88.2% of the cases (in 60% of bifurcation aneurysms).
Ischemic neurologic complications occurred in 5 (9.8%) patients and produced symptoms in 4 (7.8%) with 2 mortalities.
Transient hemorrhagic neurologic complications occurred in 1
patient. The median length of hospital stay was 4 days (Table 2).
Radiographic and Clinical Outcomes
Last imaging follow-up was performed at a median of 25.1 (IQR,
12.8e43) months. Forty-eight of 51 aneurysms had radiographic
follow-up. Last follow-up imaging modality included DSA in 18
(37.5%), MRA in 26 (54.2%), and CTA in 4 (8.3%). The primary
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.11.084
ORIGINAL ARTICLE
PAUL M. FOREMAN ET AL.
FLOW DIVERSION FOR THROMBOSED ANEURYSMS
Table 2. Continued
Table 2. Procedural and Outcomes Details
Procedures Details
n [ 51
Larger
Flow diverters
PED
40 (78.4%)
SILK
4 (7.8%)
LVIS
4 (7.8%)
eCLIPS
1 (2%)
SILK þ Leo Plus
1 (2%)
DERIVO
1 (2%)
Number of flow diverters placed
1
39 (76.5%)
2
11 (21.6%)
3
1 (2%)
Flow diverters measurements
Length (mm), median (IQR)
Diameter (mm), median (IQR)
Platelet function tests
n [ 48*
Aneurysm Imaging Follow-Up
21 (18e30)
Last radiographic follow-up
elapsed time (months), median (IQR)
DSA
18 (37.5%)
MRA
26 (54.2%)
CTA
4 (8.3%)
Occlusion status
Completely occluded
37 (77.1%)
Incompletely occluded
9 (18.8%)
No change
1 (2.1%)
Larger
1 (2.1%)
Clinical Follow-Up
n ¼ 19
Time to last follow-up (months), median (IQR)
19 (100%)
Clopidogrel responders
16 (84.2%)
Pretreatment antiplatelet medication
Aspirin 325 mg þ clopidogrel 75 mg
48 (94.1%)
Aspirin 81 mg þ ticagrelor 180 mg
3 (5.9%)
25.1 (12.8e43)
Follow-up modality
4.2 (3.7e4.9)
Aspirin responders
3 (6.3%)
Last imaging follow-up
n [ 48*
24.3 (11.7e43.3)
mRS at last follow-up
0
33 (68.8%)
1e2
10 (20.8%)
3e5
3 (6.2%)
6
2 (4.2%)
Pretreatment anticoagulation
8 (15.7%)
Post-treatment anticoagulation
24 (47.1%)
Improved
15 (31.9%)
Adjunct coiling
16 (31.4%)
No change
28 (58.3%)
Balloon angioplasty
12 (23.5%)
Post-treatment ischemic complications
Symptomatic
5 (9.8%)
mRS change (pretreatment to last follow-up)
Worse
Cranial nerve dysfunction
5 (10.4%)
n ¼ 27
4 (7.8%)
Resolved
8 (29.6%)
Post-treatment hemorrhagic complications
1 (2%)
Improved
5 (18.5%)
Hospital length of stay (days), median (IQR)
4 (1e7)
No change
11 (40.7%)
Aneurysm Imaging Follow-Up
n [ 48*
First imaging follow-up
First radiographic follow-up elapsed
time (months), median (IQR)
Worse
Final disposition
3.8 (2.2e6.7)
Follow-up modality
DSA
21 (43.7%)
MRA
27 (56.3%)
Occlusion status
Completely occluded
22 (45.8%)
Incompletely occluded
18 (37.5%)
No change
5 (10.4%)
Continues
WORLD NEUROSURGERY 135: e164-e173, MARCH 2020
3 (11.1%)
n ¼ 48
Home
44 (91.7%)
Rehab
1 (2.1%)
Nursing home
1 (2.1%)
Dead
2 (4.2%)
CTA, computed tomography angiography; DSA, digital subtracted angiography; IQR,
interquartile range; MRA, magnetic resonance angiography; mRS, modified Rankin
scale; PED, Pipeline Embolization Device.
*Missing data for 2 patients.
outcome of complete occlusion at last radiographic follow-up was
achieved in 37 (77.1%) and incomplete occlusion in 9 (18.8%); 1
aneurysm was stable and 1 aneurysm increased in size. Six (11.8%)
www.journals.elsevier.com/world-neurosurgery
e167
ORIGINAL ARTICLE
PAUL M. FOREMAN ET AL.
FLOW DIVERSION FOR THROMBOSED ANEURYSMS
Figure 1. A 69-year-old woman with a partially thrombosed right ICA
aneurysm. (A) Pretreatment CTA. (B) Pretreatment DSA. (C) Follow-up
DSA at 6 months demonstrates continued aneurysm filling. (D) Follow-up
CTA at 20 months demonstrates continued aneurysm filling. (E) Follow-up
aneurysms were retreated with flow diverters after their initial
treatment with a flow diverter, whereas 1 aneurysm (2%) was
coiled (Figures 1 and 2; Table 2).
When univariate analysis was used to compare the group of
patients with complete aneurysm occlusion at last follow-up with
those without complete aneurysm occlusion, the complete occlusion group was significantly younger (median, 55 vs. 70 years,
P ¼ 0.015), less likely to have multiple aneurysms (29.7% vs.
72.7%, P ¼ 0.01), and more likely to be current smokers (40.5% vs.
9.1%, P ¼ 0.008). There was also a trend toward higher rates of
adjunctive coiling in patients with complete aneurysm occlusion
(21.6% vs. 9.1%, P ¼ 0.09). Similarly, procedural complete neck
coverage was obtained in 72.7% of the nonoccluded versus 91.9%
in the occluded cohort (P ¼ 0.09) (Table 3).
Last clinical follow-up was performed at a median of 24.3 (11.7e
43.3) months. Forty-eight of 50 patients had clinical follow-up.
Secondary outcomes included clinical status (functional
outcome, CN outcome, and disposition) and complication rates at
last follow-up. Functional status as measured by the mRS, as
compared with pretreatment, improved in 15 (31.9%), remained
e168
www.SCIENCEDIRECT.com
CTA at 32 months demonstrates continued aneurysm filling. CTA,
computed tomography angiography; DSA, digital subtracted angiography;
ICA, internal carotid artery.
stable in 28 (58.3%), and worsened in 5 (10.4%). Of the 27 aneurysms causing pretreatment CN dysfunction, 8 (29.6%)
resolved, 5 (18.5%) improved, 11 (40.7%) were stable, and 3
(11.1%) worsened. Forty-four (91.7%) patients had a final disposition of their own home, 1 to a rehabilitation center, 1 to a
nursing home, and 2 (4.2%) died from ischemic complications (1
immediate, 1 delayed) (Table 2).
Comparison of Aneurysms Based on Percentage of Pretreatment
Thrombosis
Pretreatment aneurysm thrombosis of 50% was found in 18
(35.3%) of aneurysms. Of these 18 aneurysms, 12 (66.7%) were
saccular in morphology. The median greatest dimension of
contrast filling was 17 (IQR, 12.5e30) mm with a median greatest
axial diameter on MRI or CT of 17 (IQR, 14e28) mm. When
compared with aneurysms with <50% thrombosis, pretreatment
CN dysfunction was less common (33.3% vs. 60.6%, P ¼ 0.063)
and complete occlusion at last radiographic follow-up was
significantly less common (58.8% vs. 87.1%, P ¼ 0.026). There
was a trend toward better functional outcome (mRS 2) at last
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.11.084
ORIGINAL ARTICLE
PAUL M. FOREMAN ET AL.
Figure 2. A 77-year-old woman with partially thrombosed right ACA
aneurysm. (A) Pretreatment DSA. (B) DSA after adjunctive coiling. (C) DSA
after deployment of a PED in the A1 segment of the ACA, spanning the
neck of the aneurysm. (D) Follow-up DSA at 6 months demonstrates
follow-up in patients with <50% pretreatment aneurysm thrombosis (96.8 vs. 82.4, P ¼ 0.08) (Table 4).
Multivariable Predictors of Incomplete Occlusion
Based on potential predictors on univariate screening (Table 3), a
multivariable logistic regression model was constructed to identify
predictors of incomplete occlusion, controlling for amount of
aneurysmal thrombosis (>50% vs. <50%), age, maximal
diameter, smoking history, and procedural complete neck
coverage. Age (1-year increments) maintained its independent
significance (P ¼ 0.05), whereas amount of aneurysmal thrombosis trended in the same direction but lost independent significance (P ¼ 0.09), likely in the setting of a small sample size being
underpowered to detect significance. Similarly, 1-mm increments
increase in maximal diameter trended toward incomplete occlusion but did not reach statistical significance (P ¼ 0.06). Of note,
the results of the multivariable model should be interpreted
WORLD NEUROSURGERY 135: e164-e173, MARCH 2020
FLOW DIVERSION FOR THROMBOSED ANEURYSMS
complete aneurysm occlusion. (E) 3D reconstruction of DSA demonstrates
complete aneurysm occlusion. ACA, anterior cerebral artery; DSA, digital
subtracted angiography; PED, Pipeline Embolization Device.
tentatively given the small sample size and even smaller number of
events (i.e., incomplete occlusion), which can limit the stability of
the model. We tried to avoid overfitting by including 5 covariates
only for 50 observations (following the rule of thumb of minimum
of 10 subjects for each model covariate, while trying to account for
interactions and collinearity between the variables). However, we
had to eliminate other literature-known potential important predictors such as gender, fusiform morphology, and circulation
(anterior vs. posterior) (Table 5).
DISCUSSION
This study represents the largest series of PTIA treated with
placement of flow diverting stents to date. Inclusion of 50 patients
across 6 cerebrovascular treatment centers highlights the rarity of
the diagnosis. Thromboembolic complications occurred in 9.8%,
with a fatal ischemic stroke in 2 patients. Complete aneurysm
www.journals.elsevier.com/world-neurosurgery
e169
ORIGINAL ARTICLE
PAUL M. FOREMAN ET AL.
FLOW DIVERSION FOR THROMBOSED ANEURYSMS
Table 3. Comparison According to Last Follow-Up Aneurysm Occlusion Status
Complete Occlusion, 37 (77.1%)
Incomplete Occlusion, 11
(22.9%)
P Value
Age (years), median (IQR)
55 (46e63)
70 (62e70)
0.015
Female sex
23 (62.2%)
7 (63.6%)
0.9
Saccular
24 (64.9%)
8 (72.7%)
0.47
Fusiform/dolichoectatic
12 (32.4%)
2 (18.2%)
Variable
Morphology
1 (2.7%)
1 (9.1%)
Filling size (greatest dimension of contrast filling) (mm)*
Multilobulated
12 (8e17)
14 (11e20)
0.49
Axial size (greatest diameter on axial MRI or CT including thrombosed portion) (mm)*
16 (12e22)
22.5 (17e28)
0.065
27 (73%)
4 (36.4%)
0.026
Pretreatment aneurysmal thrombosis
<50%
>50%
Multiple aneurysms
Family history of aneurysms
10 (27%)
7 (63.6%)
11 (29.7%)
8 (72.7%)
0.01
1 (2.7%)
1 (9.1%)
0.35
19 (51.4%)
5 (45.5%)
0.008
Smoking history
Never smokers
Past smokers
3 (8.1%)
5 (45.5%)
15 (40.5%)
1 (9.1%)
19 (51.4%)
6 (54.6%)
0.85
None
32 (86.5%)
10 (90.9%)
0.7
Endovascular treatment
Current smokers
Cranial nerve dysfunction at presentation
Previous treatment history
5 (13.5%)
1 (9.1%)
Adjunct coiling
8 (21.6%)
1 (9.1%)
0.09
Complete neck coverage after deployment
34 (91.1%)
8 (72.7%)
0.09
Outcomes
Time to last imaging
follow-up (months), median (IQR)
25.4 (11.9e42.1)
24.8 (14.5e49.5)
0.45
2
35 (94.6%)
9 (81.8%)
>2
2 (5.4%)
2 (18.2%)
0 (0%)
6 (54.5%)
<0.001
3 (8.1%)
1 (9.1%)
0.92
0 (0%)
0 (0%)
NA
mRS on last follow-upy
Retreatment
Post-treatment ischemic complications
Post-treatment hemorrhagic complications
0.18
Bold value denotes statistical significance; i.e., P < 0.05.
CT, computed tomography; IQR, interquartile range; MRI, magnetic resonance imaging; mRS, modified Rankin scale.
*Measurements: median (IQR); P value: Mann-Whitney test.
yMissing data for 2 patients.
occlusion was observed in 77.1% of aneurysms at a median followup of 25.1 months. Aneurysms with 50% thrombosis noted on
pretreatment imaging were associated with lower rates of complete occlusion, despite losing significance on multivariable likely
to underpowering. Interestingly, of all the 9 ruptured aneurysms
e170
www.SCIENCEDIRECT.com
were <50% thrombosed (27.3% vs. 0%, P ¼ 0.015), suggesting
that risk of rupture might be independent of the amount of
thrombus harbored in the aneurysm, despite this group having
potentially more favorable angiographic outcome in response to
flow diversion.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.11.084
ORIGINAL ARTICLE
PAUL M. FOREMAN ET AL.
FLOW DIVERSION FOR THROMBOSED ANEURYSMS
Table 4. Comparison According to Pretreatment Aneurysm Thrombosis
‡50%, 18 (35.3%)
<50%, 33 (64.7%)
P Value
Age (years), median (IQR)
65 (59e73)
53 (46e61)
0.01
Female sex
13 (72.2%)
18 (54.6%)
0.67
Saccular
12 (66.7%)
21 (63.6%)
0.8
Fusiform/dolichoectatic
5 (27.8%)
11 (33.3%)
Multilobulated
1 (5.6%)
1 (3%)
17 (12.5e30)
15 (9e17)
0.86
17 (14e28)
18 (12e22)
0.56
Multiple aneurysms
7 (38.9%)
12 (36.4%)
0.86
Family history of aneurysms
1 (5.6%)
1 (3%)
0.65
Never smokers
10 (55.6%)
16 (48.5%)
0.8
Past smokers
3 (16.7%)
5 (15.2%)
Current smokers
5 (27.8%)
12 (36.7%)
6 (33.3%)
20 (60.6%)
0.063
0 (0%)
9 (27.3%)
0.015
17 (94.4%)
27 (81.8%)
0.2
1 (5.6%)
6 (18.2%)
4 (22.2%)
12 (36.4%)
Variable
Morphology
Filling size (greatest dimension of
contrast filling) (mm)*
Axial size (greatest diameter on axial
MRI or CT including thrombosed portion) (mm)*
Smoking history
Cranial nerve dysfunction at presentation
Ruptured at presentation
Previous treatment history
None
Endovascular treatment
Adjunct coiling
0.3
Outcomes
Time to last imaging follow-up
(months), median (IQR)
24.3 (11.8e33.1)
28.5 (13.6e46.5)
0.2
Complete occlusion
10 (58.8%)
27 (87.1%)
0.026
Incomplete occlusion
7 (41.2%)
4 (12.9%)
2
14 (82.4%)
30 (96.8%)
>2
3 (17.6%)
1 (3.2%)
Occlusion at last imaging follow-upy
mRS on last follow-upy
0.08
Retreatment
4 (22.2%)
3 (9.1%)
0.19
Post-treatment ischemic complications
2 (11.1%)
3 (9.1%)
0.81
0 (0%)
1 (3%)
0.46
Post-treatment hemorrhagic complications
n (%); P value: c2 test.
Bold values: P value <0.05.
CT, computed tomography; IQR, interquartile range; MRI, magnetic resonance imaging; mRS, modified Rankin scale.
*Measurements: median (IQR); P value: Mann-Whitney test.
yMissing data for 2 patients.
The pathophysiology of PTIA is incompletely understood, yet
believed distinct from classic “berry aneurysms” with abluminal
factors playing an important role in their development and
WORLD NEUROSURGERY 135: e164-e173, MARCH 2020
growth.9
Adventitial
inflammation
with
release
of
proinflammatory factors into the arterial wall media leads to
degradation of the extracellular matrix and elastic lamina, and
www.journals.elsevier.com/world-neurosurgery
e171
ORIGINAL ARTICLE
PAUL M. FOREMAN ET AL.
FLOW DIVERSION FOR THROMBOSED ANEURYSMS
Table 5. Logistic Regression Model for Odds of PTIA Incomplete Occlusion After Flow Diversion, Controlling for Preaneurysmal
Thrombosis, Age, Size, Smoking, and Procedural Neck Coverage
Variable
OR
95% CI
P Value
Pretreatment aneurysmal thrombosis (>50% vs. <50%)
3.5
0.83e20.8
0.09
Age, 1-year increments
1.08
1.000e1.16
0.05
Maximal diameter, 1-mm increments
1.07
0.995e1.15
0.06
Smoking
2.1
0.37e12.1
0.4
Procedural complete neck coverage
0.16
0.017e1.47
0.1
Bold value denotes statistical significance; i.e., P < 0.05.
CI, confidence interval; OR, odds ratio; PTIA, partially thrombosed intracranial aneurysms.
initiates the multistep process of aneurysm formation.9 Recurrent
microdissections of the weakened pathogenic arterial wall then
result in subadventitial hemorrhage and intramural thrombus
with subsequent formation of intraluminal thrombus.1,10 Relative
contributions of the vasa vasorum, perfused intrathrombotic
clefts, and proliferating intrathrombotic endothelial channels to
aneurysm growth and symptomatology are debated.1,11-14 The
unique pathophysiology of PTIA carries potential therapeutic implications and could underlie the unsatisfactory results of surgical
reconstruction and endovascular embolization with detachable
coils.
Surgical management of PTIA involves clip reconstruction with
or without aneurysm thrombectomy and parent artery/aneurysm
occlusion with or without arterial bypass. These cases present
unique technical challenges due to intra-aneurysmal thrombus
and are associated with significant morbidity and mortality, even
in experienced hands.2 Cases amendable to direct surgical
clipping produce the best surgical results.2 However, defining
the lesional arterial segment intraoperatively remains
challenging and can lead to inclusion of pathogenic vessel
segments within the reconstructed parent artery, increasing the
rate of recurrence.10
Standard endovascular treatment options for PTIA are less
technically challenging, and include coil embolization and
parent artery occlusion after balloon test occlusion. Selective
coil embolization of these aneurysms has yielded disappointing
results with recanalization rates as high as 78%, often due to
coil migration into the intraluminal thrombus necessitating
retreatment5,6; continued aneurysm growth resulting in a high
mortality rate was also reported.6 Although continued
aneurysm growth after parent artery occlusion has been
reported,13 strategies that include parent artery occlusion have
yielded superior results when compared with strategies that
address only the perfused aneurysmal sac.6,10 Parent artery
occlusion resulted in significantly higher rates of neurologic
improvement, aneurysm size reduction, and durable aneurysm
occlusion.6,10 Despite encouraging results, parent artery
occlusion is not always tolerated and can be precluded by
aneurysm location.
Endovascular parent artery reconstruction with flow diverting
stents is a promising treatment modality for PTIA. The technique
involves placement of the flow diverter within the parent artery
e172
www.SCIENCEDIRECT.com
avoiding potential hazards of entering or manipulating the
aneurysm. In addition, aneurysm occlusion is ultimately achieved
by endothelialization of the stent across the neck of the aneurysm, therefore overcoming the recanalization rates seen with
detachable coils. Safety and efficacy of flow diversion for the
treatment of large and giant aneurysms has been previously
demonstrated,15,16 with 1- and 5-year angiographic occlusion rates
of 87% and 95%, respectively, in the Pipeline for Uncoilable or
Failed Aneurysm Trial17; however, PITA constituted only 15.7% of
the Pipeline for Uncoilable or Failed Aneurysm Trial population,
and the outcomes were not specifically assessed in this subgroup.
Resolution of radiographic mass effect with improvement of CN
compression syndromes, including patients with large and giant
PTIA, has also been reported.18,19 The present series observed a
complete occlusion rate of 77% over a median follow-up of 25
months, with 45% of patients experiencing improvement of
pretreatment CN dysfunction. Underlying differences in the
pathophysiology of PTIA potentially account for the reduced efficacy of flow diversion when compared with other series. Postmortem histopathology of large aneurysms treated with flow
diversion has supported the theory that intra-aneurysmal
thrombus induces transient destabilization of the aneurysm
wall via mechanisms of hypoxia, inflammation, and enzymatic
degredation.20 Erythrocyte-rich thrombus, in particular, is known
to promote an inflammatory milieu of lytic enzymes produced by
activated leukocytes.21 This thrombus-initiated inflammatory
process potentially retards progressive endothelialization of the
flow diverting stent leading to higher rates of incomplete occlusion in aneurysms with a large amount of pretreatment
thrombus. Long-term follow-up is necessary to determine if the
observed trend of progressive occlusion over time will be seen in
PTIA.
Limitations
The present study is limited by its retrospective design, nonstandardized follow-up, and lack of independently adjudicated
outcomes. The retrospective design of the study allows for patient
selection bias with the decision to treat a particular PTIA left to the
discretion of the treating physician. Lack of standardized followup is of particular interest with flow diversion treatment of aneurysms due to the progressive, time-dependent nature of aneurysm occlusion. Retrospective studies that lack independent
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.11.084
ORIGINAL ARTICLE
PAUL M. FOREMAN ET AL.
FLOW DIVERSION FOR THROMBOSED ANEURYSMS
adjudication of outcomes tend to overestimate safety and efficacy
in surgical and interventional series. The inclusion of cases from 5
international centers allowed us to amass a modest number of
cases for a rare disease process and improves the generalizability
of the study. The broad inclusion criteria allowed for a practical
and comprehensive evaluation of flow diversion for the treatment
of PTIA.
CONCLUSIONS
Flow diversion treatment of PTIA carries adequate efficacy along
with a reasonable safety profile. Aneurysms with 50% pretreatment thrombosis were associated with lower rates of complete
occlusion. The unique pathophysiology of PTIA development and
growth is hypothesized to contribute to the reduced efficacy when
compared with treatment of “berry aneurysms.” Considering the
challenges of treating large and giant partially thrombosed
aneurysms, the results support the use of flow diverting stent for
these aneurysms.
ACKNOWLEDGMENTS
Paul M. Foreman contributed to conceptualization, methodology
writing, and reviewing; Mohamed M. Salem to methodology,
writing, software, and formal analysis; Christoph J. Griessenauer
to methodology and data curation; Adam A. Dmytriw, Carmen
Parra-Farinas, Patrick Nicholson, Nicola Limbucci, Anna Luisa
Kühn, and Ajit S. Puri to data curation; Leonardo Renieri to
visualization and reviewing; Sergio Nappini to visualization,
reviewing, and resources; Kimberly P. Kicielinski and Alejandro
Bugarini to reviewing and data curation; Vitor Mendes Pereira to
editing; Thomas R. Marotta and Clemens M. Schirmer to investigation and supervision; Christopher S. Ogilvy to conceptualization and resources; and Ajith J. Thomas contributed to
conceptualization, supervision, and resources.
delayed aneurysm rupture after flow-diversion
treatment. AJNR Am J Neuroradiol. 2011;32:20-25.
9. Krings T, Piske RL, Lasjaunias PL. Intracranial
arterial aneurysm vasculopathies: targeting the
outer vessel wall. Neuroradiology. 2005;47:931-937.
17. Becske T, Brinjikji W, Potts MB, et al. Long-term
clinical and angiographic outcomes following
Pipeline Embolization Device treatment of complex internal carotid artery aneurysms: five-year
results of the Pipeline for Uncoilable or Failed
Aneurysm Trial. Neurosurgery. 2017;80:40-48.
10. Yang K, Park JC, Ahn JS, Kwon DH, Kwun BD,
Kim CJ. Characteristics and outcomes of varied
treatment modalities for partially thrombosed
intracranial aneurysms: a review of 35 cases. Acta
Neurochir (Wien). 2014;156:1669-1675.
18. Szikora I, Marosfoi M, Salomvary B, Berentei Z,
Gubucz I. Resolution of mass effect and
compression symptoms following endoluminal
flow diversion for the treatment of intracranial
aneurysms. AJNR Am J Neuroradiol. 2013;34:935-939.
3. Guresir E, Wispel C, Borger V, Hadjiathanasiou A,
Vatter H, Schuss P. Treatment of partially
thrombosed intracranial aneurysms: single-center
series and systematic review. World Neurosurg.
2018;118:e834-e841.
11. Schubiger O, Valavanis A, Wichmann W. Growthmechanism of giant intracranial aneurysms;
demonstration by CT and MR imaging. Neuroradiology. 1987;29:266-271.
19. Zanaty M, Jabbour PM, Bou Sader R, et al. Intraaneurysmal thrombus modification after flowdiversion. J Clin Neurosci. 2015;22:105-110.
4. Cho YD, Park JC, Kwon BJ, Hee Han M. Endovascular treatment of largely thrombosed saccular
aneurysms: follow-up results in ten patients.
Neuroradiology. 2010;52:751-758.
12. Nagahiro S, Takada A, Goto S, Kai Y, Ushio Y.
Thrombosed growing giant aneurysms of the
vertebral artery: growth mechanism and management. J Neurosurg. 1995;82:796-801.
5. Kim SJ, Choi IS. Midterm outcome of partially
thrombosed intracranial aneurysms treated with
guglielmi detachable coils. Interv Neuroradiol. 2000;
6:13-25.
13. Iihara K, Murao K, Sakai N, et al. Continued
growth of and increased symptoms from a
thrombosed giant aneurysm of the vertebral artery
after complete endovascular occlusion and trapping: the role of vasa vasorum. Case report.
J Neurosurg. 2003;98:407-413.
REFERENCES
1. Krings T, Alvarez H, Reinacher P, et al. Growth
and rupture mechanism of partially thrombosed
aneurysms. Interv Neuroradiol. 2007;13:117-126.
2. Lawton MT, Quinones-Hinojosa A, Chang EF,
Yu T. Thrombotic intracranial aneurysms: classification scheme and management strategies in 68
patients. Neurosurgery. 2005;56:441-454 [discussion: 441-454].
6. Ferns SP, van Rooij WJ, Sluzewski M, van den
Berg R, Majoie CB. Partially thrombosed intracranial aneurysms presenting with mass effect:
long-term clinical and imaging follow-up after
endovascular treatment. AJNR Am J Neuroradiol.
2010;31:1197-1205.
7. Chow M, McDougall C, O’Kelly C, Ashforth R,
Johnson E, Fiorella D. Delayed spontaneous
rupture of a posterior inferior cerebellar artery
aneurysm following treatment with flow diversion: a clinicopathologic study. AJNR Am J Neuroradiol. 2012;33:E46-E51.
8. Kulcsar Z, Houdart E, Bonafe A, et al. Intraaneurysmal thrombosis as a possible cause of
14. Yasui T, Sakamoto H, Kishi H, et al. Rupture
mechanism of a thrombosed slow-growing giant
aneurysm of the vertebral artery—case report.
Neurol Med Chir (Tokyo). 1998;38:860-864.
15. Becske T, Kallmes DF, Saatci I, et al. Pipeline for
Uncoilable or Failed Aneurysm: results from a
multicenter clinical trial. Radiology. 2013;267:
858-868.
16. Nelson PK, Lylyk P, Szikora I, Wetzel SG,
Wanke I, Fiorella D. The Pipeline Embolization
Device for the intracranial treatment of aneurysms
trial. AJNR Am J Neuroradiol. 2011;32:34-40.
WORLD NEUROSURGERY 135: e164-e173, MARCH 2020
20. Chyatte D, Bruno G, Desai S, Todor DR. Inflammation and intracranial aneurysms. Neurosurgery.
1999;45:1137-1146 [discussion: 1146-1137].
21. Turowski B, Macht S, Kulcsar Z, Hanggi D,
Stummer W. Early fatal hemorrhage after endovascular cerebral aneurysm treatment with a flow
diverter (SILK-Stent): do we need to rethink our
concepts? Neuroradiology. 2011;53:37-41.
Conflict of interest statement: M. M. Salem is a former
recipient of research financial support from Medtronic Inc.,
Cerebrovascular Group. The rest of the authors have no
personal, financial, or institutional interest with regard to
the authorship and/or publication of this manuscript.
Received 28 August 2019; accepted 14 November 2019
Citation: World Neurosurg. (2020) 135:e164-e173.
https://doi.org/10.1016/j.wneu.2019.11.084
Journal homepage: www.journals.elsevier.com/worldneurosurgery
Available online: www.sciencedirect.com
1878-8750/$ - see front matter ª 2019 Published by Elsevier
Inc.
www.journals.elsevier.com/world-neurosurgery
e173