ORIGINAL RESEARCH

INTERVENTIONAL

How Flow Reduction Influences the Intracranial Aneurysm

Occlusion: A Prospective 4D Phase-Contrast MRI Study
O. Brina, P. Bouillot, P. Reymond, A.S. Luthman, C. Santarosa, M. Fahrat, K.O. Lovblad, P. Machi,
B.M.A. Delattre, V.M. Pereira, and M.I. Vargas

ABSTRACT
BACKGROUND AND PURPOSE: Flow-diverter stents are widely used for the treatment of wide-neck intracranial aneurysms. Various
parameters may influence intracranial aneurysm thrombosis, including the flow reduction induced by flow-diverter stent implanta-
tion, which is assumed to play a leading role. However, its actual impact remains unclear due to the lack of detailed intra-aneurys-
mal flow measurements. This study aimed to clarify this relationship by quantitatively measuring the intra-aneurysmal flow using
4D phase-contrast MR imaging.

MATERIALS AND METHODS: We acquired prospective pre- and post-stent implantation 4D phase-contrast MR imaging data of a
consecutive series of 23 patients treated with flow-diverter stents. Velocity field data were combined with the intraprocedural 3D
angiogram vessel geometries for precise intracranial aneurysm extraction and partial volume correction. Intra-aneurysmal hemody-
namic modifications were compared with occlusion outcomes at 6 and 12 months.

RESULTS: The averaged velocities at systole were lower after flow-diverter stent implantation for all patients and ranged from
21.7 6 7.1 cm/s before to 7.2 6 2.9 cm/s after stent placement. The velocity reduction was more important for the group of
patients with aneurysm thrombosis at 6 months (68.8%) and decreased gradually from 66.2% to 55% for 12-month thrombosis and
no thrombosis, respectively (P = .08).

CONCLUSIONS: We propose an innovative approach to measure intracranial flow changes after flow-diverter stent implantation.
We identified a trend between flow reduction and thrombosis outcome that brings a new insight into current understanding of
the flow-diversion treatment response.

ABBREVIATIONS: CFD 4 computational fluid dynamics; 3DRA 4 3D rotational angiography; FDS 4 flow-diverter stent; IA 4 intracranial aneurysm;
PCMR 4 phase-contrast MRI; PVRR 4 proportional velocity-reduction ratio; QICA 4 ICA systolic flow rates ratio; QICA 4 ICA mean flow rates; Velan 4 intra-
aneurysmal velocity; VENC 4 velocity encoding

F low-diverter stents (FDSs) are a widely used option for the

endovascular treatment of large-neck intracranial aneurysms
(IAs).1-4 The high density of stent struts across the IA neck
symptomatic lesions due to the progressive decompression of
surrounding tissues as the aneurysm shrinks.8 In addition, com-
pared with other endovascular procedures, FDS implantation
dampens the intrasaccular flow and promotes a progressive yields a higher rate of complete and permanent aneurysm exclu-
thrombosis of the cavity to ultimately exclude the aneurysm from sion, considering both residual and complete recanalization.2,9,10
the circulation.5,6 FDSs also have other advantages such as vessel Furthermore, from an interventional point of view, FDS implan-
wall remodeling of the parent artery, often altered in large-neck tation procedures are rapid and avoid the risky penetration of
ICA aneurysms,7 as well as the improvement of outcomes for the aneurysm sac with embolization material, though the deliv-
ery of the device requires appropriate training.11 However, the
physiologic mechanisms leading to aneurysmal occlusion are
Received May 21, 2019; accepted after revision September 20.
From the Divisions of Neuroradiology (O.B., P.R., A.S.L., C.S., K.O.L., P.M., V.M.P.,
M.I.V.) and Radiology (B.M.A.D.), Geneva University Hospitals, University of Geneva, Please address correspondence to Maria Isabel Vargas, MD, Geneva University
Geneva, Switzerland; Department of Quantum Matter Physics (P.B.), University of Hospitals, Division of Diagnostic and Interventional Neuroradiology, Rue Gabrielle-
Geneva, Geneva, Switzerland; Laboratory for Hydraulic Machines (M.F.), Ecole Perret-Gentil 4, 1211 Genève 14; e-mail: maria.i.vargas@hcuge.ch
Polytechnique Fédérale de Lausanne, Lausanne, Switzerland; Division of
Neuroradiology (V.M.P.), Department of Medical Imaging (V.M.P.), and Division of Indicates open access to non-subscribers at www.ajnr.org
Neurosurgery (V.M.P.), Department of Surgery, Toronto Western Hospital, Indicates article with supplemental on-line appendix and table.
University Health Network, Toronto, Ontario, Canada.
Indicates article with supplemental on-line photos.
This work was supported by Swiss National Science Foundation grants (SNF 32003B
160222 and SNF 320030 156813). http://dx.doi.org/10.3174/ajnr.A6312

AJNR Am J Neuroradiol 40:2117–23 Dec 2019 www.ajnr.org 2117

complex, and many factors such as hemodynamics, antiaggrega- up for device sizing and to determine the working projection
tion therapy individual responses, and biologic factors are associ- views. Given its high spatial resolution and enhanced contrast,
ated with them. This complexity may translate into posttreatment 3DRA was also used during postprocessing to segment the vessel
rupture12,13 and delayed aneurysm occlusion, thus resulting in geometry and separate the aneurysm from the parent vessel. At
patients with a suboptimally treated lesion.9,10 It is hypothesized the end of the procedure, a contrast-enhanced conebeam CT was
that the flow reduction after FDS implantation has a strong performed to verify the apposition of the device to the vessel wall.
impact on occlusion likelihood and may play a role in delayed
rupture, but this relationship is yet to be established. Data Postprocessing
At present, the quantitative assessment of aneurysmal flow Although 4D-PCMR provides the 3D velocity vector field over
changes relies on 3 main methods: 1) video densitometry based on time in the entire acquisition volume, preliminary postprocessing
DSA time-series,14,15 2) blood flow modeling using computational steps are required to prepare the raw datasets for the calculation
fluid dynamics (CFD),16-18 and 3) 2D and 4D phase-contrast MR of the velocity field within the vessel lumen. Thus, we combined
imaging (4D-PCMR). The latter presents the major advantage of the 4D-PCMR velocities with the vessel geometric information
being the only quantitative method for the in vivo measurement provided by the 3DRA. More details are provided in the On-line
of 3D velocity fields over time, and some neurovascular disorders Appendix and Bouillot et al.24
have been addressed with this technique, such as arteriovenous
malformation hemodynamics.19 Regarding IAs treated with FDSs, Proportional Velocity-Reduction Ratio Calculation
only a limited number of in vitro studies20 have been performed, Aneurysmal flow modification induced by the FDS implantation was
and even fewer in vivo studies have included a maximum of 10 quantified with the proportional velocity-reduction ratio (PVRR).22
patients. This lack of data has prevented determination of any cor- We focused on the systolic phase with the highest velocity range,
relation with the resultant treatment outcomes.21,22 minimizing the impact of the measurement uncertainties affecting
The sparse literature available may be due to the stent- low velocities. The PVRR expresses the rate of intra-aneurysmal ve-
induced metallic artifacts, which do not make 4D-PCMR intui- locity reduction normalized with the ICA flow rate as follows:
tively indicated for post-treatment acquisitions. However, we Velan QICA;Pre
have shown that these metallic artifacts are limited to the lumen PVRR ¼ 1   ;
Velan QICA;Post
of the stent without extension to the aneurysm bulge, which
guarantees the accuracy of the velocities measured in the aneu- Where Velan,Pre and Velan,Post are intra-aneurysmal velocities
rysm volume.23 On the basis of these results, we aimed to mea- averaged spatially at the systolic time-step before and after FDS
sure pre- and post-FDS velocity fields with 4D-PCMR in a implantation, respectively. The normalization with the ICA sys-
prospective patient cohort and to correlate hemodynamic tolic flow rates ratio, QICA,Pre/QICA,Post, aimed to correct for the
changes with aneurysm occlusion outcomes at 6 and 12 months. potential differences in physiologic conditions between the 2 MR
imaging examinations. The PVRR was calculated using the fol-
MATERIALS AND METHODS lowing methodology for aneurysm velocity Velan and ICA flow-
Patient Selection rate extraction QICA:
We prospectively included patients with unruptured saccular ICA
aneurysms treated with FDS only. We excluded patients with Aneurysm Velocities (VELAN). The aneurysm was disconnected
partly thrombosed aneurysms. Our treatment strategy consisted from the parent vessel by removing the mesh cells of the circulat-
of placing a single layer of flow diverter, without coil association. ing volume located within a diameter of 1 mm larger than the
The study was approved by Geneva University Hospitals institu- actual parent vessel diameter. This exclusion criterion defined
tional ethics committee (NEC 07–056). All patients provided writ- from the vessel center line (Fig 1A.2) aimed to include the outer
ten informed consent. bounds of the stent struts covering the neck. When an artery ori-
ginated from the aneurysm, elements were interactively removed
MR Imaging Acquisition to disconnect the vessel from its origin. Once extracted, the veloc-
MR imaging examinations were performed the day before and ities of the aneurysm volume at the systolic phase were averaged
within 48 hours following the stent implantation procedure with spatially to compute the Velan. We excluded patients with aneur-
an Ingenuity TF PET/MR (Achieva 3 T TX series MR imaging ysms presenting more than half of their volume with unreliable
system; Philips Healthcare, Best, the Netherlands) and an 8-chan- velocities with Velan , 7.7 cm/s before FDS implantation. This
nel head coil. The circle of Willis was imaged using a 3D-TOF threshold corresponds to the measurement error previously iden-
sequence on which the 4D-PCMR slices were positioned in a sag- tified and reported in Pereira et al22 and Bouillot et al.23
ittal/oblique orientation to cover both the aneurysm and its adja-
cent parent vessel. The acquisition parameters are detailed in the Flow Rates (QICA). As described in Bouillot et al,24 measurement
On-line Appendix.22,23 planes, each separated by 2 mm, were automatically placed or-
thogonal to the vessel center line within a user-selected distance
Endovascular Procedure corresponding to the C2–3 segment (Fig 1C.1). In each plane,
Procedures were performed with a biplane angiosuite (Allura the 3D-PCMR data velocities were interpolated within the
FD20; Philips Healthcare). 3D rotational angiography (3DRA) boundaries of the vessel provided by the 3DRA (Fig 1C.2). The
was systematically acquired as part of the preimplantation work- instantaneous flow rate was computed after partial volume
2118 Brina Dec 2019 www.ajnr.org
FIG 1. Upper row, Aneurysm extraction for pre- and poststent acquisitions (A.1 and B.1), their 3D velocity fields (A.3 and B.3), and their systolic
Velan values, respectively. The extraction method is illustrated in A.2, where the aneurysm bulge is isolated from the parent artery by removing
all the mesh cells within a diameter Dvessel þ1 mm around the center line. B.1 and B.2, The conebeam CT of the implanted Silk (radio-opaque
markers segmented in blue) is registered with the 3DRA geometry to show its clear separation from the extracted aneurysm. Lower row, QICAs
(C.3) measured on orthogonal planes positioned along the vessel center line (C.1), with partial volume correction (C.2) as described in Bouillot
et al.24

correction and subsequently averaged over the measurement RESULTS

planes (Fig 1C.3). We included 28 consecutive patients from January 2012 to
December 2017. All patients were successfully treated with the
following FDSs: Silk (Balt Extrusion, Montmorency, France)
Qualitative Evaluation of the Intra-Aneurysmal Velocities
(n = 7); Pipeline Embolization Device (PED; Covidien, Irvine,
Pre- and poststent flow patterns of each patient were qualitatively
California) (n = 17); and the Flow-Redirection Endoluminal
evaluated by means of streamlines using Paraview software
Device (FRED; MicroVention, Tustin, California) (n = 4). Five
(http://paraview.org).
patients received 2 devices to either extend the coverage length
or improve the wall apposition of the stent at the landing zone.
Follow-Up and Statistical Analysis Pre- and post-4D-PCMR sequences were successfully acquired
Patient follow-up was performed by MR imaging examinations for all patients, representing 56 MR imaging examinations. Two
at 6 and 12 months after FDS implantation. At 12 months, an patients were excluded from the study because follow-up imag-
angiogram was obtained to either confirm the complete throm- ing was not complete at 6 or 12 months. Three patients present-
bosis of the aneurysm or assess the need for retreatment in ing with .50% of their aneurysmal volumes with Velan below
the case of absent or incomplete thrombosis. Imaging records 7.7 cm/s were excluded, including 1 patient with a double stent.
were reviewed by an experienced neuroradiologist (M.I.V.). For the remaining 23 patients, the rate of aneurysm occlusion
Outcomes were labeled as follows: complete thrombosis at increased from 60.9% (n = 14) at 6 months to 82.6% (n = 19) at
6 months, complete thrombosis at 12 months, and no or par- 12-month follow-up. Four aneurysms remained patent at
tial thrombosis at 12 months. Differences between PVRR and 12 months. These results are slightly lower than the reported
thrombosis outcomes for the 3 groups were evaluated using occlusion rates of 73.6% and 86.8% for 6-month and 1-year fol-
the Kruskal-Wallis test. Statistical results are presented as low-up, respectively.1
mean 6 SD. Statistical analysis was performed in Matlab
R2017A (MathWorks, Natick, Massachusetts). Finally, we an- Flow-Reduction and Thrombosis Outcomes
alyzed the potential relationships between geometric charac- On average, the ICA mean flow rates (QICA) were not signifi-
teristics of the aneurysm and PVRR by measuring the cantly different before and after the procedure (Fig 2A; QICA,Pre =
volume, maximum size, aspect ratio, and neck size on 3DRA 3.54 6 0.7 mL/s and QICA,Post = 3.64 6 0.7 mL/s; P = .66). By con-
data. trast, the intra-aneurysmal systolic velocities, Velan, were reduced
AJNR Am J Neuroradiol 40:2117–23 Dec 2019 www.ajnr.org 2119
FIG 2. Descriptive statistics. A, ICA mean flow rates with no significant differences between pre- and poststent MR imaging acquisitions. B,
Intra-aneurysmal systolic velocities show a wide range before stent placement and converging toward a narrower range after flow diversion. C,
The PVRR for patients thrombosed at 6 and 12 months and not thrombosed at 12 months. Blue cross dots represent patients implanted with 2
stents. Thromb indicates thrombosis.

though not quantified, appears to be

not necessarily associated with higher
PVRRs as could be expected. No corre-
lations between the geometric character-
istics and flow reduction or occlusion
outcomes were observed (R2 = 0.24,
0.27, 0.02, 0.06 for volume, maximum
size, aspect ratio, and neck size, respec-
tively). More details are provided in the
On-line Appendix.

Qualitative Analysis of the

Velocity Vector Fields
Intra-aneurysmal velocity patterns
were modified by the stent either in
magnitude and/or direction. However,
these modifications were not related
to the thrombosis outcome as illus-
trated for 2 pairs of patients in Fig 3.
In each pair, the patients presented
with close PVVRs and flow-pattern
FIG 3. Pre-/post-FDS implantation streamlines at the systolic phase within the 3DRA geometries
behaviors, but with 6- and 12-month
(in transparent rendering) of 4 illustrative patients. The deployed stents imaged with vasoCT
(Philips Healthcare) are added on the poststent data. The upper row shows 2 patients (A and B) thrombosis times, respectively. For
both with strong modifications of the velocity patterns and flow reduction (PVRR  70%), leading patients A and B, flow patterns were
to different thrombosis outcomes. The lower row shows 2 other patients (C and D) with strongly modified by the stent,
unchanged velocity patterns and lower flow reduction (PVRR  55%), leading to different throm- namely, the location of the aneurysm
bosis outcomes.
inflow and the more diffuse aspect of
the jet (PVRRs  70%). On the other
for all patients, on average, from 21.7 6 7.1 cm/s before to 7.2 6 hand, patients C and D showed similar pre- and poststent flow
2.9 cm/s after FDS placement (Fig 2B). Of note, a wide range of patterns, but with a lower velocity magnitude (PVRRs  54%).
pretreatment intra-aneurysmal systolic velocities converged to-
ward a much narrower range after flow diversion. DISCUSSION
On average, PVRR gradually decreased from 68.8%, 66.2%, This study presents quantitative PCMR measurements of FDS-
and 55% for 6- and 12-month thrombosis times and no thrombo- induced flow changes performed in a cohort of patients. To our
sis at 12 months, respectively. This finding is consistent with a knowledge, this is the largest study comparing in vivo flow-reduc-
lower flow reduction when the occlusion is delayed. Although not tion measurements and IA thrombosis outcomes.
statistically significant (P = .08), a trend was identified between
PVRR and these 3 groups of different occlusion times (Fig 2C). Flow Reduction and Thrombosis Outcomes
In addition, all IAs treated with 2 stents were occluded at On average, the measured intra-aneurysmal velocities, Velan,pre =
6 months but were spread into the bulk of PVRR, including the 21.7 6 7.1 cm/s and Velan,post = 7.2 6 2.9 cm/s, were in agree-
lowest value. This finding indicates that the decrease of porosity, ment with those in previous studies. On the basis of CFD
2120 Brina Dec 2019 www.ajnr.org
simulations performed on 8 patients, Kulcsar et al16 reported PCMR Measurements and Flow Diversion
time-averaged velocities reduced from 6 cm/s (pre-FDS) to 3 cm/s This study was made possible due to prior investigations on intra-
(post-FDS) for large aneurysms (diameter, .10 mm) and from cranial stent–related artifacts.23 In particular, we showed that
14.5 to 8 cm/s for small aneurysms (diameter, ,10 mm). these artifacts were mainly related to the shielding effect and were
Consistent with our findings, Sindeev et al20 showed a wide range therefore restricted to the stent lumen. Furthermore, the follow-
of systolic Velan,pre = 44–7 cm/s before stent placement, with in ing recent technical developments24 were assembled in a postpro-
vitro 4D-PCMR measurements, which converged to a narrower cessing pipeline to obtain consistent and reliable PVRR and Velan
range after stent placement (Velan,post = 7.6–4 cm/s). assessment from 4D-PCMR data: 1) the combination of 4D-
Various CFD studies have reported correlations between flow PCMR data with 3DRA geometry for a precise delineation of the
changes and aneurysm thrombosis, but there has been disagree- circulating domain; 2) the partial volume correction allowing
ment regarding the hemodynamic criteria associated with fast an- unbiased ICA mean flow-rate quantification; and 3) the semiau-
eurysm occlusion. For example, Mut et al17 found that an tomatic aneurysm extraction, thus ensuring a systematic and
absolute threshold of mean aneurysm velocity (1.3 cm/s), mean user-independent selection of the volume of interest and the con-
aneurysm inflow rate (0.37 mL/s), and mean shear rate (16.3 sec- sistent inclusion of the relevant aneurysm inflow velocities close
onds1) discriminated between fast and slow occlusion times in to the neck.32 In the context of flow-diversion treatment, only a
a group of 23 aneurysms. By contrast, Kulcsar et al16 suggested few PCMR investigations have been reported. The hemodynamic
a relative aneurysm-specific velocity and wall shear stress reduc- changes in the parent vessel were measured in patients with 2D-
tion threshold associated with thrombosis. Similarly, Ouared et PCMR by Eker et al7 and MacDonald et al.33 Sindeev et al20 used
al25 found that a relative velocity reduction of at least one-third 4D-PCMR in 3 patient-specific models and found flow reduc-
was associated with durable thrombosis. The hemodynamic tions of 89% and 30%–50% for fast and delayed thrombosis out-
component is widely accepted as the driving factor in aneurysm comes, respectively. Karmonik et al21 used a combination of in
healing. This has led to a trend by manufacturers toward vitro experiments and the measurements of 3 patients, but these
decreasing stent porosity by increasing the mesh density, while were not associated with occlusion times, findings similar to
keeping reasonable navigation features and from the operator’s those of Pereira et al.22
side, by adding stent layers to achieve “sufficient” contrast agent This limited literature can probably be attributed to the in-
stagnation following subjective indicators considered prone to herent limitations of 4D-PCMR related to data acquisition:
thrombosis patterns.26 long scan duration; coarse spatial resolution regarding the size
In this study, the PVRR was gradually lower in the 6- to 12- of the IA; the unique velocity encoding (VENC) that cannot
month thrombosis and no-thrombosis groups. This finding is cover the large range of involved velocities; and the low tempo-
consistent with a diminished flow reduction for delayed occlu- ral resolution, which smooths out the peak systolic velocities.
sions. However, the small PVRR differences among the 3 groups Additionally, 4D-PCMR cannot provide mechanical loads,
put in perspective the role of flow reduction as a driving parame- such as wall shear stress obtained with CFD simulations.
ter in the long-term occlusion of aneurysms treated by FDS. However, simulations have their own restrictions: The patient
Other parameters should be also considered for a comprehensive flow conditions are usually unknown; the non-Newtonian
understanding of IA thrombosis as suggested by the following behavior of the flow is rarely taken into account, though non-
studies: Paliwal et al18 showed that the average velocity reduction negligible for low velocities and recirculation areas as in flow
was not different between successful (52.4%) and unsuccessful diversion34; and the virtual stent hardly replicates the actual
(49.2%) treatments in 15 patients. Similarly, Berg et al27 studied 2 procedure deployment and its related vessel-geometry modifi-
morphologically equivalent carotid-ophthalmic aneurysms pre- cations.35 In comparison, 4D-PCMR has the great advantage of
senting with completely different outcomes (3-month occlusion providing direct in vivo flow measurement readily available in
for one and 3 additional layers required for the other) and found clinical settings and already routinely applied for the hemody-
opposite hemodynamic changes. Furthermore, histologic studies namic assessment of cardiac disorders.36 In the context of in-
suggest that neck endothelialization plays an important role in tracranial measurements, further improvements are needed to
the healing process, highlighting the importance of stent wall address the spatial resolution issues, while reducing the scan-
apposition to promote the tissue growing across the neck.28 ning time.
Nevertheless, it remains unclear whether the aneurysm thrombo-
sis, the neck endothelialization, or both are dominant factors for Clinical Relevance
occlusion. Most interesting, Kadirvel et al29 suggested that long- The PVRRs of patients implanted with 2 stents were homogene-
term occlusion occurred only as a result of neck covering, charac- ously distributed along the PVRR range (Fig 2C), with 1 having
terized by a contiguous layer of endothelial cells overlying a even the lowest value. Even if all the patients with double layers
smooth-muscle cell substrate. This suggestion could bring new had occlusion at 6 months, the absence of a relationship between
insights for manufacturers, researchers, and clinicians with impli- multilayer implantations and higher PVRRs suggests that the
cations for device development (Marosfoi et al30 showed that tem- placement of additional devices does not necessarily increase the
poral and spatial endothelial growth was related to stent design), flow diversion. These results are in line with those of Chalouhi
adjunctive medications (Li et al31 showed that intravenous injec- et al,37 who demonstrated similar occlusion rates for single and
tion of recombinant human SDF-1-a accelerated re-endotheliali- multiple PED FDSs. Moreover, they showed that the placement
zation of the stent), and dual-antiplatelet therapy. of additional stent layers added only morbidity with a 3-fold
AJNR Am J Neuroradiol 40:2117–23 Dec 2019 www.ajnr.org 2121
complication rate. In our study, the wide range of velocities FDS and 4D-PCMR we confirmed that this is a valid technique to
before stent placement (probably related to the wide range of an- characterize IA flow changes with regard to clinical outcomes fol-
eurysm sizes and shapes) was dampened in a narrower range af- lowing FDS implantation. PVRR is a promising indicator for a
ter stent implantation, independent of the initial conditions. more comprehensive understanding of the FDS treatment
From a clinical point of view, this is relevant information for response.
interventionists to potentially avoid adding unnecessary stent
layers. Disclosures: Vitor Mendes Pereira—UNRELATED: Consulting Fee or Honorarium:
Medtronic and Stryker, Comments: Proctor and Steering Committee for the
PREMIER and EVOLVE studies.* *Money paid to the institution.
Limitations
Our study has some limitations. First, whereas a range of VENC
values has been used in the literature, we chose a VENC of REFERENCES
1. Becske T, Kallmes DF, Saatci I, et al. Pipeline for uncoilable or failed
80 cm/s in accordance with Markl et al38 for intracranial vessel
aneurysms: results from a multicenter clinical trial. Radiology
measurements. Other 4D-PCMR studies used, VENCs of 2013;267:858–68 CrossRef Medline
120 cm/s,39 100 cm/s,20 and a range of 60–80 cm/s21 for pre- and 2. Briganti F, Leone G, Marseglia M, et al. Endovascular treatment of
poststent acquisitions. Our main limitation, with a potential cerebral aneurysms using flow-diverter devices: a systematic
impact on the results, was the choice of the poststent VENC, review. Neuroradiol J 2015;28:365–75 CrossRef Medline
3. Brinjikji W, Murad MH, Lanzino G, et al. Endovascular treatment
which led us to decrease this parameter from 80 cm/s (the first
of intracranial aneurysms with flow diverters: a meta-analysis.
17 patients) to 40 cm/s (in the remaining 6 patients) to capture Stroke 2013;44:442–47 CrossRef Medline
the low velocities more accurately at the expense of aliasing arti- 4. Walcott BP, Stapleton CJ, Choudhri O, et al. Flow diversion for the
facts. Furthermore, to rely on the aneurysmal velocities pre- treatment of intracranial aneurysms. JAMA Neurol 2016;73:1002–
sented in this study, we used the cutoff published in Pereira et 08 CrossRef Medline
5. Bouillot P, Brina O, Ouared R, et al. Particle imaging velocimetry
al22 and Bouillot et al23 to exclude patients with .50% of their
evaluation of intracranial stents in sidewall aneurysm: hemody-
prestent velocities below this value, assuming that poststent namic transition related to the stent design. PLoS One 2014;9:
measurements would be severely biased. For some patients, the e113762 CrossRef Medline
proportion of aneurysm volume below this threshold could 6. Bouillot P, Brina O, Ouared R, et al. Hemodynamic transition
have been larger than 50% after stent placement, thus reflecting driven by stent porosity in sidewall aneurysms. J Biomech 2015;
48:1300–09 CrossRef
very low velocities or nearly stagnant flows. This sensitive ac- 7. Eker OF, Boudjeltia KZ, Jerez RA, et al. MR derived volumetric flow
quisition parameter would need to be refined by using in vitro rate waveforms of internal carotid artery in patients treated for
ground truth measurements, such as particle imaging velocime- unruptured intracranial aneurysms by flow diversion technique. J
try, especially for poststent measurements. Cereb Blood Flow Metab 2015;35:2070–79 CrossRef Medline
Second, poststent MR imaging measurements were performed 8. Szikora I, Marosfoi M, Salomvary B, et al. Resolution of mass effect
and compression symptoms following endoluminal flow diver-
shortly after the procedure and did not reflect the entire IA sion for the treatment of intracranial aneurysms. AJNR Am J
thrombosis evolution influencing both the flow conditions and Neuroradiol 2013;34:935–39 CrossRef Medline
the neck endothelialization as shown in vitro by Gester et al.40 9. Chalouhi N, Tjoumakaris S, Starke RM, et al. Comparison of flow
Consequently, our statements should be weighed carefully, and diversion and coiling in large unruptured intracranial saccular
further studies should be considered to monitor aneurysmal ve- aneurysms. Stroke 2013;44:2150–54 CrossRef Medline
10. Di Maria F, Pistocchi S, Clarencon F, et al. Flow diversion versus
locity modification during follow-up imaging. Third, our results standard endovascular techniques for the treatment of unruptured
presented here (based on systolic velocity reduction) are limited carotid-ophthalmic aneurysms. AJNR Am J Neuroradiol 2015;36:
among other relevant flow parameters in flow diversion, such as 2325–30 CrossRef Medline
wall shear stress, the stagnation zone, and residence time. The in- 11. Delgado Almandoz JE, Kayan Y, Tenreiro A, et al. Clinical and
angiographic outcomes in patients with intracranial aneurysms
herent limitations of the PCMR technique (spatial and temporal
treated with the Pipeline embolization device: intra-procedural
resolution) must be considered to accurately resolve near-wall technical difficulties, major morbidity, and neurological mortality
velocities and low velocity ranges. Further research is needed to decrease significantly with increased operator experience in device
improve PCMR measurements to allow a reliable computation of deployment and patient management. Neuroradiology 2017;59:
these hemodynamic parameters. Finally, the small number of 1291–99 CrossRef Medline
12. Ikeda H, Ishii A, Kikuchi T, et al. Delayed aneurysm rupture due to
patients, especially for 12-month thrombosis time and thrombo-
residual blood flow at the inflow zone of the intracranial paraclinoid
sis absence, may have limited the emergence of a significant rela- internal carotid aneurysm treated with the Pipeline Embolization
tionship between PVRR and thrombosis time. The inaccuracy of Device: histopathological investigation. Interv Neuroradiol 2015;21:
low-velocity measurements could have also affected the low 674–83 CrossRef Medline
PVRRs. A study with a larger sample size would help in confirm- 13. Kulcsar Z, Houdart E, Bonafe A, et al. Intra-aneurysmal thrombosis
as a possible cause of delayed aneurysm rupture after flow-diver-
ing the PVRR as a potential predictor for fast and delayed
sion treatment. AJNR Am J Neuroradiol 2011;32:20–25 CrossRef
thrombosis. Medline
14. Pereira VM, Bonnefous O, Ouared R, et al. A DSA-based method
CONCLUSIONS using contrast-motion estimation for the assessment of the intra-
aneurysmal flow changes induced by flow-diverter stents. AJNR
This study showed a trend between IA occlusion time and veloc- Am J Neuroradiol 2013;34:808–15 CrossRef Medline
ity reductions measured with 4D-PCMR among a cohort of 15. Chien A, Vinuela F. IS FlowMap, a novel tool to examine blood
patients treated with FDSs. Thanks to previous research on the flow changes induced by flow diverter stent treatment: initial

2122 Brina Dec 2019 www.ajnr.org

 experiences with Pipeline cases. J Neurointerv Surg 2013;5(Suppl 3): 27. Berg P, Saalfeld S, Janiga G, et al. Virtual stenting of intracranial
iii43–47 CrossRef Medline aneurysms: a pilot study for the prediction of treatment success
16. Kulcsar Z, Augsburger L, Reymond P, et al. Flow diversion treat- based on hemodynamic simulations. Int J Artif Organs 2018;41:698–
ment: intra-aneurismal blood flow velocity and WSS reduction 705 CrossRef Medline
are parameters to predict aneurysm thrombosis. Acta Neurochir 28. Rouchaud A, Ramana C, Brinjikji W, et al. Wall apposition is a key
(Wien) 2012;154:1827–34 CrossRef Medline factor for aneurysm occlusion after flow diversion: a histologic
17. Mut F, Raschi M, Scrivano E, et al. Association between hemody- evaluation in 41 rabbits. AJNR Am J Neuroradiol 2016;37:2087–91
namic conditions and occlusion times after flow diversion in cere- CrossRef Medline
bral aneurysms. J Neurointerv Surg 2015;7:286–90 CrossRef Medline 29. Kadirvel R, Ding YH, Dai D, et al. Cellular mechanisms of aneurysm
18. Paliwal N, Damiano RJ, Davies JM, et al. Association between hemo- occlusion after treatment with a flow diverter. Radiology 2014;
dynamic modifications and clinical outcome of intracranial aneur- 270:394–99 CrossRef Medline
ysms treated using flow diverters. Proc SPIE Int Soc Opt Eng 2017; 30. Marosfoi M, Langan ET, Strittmatter L, et al. In situ tissue engineer-
10135 CrossRef Medline ing: endothelial growth patterns as a function of flow diverter
19. Wu C, Ansari SA, Honarmand AR, et al. Evaluation of 4D vascular design. J Neurointerv Surg 2017;9:994–98 CrossRef Medline
flow and tissue perfusion in cerebral arteriovenous malformations: 31. Li Z, Zhao R, Fang X, et al. Recombinant human SDF-1alpha
influence of Spetzler-Martin grade, clinical presentation, and administration accelerates aneurysm neck reendothelialization in
AVM risk factors. AJNR Am J Neuroradiol 2015;36:1142–49 rabbit saccular aneurysm after flow diverter treatment. Acta
CrossRef Medline Biochim Biophys Sin (Shanghai) 2017;49:246–53 CrossRef Medline
32. Qiu T, Jin G, Bao W, et al. Intercorrelations of morphology with
20. Sindeev S, Arnold PG, Frolov S, et al. Phase-contrast MRI versus nu-
hemodynamics in intracranial aneurysms in computational fluid
merical simulation to quantify hemodynamical changes in cere-
dynamics. Neurosciences (Riyadh) 2017;22:205–12 CrossRef Medline
bral aneurysms after flow diverter treatment. PLoS One 2018;13:
33. MacDonald ME, Dolati P, Mitha AP, et al. Hemodynamic altera-
e0190696 CrossRef Medline
tions measured with phase-contrast MRI in a giant cerebral aneu-
21. Karmonik C, Anderson JR, Elias S, et al. Four-dimensional phase
rysm treated with a flow-diverting stent. Radiol Case Rep 2015;10:
contrast magnetic resonance imaging protocol optimization using
1109 CrossRef Medline
patient-specific 3-dimensional printed replicas for in vivo imaging
34. Frolov SV, Sindeev SV, Liepsch D, et al. Experimental and CFD flow
before and after flow diverter placement. World Neurosurg 2017;
studies in an intracranial aneurysm model with Newtonian and
105:775–82 CrossRef Medline
non-Newtonian fluids. Technol Health Care 2016;24:317–33 CrossRef
22. Pereira VM, Brina O, Delattre BM, et al. Assessment of intra-aneur- Medline
ysmal flow modification after flow diverter stent placement with 35. Bouillot P, Brina O, Yilmaz H, et al. Virtual-versus-real implanta-
four-dimensional flow MRI: a feasibility study. J Neurointerv Surg tion of flow diverters: clinical potential and influence of vascular
Surg 2015;7:913–19 CrossRef Medline geometry. AJNR Am J Neuroradiol 2016;37:2079–86 CrossRef
23. Bouillot P, Brina O, Delattre BMA, et al. Neurovascular stent arti- Medline
facts in 3D-TOF and 3D-PCMR: influence of stent design on flow 36. Stankovic Z, Allen BD, Garcia J, et al. 4D flow imaging with MRI.
measurement. Magn Reson Med 2019;81:560–72 CrossRef Medline Cardiovasc Diagn Ther 2014;4:173–92 CrossRef Medline
24. Bouillot P, Delattre BMA, Brina O, et al. 3D phase contrast MRI: 37. Chalouhi N, Tjoumakaris S, Phillips JL, et al. A single Pipeline embo-
partial volume correction for robust blood flow quantification in lization device is sufficient for treatment of intracranial aneurysms.
small intracranial vessels. Magn Reson Med 2018;79:129–40 CrossRef AJNR Am J Neuroradiol 2014;35:1562–66 CrossRef Medline
Medline 38. Markl M, Frydrychowicz A, Kozerke S, et al. 4D flow MRI. J Magn
25. Ouared R, Larrabide I, Brina O, et al. Computational fluid dynamics Reson Imaging 2012;36:1015–36 CrossRef Medline
analysis of flow reduction induced by flow-diverting stents in in- 39. Anderson JR, Klucznik R, Diaz O, et al. Quantification of velocity
tracranial aneurysms: a patient-unspecific hemodynamics change reduction after flow diverter placement in intracranial aneurysm:
perspective. J Neurointerv Surg 2016;8:1288–93 CrossRef Medline an ex vivo study with 3D printed replicas. Conf Proc IEEE Eng Med
26. O’Kelly CJ, Krings T, Fiorella D, et al. A novel grading scale for the Biol Soc 2015;2015:7300–03 CrossRef Medline
angiographic assessment of intracranial aneurysms treated using 40. Gester K, Luchtefeld I, Busen M, et al. In vitro evaluation of intra-
flow diverting stents. Interv Neuroradiol 2010;16:133–37 CrossRef aneurysmal, flow-diverter-induced thrombus formation: a feasibil-
Medline ity study. AJNR Am J Neuroradiol 2016;37:490–96 CrossRef Medline

AJNR Am J Neuroradiol 40:2117–23 Dec 2019 www.ajnr.org 2123