Keywords first-line strategy - hyperdense vessel sign
Introduction
Currently, the role of mechanical thrombectomy (MT) as the standard of care in acute
ischemic stroke (AIS) consequent to a large vessel occlusion (LVO) is well established.[1 ] With functional outcome in LVO being profoundly time dependent, achieving first-pass
recanalization (FPR) holds paramount importance.[2 ]
[3 ] Currently there is no robust evidence citing significant difference in terms of
efficacy or safety between the first-line stent retriever (SR) and contact aspiration
(CA) techniques.[4 ] Though recent studies have suggested that red blood cell (RBC)-rich clot shows better
recanalization using SR compared with the CA as the first-line technique,[5 ] the debate as to the best first-line approach remains unsettled. The currently available
data, as a guide to decision making in selecting an optimal first-line MT technique
(SR or CA) based on noncontrast computed tomography (NCCT) imaging appearances is
at the best sparse.
In this study, we evaluated the influence hyperdense vessel sign (HVS) observed on
NCCT has in selecting an optimal individualized first-line approach for MT, with the
intention of increasing FPR.
Material and Methods
This retrospective study was performed at a tertiary care center equipped with a “comprehensive
stroke” protocol.
Patient Selection and Characteristics
All the clinical and imaging data pertaining to patients with acute LVO who underwent
MT using SR or CA approach from June 2017 to May 2020 was reviewed. Patients less
than 18 years of age, occlusions involving middle cerebral artery (MCA) distal to
M1 bifurcation, vertebrobasilar circulation, and tandem lesions were excluded from
the study. Also, patients who underwent MT employing Solumbra technique were not included
in the analysis. A total of 52 patients fulfilled the inclusion criteria and were
included in the final analysis.
Workflow for Acute Stroke at Our Tertiary Care Center
A stroke neurologist performed an initial assessment, primarily based on the National
Institutes of Health Stroke Scale (NIHSS) score. The prethrombectomy evaluation to
assess presence of bleed, extent of infarction, and vessel status was done using NCCT
with a triple-phase CT angiogram (CTA) for neck and intracranial vessels. Cases with
wake-up stroke and late presentations (beyond 6 hours from ictus) underwent MT, provided
CT perfusion demonstrated significant mismatch and hence salvageability. If NIHSS
score was > 4, imaging ruled out intracranial bleed, and CTA confirmed a LVO, the
patient was shifted to interventional laboratory for MT procedure. The intravenous
thrombolysis was initiated for patients presenting within 4.5 hours of symptom onset
at the discretion of attending stroke physician, after ruling out conventional contraindications
to thrombolysis. All procedures were performed by either one or two interventional
neuroradiologists, using a biplane angiography system. The upfront approach at our
center is SR first technique with adoption of CA technique consequent to three failed
SR attempts to achieve modified treatment in cerebral infarction (mTICI) score 2b/3
recanalization. The CA technique employed placement of ACE 68 (large bore aspiration
catheter) at the face of the clot and suction aspiration using a Penumbra MAX aspiration
pump. The decision for conscious sedation or general anesthesia is individualized
(based on patient's clinical condition) and at the discretion of interventional neuroradiologist.
Retrospective Evaluation of Images
All the NCCT images of 52 patients were evaluated by two neuroradiologists to assess
presence or absence of the HVS in the region of LVO. The CTA data was perused to correlate
the filling defect with the HVS on NCCT images, thus confirming the clot site. Initially
the neuroradiologists were blinded to each other's imaging conclusions. Subsequently,
following consensus meeting between the two neuroradiologists, interobserver agreement
for HVS was reached in all cases (kappa: 1).
Angiographic data and follow-up NCCT were reviewed to identify occlusion site, average
number of SR passes needed to achieve mTICI 2b/3, mTICI 2b/3 reperfusion after SR
strategy, rate of FPR, early recanalization, rescue therapy (switch over to CA), mTICI
2b/3 reperfusion at end of procedure (following CA strategy), incidence of distal
embolization using SR technique, and incidence of any or symptomatic intracerebral
hemorrhage (ICH).
Postprocedure mTICI grade of ≥ 2b was defined as successful recanalization. Early
recanalization was defined as recanalization with maximum of two passes of SR. Symptomatic
ICH was defined as any intracranial hemorrhage associated with ≥ 4 points increase
on NIHSS score at 24 hours (Electronic Consolidated Automated Support Criteria).
Statistical Analysis
Quantitative variables were expressed as mean and standard deviation. Qualitative
variable were expressed as frequency and percentage. Comparison of continuous variables
between two groups was analyzed by Student's t -test. Comparison of continuous variable among more than two categories was analyzed
by analysis of variance. Association between qualitative variables was analyzed by
chi-square test. A p -value of < 0.05 was considered statistically significant. Interobserver agreement
for HVS was calculated using kappa statistics. Data were entered in Microsoft Excel
and data analysis was performed using SPSS version 22.0.
Results
Of the 52 patients included, 28 and 24 were randomized to HVS (+) and HVS (−) cohort,
respectively. Excellent interobserver agreement for HVS was noted using kappa statistics
(kappa = 1). All but three patients underwent MT with onset of symptoms being within
6 hours. These three patients (2 HVS+ and 1 HVS–) presented between 6 and 10 hours
of symptoms onset and underwent MT following demonstration of brain tissue salvageability
on CT perfusion. MT was performed under conscious sedation in 26 (92.9%) of HVS (+)
and 22 (91.7%) of HVS (−) cohort, respectively. Two cases underwent the procedure
under general anesthesia in each group.
Occlusion involving the terminal internal carotid artery was documented in three patients
in each cohort, but affected M1 MCA in 25 (89.3%) and 21 (87.5%) of HVS (+) and HVS
(−) cohort, respectively. Intravenous tissue plasminogen activator was administered
in 13 (46.4%) and 12 (50%) of HVS (+) and HVS (−) group, respectively. First-pass
mTICI 2b/3 recanalization was achieved using SR in 14 (50%) of HVS (+) and 5 (20.9%)
of HVS (−) cohorts. Early mTICI 2b/3 recanalization was achieved using SR in 18 (64.2%)
of HVS (+) and 9 (37.5%) of HVS (−) cohorts. A total of 7 patients (25%) from HVS
(+) needed the rescue therapy (switched over to CA technique) whereas 12 patients
(50%) from HVS (−) needed adoption of CA technique consequent to three failed SR attempts
to achieve mTICI 2b/3 recanalization. Overall, the CA technique was employed in 19
patients. Following rescue therapy with CA, mTICI 2b/3 recanalization was achieved
in 2/7 (28.6%) and 6/12 (50%) of HVS (+) and HVS (−) group, respectively. The rate
of distal embolization was 2 (7.1%) in the susceptibility vessel sign (SVS) (+) compared
with 4 (16.7%) in the SVS (−) group ([Table 1 ]).
Table 1
Comparison in angiographic outcomes according to hyperdense vessel sign (HVS)
Characteristics
HVS positive (n = 28)
n (%)
HVS negative
(n = 24)
n (%)
p -Value
Initial NIHSS score, mean (SD)
13.2 (2.7)[a ]
14.4 (3.1)[a ]
0.142
ASPECTS, mean (SD)
6.2 (1.3)[a ]
6.0 (1.1)[a ]
0.556
Occlusion site
0.841
-ICA terminus
3/28 (10.7)
3/24 (12.5)
-M1 MCA
25/28 (89.3)
21/24 (87.5)
IV t-PA infusion
13/28 (46.4)
12/24 (50.0)
0.797
Anesthesia
0.872
Conscious sedation
26/28 (92.9)
22/24 (91.7)
GA
2/28 (7.1)
2/24 (8.3)
SR passes needed to achieve mTICI 2b/3:
1
14/28 (50.0)
5/24 (20.9)
0.029
2
4/28 (14.3)
4/24 (16.7)
0.812
3
3/28 (10.7)
3/24 (12.5)
0.841
First-pass recanalization
14/28 (50.0)
5/24 (20.9)
0.029
Early recanalization (≤ 2 passes)
18/28 (64.2)
9/24 (37.5)
0.054
mTICI 2b/3 reperfusion after SR strategy
21/28 (75.0)
12/24 (50.0)
0.062
Rescue therapy[b ]
7/28 (25.0)
12/24 (50.0)
0.062
mTICI 2b/3 reperfusion post rescue strategy
2/7 (28.6)
6/12 (50.0)
0.075
Overall mTICI 2b/3 reperfusion at end of procedure
23/28 (82.1)
18/24 (75%)
0.530
Distal embolization
2/28 (7.1)
4/24 (16.7)
0.284
ICH at 24 h
2/28 (7.1)
3/24 (12.5)
0.514
Symptomatic ICH
1/28 (3.6)
1/24 (4.2)
0.911
Abbreviations: CA, contact aspiration; GA, general anesthesia; ICA, internal carotid
artery; ICH, intracerebral hemorrhage; IV, intravenous; MCA, middle cerebral artery;
mTICI, modified treatment in cerebral infarction score; NIHSS, National Institutes
of Health Stroke Scale; SD, standard deviation; SR, stent retriever; t-PA, tissue
plasminogen activator.
a Values indicates mean (SD), rest of values in second and third columns represent
numbers (percentage).
b Contact aspiration following three failed attempts at recanalization with SR.
The representative cases of HVS (+) and HVS (−) cohort, managed with SR technique
and rescue therapy are depicted in [Figs. 1 ] and [2 ], respectively.
Fig. 1 Successful recanalization using stent retriever (SR) in patient with hyperdense vessel
sign (HVS). (A ) Noncontrast computed tomography (NCCT) image shows presence of HVS in left middle
cerebral artery (MCA) (arrow). (B ) CT angiogram and (C ) baseline angiography confirmed complete occlusion (arrow in B and C ). (D ) SR first strategy employed, following navigation of an appropriate microcatheter
over a microwire across the occluded segment. Looped microwire tip parked in M2 segment
of MCA (two-headed arrow). (E ) Final angiography demonstrates successful recanalization (arrow with double stroke).
Fig. 2 Successful recanalization following rescue therapy in patient with absent hyperdense
vessel sign (HVS). (A ) Noncontrast computed tomography (NCCT) image shows absence of HVS (arrow). (B ) CT angiogram and (C ) baseline angiography demonstrates complete occlusion of M1 segment of right middle
cerebral artery (MCA) (arrow in B and C ). (D ) Contact aspiration attempted consequent to three failed attempts with stent retriever.
Depicts tip of ACE 68 aspiration catheter at face of the clot (two-headed arrow).
(E ) Final angiography shows successful recanalization (arrow with double stroke).
FPR was observed using SR in 50% of HVS (+) and 20.9% of HVS (−), a statistically
significant difference (p = 0.029) between the two cohort. The early recanalization (≤ 2 SR passes) was documented
in 64.2% of HVS (+) and 37.5% of HVS (−), a near statistically significant difference
(p = 0.054) between the two cohort. The rate of rescue therapy was much lower in patients
demonstrating HVS on NCCT, compared with those in the SVS (−) cohort (p = 0.062) ([Table 1 ] and [Fig. 3 ]). Also, the incidence of distal embolization was higher in the SVS (−) than in SVS
(+) cohort (16.7 vs. 7.1%). No statistically significant difference was noted in the
age, comorbidities, NIHSS and ASPECTS score, use of type of anesthesia, intravenous
thrombolysis, and incidence of any or symptomatic ICH between HVS (+) and HVS (−)
cohorts ([Table 1 ]).
Fig. 3 Bar chart representing angiographic recanalization outcomes with stent retriever
(SR) and rescue therapy (contact aspiration consequent to three failed SR attempts)
strategies between the two cohort. HVS (+), hyperdense vessel sign present; HVS (−),
hyperdense vessel sign absent.
Discussion
Consequent to demonstration of clinical efficacy of MT unequivocally by the multiple
landmark randomized clinical trials, presently its role as the standard of care for
LVO in AIS is well established.[1 ]
[6 ]
[7 ]
[8 ]
[9 ]
Although functional outcome for LVO with MT is definitely superior to intravenous
thrombolysis, it is profoundly time dependent.[2 ]
[3 ] Considering therefore the need of early recanalization, FPR is emerging as a key
concept in management of AIS.[10 ]
[11 ] Recent studies emphasize the importance of first-pass effect and documented it as
an independent prognostic factor in achieving better functional outcome in LVO.[3 ]
[11 ]
[12 ]
[13 ] Also, recent literature emphasize the quality of reperfusion to be an important
predictor of good clinical outcome, with progressively better clinical outcomes seen
from mTICI 2b to 3.[14 ]
[15 ] The multiple passes not only lead to clot compaction, increased time to recanalization,
but also to intimal injury and increase in procedure-related complications.[11 ]
[16 ]
[17 ]
[18 ]
[19 ]
[20 ]
Therefore, the interventionist must apply a strategy which has the highest likelihood
for achieving FPR, based on individual's experience and available literature.
The presence of HVS on NCCT and SVS on gradient-recalled-echo magnetic resonance imaging
(MRI) suggests RBC-rich clot, whereas its absence favors a fibrin-predominant occlusive
thrombi.[21 ]
[22 ]
[23 ]
[24 ] Also, good concordance between HVS and SVS has been documented in the literature.[21 ]
The presence of SVS, considered to be due to T2-shortening effect of intracellular
deoxyhemoglobin (acute stage of RBC clot), is documented to have a direct correlation
with successful recanalization using SR.[25 ]
[26 ]
[27 ] Although studies correlating SVS with clot composition and outcomes (angiographic
and clinical) following SR are available, the literature comparing HVS (on NCCT) with
angiographic outcome using SR or CA strategy is scarce. Also, CT being a more commonly
undertaken investigation in clinical setting of anterior circulation AIS, a first-line
strategy (SR or CA) based on clot appearances on CT is more pragmatic and hence warranted.
HVS in an appropriate clinical context, is a highly specific, and an early sign of
AIS on NCCT.[28 ] In our study, the two neuroradiologists had an almost perfect interobserver agreement
(kappa = 1) for HVS.
Brinjikji et al, based on their systematic review, concluded that clots with a higher
mean thrombus Hounsfield unit were more likely to be RBC-rich and showed better angiographic
outcome compared with those with lower values.[29 ] Similarly, in our study HVS presence was observed to be an independent predictor
of successful, first pass, as well as early recanalization (≤ 2 SR passes). Such results
are biologically plausible with recent data demonstrating improved entrapment of RBC-rich
clot in a SR.[12 ] The favorable recanalization in the HVS (+) cohort noted in our study is in accordance
with early recanalization observed in those with SVS sign on MRI.[25 ]
[26 ]
[27 ]
[30 ]
[31 ] Our results support the premise that RBC-rich clots manifest as HVS and SWS on plain
CT and MRI, respectively, and show early recanalization using SR.
A few studies have examined the association between CT attenuation of clot and angiographic
outcomes, but not the subgroup analysis to assess and compare efficiency of SR versus
CA in those with absent HVS.[32 ]
[33 ]
[34 ] Similar to our results, Froehler et al documented successful recanalization in 79%
of patients with HVS, but in only 36% without HVS.[34 ] However, the thrombectomy in this study was limited to MERCI retriever with no option
to switch over to CA (rescue therapy) in those with failed recanalization.
We documented that first-pass mTICI 2b/3 recanalization was achieved using SR in 14
(50%) of HVS (+), but in only 5 (20.9%) of HVS (−) cohort, plausible explanation being
the tendency of the fibrin-rich clot to roll out of the SR during retrieval.[30 ]
[35 ] In our study, following rescue therapy with CA (consequent to three failed SR attempts),
mTICI 2b/3 recanalization was achieved in 6 (50%) of HVS (−) cohort, with maximum
of three CA passes. This observation highlights the effectiveness of aspiration technique
in this subgroup of patients.
The incidence of distal embolization was 7.1% in SVS (+) compared with 16.7% in SVS
(−) cohort, explained by the higher number of passes needed in the latter.
Although, a recent systematic review and meta-analysis by Boulanger et al suggest
no significant difference in terms of efficacy or safety between first-line SR and
CA techniques,[4 ] our study highlights that higher FPR can be achieved if individualized first-pass
strategy is employed (based on the appearance of clot on NCCT), instead of a generalized
SR first-pass approach.
Some of the limitations of our study include the relatively small number of cases
treated, a single-center experience, and the lack of direct analysis of thrombi histologically.
Although our study showed high rate of successful recanalization following the rescue
therapy with CA (consequent to three failed attempts with SR) in patients with absence
of HVS, we cannot definitely comment on the efficacy a CA-based first-line strategy
would have yielded if it was applied as the upfront approach for those demonstrating
HVS. Further studies employing CA as the first-line strategy are needed to effectively
comment on the efficacy the CA technique would have in those with HVS (+). However,
being a single-center study, the protocol of adhering to SR as the first strategy
in all cases added homogeneity to the cohort.
To our knowledge, there is no other CT-based study, evaluating the influence “HVS”
has in selecting an appropriate first approach (ST vs. CA) for MT, with intention
of increasing FPR.
Conclusion
The good angiographic recanalization achieved following adoption of rescue therapy
with CA technique (consequent to three failed attempts with SR) in the patients demonstrating
absence of HVS, support the premise that CA may be considered as the first-line technique
in such subset of patients. Our study highlights that higher FPR can be achieved if
individualized first-pass strategy is employed (based on the appearance of clot on
NCCT), instead of a generalized SR first-pass approach. This CT imaging-based strategy,
formulated with our single tertiary center experience, is a step closer to achieving
the primary angiographic goal of FPR.
