MASARYK UNIVERSITY
FACULTY OF MEDICINE
DIAGNOSTICS AND RECANALIZATION
TREATMENT OF ISCHEMIC STROKE
HABILITATION THESIS
in Neurology
(Collection of previously published scholarly works)
2021 Ondřej Volný, MD, Ph.D.
I hereby declare that I wrote this habilitation thesis on my own, using the relevant resources listed
in the references.
……………………..
Signature
Obsah
Acknowledgement........................................................................................................................... 5
Abstract ........................................................................................................................................... 6
Abstrakt ........................................................................................................................................... 7
Section 1 – General Overview ........................................................................................................ 8
Pathophysiology of Ischemic Stroke..................................................................................................................... 8
Section 2 – Detection of Early Ischemic Changes and Utility of Time-variant Multiphase CT
Angiography Color Maps in Acute Anterior Circulation Stroke due to Large Vessel Occlusion
....................................................................................................................................................... 10
Assessment of Early Ischemic Changes.............................................................................................................. 10
ASPECT Score.................................................................................................................................................... 11
Automated Imaging Tools for ASPECT Scoring................................................................................................ 12
Multiphase CTA and CTP in Determination of Irreversibly Damaged Brain Tissue (Core).............................. 12
Imaging Paradigms and Randomized Controlled Trials ..................................................................................... 14
Section 2.1 Detection of ischemic changes on baseline multimodal computed tomography: expert
reading vs. Brainomix and RAPID software .......................................................................................15
Section 2.2 – Utility of Time-Variant Multiphase CTA Color Maps in Outcome Prediction for
Acute Ischemic Stroke Due To Anterior Circulation LVO................................................................35
Section 3 – Endovascular Treatment of Acute Ischemic Stroke................................................. 45
Section 3.1 – Mechanical Thrombectomy Performs Similarly in Real-World Practice: A 2016
Nationwide Study from the Czech Republic........................................................................................46
Section 3.2 – Thrombectomy vs. Medical Management in Low NIHSS Acute Anterior
Circulation Stroke ................................................................................................................................57
Conclusions................................................................................................................................... 74
List of Abbreviations..................................................................................................................... 75
List of Figures............................................................................................................................... 77
List of Tables................................................................................................................................. 78
Annex 1 ......................................................................................................................................... 79
VOLNY O., P. CIMFLOVA, T.-Y. LEE, B. K. MENON and C. D. D’ESTERRE. Permeability surface area
product analysis in malignant brain edema prediction – A pilot study. Journal of the Neurological Sciences
[online]. 2017, 376, 206–210. ISSN 0022-510X................................................................................................. 79
Annex 2 ......................................................................................................................................... 85
OSPEL J. M., O. VOLNY, W. QIU, M. NAJM, N. KASHANI, M. GOYAL a B. K. MENON. Displaying
Multiphase CT Angiography Using a Time-Variant Color Map: Practical Considerations and Potential
Applications in Patients with Acute Stroke. American Journal of Neuroradiology [online]. 2020, 41(2), 200–
205. ISSN 0195-6108. ......................................................................................................................................... 85
Annex 3 ......................................................................................................................................... 92
VANICEK J., P. CIMFLOVA, M. BULIK, J. JARKOVSKY, V. PRELECOVA, V. SZEDER a O. VOLNY
(senior author). Single-Centre Experience with Patients Selection for Mechanical Thrombectomy Based on
Automated Computed Tomography Perfusion Analysis-A Comparison with Computed Tomography Perfusion
Thrombectomy Trials. Journal of Stroke & Cerebrovascular Diseases [online]. 2019, 28(4), 1085–1092.
ISSN 1052-3057. ................................................................................................................................................. 92
Annex 4 ....................................................................................................................................... 101
VOLNY O., M. BAR, A. KRAJINA, P. CIMFLOVA, L. KASICKOVA, R. HERZIG, D. SANAK, O.
SKODA, A. TOMEK, D. SKOLOUDIK, D. VACLAVIK, J. NEUMANN, M. KOCHER, M. ROCEK, R.
PADR, F. CIHLAR a R. MIKULIK. A Comprehensive Nationwide Evaluation of Stroke Centres in the Czech
Republic Performing Mechanical Thrombectomy in Acute Stroke in 2016. Ceska a Slovenska Neurologie
a Neurochirurgie [online]. 2017, 80(4), 445–450. ISSN 1210-7859................................................................. 101
Annex 5 ....................................................................................................................................... 108
KÖCHER M., D. SANAK, J. ZAPLETALOVA, F. CIHLAR, D. CZERNY, D. CERNIK, P. DURAS, L.
ENDRYCH, R. HERZIG, J. LACMAN, M. LOJIK, S. OSTRY, R. PADR, V. ROHAN, M. SKORNA, M.
SRAMEK, L. STERBA, D. VACLAVIK, J. VANICEK, O. VOLNY and A. TOMEK. Mechanical
Thrombectomy for Acute Ischemic Stroke in Czech Republic: Technical Results from the Year 2016.
Cardiovascular and Interventional Radiology [online]. 2018, 41(12), 1901–1908. ISSN 0174-1551.............. 108
5
Acknowledgement
I would like to express my sincere gratitude to all my motivating mentors (associate prof. Bijoy
K. Menon, prof. Michael D. Hill, prof. Andrew M. Demchuk, prof. Mayank Goyal and prof. Robert
Mikulík), whom taught me the basics of clinical research, statistics and strokology.
Special thanks to my challenge driven research collaborators dr. Petra Cimflová, dr. Johanna
M. Ospel and dr. Charlotte Zerna for the countless spent hours in darkroom and deep discussions
about our research projects and papers.
Finally, I would like to express my heartfelt appreciation to my whole family, my amazing
and always supportive wife, Michaela, and our adorable son, Richard. This work is dedicated
to you!
6
Abstract
Stroke is a leading cause of adult disability worldwide. Clinical trials studying the reperfusion
therapies (intravenous thrombolysis and mechanical thrombectomy) in acute stroke are of proven
benefit. Speed and accuracy of diagnosis and interpretation of neuroimaging is critical
before the treatment is initiated.
This habilitation thesis is divided into four main sections. The first section summarizes briefly
stroke epidemiology and pathophysiology. The second section is devoted to modern neuroimaging
tools in acute ischemic stroke with a special emphasis put on the visual and automated detection
of early ischemic changes on baseline neuroimaging and clinical utility of the multiphase CTA.
The third section is devoted to endovascular treatment (mechanical thrombectomy) of acute
ischemic stroke. Its first part focuses on comparisons of nationwide endovascular data
from the Czech Republic with a meta-analysis HERMES (Highly Effective Reperfusion evaluated
in Multiple Endovascular Stroke Trials). Its second part presents an observational multicenter
study assessing the effectiveness and safety of endovascular treatment versus best medical
management in patients with CTA detected large vessel occlusion in anterior circulation
and mild neurological deficit using recent data from comprehensive datasets and propensity
score matching. The last section contains full texts of original publications (annexes) related
to the topic of this habilitation thesis.
7
Abstrakt
Cévní mozková příhoda (CMP) je celosvětově hlavní příčinou invalidity dospělých.
Randomizované studie zkoumající reperfúzní terapii (intravenózní trombolýza a mechanická
trombektomie) u akutní CMP prokázaly její jednoznačný přínos (benefit) na výsledný klinický
stav. Rychlost a přesnost diagnostiky a interpretace neurozobrazování je rozhodující
před zahájením rekanalizační léčby.
Tato habilitační práce je rozdělena do čtyř hlavních částí. První část stručně shrnuje epidemiologii
a patofyziologii CMP. Druhá část je věnována moderním neurozobrazovacím nástrojům s důrazem
na automatickou detekci časných ischemických změn na vstupním neurozobrazení a klinický
význam multifázické CTA. Třetí část je věnována endovaskulární léčbě (mechanické
trombektomii) CMP. Její první oddíl je zaměřen na srovnání národních EVT dat za ČR s metaanalýzou
HERMES (Highly Effective Reperfusion evaluated in Multiple Endovascular Stroke
Trials). Druhý oddíl představuje observační multicentrickou studii hodnotící účinnost a bezpečnost
endovaskulární léčby ve srovnání se standardní léčbou u pacientů s prokázaným uzávěrem velké
mozkové tepny v přední mozkové cirkulaci a s nízkým neurologickým deficitem za použití
tzv. propensity score matching analýzy. Poslední část (přílohy) obsahuje plné texty originálních
prací souvisejících s tématem habilitační práce.
8
Section 1 – General Overview
Stroke represents a major cause of disability and is the second leading cause of death worldwide
with an incidence of about 17 million per year. Improvements in primary prevention and lifestyle
changes have led to a decreased incidence of age-adjusted stroke. Nevertheless, the overall number
of strokes cases has been increasing and is expected to accelerate over the coming decades because
of the aging in the population. It is predicted that stroke will account for 6.2% of the total burden
of illness in developed countries by 2020. Without major advances in prevention, acute
management and treatment, as well as in post-stroke rehabilitation, the burden and cost
of this disease will considerably increase. (1) (2) (3)
Pathophysiology of Ischemic Stroke
Acute occlusion of a cerebral artery or arteriole leads to an immediate decrease in arterial blood
flow in a particular vascular territory. Large vessel occlusion (LVO) caused by larger clots
are associated with more severe neurological deficits than occlusions of more distal and smaller
arteries. Immediately after the occlusion, cerebrovascular and systemic compensatory mechanisms
are activated: acute stress reaction, blood pressure increase, recruitment of collaterals, etc., in order
to maintain the sufficient perfusion. Blood flow reduction may be partially compensated
by cerebral collateral circulation at different levels, mostly at the level of leptomeningeal (pial)
collaterals. If the blood flow is above 20 ml/100 g per min (40% of a normal flow), cerebrovascular
autoregulatory mechanisms lead to increased oxygen extraction. Below this level of blood flow,
the neurotransmission will cease, and neurologic symptoms will manifest in correlation
to the affected brain areas. Neurons are able to survive for minutes, but if sufficient blood flow
is not restored they die at an average of 1.9 million nerve cells per minute. (4) The processes
of ischemic and apoptotic changes are dynamic and occur within the next few hours/days.
If the vessel is not opened and brain perfusion is not restored, the failure of these compensatory
mechanisms beside the critical decrease in arterial perfusion lead to hypoxia progressing
into ischemia, neuronal death, and eventually developing infarct. (5) (6)
Larger infarcts are associated with more severe deficits and worse functional outcome,
e.g. infarction in the whole middle cerebral artery (MCA) territory has a poor prognosis, with only
9
5% functional independence after the first year in comparison to the M2 territory infarction.
The urgent priority in acute ischemic stroke treatment is to reopen the occluded artery or arteries
(recanalization), because early recanalization improves the short-term and the long-term outcomes.
(7) (8)
10
Section 2 – Detection of Early Ischemic Changes and Utility of Timevariant
Multiphase CT Angiography Color Maps in Acute Anterior
Circulation Stroke due to Large Vessel Occlusion
Acute ischemic stroke (AIS) due to LVO is a highly time-critical disease. In 2015, endovascular
treatment (EVT) became the standard of care for LVO stroke patients presenting within 6 hours
from symptom onset. (9) EVT represents a powerful treatment, but its effect is time-dependent.
The overarching goal when performing EVT is therefore to treat the patient as fast as possible;
however, not all patients will benefit from EVT, and screening the eligible patients is crucial
in the selection process, in order not to cause harm. (10)
Currently, patients’ selection is based on two main pillars: clinical characteristics (e.g. severity
of neurological deficit, premorbid functional status, time of symptom onset) and brain imaging.
Brain computed tomography (CT) is the most widely used and widespread modality for stroke
imaging; non-contrast CT (NCCT), CT angiography (CTA), and CT perfusion (CTP) represent
important imaging tools, complementary to clinical examination and patient’s history, aiding
in the diagnosis and the decision-making of the subsequent therapy.
Assessment of Early Ischemic Changes
Early ischemic changes tend to develop in the first hours after stroke symptoms onset. In order
to improve the rater detection of early ischemic changes on baseline NCCT, it is important to set
up the so-called hard brain window. It assists the rater in differentiation of subtle changes
in the grey and white matter density (window width [WW] 35 to 40 Hounsfield units [HU],
window level [WL] 35-40 HU versus standard brain window WW 80 HU, WL 40 HU), Figure 1.
11
Figure 1: Comparison of the standard brain window and “hard brain” window
Legend: upper row (WW 80, WL 40), lower row = hard brain window (WW 40, WL 40) for proper assessment of early
ischemic changes (marked with red ovals).
ASPECT Score
The most clinically used and validated score for assessing the extent of early ischemic changes
in the anterior circulation is the Alberta Stroke Program Early CT Score (ASPECTS). The MCA
territory is divided into 10 regions. For each region affected by early ischemic changes, one point
is subtracted from a total score of 10 including the following areas: (A) at basal ganglia level
[caudate (C), lentiform (L), internal capsule (IC), insula (I) and M1 to M3 territory]; (B) the M4
to M6 cortical regions at the supraganglional level. The ASPECTS is used as an imaging selection
tool for EVT and it was proven to be a significant predictor of functional outcome. (11) (12) (13)
On the one hand, it represents a validated grading system, and on the other and, the inter-rater
reliability is limited. There is a trend to work on developing reliable and accurate software tools
to help stroke physicians with the scan assessment and decision making.
12
Automated Imaging Tools for ASPECT Scoring
Neuroimaging interpretation in acute stroke requires some level of expertise, hence it might cause
some time delays among less-experienced physicians (stroke clinician needs simple, quick
and accurate imaging tools). The e-ASPECTS software (Brainomix, Oxford, U.K.) is a fullyautomated
ASPECT scoring tool for NCCT, which has previously demonstrated a scoring
at an expert level. The advantage of e-ASPECTS is its potential to eliminate the inter-rater
variability. Automated post-processing with RAPID software was used as an accurate prediction
tool for irreversibly damaged tissue (infarct core), and tissue at risk of infarction (penumbra).
Section 2.1 is devoted to this subject in detail. The main aim was to compare the assessment of early
ischemic changes by the expert reading and by the available automated software for NCCT
and CTP and ultimately demonstrate the accuracy for the final infarct prediction.
Multiphase CTA and CTP in Determination of Irreversibly Damaged Brain Tissue (Core)
Baseline neuroimaging is also used to determine how much of brain tissue is already irreversibly
damaged, since in patients with extensive early ischemic changes, tissue is unlikely to be salvaged
by EVT, and the risk of reperfusion injury (futile recanalization) is higher. There are multiple ways
of identifying irreversibly damaged tissue. The most commonly used imaging techniques are CTP
and multiphase CTA (mCTA). Both techniques have been successfully used for EVT patient
selection in randomized controlled trials, and both have their advantages and disadvantages: CTP
maps can be quickly and easily interpreted even with limited imaging experience, because
the color-coded display format is a clear visual indicator of pathology. (2) (14) (15) (16)
On the other hand, CTP is susceptible to patient motion and post-processing artifacts
and generating post-processed maps takes some time. mCTA is more robust against patient motion
and requires less contrast and radiation dose. It is equally as reliable as CTP as an EVT selection
and outcome prediction tool, and because it can easily be implemented without any additional
technical requirements, it is particularly attracts smaller hospitals and places in which
it is not possible to afford additional hardware and/or software. (17) However, the standard display
format for mCTA consists of 3 separate gray scale images of the cerebral vessels (Figure 2),
and evaluating the collaterals requires the reader to assess all three of them simultaneously.
Our research team have recently described a new color-coded mCTA display format, in which all
3 mCTA series are consolidated in a single color-coded map, thereby potentially facilitating
13
and improving mCTA interpretation, (Figure 2) and (ANNEX 1 and ANNEX 2). mCTA colormaps
therefore constitute a good alternative to facilitate interpretation of collateral status until fully
automated collateral assessment becomes routinely available, particularly for less-experienced
readers. Section 2.2 is devoted to this subject in detail.
Figure 2: Conventional and color-based collateral scoring
Legend: Toprow – good collaterals. Most collaterals are opacified on the first mCTA phase and appear red on the colormap,
which is consistent with a zero-phase delay. The vessel extent is nearly identical to the contralateral side. Middle
row: intermediate collaterals. Most collaterals are opacified on the second mCTA phase and appear green on the colormap,
which is consistent with a one-phase delay. The vessel extent is slightly reduced compared to the contralateral
side. Bottom row: poor collaterals. The few visible collaterals are mostly opacified on the third mCTA phase and appear
blue on the color-map, which is consistent with a two-phase delay. The vessel extent is markedly reduced compared
to the contralateral side.
14
Imaging Paradigms and Randomized Controlled Trials
The vanguard trials that established efficacy of EVT in patients with AIS used various imaging
criteria for patient selection. These ranged from simple paradigms like NCCT and single-phase
CTA in the MR CLEAN and the RESILIENT trial, collateral imaging using mCTA in the ESCAPE
trial and to multimodal MRI protocol in the THRACE trial. CTP was used exclusively
in the EXTEND IA trial, a mixture of advanced imaging was used in the SWIFT PRIME trial (some
mCTA and some CTP) and the late window DAWN and DEFUSE-3 trials had NCCT and CTP
paradigms.
Each approach has its advantages and its drawbacks but the lack of standardization of imaging
paradigms globally results in some misunderstanding about what types of patients are actually
being enrolled into the studies. This makes comparison of patient populations and treatment effects
difficult, but it also constitutes an opportunity to compare different imaging paradigms and try
to find an optimal imaging approach; one that provides just enough information for treatment
decision-making and outcome prediction, without delaying or compromising treatment by either
obtaining unnecessary information or excluding patients whom may have benefited from treatment.
Of all the imaging paradigms in use in patients with acute stroke, NCCT with single phase CTA
is the simplest and arguably the fastest, but reliability of assessment of the extent of early ischemic
changes is low, particularly among less-experienced physicians. Pial collateral status assessment
has high specificity if the collaterals are good on a single-phase CTA but poor collateral filling
could be a false result due to delay in timing of the contrast bolus with consequent arterial filling.
Both CTP and mCTA provide time-resolved images that try to address the issue of mistimed bolus
contrast influencing assessment of contrast enhanced CT in patients with acute stroke. In CTP,
the brain is continuously scanned over 45 – 90 seconds, while in mCTA, three scans cycles
are performed over 16-20 seconds. Due to its higher temporal resolution, the information content
in CTP images is higher when compared to mCTA, but this comes at the cost of lower spatial
resolution, higher motion susceptibility and requirement for algorithm based image postprocessing.
15
Section 2.1 Detection of ischemic changes on baseline multimodal computed
tomography: expert reading vs. Brainomix and RAPID software
Based upon: CIMFLOVA, Petra, Ondrej VOLNY*( joint first author)*, Robert MIKULIK, Bohdan TYSHCHENKO,
Silvie BELASKOVA, Jan VINKLAREK, Vladimir CERVENAK, Tomas KRIVKA, Jiri VANICEK a Antonin
KRAJINA. Detection of ischemic changes on baseline multimodal computed tomography: expert reading
vs. Brainomix and RAPID software. Journal of Stroke & Cerebrovascular Diseases [online]. 2020, 29(9), 104978.
ISSN 1052-3057.
Abstract
Background – The aim of the study was to compare the assessment of early ischemic changes
by expert reading and available automated software for NCCT and CTP on baseline multimodal
imaging and demonstrate the accuracy for the final infarct prediction.
Methods – Early ischemic changes were measured by ASPECTS on the baseline neuroimaging
of consecutive patients with anterior circulation ischemic stroke. The presence of early ischemic
changes was assessed a) on NCCT by two experienced raters, b) on NCCT by e-ASPECTS,
and c) visually on derived CT perfusion maps (CBF<30%, Tmax>10s). Accuracy was calculated
by comparing presence of final ischemic changes on 24-hour follow-up for each ASPECTS region
and expressed as sensitivity, specificity, positive predictive value (PPV), and negative predictive
value (NPV). The sub-analysis for patients with successful recanalization was conducted.
Results – Of 263 patients, 81 fulfilled inclusion criteria. Median baseline ASPECTS was 9 for all
tested modalities. Accuracy was 0.76 for e-ASPECTS, 0.79 for consensus, 0.82 for CBF<30%,
0.80 for Tmax>10s. e-ASPECTS, consensus, CBF<30%, and Tmax>10s had sensitivity 0.41, 0.46,
0.49, 0.57, respectively; specificity 0.91, 0.93, 0.95, 0.91, respectively; PPV 0.66, 0.75, 0.82, 0.73,
respectively; NPV 0.78, 0.80, 0.82, 0.83, respectively. Results did not differ in patients
with and without successful recanalization.
Conclusion – This study demonstrated high accuracy for the assessment of ischemic changes
by different CT modalities with the best accuracy for CBF<30% and Tmax>10s.
The use of automated software has a potential to improve the detection of early ischemic changes.
16
Introduction
The ASPECTS quantifies the extent of early ischemic changes in the MCA territory on baseline
NCCT scans. (18) It has been proven to be a significant predictor of clinical outcome in patients
with AIS in the anterior circulation. (12) (13) It is also used to select patients for EVT. (11)
It represents a validated grading system, but the inter-rater variability has been questioned.
Even experienced clinicians show only a 39% agreement in the identification of ischemic changes
on NCCT involving more than one-third of the MCA territory. (19) Hence, there is a trend
to develop reliable software tools to help stroke physicians in acute scan reading and subsequent
decision making. (10) (16)
The e-ASPECTS software (Brainomix, Oxford, UK) is a fully-automated ASPECT scoring tool
for NCCT, which has previously demonstrated a scoring on expert level. The advantage of eASPECTS
is its potential to eliminate the inter-rater variability. (20) (21) (22) CTP has a potential
to discriminate between irreversibly damaged tissue, infarct core, and tissue at risk of infarction,
penumbra. (23) (24) It has been demonstrated that visual applying of ASPECTS into CTP
parametric maps has a strong correlation with good clinical outcome (defined as modified Rankin
scale [mRS] 0-2), with a prognostic value greater than NCCT ASPECTS. (25) (26) (27) (28)
All previous studies have shown the highest correlation of good clinical outcome with cerebral
blood volume (CBV) ASPECTS.
The most accurate prediction of irreversibly ischemic changes by automatic software postprocessing
with RAPID was shown for relative cerebral blood flow (CBF) less than 30%
in comparison to the mean CBF of normally perfused brain parenchyma. (29) (30) This threshold
was used in the randomized controlled trials with patient selection based on perfusion mismatch
(SWIFT-PRIME, EXTEND-AI, DAWN, DEFUSE III) to define ischemic core. (31) (15) (32) (14)
Severe hypoperfusion has been associated with irreversible necrosis of the ischemic lesion even
after reperfusion. (33) In the DEFUSE and EPITHET meta-analysis, large regions of severe delay
(>10 s) have been associated with poor outcome after reperfusion. (34) This finding suggests
that higher Tmax may identify tissue with more severely reduced cerebral blood flow,
which may have a substantial impact on the evolution of the acute ischemic lesion.
17
The main aim of our study was to evaluate how accurate the different CT modalities
with and without software processing (consensus reading, e-ASPECTS, CBF<30%, Tmax>10s)
assess early ischemic changes at baseline and what is their accuracy for final ischemia prediction.
Methods
Patient selection
Ethical approval was obtained from the local Institutional Review Boards (the Boards waived
the need for patient consent). All patients with symptoms of AIS and no history of contrast allergy
routinely underwent NCCT, mCTA from the aortic arch to the vertex and CTP in our
institution.(17) If the diagnosis of AIS was confirmed by this neuroimaging protocol, NCCT
was repeated within 24-32 hours to determine the extent and location of ischemia and diagnose
potential complications such as hemorrhagic transformation. Radiological data of consecutive
patients from March 2017 to September 2017 presenting with symptoms of AIS in the anterior
circulation within 6 hours of last seen normal (symptom onset) were retrospectively reviewed.
This time period was chosen in order to compare the reliability of the detection of early ischemic
changes while the software for automatic detection of early ischemic changes was implemented
into our institutional system. Inclusion criteria were: 1) availability of baseline NCCT
with automatic software analysis, baseline CTP and follow-up 24-hour NCCT. Exclusion criteria
were: 1) evidence of any intracranial hemorrhage or non-ischemic lesion, 2) negative findings
on baseline diagnostic imaging and no ischemic changes on follow-up CT. We defined patients
with successful reperfusion/recanalization angiografically as Thrombolysis in Cerebral Infarction
(TICI) 2b-3 in patients treated with EVT or as >40% decrease in the 24-hour NIHSS score
in patients treated with intravenous thrombolysis (IVT) only. (35) Sub-analysis of this subgroup
was conducted.
Imaging Protocol
The imaging protocol set up in our stroke center combines NCCT, mCTA and CTP
and both software programs were available during the study period for automatic analysis
(Brainomix for NCCT and RAPID for CTP). NCCT was acquired on a multi-detector scanner
(120kV, 328 mAs (419mAs/slice), Brilliance iCT 256; Philips Healthcare, Cleveland, OH)
with a section thickness of 0.9mm and an image reconstruction of 3mm. For the CTP protocol,
18
40 ml of contrast agent (Iomeron 300; Mallinckrodt Pharmaceuticals; Dublin, Ireland) was power
injected at 5 ml/s followed by a saline chase of 50 ml at 5 ml/s. Sections of 8cm thickness
were acquired at 10 mm slice thickness. Scanning began after a delay of 5s from contrast injection
in every 1.8s for 75s. After 24 hours, a NCCT was acquired for final infarct delineation
in all patients.
Image Processing
NCCT scans were automatically analysed by the e-ASPECTS software (version 6.0, Brainomix,
Oxford, UK). The e-ASPECTS software is a standardized, fully-automated, CE mark-approved
ASPECTS scoring tool for NCCT, which has previously demonstrated scoring on an expert level.
The e-ASPECTS software is based on a combination of advanced image-processing and machinelearning
algorithms. Its scoring module operates on the standardized 3D images, classifying signs
of ischemic damage and assigning them to ASPECTS regions. CT perfusion studies were postprocessed
using the RAPID software (iSchemaView, Menlo Park, CA, USA) to generate perfusion
maps of cerebral blood flow (CBF), cerebral blood volume (CBV), mean transit time (MTT),
and time to the maximum of the residue function (Tmax). The RAPID software also automatically
segmented and calculated volumes of the ischemic core (relative regional CBF<30%)
and the critically hypoperfused tissue (Tmax>6s). (29)
Image review
Early ischemic changes were assessed on baseline NCCT by two experienced readers (a consultant
neuroradiologist, PC, and a stroke neurologist, OV)* using the ASPECTS score defined by Barber
et al. previously, blind to the results of the e-ASPECTS analysis, as well as to other baseline
imaging modalities and follow-up NCCT. (13) Automatic segmentations of ASPECTS regions
on e-ASPECTS derived scans were visually checked to avoid any severe inaccuracy. Otherwise,
the given e-ASPECTS score were not modified and the original e-ASPECTS was noted. CTP maps
were superposed on the CT-ASPECTS template and visually assessed by an experienced reader
(PC). Ischemic changes on CTP maps were evaluated using the ASPECTS as follows:
1) on the CBF map as the area with CBF<30 % when compared to the contralateral hemisphere
and 2) on the Tmax map as the area with Tmax>10s delay in the maximum contrast filling
within the region of interest when compared to the contralateral hemisphere (Figure 3). The reader
19
was blind to findings on NCCT, perfusion baseline scans available in the summary of RAPID
analysis were visually controlled to exclude any false positive CTP findings (e.g. chronic
infarction). The final infarction was assessed on a 24-hour follow-up NCCT with consensus
of the two readers (PC, OV), during a different session, one month after the previous assessment
of the early ischemic changes on baseline NCCT. To support the reliability of the consensus
reading, two radiologists (VC, TK) evaluated ASPECTS of 40 random admission NCCT
and 40 follow-up scans (of different patients). The inter-rater agreement with the consensus
was counted using weighted kappa (kw) and Krippendorf’s alfa (a). The moderate agreement
between raters was demonstrated for baseline NCCT (kw=0.53-0.54; a=0.72) and good to excellent
agreement for follow-up imaging (kw=0.78-0.88; a=0.94)**. *PC and OV have 6-year experience
with stroke imaging evaluation, 5-year experience in comprehensive stroke centre and both
were trained in ASPECTS reading as members of the Calgary Stroke Program.
Statistical Analysis
Clinical and imaging baseline characteristics were summarized using descriptive statistics.
The accuracy, sensitivity, specificity, positive predictive value (PPV) and negative predictive value
(NPV) were calculated for particular ASPECTS regions (81 patients x 10 ASPECTS regions)
at baseline imaging (e-ASPECTS, expert consensus reading, CBF<30%, Tmax>10s)
in comparison with ASPECTS regions at the follow-up CT. The Bland-Altman plots
were calculated to compare the differences between each baseline imaging method and follow-up
ASPECTS. The sensitivity analysis of pooled data for the group with determined successful
reperfusion/recanalization was conducted; clinical and imaging baseline characteristics
were summarized using descriptive statistics and compared to the group with non-determined
recanalization/reperfusion using Wilcoxon rank sum test; the accuracy, sensitivity, specificity,
PPV, NPV as well as Bland-Altman plots were calculated. To compare the two subgroups,
we calculated residuals between follow-up ASPECTS and each baseline ASPECT score method
(e-ASPECTS, expert consensus reading, CBF<30%, Tmax>10) and analysed these residuals using
Wilcoxon rank sum test. This study provides hierarchically structured data with 3 levels: subject
ID, imaging modalities (e-ASPECTS, expert consensus reading, CBF<30%, Tmax>10s,
and follow-up ASPECTS); and ASPECTS regions (M1-M6, Insula, Lentiform, Capsula, Caudate).
We regarded regions as a fixed effect. The generalized estimating equation accommodating
20
clustering at the subject ID level was used (PROC GENMOD; SAS Institute Inc, Cary, NC). LSmeans
estimates of fixed effect “region” computed from generalized mixed model were graphically
illustrated. All analyses were performed in Stata 16.1 (StataCorp LLC, College Station, TX, USA)
and SAS 9.3 (SAS Institute, Cary, NC, USA).
Results
Baseline scans of 263 patients were retrospectively reviewed; 16 patients with intracranial
hemorrhage and 166 patients with either negative findings on all imaging modalities or missing
follow-up imaging were excluded. Overall, 81 patients met all the criteria and were included
into the analysis. Mean age was 70 years (standard deviation [SD] 14 years, range 30-92 years),
38 (46,9%) were women. Median baseline NIHSS was 9 (interquartile range [IQR]=4–17).
The median time interval from symptom onset to CT was 156 mins (IQR=71-220);
there were 12 patients with the unknow time of symptom onset or wake-up stroke. Median baseline
ASPECTS was 9 for all tested modalities (IQR=8-10 for e-ASPECTS, IQR=7-10 for consensus,
IQR=7-10 for CBF<30%, IQR 6-10 for Tmax>10s, median ASPECTS on follow-up NCCT was 8,
IQR=5-9), left hemisphere was affected in 44 cases (54.3%). Fifty patients received intravenous
thrombolysis and 19 patients had mechanical thrombectomy. Reperfusion was achieved
in 11 patients in the mechanical thrombectomy group and in 22 patients in the IVT group, the data
from mechanical thrombectomy and intravenous thrombolysis groups were pooled for further
analysis (as determined recanalization).
Accuracy of baseline ASPECTS and follow-up ASPECTS was 0.76 for e-ASPECTS, 0.79
for expert consensus, 0.81 for CBF<30% and 0.8 for Tmax>10s. Sensitivity and specificity were
0.41 and 0.91 for e-ASPECTS; 0.46 and 0.93 for expert consensus; 0.49 and 0.95 for CBF<30%;
0.57 and 0.91 for Tmax>10s respectively. PPV and NPV were 0.66 and 0.78 for e-ASPECTS; 0.75
and 0.8 for expert consensus; 0.82 and 0.81 for CBF<30%; 0.73 and 0.83 for Tmax>10s,
respectively, Figure 4 and Table 1. Bland-Altman plots comparing differences in scores
of baseline ASPECTS and follow-up ASPECTS are demonstrated in Figure 5. The mean
difference between e-ASPECTS and follow-up was -1.16 ± 2.52 (median undercall of ASPECTS
was -1), expert consensus and follow-up -1.16 ± 2.23 (median undercall was -1), CBF<30%
and follow-up -1.15 ± 1.77 % (median undercall was -1), and Tmax>10s and follow-up -0.59 ±
21
1.86 (median undercall was 0). The ASPECTS was rated as lower on baseline imaging in 15/81
cases for e-ASPECTS, 11/81 for expert consensus, 6/81 for CBF<30, and in 15/81 cases
for Tmax>10s.
Sensitivity analysis
Clinical and imaging baseline characteristics for patients with determined successful recanalization
were not significantly different in comparison to the subgroup of patients with non-determined
recanalization (Table 2). The results of the subgroup analysis in patients with successful
reperfusion/recanalization are graphically demonstrated in Figure 6 (Table 3). Accuracy
of baseline ASPECTS and follow-up ASPECTS was 0.79 for e-ASPECTS, 0.81 for expert
consensus, 0.83 for CBF<30% and 0.82 for Tmax>10s. Sensitivity and specificity were 0.51
and 0.90 for e-ASPECTS; 0.53 and 0.92 for expert consensus; 0.55 and 0.94 for CBF<30%; 0.66
and 0.89 for Tmax>10s, respectively. PPV and NPV were 0.67 and 0.82 for e-ASPECTS; 0.73
and 0.83 for expert consensus; 0.77 and 0.84 for CBF<30%; 0.7 and 0.87 for Tmax>10s,
respectively.
Bland-Altman plots for the subgroup analysis comparing differences in scores of baseline
ASPECTS and follow-up ASPECTS are demonstrated in Figure 7. The mean difference
between e-ASPECTS and follow-up was -0.70 ± 2.48 (median undercall of ASPECTS was -1),
expert consensus and follow-up -0.79 ± 2.33 (median undercall was 0), CBF<30% and follow-up
-0.82 ± 1.77 (median undercall was -1), and Tmax>10s and follow-up -0.21 ± 1.74 (median
undercall was 0). The ASPECTS was lower on baseline imaging in 7/33 cases for e-ASPECTS,
6/33 for expert consensus, 5/33 for CBF<30% , and in 9/33 cases for Tmax>10s.
There was no significant difference between residuals of the follow-up ASPECTS
and each baseline ASPECTS for the two subgroups (determined recanalization versus nondetermined
recanalization group), the median undercall of baseline ASPECT scores was -1 point
in comparison to the follow-up ASPECTS for the baseline methods in the subgroup with nondetermined
recanalization and for CBF<30% and e-ASPECTS in the subgroup with determined
recanalization. There was a trend observed for Tmax>10s that show a higher precision
in the subgroup with determined successful recanalization with the median undercall of 0 points.
(Table 4). Results from generalized mixed model are illustrated in Figure 8.
22
Figure 3: Comparison of CT imaging modalities and evaluation of early ischemic changes
Legend: Baseline ASPECTS was assessed as follows: 10 points on NCCT by expert reading (A), 9 (lentiform)
by automatic e-ASPECTS (B) and 6 (M2, M3, insula, lentiform) on Tmax>10s CTP maps (C). Follow-up CT (D)
shows the ischemic changes within insula, lentiform and M5 (ASPECTS 7); and hemorrhagic transformation within
the right insula.
23
Figure 4: Accuracy, sensitivity, specificity, positive predictive value, negative predictive value
of baseline ASPECTSs evaluated by e-ASPECTS, consensus (expert reading), CBF<30%
and Tmax>10s)
Legend: ACC – accuracy; TPR – true positive value/sensitivity, TNR – true negative value/specificity, PPV – positive
predictive value, NPV – negative predictive value
24
Figure 5: Bland-Altman plots
Legend: Bland-Altman plots illustrating the level of agreement between the baseline and follow-up ASPECTS for
different baseline CT modalities and means of evaluation (software vs. expert reading). Solid line indicates the mean
difference between the baseline and follow-up, dashed lines indicate the limits of the agreement.
25
Figure 6: Subgroup analysis of patients with successful reperfusion – accuracy, sensitivity,
specificity, positive predictive value and negative predictive value for baseline assessment (eASPECTS,
consensus, CBF<30% and Tmax>10s) and final ischemic changes on follow-up
NCCT
Legend: ACC – accuracy; TPR – true positive value/sensitivity, TNR – true negative value/specificity, PPV – positive
predictive value, NPV – negative predictive value
26
Figure 7: Bland-Altman plots for the subgroup analysis of patients with successful
recanalization/reperfusion (MT and IVT group pooled data)
Legend: Bland-Altman plots compare the baseline ASPECTS and follow-up ASPECTS for baseline CT modalities
and different means of assessment. Solid line indicates the mean difference between baseline and follow-up, dashed
lines indicate the limits of the agreement.
27
Figure 8: Least square means estimates of fixed effect “region” computed from generalized
mixed model
Legend: The least square means (Ls-means) estimates were computed from a generalized mixed model (fixed effect
was ASPECTS region). Follow-up NCCT was used as a reference grid. Results expressed on a logit scale demonstrate
the highest agreement for final ischemia in insula and M5 region regardless the used CT modality and scoring approach.
The lowest odds were demonstrated for internal capsule, which also showed the highest variability in scoring.
28
Table 1: Accuracy, sensitivity, specificity, PPV and NPV of baseline ASPECTS (e-ASPECTS,
consensus, CBF<30% and Tmax>10s) vs. follow-up imaging
Accuracy Sensitivity Specificity PPV NPV
e-ASPECTS vs. follow-up 0.76 0.41 0.91 0.66 0.78
Consensus vs. follow-up 0.79 0.46 0.93 0.75 0.8
CBF<30% vs. follow-up 0.81 0.49 0.95 0.82 0.81
Tmax>10 s vs. follow-up 0.80 0.57 0.91 0.73 0.83
Legend: ASPECTS = Alberta Stroke Program Early CT Score; CBF = cerebral blood flow; NPV = negative predictive
value; PPV = positive predictive value; Tmax = time to maximum.
29
Table 2: Comparison of patient baseline characteristics for the whole dataset (n=81)
and a subgroup of patients with determined successful recanalization (n=33)
Dataset
N= 81
Recanalization
subgroup N=33
P-value*
Female sex – n (%) 38 (46.9) 13 (39.4) 0.26
Age – median (IQR) 71 (62 – 81) 69 (62 – 80) 0.54
Affected left side – n (%) 44 (54.3) 18 (54.55) 0.97
Baseline NIHSS – median (IQR) 9 (4 – 17)1
14 (7 – 17.5)1
0.002
Baseline ASPECTS – median (IQR) 9 (7 – 10) 9 (6.5 – 10) 0.41
Onset to baseline CT in min – median
(IQR)
156 (71-220)2
110 (67-216)2
0.41
Legend: IQR = interquartile range, NIHSS = National Institutes of Health Stroke Scale, ASPECTS = Alberta Stroke
Program Early CT Score, EVT = endovascular treatment
*Derived from Wilcoxon rank sum test for two subgroups – determined successful recanalization versus nondetermined
recanalization
1
computed for n=79 and subgroup n=32
2
computed for n=69 and subgroup n=31
30
Table 3: Subgroup analysis of patients with successful reperfusion/recanalization – accuracy,
sensitivity, specificity, PPV and NPV of baseline ASPECTSs (e-ASPECTS, consensus,
CBF<30% and Tmax>10s) vs. follow-up imaging
Accuracy Sensitivity Specificity PPV NPV
e-ASPECTS vs. follow-up 0.79 0.51 0.93 0.67 0.82
Consensus vs. follow-up 0.81 0.53 0.92 0.73 0.83
CBF<30% vs. follow-up 0.83 0.55 0.94 0.77 0.84
Tmax>10 s vs. follow-up 0.82 0.66 0.89 0.69 0.87
Legend: ASPECTS = Alberta Stroke Program Early CT Score; CBF = cerebral blood flow; NPV = negative predictive
value; PPV = positive predictive value; Tmax = time to maximum.
31
Table 4: Comparison of residuals for the follow-up ASPECTS and the baseline ASPECTS
for the subgroups of patients with determined successful recanalization versus nondetermined
recanalization
Determined
recanalization
n=33
Non-determined
recanalization
n=48
p-value*
e-ASPECTS – median (IQR) -1 (-3 – 0) -1 (-1 – 0) 0.54
Consensus – median (IQR) -1 (-2.5 – 0) 0 (-2 – 0) 0.15
CBF <30% – median (IQR) -1 (-2 – 0) -1 (-2 – 0) 0.26
Tmax >10s – median (IQR) -1 (-2 – 0) 0 (-1 – 1) 0.03
Legend: ASPECTS = Alberta Stroke Program Early CT Score; CBF = cerebral blood flow; Tmax = time to maximum.
*Derived from Wilcoxon rank sum test
32
Discussion
In this study we demonstrated high sensitivity and specificity for detection of acute ischemic
changes for CT imaging modalities including assessment of acute ischemic changes
by experienced readers and clinically available software. Unlike in previous studies, we have
focused on CTP parameters representing either ischemic core (CBF<30%) or severely
hypoperfused tissue (Tmax>10s), parameters that were not analyzed previously in the perspective
of ASPECT scoring. CBF <30% is nowadays widely accepted to represent the ischemic core
with the high sensitivity and specificity and with low overestimation of the core, that could result
in unwarranted exclusion of patients who could benefit from reperfusion. (29) In contrast
to Tmax>6s, which is used to define penumbra, we evaluated a more severe delay, Tmax>10s,
representing the critically hypoperfused tissue, which is associated with irreversible necrosis
of the ischemic lesion even after reperfusion. (33)
The highest specificity was observed for CTP parameter, rCBF<30%, assessed visually on CTP
maps processed by RAPID software. This CTP parameter also showed the highest positive
predictive value for final ischemic changes. Moreover, the CTP parameter of Tmax delay >10s,
representing a severe hypoperfusion, showed the highest sensitivity and high accuracy
for prediction of final ischemic lesion (within the whole dataset as well as in the subgroup analysis
of successful reperfusion/recanalization). Tmax delay >10s was studied previously –
the association of large Tmax>10s lesion and malignant MCA profile was showed in previous
studies. (30) (36) Tmax volumes at a delay of >8s and >10s were strongly correlated with clinical
outcome. (37) Our findings support the importance of this parameter in the detection of irreversible
ischemic changes on baseline neuroimaging. We demonstrated that both CBF <30% and Tmax>10s
have high accuracy in detection of early ischemic changes as shown previously for CBV and these
changes could be easily assessed on the derived perfusion maps from RAPID analysis. (25) (26)
(27) (28)
The Blant-Altman plots showed the lowest difference in baseline ASPECTS and follow-up
ASPECTS for Tmax>10s. The other baseline methods showed similar differences in baseline
and follow-up ASPECTS with the median undercall of the baseline score of 1 point (these findings
did not differ when we analyzed the residuals between follow-up ASPECTS and baseline
33
ASPECTS for the subgroups with determined/non-determined recanalization). The CBF <30%
and Tmax>10s also demonstrated the lowest data dispersion for baseline and follow-up ASPECTS.
This indicates that these perfusion parameters may represent irreversibly affected tissue with higher
accuracy in comparison to detectable changes on baseline NCCT. Nevertheless, the semiautomated
analysis showed similar results with expert reading. This finding suggests a comparable
diagnostic value of the software evaluation and expert reading in the acute stroke management.
Although e-ASPECTS showed the lowest accuracy and sensitivity among the tested baseline
methods, the accuracy of 0.76 could still be considered as good, the sensitivity analysis also did not
show any significant difference between baseline methods for the tested subgroups.
The comparable findings for e-ASPECTS and other studied imaging methods implicates
the benefit of software evaluation for less experienced readers.
We observed a certain level of variability in assessment of particular ASPECTS regions.
The highest odds for agreement in evaluation of baseline ischemic changes and final ischemia
was demonstrated for insula regardless the baseline imaging modality and the way of ASPECT
scoring. It was demonstrated previously that the insular ribbon sign represented a very early
ischemic change in the middle cerebral artery strokes. (38) Contrarily, the lowest odds
for agreement between multimodal baseline and follow-up imaging was observed in the internal
capsule. That might be explained by difficulties in the visual assessment of hypoattenuation
within this region as the internal capsule is naturally less hypodense on NCCT. (39) This small
subcortical region might also be challenging to be distinguished on the CTP maps as hypoperfused.
Additionally, there was low variability demonstrated for all cortical ASPECTS regions, caudate
and lentiform. It may reflect that early ischemic changes of the insula are easy to detect even
with the low experience, but assessment of early ischemic changes within the internal capsule
might be problematic also for experienced readers.
We are aware of some limitations of this study. First of all, this was a single center observational
study. The patients were not selected according to the recanalization rate. Information
about the recanalization status was available only in patients indicated to mechanical
thrombectomy. Nevertheless, due to a limited (6-months) period when the e-ASPECTS software
was available at our institution, we decided to include all patients meeting our inclusion criteria
34
regardless of the treatment or recanalization status. We also did not focus on the correlation
of ASPECTS and final clinical outcome, as this relationship has been studied in other work.(40)
The main purpose of this work was to evaluate the accuracy of ASPECTS assessment on baseline
multimodal imaging. We are also aware of a possible misinterpretation of particular regions caused
by visual application of ASPECTS regions into the CTP maps processed by RAPID software.
At the time of the patient recruitment, the RAPID CTP software presented only the volumes
of impaired tissue perfusion (not co-registered within the ASPECTS regions). The automatic
segmentation of ASPECTS regions on e-ASPECTS scans also has its limitations and beside
the visual control to avoid any severe inaccuracy we did not tend to correct the automatic
segmentation and given ASPECTS scoring as we aimed to test the accuracy of commercially
available version of the software.
There are a few potential pitfalls in regard of the detection of acute ischemic changes
with automatic analysis. There might be a false positive finding on CTP maps in patients
with a subacute or chronic infarction. The RAPID software automatically segments and removes
areas with very low CBF, such as CSF spaces and other extra-parenchymal tissue, so in most cases
subacute/chronic infarction is also excluded. Another known pitfall is that CTP maps do not display
an infarcted area if the reperfusion was achieved ahead of the imaging, even though
there is evidence of the infarction on NCCT. These potential pitfalls highlight the necessity
of a visual control of CTP derived maps with NCCT or other available imaging as well as a control
of the correct placement of arterial input function and venous output function.
35
Section 2.2 – Utility of Time-Variant Multiphase CTA Color Maps in Outcome
Prediction for Acute Ischemic Stroke Due To Anterior Circulation LVO
Based upon: OSPEL, Johanna M., Petra CIMFLOVA, Ondrej VOLNY, Wu QIU, Moiz HAFEEZ, Arnuv
MAYANK, Mohamed NAJM, Kevin CHUNG, Nima KASHANI, Mohammed A. ALMEKHLAFI, Bijoy K. MENON
a Mayank GOYAL. Utility of Time-Variant Multiphase CTA Color Maps in Outcome Prediction for Acute Ischemic
Stroke Due to Anterior Circulation Large Vessel Occlusion. Clinical Neuroradiology [online]. 2020. ISSN 1869-1439.
Abstract
Background – mCTA is an established tool for EVT decision-making and outcome prediction
in acute ischemic stroke. We aimed to determine whether mCTA-based prediction of clinical
outcome and final infarct volume can be improved by assessing collateral status on time-variant
mCTA color maps rather than using a conventional mCTA display format.
Methods – Patients from the PRove-IT cohort study with anterior circulation LVO were included
in this study. Collateral status was assessed with a three-point scale using the conventional display
format. Collateral extent and filling dynamics were then graded on a three-point scale using timevariant
mCTA color-maps. Multivariable logistic regression was performed to determine
the association of conventional collateral score, color-coded collateral extent and color-coded
collateral filling dynamics with good clinical outcome and final infarct volume.
Results – A total of 285 patients were included in the analysis and 53% (152/285) of the patients
achieved a good outcome. Median infarct volume on follow-up was 12.6 ml. Color-coded collateral
extent was significantly associated with good outcome (adjusted odds ratio [adjOR] 0.53, 95%
confidence interval [CI]:0.36–0.77) while color-coded collateral filling dynamics (adjOR 1.30
[95% CI:0.88–1.95]) and conventional collateral scoring (adjOR 0.72 [95%CI:0.48–1.08]) were
not. Both color-coded collateral extent (adjOR 2.67 [95%CI:1.80–4.00]) and conventional
collateral scoring (adjOR 1.84 [95%CI:1.21–2.79]) were significantly associated with follow-up
infarct volume, while color-coded collateral filling dynamics were not (adjOR 1.21 [95%CI:0.83–
1.78]).
Conclusion – Collateral extent assessment on time-variant mCTA maps improved prediction
of good outcome and has similar value in predicting follow-up infarct volume compared
to conventional mCTA collateral grading.
36
Introduction
Acute ischemic stroke due to LVO is a highly time-critical disease. In 2015, EVT became
the standard of care for LVO strokes presenting within 6 hours from symptom onset.
There are many ways of identifying irreversibly damaged tissue; the most commonly used imaging
techniques are CT perfusion and mCTA. Both techniques have been successfully used for EVT
patient selection in randomized controlled trials. (14) (15) (2) (16) We have recently described
a new color-coded mCTA display format, in which all 3 mCTA series are consolidated in a single
color-coded map, thereby potentially facilitating and improving mCTA interpretation. (41)
The purpose of this study was to compare prediction of clinical outcome and final infarct volume
in acute ischemic stroke due to LVO using a conventional mCTA display format vs. time-variant
color maps.
Methods
Patient Population
The Precise and Rapid Assessment of Collaterals Multiphase CTA in the Triage of Patients
with Acute Ischemic Stroke for IA Therapy (Prove-IT) study was a prospective multicenter cohort
study that enrolled 464 patients who presented with symptoms consistent with AIS
(NCT02184936). Patients were eligible for the study if they presented to the emergency
department with symptoms consistent with AIS, were older than 18 years, and mCTA and CTP
were performed within 12 hours of symptom onset and before recanalization therapy. We included
patients with anterior circulation LVO (internal carotid artery, M1 or proximal M2 MCA
occlusions). Enrolled patients in who baseline NCCT and mCTA images were incomplete
or not interpretable were excluded from this analysis.
Image Acquisition
NCCT and mCTA: NCCT was acquired with 5 mm slice thickness. The mCTA scans consisted
of three phases, with arch to vertex coverage in the first (arterial) and skull base to vertex coverage
the second (peak venous) and third (late venous) phases. Detailed mCTA acquisition parameters
have been published previously.(17) Axial images with 1 mm overlap and multiplanar axial,
coronal and sagittal reconstructions with 3 mm thickness, 1 mm intervals and 1 mm overlap
for the first phase were then generated as well as axial maximum intensity projections (MIPs)
37
for all three phases. Time variantmCTA color maps were generated with the FastStroke research
prototype (GE Healthcare, Milwaukee, WI, USA) and displayed as axial, coronal, sagittal,
and oblique MIP reformations. Color-coding of the collaterals is hereby based on a per-patient
adaptive threshold technique; vessels with maximum enhancement in the pre-venous phase
are displayed in red, those with maximum enhancement in the peak venous phase and late venous
phase are displayed in green and blue, respectively (Figure 2). (41)
Image Interpretation
All images were assessed in a consensus reading (by a neurologist OV and neuroradiologist JO).
ASPECTS was scored on 5 mm reconstructed axial unenhanced NCCT images. Occlusion site
was determined on axial mCTA MIP images and was reported as either terminal internal carotid
artery, M1 segment or proximal M2 segment. We decided to include the proximal M2 segment
in our analysis, since most physicians consider proximal M2 occlusions as LVO and as appropriate
target lesions for EVT. (42) Proximal M2 occlusions were hereby defined as Sylvian segment
M2 occlusions located within 1 cm from the MCA bifurcation.
Conventional mCTA Collateral Grading: The delay and extent of collateral filling was graded
on axial MIPs of the three mCTA phases. A trichotomized collateral scale as used in the ESCAPE
(2) and ESCAPE NA1 (43) trials was applied:
Poor collaterals: no or only few vessels visible in any phase within the occluded vascular territory
compared to the asymptomatic contralateral hemisphere.
Intermediate collaterals: delay of two phases in filling in of peripheral vessels or a one-phase delay
and some ischemic regions with only few or no vessels compared to the asymptomatic contralateral
hemisphere.
Good collaterals: no delay or 1 phase delay in filling of peripheralvessels with identicalor increased
prominence of vessels compared to the asymptomatic contralateral hemisphere.
Collateral Grading on Time-variant Color Maps: Both delay and extent of collateral filling were
graded on a trichotomized scale on axial color-coded MIPs. In other words, in the color-map based
assessment, two separate scores were applied for collateral filling dynamics and collateral extent,
38
as opposed to conventional mCTA scoring, in which both of these factors were graded in a single
score.
Collateral extent was graded as follows:
Normal or almost normal extent (>90%) of visible vessels within the occluded vascular territory
compared to the contralateral hemisphere.
Vessel extent 50–90% within the occluded vascular territory compared to the contralateral
hemisphere.
Vessel extent <50% within the occluded vascular territory compared to the contralateral
hemisphere.
Collateral delay was graded as follows:
Predominantly no delay (most vessels are displayed in red) within the occluded vascular territory.
Predominantly 1 phase delay (most vessels are displayed in green) within the occluded vascular
territory.
Predominantly 2 phase delay or no collaterals (most vessels are displayed in blue/no vessels visible
at all)within the occluded vascular territory.
Follow-up infarct volumes were measured by summation of manual planimetric demarcation
of infarct on axial NCCT or diffusion-weighted imaging magnetic resonance imaging (DWI-MRI)
follow-up imaging at 24-32 h.
Interrater Agreement
To determine interrater agreement for scoring of collateral extent and filling dynamics on timevariant
mCTA color maps, a neuroradiologist and a medical student independently reviewed
30 cases in 2 separate reading sessions with a 1-week break between the sessions
(session 1: conventional collateral scoring, session 2: assessment of collateral extent and filling
dynamics on time-variant color maps). The readers had access to the site of occlusion, age
and baseline National Institutes of Health Stroke Scale (NIHSS), but they were blinded to all other
baseline information and patient outcomes.
39
Statistical Analysis
Patient baseline characteristics were described using descriptive statistics. Univariable
and multivariable logistic regression was used to determine the association of conventional
and color-coded collateral scores and a) good outcome, defined as mRS 0–2 at 90 days (primary
outcome), and b) follow-up infarct volume (secondary outcome). Follow-up infarct volume
was hereby included in the models as binary variable (infarct volume below or equal to vs.
above the median infarct volume in the study sample). Information loss across models
was compared using the Akaike and Bayesian information criteria (AIC, BIC) and the area
under the curve (AUC). Adjustment was performed for patient age, sex and baseline NIHSS.
Since the follow-up imaging modality could influence follow-up infarct volume measurements;
sensitivity analysis was performed for follow-up infarct volume as dependent variable for patients
with NCCT vs. DWI-MRI follow-up imaging. Interrater agreement was assessed using the Kappa
statistics. All statistical tests were two-sided and conventional levels of significance (alpha=0.05)
were used for interpretation. All analysis was performed using Stata 15.1 (Stata Corp LLC, College
Station, TX, USA).
Results
Out of 464 patients 285 were included in the analysis. When using the trichotomized grading system
on conventional display format, 60.7% (173/285) patients had good collaterals, 30.2% (86/285)
had intermediate and 9.1% (26/285) poor collaterals. Collateral extent on time-variant color maps
was normal or almost normal in 50.9% (145/285) patients, a collateral extent of 50–90% compared
to the contralateral hemisphere was seen in 34.0% (97/285), and a collateral extent of less than 50%
compared to the contralateral hemisphere in15.1% (43/285). When using time-variant color maps,
there was mostly no delay in 14.4% (41/285), mostly a one-phase delay in 56.5% (161/285)
and mostly a two-phase delay in 29.1% (83/285).
Collateral Grading and Clinical Outcome
Overall, 53.3% of patients (152/285) achieved a good outcome at 90 days. Table 5 shows
unadjusted and adjusted measures of effect size for the association of conventional collateral grade,
color-coded collateral extent, color-coded collateral filling dynamics and good clinical outcome,
as well as the respective AIC, BIC, and AUC values for the multivariable models.
40
Collateral Grading and Follow-up Infarct Volume
Infarct volume was available for 93.0% (265/285) patients. Median final infarct volume was 12.6
ml (IQR 1.7–49.2). Table 6 shows unadjusted and adjusted measures of effect size
for the association of conventional collateral score, color-coded collateral filling dynamics, colorcoded
collateral extent and follow-up infarct volume (infarct volume equal to or below vs.
above the median infarct volume), as well as the respective AIC, BIC, and AUC values
for the multivariable models. Interrater agreement for color-coded grading of collateral filling
dynamics and collateral extent was substantial (Kappa=0.69 and 0.74, respectively).
41
Table 5: Association of conventional and color-map based collateral grade and good clinical
outcome (N = 285)
Collateral score Unadjusted OR
(95% CI)
Adjusted OR a
(95% CI)
AIC b
BIC b
AUC b
(95% CI)
Conventional score 0.62 (0.43–0.89) 0.72 (0.48–1.08) 350.0 368.2 0.74
(0.69–0.80)
Color-map based collateral
extent
0.54 (0.39–0.75) 0.53 (0.36–0.77) 340.9 359.2 0.76
(0.70–0.81)
Color-map based filling
delay
1.21 (0.84–1.74) 1.30 (0.88–1.95) 350.8 369.1 0.74
(0.68–0.80)
Legend: OR odds ratio, 95% CI 95% confidence interval, AIC Akaike information criterion, BIC Bayesian information
criterion, AUC area under the curve
aAdjusted for patient age, sex and baseline National Institutes of Health Stroke Scale
bDerived from adjusted models
42
Table 6: Association of conventional and color-map based collateral grade and final infarct
volume (N = 265)
Collateral score Unadjusted OR
(95% CI)
Adjusted OR a
(95% CI)
AIC b
BIC b
AUC b
(95% CI)
Conventional score 2.05 (1.37–3.07) 1.84 (1.21–2.79) 354.2 372.1 0.67
(0.60–0.73)
Color-map based
collateral extent
2.99 (2.04–4.40) 2.67 (1.80–4.00) 336.4 354.3 0.72
(0.66–0.78)
Color-map based filling
delay
1.29 (0.89–1.87) 1.21 (0.83–1.78) 361.7 379.6 0.63
(0.57–0.70)
Legend: Final infarct volume was coded as a binary variable in this analysis (volume below vs. above the median
infarct volume)
OR odds ratio, 95% CI 95% confidence interval, AIC Akaike information criterion, BIC Bayesian information criterion,
AUC area under the curve
aAdjusted for patient age, sex and baseline National Institutes of Health Stroke Scale.
bDerived from adjusted models
43
Discussion
Our study has the following main findings: 1) color-coded mCTA grading of collateral extent
improves prediction of good outcome at 90 days, and its performance in predicting follow-up
infarct volume is similar compared to conventional collateral grading, 2) color-coded mCTA
grading of collateral filling dynamics performs worse than conventional collateral grading and 3)
interrater agreement for color-coded mCTA grading of collateral extent and filling dynamics
is substantial.
Assessing collateral status on mCTA using a conventional display format, i.e. three separate series
that are usually linked by the reader and then assessed in conjunction, takes both collateral filling
dynamics and extent into account. (44) When using time-variant mCTA color maps, collateral
extent and filling dynamics are graded separately. When color-coded collateral extent was used
to predict good outcome and follow-up infarct volume in our study, information loss was lower
and discrimination better compared to conventional mCTA collateral scoring and color-coded
scoring of filling dynamics. These results potentially indicate that collateral extent reflects tissue
viability more accurately compared to collateral filling dynamics. In a previous study, d’Esterre
et al. assessed collateral extent and filling dynamics on conventional mCTA images and found
that the former was not independently associated with follow-up infarction, while washout,
a parameter that partly reflects filling dynamics, was associated with follow-up infarction. (45)
The apparently contradictory findings between their study and our study could be explained
by the fact that in the current study, the entire hemisphere was assessed, while d’Esterre et al.
evaluated brain tissue per ASPECTS region. Both the current study and the study by d’Esterre et al.
relied on visual assessment of collaterals, which will always be subject to some degree of interrater
variability. This variability could also explain the different results. Automation of collateral scoring
could mitigate this problem, but the automated assessment would have to be available
instantaneously, and integration of thetechnology into routine clinical practice will take some time.
(46) Software to generate time-variant mCTA maps on the other hand is already available,
and the color-maps can be generated within a few seconds. mCTA color-maps therefore constitute
a good alternative to facilitate interpretation of collateral status until fully automated collateral
assessment becomes routinely available, particularly for less experienced readers. Indeed,
when comparing a non-expert to an expert reader, interrater agreement for color-map based
44
collateral grading in our study was substantial. Agreement was higher for color-map based grading
of collateral extent compared to filling dynamics. This suggests that the latter is more challenging,
which could be the reason for the lacking association of collateral filling dynamics and clinical
outcome/follow-up infarct volume. The predictive utility of conventional collateral assessment,
while it was still good overall, was slightly lower when compared to color-map based grading
of collateral extent. It is possible that complications that happened after treatment in the 3-month
follow-up period have influenced the association with clinical outcomes, while the efficacy
of treatment (either EVT or IVT) might have influenced the association of collateral grade
and follow-up infarct volumes, although the latter two points would in theory affect
both conventional and color-map based collateral grading. The exact reasons for the differences
in predictive power remain therefore unknown at the time being.
Limitations
Our study has several limitations: First, assessing infarct volumes on NCCT can be challenging,
since the infarct is often not clearly demarcated. Second, we restricted our analysis to patients
withLVO (including proximal M2 occlusions); our findings can thus not be generalized to more
distal occlusion sites. Third, reperfusion status is an important predictor of infarct volume
and outcome, but since vascular imaging was not available in all patients, we could not stratify
our analysis by reperfusion status. Fourth, recanalization data were missing in a relatively large
number of patients, partly because it was impractical to obtain follow-up vascular imaging in many
local institutional settings, and partly because it does not have a therapeutic consequence in the vast
majority of cases. Fifth, we showed that color-map based assessment of collateral extent
is significantly associated with good outcome and infarct volume in LVO patients, but we could
not assess in our study whether and how this alters clinical decision-making. Doing so would
warrant a diagnostic randomized controlled trial. Such trials generally require very large sample
sizes and are difficult to conduct for various reasons. (47)
45
Section 3 – Endovascular Treatment of Acute Ischemic Stroke
The treatment of AIS has undergone very dramatic changes in last decade. Randomized trials
demonstrated that EVT represents a highly effective and safe treatment. In order to confirm
the broad applicability of EVT in the anterior circulation LVO strokes and to establish the treatment
effect at a national level, we compared the Czech EVT data with the patient-level HERMES metaanalysis
pooling data from the five randomized controlled trials (MR CLEAN, ESCAPE,
REVASCAT, SWIFT PRIME, and EXTEND-IA). Section 3.1 is devoted to this topic in detail.
EVT represents a standard of care for AIS due to LVO, but level 1A guideline
recommendations for EVT are currently restricted to LVO patients with NIHSS ≥ 6.
AIS with NIHSS ≤ 6 is routinely considered as “mild” and/or “non-disabling”. However,
one in four patients with low baseline NIHSS suffer early neurologic deterioration. The aim
of our observational multicenter study was to assess the effectiveness and safety of EVT
versus best medical management in patients with CTA detected LVO in the anterior circulation
and NIHSS ≤ 6 using recent data from comprehensive datasets and propensity scorematching.
Section 3.2 is devoted to this topic in detail.
Additionally, the ANNEX 3 is dedicated to our single-centre experience with patients’ selection
for EVT based on automated CTP analysis and it also compares our results with CTP-based
randomized controlled trials. The ANNEX 4 represents the first comprehensive nationwide
questionnaire-based evaluation of all stroke centres in the Czech Republic performing EVT
in 2016. The ANNEX 5 summarizes the technical EVT results from the year of 2016.
46
Section 3.1 – Mechanical Thrombectomy Performs Similarly in Real-World
Practice: A 2016 Nationwide Study from the Czech Republic
Based upon: VOLNY, Ondrej, Antonin KRAJINA, Silvie BELASKOVA, Michal BAR, Petra CIMFLOVA, Roman
HERZIG, Daniel SANAK, Ales TOMEK, Martin KOCHER, Miloslav ROCEK, Radek PADR, Filip CIHLAR,
Miroslava NEVSIMALOVA, Lubomir JURAK, Roman HAVLICEK, Martin KOVAR, Petr SEVCIK, Vladimir
ROHAN, Jan FIKSA, Bijoy K. MENON a Robert MIKULIK. Mechanical thrombectomy performs similarly in real
world practice: a 2016 nationwide study from the Czech Republic. Journal of Neurointerventional Surgery [online].
2018, 10(8), 741–745. ISSN 1759-8478.
Abstract
Background – Randomized clinical trials have proven EVT to be a highly effective and safe
treatment in acute stroke. The purpose of this study was to compare EVT data from the Czech
Republic (CR) with data from the HERMES meta-analysis.
Methods – Available nationwide data for the CR from the year 2016 from the SITS-TBY registry
on patients with terminal internal carotid artery (ICA) and/or MCA occlusions were compared
with data from the HERMES pooled dataset. CR and HERMES patients were comparable in age,
sex and baseline NIHSS scores.
Results – From a total of 1,053 EVTs performed in the CR, 845 (80%) were reported in the SITSTBY.
From these, 604 (72%) were included in this study. Occlusion locations were as follows
(CR vs. HERMES): ICA 22% vs. 21% (p=0.16), M1 MCA 62% vs. 69% (p=0.004), M2 MCA
16% vs. 8% (p<0.0001). Intravenous thrombolysis was given in 76% vs. 83% patients (p=0.003).
Median onset-to-reperfusion times were comparable: 232 vs. 285 min (p=0.66). A modified TICI
score of 2b/3 was achieved in 74% (433/584) vs. 71% (390/549) of the patients (OR=1.17,
95%CI=0.90-1.5, p=0.24). There was no statistically significant difference in percentage of PH
type 2 (OR=1.12, 95%CI=0.66-1.90, p=0.68). A modified Rankin scale score of 0-2 at 3 months
was achieved in 48% (184/268) vs. 46% (291/633) of the patients (OR=0.92, 95% CI=0.71-1.18,
p=0.48).
Conclusions – Data on efficacy, safety and logistics of EVT from the CR is similar to data
from the HERMES collaboration.
47
Introduction
Recent randomized trials have demonstrated that EVT with second-generation neurothrombectomy
devices represents a highly effective and safe treatment for patients with AIS due to occlusion/s
in the anterior cerebral circulation. (48) (2) (31) (49) (15) Data from the HERMES meta-analysis
have proven the degree of benefit of this procedure to be substantial; for every 100 patients treated,
38 will have a less disabled outcome than those who receive the best possible medical treatment,
and 20 more will achieve functional independence (defined as a mRS < 2). (10)
Nevertheless, several limitations have emerged concerning the methodological background
of the HERMES trial, including the fact that all the data from this study come from high-volume
comprehensive stroke centres with the highest expertise in endovascular treatment of AIS
in the world.
In order to confirm a broad applicability of EVT into a real-world clinical practice and to establish
the treatment effect of EVT at a national level, we compared the available neurothrombectomy data
from the Czech Republic (CR) prospectively collected in the Safe Implementation of Treatments
in Stroke – Thrombectomy (SITS-TBY) registry with data from the HERMES meta-analysis.
We hypothesised that the safety and effectiveness of EVT in real-world clinical practice
at a nationwide level are comparable with the results of the randomized control trials.
48
Methods
Nationwide data from the CR for the year 2016 were collected from the SITS-TBY registry. SITS
(Safe Implementation of Treatments in Stroke) is a non-profit, free to use, research-driven,
independent, international collaboration founded in the Karolinska Institute in Sweden. SITS
was set up as an initiative to provide safe implementation of stroke treatments in routine clinical
practice. It offers a platform for collecting high quality stroke data in over 1600 stroke centres.
The registry is internet-based, which allows rapid data entry and retrieval and allows centres
to compare their own treatment results on both a national and global scale. Data concerning
patients with pre-treatment mRS scores of less than 2 and occlusions in the ICA and/or MCA
treated with second-generation neurothrombectomy devices ± IV tPA (if eligible) were compared
with the data from HERMES meta-analysis.
Demographic details, vascular risk factors and NIHSS scores (range 0-42, with higher scores
indicating severe stroke) were collected for each patient. There was no upper age limit. All patients
underwent NCCT followed by CTA from the aortic arch to the vertex to document LVO. Treatment
decisions were made in comprehensive stroke centres in accordance with the current European
guidelines. (50)
Outcomes included: NIHSS score at 24 hours after stroke onset, median change in NIHSS score
from baseline to 24 hours, parenchymal haematoma type 2 (PH 2) according to the SITS-MOST
criteria (Safe Implementation of Thrombolysis in Stroke-Monitoring Study) and achievement
of a mRS score of 0-2 at 90 days after stroke onset. Technical efficacy was assessed by counting
the number of cases where a mTICI scale score of 2b or 3 was achieved (corresponding
to reperfusion of at least 50% of the affected vascular territory). Reported times included: onset
to reperfusion time, onset to groin time and groin to reperfusion time.
Ethics approval was obtained from the local institutional review boards and written informed
consent was obtained from all patients.
49
Statistical analysis
Categorical variables are presented as absolute values and percentages, and continuous variables
as mean and SD if symmetrically distributed, or otherwise as median and IQR. We used the MannWhitney
test to compare continuous and ordinal variables because we did not have access
to the raw HERMES data. We used the Chi-squared test to compare categorical variables. In order
to determine whether it was possible to use the above-mentioned tests and that the effects
were not due to chance alone we first calculated the intra-cluster correlation coefficient (ICC)
for data of each centre by using Proc Mixed. All the tests were two-sided and significance
was defined as a p-value of 0.05. Statistical analyses were obtained using SAS 9.3 software (SAS
Institute, Cary, NC).
Results
Fourteen out of the 15 comprehensive stroke centres in the CR in 2016 reported data to the SITSTBY
registry. In this year 1,053 EVTs were performed in the CR. The smallest number
of procedures performed by one centre was 17 and the largest was 136. Three centres performed
more than 100 procedures, six centres performed between 50 and 100 procedures
and six performed under 50 procedures per year (17, 34, 34, 34, 43, and 46). (51) Eight hundred
and forty-five patients were reported to the registry. Incompleteness of data and/or patients who
did not meet inclusion criteria led to exclusion of 241 patients. Therefore, the final dataset consisted
of 604 (72%) of the total number of patients, representing 57% of all EVT cases performed in 2016
in the CR (with a pre-treatment mRS score of less than 2 and occlusion of the terminal ICA
and/or MCA treated with second-generation neurothrombectomy devices ± IV tPA). Seventy per
cent of these EVTs were mothership procedures and 30% were drip and ship procedures. CR
and HERMES patients were comparable in age, sex, and baseline NIHSS. More CR patients
were hypertensive and had diabetes mellitus and fewer patients were smokers. Fewer patients
in the SITS-TBY cohort were treated with IV tPA: 76% vs. 83% (p=0.003). Occlusion locations
were as follows (CR vs. HERMES): terminal ICA 22% vs. 21% (p=0.16), M1 MCA 62% vs. 69%
(p=0.004), M2 MCA 16% vs. 8% (p<0.0001). The median times from onset to reperfusion
were comparable: 232 vs. 285 minutes (p=0.66); the median groin-to-reperfusion times
were 58 vs. 63 minutes.
50
Modified TICI 2b/3 was achieved in 74% (433/584) vs. 71% (390/549) of patients, OR 1.17 (95%
CI=0.90-1.51), p=0.24. CR and HERMES patients did not differ in the median change in NIHSS
score from baseline to 24 hours. There was no difference in percentage of PH 2 (5.7 vs. 5.1%),
OR 1.12 (95% CI=0.66-1.90), p=0.68. A modified Rankin scale score of 0-2 at 3 months
was achieved in 48% (184/268) vs. 46% (291/633) of patients, OR 0.92 (95% CI =0.71-1.18),
p=0.48; Figure 9. The ICC for the mRS scores was estimated to be 0.023 and the ICC for NIHSS
at 24 hours was 0.058, indicating only negligible differences among the centres.
51
Table 7: Comparison of Czech SITS-TBY and HERMES data on demographic
characteristics, past medical history, clinical and radiological characteristics, treatment
details and outcomes
Demographic characteristics Czech Republic
cohort
(n=604)
HERMES
(n=634)
p
Median age, years 71 (63-79) 68 (57-77) 0.44
Sex, women 307 (51%) 304 (48%) 0.32
Past medical history
Hypertension 442 (73%) 352 (56%) <0.000
1
Diabetes mellitus 159 (26%) 82 (13%) <0.000
1
Atrial fibrillation 209 (35%) 209 (33%) 0.56
Smoking (recent or current) 90 (15%) 194 (31%) <0.000
1
Clinical characteristics
Baseline NIHSS score 15 (11-18) 17 (14-20) 0.66
Imaging characteristics
ASPECTS on baseline CT not available 9 (7-10) Intracranial
clot location:
Terminal internal carotid artery 136 (22%) 122 (19%) 0.16
M1 segment middle cerebral
artery
372 (62%) 439 (69%) 0.004
M2 segment middle cerebral
artery
97 (16%) 51 (8%) <0.000
1
Extracranial ICA (tandem
lesions)
55 (10%) 61 (10%) 0.987
52
Treatment details
and procedural times (min)
Treatment with intravenous
alteplase
460 (76%) 526 (83%) 0.003
Onset-to-reperfusion time 232
(152-320);
467 patients
285
(210-362)
0.66
Onset-to-groin time 175
(100-767);
426 patients
238
(180-302)
0.66
Groin-to-reperfusion time 58
(44-208);
463 patients
63 Outcomes
and safety:
Modified TICI 2b/3 433/584 (74%) 390/549 (71%) 0.24
NIHSS at 24 hours 8 (4-17) 10.4 (8.7) Median
change in NIHSS score
from baseline to 24 h
-5(-10 to 0) -7 (-12 to -1) 0.66
Parenchymal haematoma type 2 5.7% (26/460) 5.1% (32/629) 0.68
mRS 0-2 after 3 months 48% (184/382) 46% (291/633) 0.48
Legend: Data are median (IQR), n (%), or mean (SD). NIHSS = National Institutes of Health Stroke Scale. ASPECTS
= Alberta Stroke Program Early CT Score. ICA = internal carotid artery. mTICI = modified Thrombolysis in Cerebral
Infarction Score. mRS = modified Rankin scale.
53
Figure 9: Modified Rankin Scale scores at 90 days
Legend: Distribution of scores at 90 days in per cents for the SITS-TBY and HERMES groups.
10
19
17
16
19
12
17
10
16
5
6
11
15
25
HERMES
SITS-TBY
mRS at 90 days
0 1 2 3 4 5 6
54
Discussion
Our analysis of the available national data on mechanical thrombectomy from the SITS-TBY
registry confirms the applicability of neurothrombectomy in the real-world clinical practice.
As two thirds of the patients from the SITS-TBY registry had complete data, our analysis provides
a relatively accurate population-based snapshot of the efficacy of EVT, comparable
to the randomized control trials and previously published real-world thrombectomy experiences.
STRATIS, the largest prospective multicenter (55 US centers) non-randomized registry including
patients undergoing EVT with the Solitaire device, demonstrated that this procedure can be safely
and efficaciously performed in the community setting. (52) Before the revelation of the STRATIS
results, published data concerning EVT was limited, arising from mostly single-center studies
with relatively low numbers of patients. (53) (54) One of the strengths of our study is that
we included nationwide patient data from 14 comprehensive stroke centers from the year 2016,
thus after the publication of the randomized control trials.
A total of 1,053 EVTs with second-generation neurothrombectomy devices were performed
in the CR in 2016. If we take into the account the estimated incidence of acute ischemic stroke
in the CR (211 per 100,000 inhabitants according to a recent stroke epidemiology survey)
then approximately 5% of all acute ischemic stroke patients were treated with EVT in 2016. (55)
This relatively high proportion of thrombectomy patients is consistent with previously reported
data on EVT eligibility, reflecting the high level of organization of acute stroke care in the CR.
(56) (57) (58)
The Institute for Health Information and Statistics of the CR has been collecting medical
information for all patients admitted to all hospitals since 1992. All medical facilities are required
by law to register all admissions and discharges and according to the Guidelines of the Czech
Stroke Society, every patient with a diagnosis of stroke should be hospitalized and receive
appropriate care in a specialized stroke unit. Since September 2016 the Czech Stroke Society have
been monitoring EVTs performed in the CR. Every three months they report data on the number
of EVTs performed per month and per center, the percentage of door-to-needle times under 30 min
in IVT eligible patients, and procedural times, including median onset-to-groin time, median door-
55
to-groin time and median groin-to-reperfusion. Summary data reports including the parameters
mentioned above are sent to all stroke centers and physicians involved in stroke care (neurologists
and interventional radiologists). This serves as an important tool for improving acute stroke care
logistics and reducing procedural times.
Randomized trials have demonstrated that workflow speeds are strongly associated with better
functional outcomes, thus the reduction of procedural times should be targeted in every-day clinical
practice. (59) (60) (61) Available onset-to-reperfusion and onset-to-groin times in our analysis
(467 and 426 patients, respectively) are consistent with the HERMES meta-analysis, suggesting
a good pre-hospital and in-hospital management of EVT candidates across the CR.
A robust predictor of functional outcome is successful reperfusion, defined as achievement
of mTICI score of 2b or 3. Among the 549 HERMES patients who underwent EVT intervention
and had mTICI scores documented, substantial reperfusion was achieved in 390 (71%) patients,
which is consistent with our data; 433 out of 584 patients had mTICI scores of 2b or 3 (74%).
In terms of the effect on improvement of neurological status, comparable the median change
in NIHSS from baseline to 24 hours achieved in the Czech cohort was similar to the HERMES
cohort, indicating a comparable clinical effect even in the non-randomized registry. In terms
of safety, the proportion of patients with symptomatic PH type 2 was also similar (5.7 vs. 5.1%).
Modified Rankin scale scores at 3 months as a marker of long-term disability was available in only
60% of our patients. The proportion of available patients with a good clinical outcome (mRS scores
of 0-2) was comparable (48 vs. 46%). Nevertheless, it is important to be aware of the proportion
of missing long-term outcome data. On the other hand, comparable short-term outcomes mentioned
above (NIHSS scores at 24 hours and median change in NIHSS from baseline to 24 hours) indicate
a positive effect of the procedure on early neurological recovery.
Although cerebral and vascular imaging is done in all IVT and EVT eligible patients
in the CR before treatment decision making, ASPECT scores are not available in the SITS-TBY
registry. (51) There are only two randomized controlled trials on mechanical thrombectomy
in acute stroke published in which baseline cerebral imaging data were not used to select patients
on the basis of the size of ischemic territory as determined by ASPECTS (THRACE
56
and MR CLEAN). (48) (62) According to the current European guidelines and results
of our questionnaire survey run in January 2017, we can estimate that a majority of patients treated
with EVT in the CR in 2016 had ASPECTS > 5 and were treated within 8 hours from stroke onset.
(50) (51)
From demographic standpoint, HERMES and Czech SITS-TBY cohorts were balanced in age, sex
and history of atrial fibrillation (Table 7). The groups differed in proportion of hypertensive,
diabetic and smoking patients. Patients were comparable in the severity of admission neurological
deficit assessed by NIHSS (moderate to severe stroke). HERMES and Czech SITS-TBY
populations differed in clot locations in the M1 and M2 segments of the MCA as assessed by local
radiologists and referred to the SITS-TBY registry, indicating that more interventions occurred
for distal occlusions. Nevertheless, it is important to consider a challenge associated
with a relatively poor standardization in distinguishing between the M1 and M2 MCA segments
rated by different radiologists in our non-randomized analysis without a core lab rating of clot
locations. This limitation of M1/M2 misclassification has been discussed in HERMES
and elsewhere. (10) (63)
Our study has several limitations. We are aware that our comparisons with the clinical trials
mentioned above are merely qualitative, since our data are based on a retrospective analysis
of prospective registry data. As mTICI scoring was performed by the same attending
interventionalist, who performed the procedure, and as clinical evaluation of disability (NIHSS
and mRS) was not completely blinded, there may be a source of bias. Additionally,
there are no imaging data stored in the SITS-TBY registry for neuroimaging core lab reassessment,
thus the possibility for misclassification of M1/M2 segments of MCA on CTA
as discussed above or mTICI on final run digital subtraction angiography exists and limits
our analysis. Another limitation is missing long-term outcome data in the SITS-TBY registry.
57
Section 3.2 – Thrombectomy vs. Medical Management in Low NIHSS Acute
Anterior Circulation Stroke
Based upon: VOLNY, Ondrej, Charlotte ZERNA, Ales TOMEK, Michal BAR, Miloslav ROCEK, Radek PADR,
Filip CIHLAR, Miroslava NEVSIMALOVA, Lubomir JURAK, Roman HAVLICEK, Martin KOVAR, Petr SEVCIK,
Vladimir ROHAN, Jan FIKSA, David CERNIK, Rene JURA, Daniel VACLAVIK, Petra CIMFLOVA, Josep PUIG,
Dar DOWLATSHAHI, Alexander V. KHAW, Enrico FAINARDI, Mohamed NAJM, Andrew M. DEMCHUK, Bijoy
K. MENON, Robert MIKULIK a Michael D. HILL. Thrombectomy vs medical management in low NIHSS acute
anterior circulation stroke. Neurology [online]. 2020.
Abstract
Background – EVT is highly effective for acute ischemic stroke with LVO and moderate
to severe neurological deficits.
Objective: To undertake an effectiveness and safety analysis of EVT in patients with LVO
and NIHSS≤6 using datasets of multicentre and multinational nature.
Methods – We pooled patients with anterior circulation occlusion from three prospective
international cohorts. Patients were eligible if presentation occurred within 12 hours from last
known well and baseline NIHSS≤6. Primary outcome was mRS 0–1 at 90 days. Secondary
outcomes included neurological deterioration at 24 hours (change in NIHSSof≥ 2 points), mRS
0-2 at 90-days and 90-day all-cause mortality. We used propensity score matching to adjust
for non-randomized treatment allocation.
Results – Among 236 patients who fit inclusion criteria, 139 received EVT and 97 received
medical management. Compared to medical management, the EVT group was younger
(65 versus 72 years; p<0.001), had more proximal occlusions (p<0.001), and less frequently
received concurrent IVT (57.7% versus 71.2%; p=0.04). After propensity score matching,
clinical outcomes between the two groups were not significantly different. EVT patients
had an 8.6% (95% CI: -8.8–26.1%) higher rate of excellent 90-day outcome, despite a 22.3%
(95% CI: 3.0–41.6%) higher risk of neurological deterioration at 24 hours.
Conclusions – EVT for LVO in patients with low NIHSS was associated with increased risk
of neurological deterioration at 24 hours. However, both EVT and medical management
resulted in similar proportions of excellent clinical outcomes at 90days.
58
Introduction
Patients with AIS due to LVO usually suffer from severely disabling symptoms. (64)
However, a significant number of LVO patients present with milder symptoms. (65) EVT
is a standard of care for AIS due to LVO, but level 1A guideline recommendations for EVT
are currently restricted to LVO patients with NIHSS ≥ 6, since only a limited number
of patients with low baseline NIHSS was enrolled in the randomized controlled trials. (9) (10)
AIS with NIHSS ≤ 6 is routinely considered as “mild” and “non-disabling”. However,
one in four LVO patients with low baseline NIHSS suffer early neurologic deterioration
resulting in poorer outcome. (66) (67) (68) From a patient perspective, milder deficits
can restrict daily activities and can be devastating to their quality of life. Patients with LVO
and low baseline NIHSS often have distinct clinical, demographic, and hospital arrival
characteristics. (69) Multiple non-randomized studies have sought to evaluate the efficacy
and safety of EVT in such patients and showed mixed results. These studies were mostly
limited by their non-randomized design, small sample size, single-center experiences, varying
practice or including patients treated prior to the efficacy of EVT was proven and incorporated
into the (inter)national guidelines. (69) (70) (71) (72, 73) (74) (75) (76)
The aim of our observational multicentre study was to assess the effectiveness and safety
of EVT versus medical management in patients with LVO and NIHSS ≤ 6 using recent data
from comprehensive datasets and propensity scorematching.
59
Methods
The data that support the findings of this study are available from the corresponding authors
upon reasonable request. We retrospectively identified acute stroke patients with CTA proven
anterior circulation occlusion and admission NIHSS ≤ 6. For this purpose, we retrieved
EVT data from the Safe Implementation of Treatments in Stroke–Thrombectomy registry
(SITS-TBY) and compared them with medical management data derived
from the INTERRSeCT and PROVE-IT study.
Standard Protocol Approvals, Registrations, and Patient Consents
Permissions to analyze data for the SITS-TBY (a non-profit, quality improvement-driven,
international) registry were provided by the ethics committee of St. Anne's University
Hospital, Brno, Czech Republic; individual patient consent for the SITS-TBY registry
was not sought. The PROVE-IT study used a waiver of consent which was approved
by the Conjoint Health Research Ethics Board at the University of Calgary. Written informed
consent was provided by the patient or a surrogate for the INTERSeCT study. The reviews
of the Institutional Review Boards (IRB) for each study determined that informed consent
was not required for this current pooled analysis. The study protocol was approved
by the scientific committees of each study.
Endovascular data source
National EVT data from the CR were extracted from the population-based SITS-TBY registry
from January 2015 to December 2018 to cover the time period after the publication of positive
endovascular trials. The SITS-TBY registry represents a non-profit, research-driven,
international registry collecting data on endovascular treatment. Anonymized patient-level data
are entered at each stroke center either by a research nurse or physician at discharge or at 90days
follow-up. There has been no formal audit of the Czech SITS-TBY or global SITS-TBY
data. However, 12 (out of 15) comprehensive stroke centers participated in the registry in 2016
as part of a quality improvement program. Random hospital-level metrics reported
to the Czech Ministry of Health were cross-checked with the SITS data and showed high level
of consistency (unpublished). Additionally, a nationwide questionnaire survey run in 2016
did not show major differences in clinical practise including neuroimaging, logistics
and treatment standards in all 15 comprehensive stroke centers in the CR, for details see
60
ANNEX 4). Furthermore, since 2016, the Czech Stroke Society has been providing feedback
quarterly to all participating stroke centers on number of EVT cases and time metrics based
on the data from the registry. Patients for endovascular treatment in the CR are selected
through CT and CTA imaging.
Medical management data source
Medical management was based on the current guidelines (Canadian Stroke Best Practise
Recommendations), American Heart and Stroke Association), including IVT in patients
presenting within the first 4.5 hours from last seen normal. In patients not eligible for IVT,
an antiplatelet agent was administered on day 1, unless there was an indication for early
anticoagulation.
The two non-thrombectomy cohorts were selected from the multicentre international
observational studies: 1) Measuring Collaterals with Multi-phase CT Angiography in patients
with Ischemic Stroke (PROVE-IT, patient enrolment between July 2014 to October 2017);
and 2) Identifying New Approaches to Optimise Thrombus Characterization for Predicting
Early Recanalization and Reperfusion With IV Alteplase and Other Treatments Using Serial
CT Angiography study (INTERRSeCT, patient enrolment between March 2010 to March
2016).
PROVE-IT was a prospective multi-center cohort study of 500 consecutive patients with AIS
presenting within 12 hours of stroke symptom onset with evidence of intracranial occlusion
on routine CTA. The primary aim of this trial was to evaluate imaging selection
for thrombolysis and EVT decision-making in the setting of AIS. (77) INTERRSeCT
was a multicentre prospective cohort study that enrolled 575 patients with AIS
with intracranial thrombi documented via CTA. The study included patients with a wide range
of clinical presentations (within 12 hours from last known well), occlusion sites, and thrombus
characteristics to identify clinical and imaging variables associated with recanalization
with or without IVT. (78)
For the current study, we only included patients who were independently functioning
in the community immediately prior to their stroke (estimated baseline mRS 0-2). Patients
61
were further eligible if they presented to the emergency department with symptoms consistent
with AIS 12 hours from time last known well, baseline NIHSS≤6, and baseline CTA
with the evidence of symptomatic intracranial occlusion (ICA or MCA including
M1 and proximal M2 segments). Patients with the primary posterior circulation occlusions
wereexcluded.
Demographics, Variables, and Measurements
Information on baseline demographics, vascular risk factors (hypertension, diabetes mellitus,
dyslipidaemia, atrial fibrillation, smoking history (current/past), congestive heart failure), time
last seen normal, NIHSS score (range, 0-42, with higher scores indicating severe stroke),
occlusion location (ICA, M1, M2 MCA), prior use of anticoagulation, prior use
of antithrombotic treatment, intravenous alteplase administration (if applicable although many
low NIHSS patients are not thrombolysed), were collected. Other clinical endpoints were 24-h
NIHSS and functional outcome at 90 days measured on the mRS.
Study outcomes
Modified Rankin Scale 0–1 at 90 days was chosen as the primary outcome because patients
with mild deficits at baseline are more likely to have excellent outcomes. Secondary outcomes
were the neurological deterioration at 24 hours (defined as increase of NIHSS score
by 2 or more points)(79), mRS 0-2 at 90 days and all-cause mortality at 90days.
Missing/incomplete data
Among 236 patients, four had missing baseline NIHSS, five missing prior anticoagulation
history, six missing prior smoking history, six prior atrial fibrillation history, one missing prior
hypertension and prior dyslipidemia history; 14 missing 24-h NIHSS, 13 missing 90-days
mRS. We imputed the missing NIHSS baseline values with the group median
from the remaining available data and imputed “no” for missing binary variables.
The 13 missing values for 90-day mRS were again imputed with the median of the remaining
available data. Since all missing mRS occurred in the EVT group, we performed additional
sensitivity analysis assuming that the missing 90-day mRS values indicated that the person
did not reach a favorable outcome and was disabled/dead (i.e. worst-case scenario),
did achieve a favorable outcome (i.e. best-case scenario), and omitted cases with missing mRS
62
scores. For the 14 missing values of 24h-NIHSS (all but one from the EVT group) we assumed
that the patients had neurological deterioration (i.e. worst-case scenario).
Data about intracranial hemorrhages were missing and unverifiable in the SITS-TBY registry
and could thus not be analyzed in our current study. Since the SITS-TBY registry is a study
of implementation of thrombectomy in routine clinical practice, it was not mandated that time
metrics were being collected. Onset-to-treatment times were thus not available for our analysis.
Statistical analysis
Standard descriptive statistics were used to measure central tendency and variability
of baseline characteristics. Ordinal/continuous variables were compared by the MannWhitney
test or t-test based on their distribution. Categorical variables were compared using
the Fisher’s exact test. Since our data were not randomized, we used propensity score matching
to estimate the adjusted treatment effect of EVT compared to the best medical management,
accounting for differences in baseline variables. We used the treatment effect option
with propensity score matching in STATA version 14.2 (College Station, TX). Derivation
of standard error accounted for the fact that propensity scores are estimated rather than known.
Propensity scores were derived from a multivariable logistic regression model that calculates
the treatment probability for each subject. This model was adjusted for the following clinically
relevant baseline variables: sex, age, occlusion location, thrombolysis status, baseline mRS,
prior antithrombotic treatment and NIHSS. The propensity scores were then used to impute
the missing potential outcome (if a subject received EVT then medical management
is considered counterfactual and if the subject received medical management then EVT
is considered counterfactual) for each subject by taking the outcome of a similar subject that
received the other treatment level (or multiple subjects if there was a tie for similarity).
Similarity between subjects was based on the propensity scores. Common support
was assessed using an overlap plot and examination of mean propensity scores by treatment
group and quintiles and found to be adequate (Table 8). The treatment effect was computed
by taking the average of the difference amongst each EVT and medical management pairs,
where outcomes were either observed or derived as the counterfactual from the propensity
matching process and presented as a risk difference with 95% confidence interval (CI).
We used this method of matching and analysis for our primary, secondary and safety
63
outcomes. We visualized the unadjusted data and the results of our primary analysis using
horizontally stacked bar graphs. Sensitivity analysis was performed for the primary outcome
using the worst-case scenario, best-case scenario and by omitting cases with missing primary
outcome as described above. All tests were two-sided and the significance level
was considered as 0.05. Statistical analyses were performed using STATA version 14.2
(College Station, TX).
Results
Baseline Characteristics
Our pooled dataset resulted in 281 patients with LVO and mild symptoms. We excluded eleven
patients who were not independent at baseline, 33 patients with distal M2, M3
and no occlusion, and one patient treated with tenecteplase (TNK). This left 236 patients
for analysis; 139 received EVT and 97 medical management. The two groups had similar
baseline NIHSS, baseline mRS and baseline vascular risk factors. The EVT group was younger
(65 versus 72 years), with more proximal occlusions (50.4% M1 and 15.8% ICA versus 25.8%
M1 and 2.1 ICA), and less concurrent intravenous alteplase treatment (57.7% versus 71.2%)
as illustrated in Table 9.
Primary and secondary outcomes
Ninety-day excellent outcome (mRS 0-1) was achieved in 62.7% (n=148) of patients overall,
with no difference between the EVT and medical management group (61.9 % versus 63.9 %,
p=0.785) in unadjusted analysis. The raw distribution of mRS scores between the EVT
and medical management group at 90 days is shown in Figure 10. After propensity
score matching, patients in the EVT group had an 8.6% (95% CI: -8.8% – 26.1%) higher
chance of excellent outcome at 90 days compared to the medical management group.
The distribution of mRS scores of the EVT and medical management group at 90 days
after the propensity score matching is presented in Figure 11. The result was unchanged
in sensitivity analysis using the worst-case scenario (missing outcomes assumed
to have achieved mRS 2-6) and when cases with missing mRS were omitted (p=0.33
and p=0.250, respectively). However, assuming best-case scenario (missing outcomes actually
achieved mRS 0-1), patients in the EVT group had a 17.6% (95% CI: 0.01% – 35.4%) higher
chance of excellent outcome at 90 days compared to the medical management group.
64
Unadjusted analyses of the secondary outcomes are shown in Table 10. After propensity score
matching, patients in the EVT group had a 22.3% (95% CI: 3.0% – 41.6%) higher risk
of neurological deterioration at 24 hours compared to patients in the medical management
group. Patients in the EVT group also had a 2.2% (95% CI: -3.6% - 7.9%) higher risk of death
from any cause within the first 90 days after the index event compared to the medical
management group.
65
Table 8: Mean Propensity Scores by Quintiles and Treatment Group
Propensity Score
Quintile
No.
of observations
Mean
Propensity Score
Medical management 1 16 0.149
2 32 0.324
3 32 0.505
4 11 0.702
5 6 0.903
Thrombectomy 1 - -
2 15 0.316
3 34 0.508
4 37 0.711
5 53 0.915
66
Table 9: Baseline Characteristics before Propensity Score Matching Process
Variable Medical
Management group
N = 97
Endovascular
group
N =139
Median age in years (25% – 75%) 72 (63 – 80) 65 (55 – 75)
Sex, male, % 48.9 43.4
Occlusion site, %
Internal carotid artery
2.1 15.8
Tandem occlusion
2.1 3.6
M1 segment
25.8 50.4
Proximal M2 segment
70 30.2
Median baseline NIHSS (25% – 75%) 5 (4 – 6) 4 (3 – 6)
Baseline modified Rankin Scale, %
0
87.6 87.8
1
7.2 5.0
2
5.2 7.2
Intravenous alteplase treatment, % 71.2 57.7
Prior anticoagulation, % 9.4 10.3
Prior antithrombotic treatment, % 25.2 44.3
67
Hypertension, % 65.5 68
Diabetes mellitus, % 14.4 10.8
Dyslipidemia, % 26.6 38
Atrial fibrillation, % 23.7 28.9
Smoking current/past, % 28.9 22.3
Ischemic heart disease, % 6.5 3.1
Legend: NIHSS means National Institutes of Health Stroke Scale
68
Table 10: Unadjusted Outcome Analysis
Outcome Medical
Management
group
N = 97
Endovascular
group
N =139
Fisher’s
exact test,
p-value
modified Rankin Scale score 0-1
at 90 days, % 63.9 61.9 0.785
Neurological deterioration
at 24 hours, % 10.3 30.2 <0.001
modified Rankin Scale score 0-2
at 90 days, % 79.4 69.1 0.100
All-cause mortality
at 90 days, % 3.1 5.0 0.532
69
Figure 10: Unadjusted analysis of 90-day modified Rankin Scale shift
Legend: EVT means endovascular treatment, MM medical management, mRS modified Rankin scale.
70
Figure 11: Propensity-Score matched analysis of 90-day modified Rankin Scale shift
Legend: EVT means endovascular treatment, MM medical management, mRS modified Rankin Scale.
71
Discussion
In our study endovascular treatment and best medical management for large vessel anterior
circulation occlusion in patients presenting with low NIHSS resulted in similar proportions
of excellent functional outcome at 90 days and comparable all-cause 90-day mortality.
This outcome parity occurred despite an increased endovascular treatment risk of neurological
deterioration at 24 hours.
In keeping with our results, other smaller multicenter studies have utilized propensity score
matching and found no significant difference in the excellent functional outcome at 90-days.
(70) Although the study by Nagel et al (77 matched pairs) showed a 14.4% absolute difference
in good clinical outcome (84.4% versus 70.1%, p=0.03) defined as mRS 0-2 and an adjusted
OR of 3.1 (95% CI, 1.4-6.9) favoring immediate EVT, there was no such difference seen
for excellent outcome defined as mRS 0-1 at 90 days. (71) Additionally, the study enrolled 7%
of patients with basilar occlusions and 6% of patients with initial mRS>2 (20/300 patients).
In the study by Haussen et al, the protocol also allowed inclusion of patients with basilar
occlusions, which then made up 23% of the EVT group. (73) These and other previously
published (mostly single-center) studies have selected individual sites with variability in their
approach to patient care and differing local treatment guidelines. A more recent study
by Asdaghi et al looked at over 400 registry patients and found association of EVT
with favorable discharge outcomes and ambulatory status. (69) However, the 90-day outcomes
as well as the occlusion status and thrombus location were not documented consistently.
All of the above mentioned studies included patients who received EVT before the publication
of positive randomized control trials in 2015 and thus the incorporation of EVT as a standard
of care into national guidelines. Thus, the results of these studies might be reflective
of the heterogeneity in workflow and experience with such patients and might
not be representative of the current clinical practice.
We found no difference in all-cause 90-day mortality in our study, which is in congruence
with the smaller multicentre studies of Nagel et al and Dargazanli et al. In the study of Sarraj
et al, the patients undergoing EVT had higher mortality (8.9% versus 1.1%, p=0.03) possibly
driven by increased risk of symptomatic ICH (5.8% versus 0%). Similarly, Asdaghi et al
reported mortality of 5.2% and symptomatic ICH rates of4.5%.
72
Endovascular treatment was associated with increased risk of neurological deterioration
at 24hours (defined as ≥ 2 points increase of the NIHSS scale) in our propensity-score matched
analysis. One possible explanation might be the occurrence of symptomatic ICH
with the endovascular treatment. Neurological deterioration might have also been more
apparent on formal testing in a setting of mild initial symptoms and thus been more diligently
scored. Further, other complications of EVT including embolic events into other arterial
territories, arterial access adverse events such as haemorrhage, retroperitoneal hematoma
and pseudoaneurysm formation may impact 24-hour assessment.
Multiple studies have shown that since low NIHSS patients are generally considered too mild
for thrombolysis and EVT, up to one third end up disabled or dead at the 90-day follow-up
when left hyperacutely untreated. (80) (81) (82) (83) It is known that in stroke due to (large)
vessel occlusion, there is a clear relationship between recanalization and favorable/excellent
outcome even though our current study and various other have shown differing effect sizes
(from 8.6% to 14.4%). (71) (84) Yet, interventional treatment, whether medical
with IVT or interventional with endovascular thrombectomy has possible harm. The value
of a future randomized controlled trial in this patient cohort is thus not to show the benefit
of EVT but rather to assess if the benefit outweighs the potential harm of the treatment. Data
about intracranial hemorrhages were missing and unverifiable in the SITS-TBY registry
and could thus not be analyzed in our current study. But two previous studies in low NIHSS
strokes that have specifically measured sICH found a notable difference between the EVT
and medical management group (Sarraj et al 5.8% vs. 0% [p=0.02], Nagel et al 5% vs. 1.4%
[p=0.08]). (70) (71) However, these are sICH risks that we accept for moderate-severely
disabling stroke and the acceptable risk of sICH must be significantly less in low NIHSS strokes
to justify the risk of death and disability as a complication of EVT. Due to this uncertainty
in numbers, despite several analyses from groups around the world, a well-designed
randomized controlled trial would be able to finally answer the question about the risk-benefitratio
of EVT for low NIHSS strokes.
The strength of our study is its multicenter and multinational nature of the utilized datasets.
The national population-based EVT data were extracted from the SITS-TBY registry
73
from January 2015 to December 2018 in order to cover the time period after the publication
of positive endovascular trials and as such reflect the current clinical practice.
Our study is limited by its retrospective nature and even though we tried to account
for confounding by using advanced statistical methods, there is still a risk of residual
confounding due to unmeasured variables. For example, even though we incorporated
occlusion location into our propensity score model, we were unable to also incorporate
a measure of early ischemic changes since these data were not available in the SITS-TBY
registry. Furthermore onset-to-treatment times were not available for our analysis. Although
twelve of total 15 comprehensive stroke centers in the Czech Republic contributed to the SITSTBY
registry during the study period. Our previous study showed that over 80% of all EVT
cases in the CR were reported to the registry in 2016. (85) We also had no data
on the use of anesthesia/sedation during EVT available. Blood pressure changes during
induction of general aneasthesia may risk penumbral tissue perfusion and might thus
contribute to the neurological deterioration at 24 hours. (86) (87) The mRS, even though
one of the most commonly used clinical outcome markers in stroke and captured by the three
observational data sources we have used, lacks sensitivity at the minor disability
end of the scale and we might thus have not been able to detect a significant difference
in our primary outcome. Our matched analysis is larger than the sample size of prior studies
but still might have affected our statistical power to detect a true difference.
74
Conclusions
The first study (Section 2.1) comparing the assessment of early ischemic changes by expert reading
and available automated software for NCCT and CTP demonstrated high accuracy
for the evaluation of early ischemic changes by different CT modalities. The best accuracy
was found for the CBF<30% and Tmax>10s parameters. The use of automated software in daily
clinical practice has a potential to improve detection and extent of early ischemic changes.
The second study (Section 2.2) focused on utility of time-variant multiphase CTA color maps
brought new evidence that collateral extent assessed on the time-variant mCTA maps improved
prediction of good clinical outcome and had similar utility in predicting follow-up infarct volume
compared to conventional mCTA collateral grading as previously published.
The third study (Section 3.1) comparing EVT data from the Czech Republic with data
from the HERMES meta-analysis confirmed the applicability of EVT in a nationwide real-world
practice for a CTA proven LVOs in the anterior cerebral circulation.
Our multicenter observational post-hoc study (Section 3.2) showed that the EVT for LVO
in patients with low baseline NIHSS resulted in similar 90-day clinical outcomes compared
to the best medical management despite an increased odds of neurological deterioration
at 24 hours.
75
List of Abbreviations
ACC – accuracy
AIC – acute ischemic stroke
ASPECTS – Alberta Stroke Program Early CT Score
AUC – area under the curve
BIC – Bayesian information criteria
CBF – cerebral blood flow
CBV – cerebral blood volume
CI – confidence interval
CT – computed tomography
CTA – computed tomography angiography
CTP – computed tomography perfusion
CR – Czech Republic
DWI-MRI – diffusion-weighted imaging – magnetic resonance imaging
EVT – endovascular treatment
HU – Hounsfield units
ICA – internal carotid artery
ICC – intra-cluster correlation coefficient
ICH – intracerebral haematoma
IQR – interquartile range
IVT – intravenous thrombolysis
LVO – large vessel occlusion
Ls-means – least square means
MCA – middle cerebral artery
mCTA – multiphase CTA
MIPs – maximum intensity projections
MT – mechanical thrombectomy
mTICI – modified Thrombolysis in Cerebral Infarction
MTT – mean transit time
mRS – modified Rankin Scale
76
NCCT – non-contrast CT
NIHSS – National Institutes of Health Stroke Scale
NPV – negative predictive value
OR – odds ratio
PH – parenchymal haematoma
PPV – positive predictive value
SD – standard deviation
SITS-MOST – Safe Implementation of Thrombolysis in Stroke-Monitoring Study
SITS-TBY – Safe Implementation of Treatments in Stroke – Thrombectomy registry
TICI – Thrombolysis in Cerebral Infarction
Tmax – time to maximum
TNK – tenecteplase
TNR – true negative value
tPA – tissue plasminogen activator
TPR – true positive value
WL – window level
WW – window width
77
List of Figures
Figure 1: Comparison of the standard brain window and “hard brain” window, page 11
Figure 2: Conventional and color-based collateral scoring, page 13
Figure 3: Comparison of CT imaging modalities and evaluation of early ischemic changes, page 22
Figure 4: Accuracy, sensitivity, specificity, positive predictive value, negative predictive values
of baseline ASPECTSs evaluated by e-ASPECTS, consensus (expert reading), CBF<30%
and Tmax>10s), page 23
Figure 5: Bland-Altman plots, page 24
Figure 6: Subgroup analysis of patients with successful reperfusion – accuracy, sensitivity,
specificity, positive predictive value and negative predictive value for baseline assessment (eASPECTS,
consensus, CBF<30% and Tmax>10s) and final ischemic changes on follow-up NCCT,
page 25
Figure 7: Bland-Altman plots for the subgroup analysis of patients with successful
recanalization/reperfusion (MT and IVT group pooled data), page 26
Figure 8: Least square means estimates of fixed effect “region” computed from generalized mixed
model, page 27
Figure 9: Modified Rankin Scale scores at 90 days, page 53
Figure 10: Unadjusted analysis of 90-day modified Rankin Scale shift, page 69
Figure 11: Propensity-Score matched analysis of 90-day modified Rankin Scale shift, page 70
78
List of Tables
Table 1: Accuracy, sensitivity, specificity, PPV and NPV of baseline ASPECTS (e-ASPECTS,
consensus, CBF<30% and Tmax>10s) vs. follow-up imaging, page 28
Table 2: Comparison of patient baseline characteristics for the whole dataset (n=81)
and a subgroup of patients with determined successful recanalization (n=33), page 29
Table 3: Subgroup analysis of patients with successful reperfusion/recanalization accuracy,
sensitivity, specificity, PPV and NPV of baseline ASPECTSs (e-ASPECTS, consensus, CBF<30%
and Tmax>10s) vs. follow-up imaging, page 30
Table 4: Comparison of residuals for the follow-up ASPECTS and the baseline ASPECTS
for the subgroups of patients with determined successful recanalization versus non-determined
recanalization, page 31
Table 5: Association of conventional and color-map based collateral grade and good clinical
outcome (N = 285), page 41
Table 6: Association of conventional and color-map based collateral grade and final infarct volume
(N = 265), page 42
Table 7: Comparison of Czech SITS-TBY and HERMES data available on demographic
characteristics, past medical history, clinical and radiological characteristics, treatment details and
outcomes, page 51
Table 8: Mean Propensity Scores by Quintiles and Treatment Group, page 65
Table 9: Baseline Characteristics before Propensity Score Matching Process, page 66
Table 10: Unadjusted Outcome Analysis, page 68
79
Annex 1
VOLNY O., P. CIMFLOVA, T.-Y. LEE, B. K. MENON and C. D. D’ESTERRE. Permeability
surface area product analysis in malignant brain edema prediction – A pilot study. Journal
of the Neurological Sciences [online]. 2017, 376, 206–210. ISSN 0022-510X
80
81
82
83
84
85
Annex 2
OSPEL J. M., O. VOLNY, W. QIU, M. NAJM, N. KASHANI, M. GOYAL a B. K. MENON.
Displaying Multiphase CT Angiography Using a Time-Variant Color Map: Practical
Considerations and Potential Applications in Patients with Acute Stroke. American Journal
of Neuroradiology [online]. 2020, 41(2), 200–205. ISSN 0195-6108.
86
87
88
89
90
91
92
Annex 3
VANICEK J., P. CIMFLOVA, M. BULIK, J. JARKOVSKY, V. PRELECOVA, V. SZEDER a O.
VOLNY (senior author). Single-Centre Experience with Patients Selection for Mechanical
Thrombectomy Based on Automated Computed Tomography Perfusion Analysis-A Comparison
with Computed Tomography Perfusion Thrombectomy Trials. Journal of Stroke
& Cerebrovascular Diseases [online]. 2019, 28(4), 1085–1092. ISSN 1052-3057.
93
94
95
96
97
98
99
100
101
Annex 4
VOLNY O., M. BAR, A. KRAJINA, P. CIMFLOVA, L. KASICKOVA, R. HERZIG, D. SANAK,
O. SKODA, A. TOMEK, D. SKOLOUDIK, D. VACLAVIK, J. NEUMANN, M. KOCHER, M.
ROCEK, R. PADR, F. CIHLAR a R. MIKULIK. A Comprehensive Nationwide Evaluation
of Stroke Centres in the Czech Republic Performing Mechanical Thrombectomy in Acute Stroke
in 2016. Ceska a Slovenska Neurologie a Neurochirurgie [online]. 2017, 80(4), 445–450.
ISSN 1210-7859.
102
103
104
105
106
107
108
Annex 5
KÖCHER M., D. SANAK, J. ZAPLETALOVA, F. CIHLAR, D. CZERNY, D. CERNIK,
P. DURAS, L. ENDRYCH, R. HERZIG, J. LACMAN, M. LOJIK, S. OSTRY, R. PADR,
V. ROHAN, M. SKORNA, M. SRAMEK, L. STERBA, D. VACLAVIK, J. VANICEK, O. VOLNY
and A. TOMEK. Mechanical Thrombectomy for Acute Ischemic Stroke in Czech Republic:
Technical Results from the Year 2016. Cardiovascular and Interventional Radiology [online].
2018, 41(12), 1901–1908. ISSN 0174-1551.
109
110
111
112
113
114
115
116
117
References
1. Lopez AD, Mathers CD, Ezzati M, Jamison DT, Murray CJ. Global and regional burden
of disease and risk factors, 2001: systematic analysis of population health data. Lancet.
2006;367(9524):1747-57.
2. Goyal M, Demchuk AM, Menon BK, Eesa M, Rempel JL, Thornton J, et al.
Randomized assessment of rapid endovascular treatment of ischemic stroke. N Engl J Med.
2015;372(11):1019-30.
3. Feigin VL, Forouzanfar MH, Krishnamurthi R, Mensah GA, Connor M, Bennett DA, et al.
Global and regional burden of stroke during 1990-2010: findings from the Global Burden
of Disease Study 2010. Lancet. 2014;383(9913):245-54.
4. Saver JL. Time is brain--quantified. Stroke. 2006;37(1):263-6.
5. Astrup J, Siesjo BK, Symon L. Thresholds in cerebral ischemia - the ischemic penumbra. Stroke.
1981;12(6):723-5.
6. Menon BK, Smith EE, Modi J, Patel SK, Bhatia R, Watson TW, et al. Regional leptomeningeal
score on CT angiography predicts clinical and imaging outcomes in patients with acute anterior
circulation occlusions. AJNR Am J Neuroradiol. 2011;32(9):1640-5.
7. Smith WS, Sung G, Saver J, Budzik R, Duckwiler G, Liebeskind DS, et al.
Mechanical thrombectomy for acute ischemic stroke: final results of the Multi MERCI trial. Stroke.
2008;39(4):1205-12.
8. Mikulik R, Ribo M, Hill MD, Grotta JC, Malkoff M, Molina C, et al. Accuracy of serial National
Institutes of Health Stroke Scale scores to identify artery status in acute ischemic stroke.
Circulation. 2007;115(20):2660-5.
9. Powers WJ, Rabinstein AA, Ackerson T, Adeoye OM, Bambakidis NC, Becker K, et al.
Guidelines for the Early Management of Patients With Acute Ischemic Stroke: 2019 Update to the
2018 Guidelines for the Early Management of Acute Ischemic Stroke: A Guideline for Healthcare
Professionals From the American Heart Association/American Stroke Association. Stroke.
2019;50(12):e344-e418.
10. Goyal M, Menon BK, van Zwam WH, Dippel DW, Mitchell PJ, Demchuk AM, et al.
Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual
patient data from five randomised trials. Lancet. 2016;387(10029):1723-31.
118
11. Hill MD, Demchuk AM, Goyal M, Jovin TG, Foster LD, Tomsick TA, et al. Alberta Stroke
Program early computed tomography score to select patients for endovascular treatment:
Interventional Management of Stroke (IMS)-III Trial. Stroke. 2014;45(2):444-9.
12. Wardlaw JM, Mielke O. Early signs of brain infarction at CT: observer reliability and outcome
after thrombolytic treatment--systematic review. Radiology. 2005;235(2):444-53.
13. Barber PA, Demchuk AM, Zhang J, Buchan AM. Validity and reliability of a quantitative
computed tomography score in predicting outcome of hyperacute stroke before thrombolytic
therapy. ASPECTS Study Group. Alberta Stroke Programme Early CT Score. Lancet.
2000;355(9216):1670-4.
14. Albers GW, Marks MP, Kemp S, Christensen S, Tsai JP, Ortega-Gutierrez S, et al.
Thrombectomy for Stroke at 6 to 16 Hours with Selection by Perfusion Imaging. N Engl J Med.
2018;378(8):708-18.
15. Campbell BC, Mitchell PJ, Kleinig TJ, Dewey HM, Churilov L, Yassi N, et al.
Endovascular therapy for ischemic stroke with perfusion-imaging selection. N Engl J Med.
2015;372(11):1009-18.
16. Nogueira RG, Jadhav AP, Haussen DC, Bonafe A, Budzik RF, Bhuva P, et al.
Thrombectomy 6 to 24 Hours after Stroke with a Mismatch between Deficit and Infarct. N Engl J
Med. 2018;378(1):11-21.
17. Menon BK, d'Esterre CD, Qazi EM, Almekhlafi M, Hahn L, Demchuk AM, et al.
Multiphase CT Angiography: A New Tool for the Imaging Triage of Patients with Acute Ischemic
Stroke. Radiology. 2015;275(2):510-20.
18. Puetz V, Dzialowski I, Hill MD, Demchuk AM. The Alberta Stroke Program Early CT Score
in clinical practice: what have we learned? Int J Stroke. 2009;4(5):354-64.
19. Grotta JC, Chiu D, Lu M, Patel S, Levine SR, Tilley BC, et al. Agreement and variability in the
interpretation of early CT changes in stroke patients qualifying for intravenous rtPA therapy.
Stroke. 1999;30(8):1528-33.
20. Pfaff J, Herweh C, Schieber S, Schonenberger S, Bosel J, Ringleb PA, et al. eASPECTS
Correlates with and Is Predictive of Outcome after Mechanical Thrombectomy. AJNR
Am J Neuroradiol. 2017;38(8):1594-9.
119
21. Herweh C, Ringleb PA, Rauch G, Gerry S, Behrens L, Mohlenbruch M, et al. Performance of eASPECTS
software in comparison to that of stroke physicians on assessing CT scans of acute
ischemic stroke patients. Int J Stroke. 2016;11(4):438-45.
22. Nagel S, Sinha D, Day D, Reith W, Chapot R, Papanagiotou P, et al. e-ASPECTS software
is non-inferior to neuroradiologists in applying the ASPECT score to computed tomography scans
of acute ischemic stroke patients. Int J Stroke. 2017;12(6):615-22.
23. Murphy BD, Fox AJ, Lee DH, Sahlas DJ, Black SE, Hogan MJ, et al.
Identification of penumbra and infarct in acute ischemic stroke using computed tomography
perfusion-derived blood flow and blood volume measurements. Stroke. 2006;37(7):1771-7.
24. Wintermark M, Flanders AE, Velthuis B, Meuli R, van Leeuwen M, Goldsher D, et al.
Perfusion-CT assessment of infarct core and penumbra: receiver operating characteristic curve
analysis in 130 patients suspected of acute hemispheric stroke. Stroke. 2006;37(4):979-85.
25. Aviv RI, Mandelcorn J, Chakraborty S, Gladstone D, Malham S, Tomlinson G, et al.
Alberta Stroke Program Early CT Scoring of CT perfusion in early stroke visualization and
assessment. AJNR Am J Neuroradiol. 2007;28(10):1975-80.
26. Sillanpaa N, Saarinen JT, Rusanen H, Hakomaki J, Lahteela A, Numminen H, et al.
CT Perfusion ASPECTS in the Evaluation of Acute Ischemic Stroke: Thrombolytic Therapy
Perspective. Cerebrovasc Dis Extra. 2011;1(1):6-16.
27. Padroni M, Bernardoni A, Tamborino C, Roversi G, Borrelli M, Saletti A, et al. Cerebral Blood
Volume ASPECTS Is the Best Predictor of Clinical Outcome in Acute Ischemic Stroke:
A Retrospective, Combined Semi-Quantitative and Quantitative Assessment. PLoS One.
2016;11(1):e0147910.
28. Naylor J, Churilov L, Rane N, Chen Z, Campbell BCV, Yan B. Reliability and Utility
of the Alberta Stroke Program Early Computed Tomography Score in Hyperacute Stroke. J Stroke
Cerebrovasc Dis. 2017;26(11):2547-52.
29. Mokin M, Levy EI, Saver JL, Siddiqui AH, Goyal M, Bonafe A, et al. Predictive Value
of RAPID Assessed Perfusion Thresholds on Final Infarct Volume in SWIFT PRIME (Solitaire
With the Intention for Thrombectomy as Primary Endovascular Treatment). Stroke.
2017;48(4):932-8.
120
30. Albers GW, Goyal M, Jahan R, Bonafe A, Diener HC, Levy EI, et al.
Ischemic core and hypoperfusion volumes predict infarct size in SWIFT PRIME. Ann Neurol.
2016;79(1):76-89.
31. Saver JL, Goyal M, Bonafe A, Diener HC, Levy EI, Pereira VM, et al. Stentretriever
thrombectomy after intravenous t-PA vs. t-PA alone in stroke. N Engl J Med.
2015;372(24):2285-95.
32. Nogueira RG, Haussen DC, Dehkharghani S, Rebello LC, Lima A, Bowen M, et al.
Large Volumes of Critically Hypoperfused Penumbral Tissue Do Not Preclude Good Outcomes
After Complete Endovascular Reperfusion: Redefining Malignant Profile. Stroke. 2016;47(1):94-
8.
33. Olivot JM, Mlynash M, Inoue M, Marks MP, Wheeler HM, Kemp S, et al.
Hypoperfusion intensity ratio predicts infarct progression and functional outcome in the DEFUSE
2 Cohort. Stroke. 2014;45(4):1018-23.
34. Mlynash M, Lansberg MG, De Silva DA, Lee J, Christensen S, Straka M, et al.
Refining the definition of the malignant profile: insights from the DEFUSE-EPITHET pooled data
set. Stroke. 2011;42(5):1270-5.
35. Kharitonova T, Mikulik R, Roine RO, Soinne L, Ahmed N, Wahlgren N, et al.
Association of early National Institutes of Health Stroke Scale improvement with vessel
recanalization and functional outcome after intravenous thrombolysis in ischemic stroke. Stroke.
2011;42(6):1638-43.
36. Inoue M, Mlynash M, Straka M, Lansberg MG, Zaharchuk G, Bammer R, et al.
Patients with the malignant profile within 3 hours of symptom onset have very poor outcomes after
intravenous tissue-type plasminogen activator therapy. Stroke. 2012;43(9):2494-6.
37. Seker F, Pfaff J, Potreck A, Mundiyanapurath S, Ringleb PA, Bendszus M, et al.
Correlation of Tmax volumes with clinical outcome in anterior circulation stroke. Brain Behav.
2017;7(9):e00772.
38. Nakano S, Iseda T, Kawano H, Yoneyama T, Ikeda T, Wakisaka S. Correlation of early CT
signs in the deep middle cerebral artery territories with angiographically confirmed site of arterial
occlusion. AJNR Am J Neuroradiol. 2001;22(4):654-9.
121
39. Finlayson O, John V, Yeung R, Dowlatshahi D, Howard P, Zhang L, et al.
Interobserver agreement of ASPECT score distribution for noncontrast CT, CT angiography, and
CT perfusion in acute stroke. Stroke. 2013;44(1):234-6.
40. Sundaram VK, Goldstein J, Wheelwright D, Aggarwal A, Pawha PS, Doshi A, et al.
Automated ASPECTS in Acute Ischemic Stroke: A Comparative Analysis with CT Perfusion.
AJNR Am J Neuroradiol. 2019;40(12):2033-8.
41. Ospel JM, Volny O, Qiu W, Najm M, Kashani N, Goyal M, et al. Displaying Multiphase
CT Angiography Using a Time-Variant Color Map: Practical Considerations and Potential
Applications in Patients with Acute Stroke. AJNR Am J Neuroradiol. 2020;41(2):200-5.
42. Almekhlafi M, Ospel JM, Saposnik G, Kashani N, Demchuk A, Hill MD, et al.
Endovascular Treatment Decisions in Patients with M2 Segment MCA Occlusions. AJNR Am J
Neuroradiol. 2020;41(2):280-5.
43. Hill MD, Goyal M, Menon BK, Nogueira RG, McTaggart RA, Demchuk AM, et al.
Efficacy and safety of nerinetide for the treatment of acute ischaemic stroke (ESCAPE-NA1): a
multicentre, double-blind, randomised controlled trial. Lancet. 2020;395(10227):878-87.
44. Tan IY, Demchuk AM, Hopyan J, Zhang L, Gladstone D, Wong K, et al. CT angiography clot
burden score and collateral score: correlation with clinical and radiologic outcomes in acute middle
cerebral artery infarct. AJNR Am J Neuroradiol. 2009;30(3):525-31.
45. d'Esterre CD, Trivedi A, Pordeli P, Boesen M, Patil S, Hwan Ahn S, et al. Regional Comparison
of Multiphase Computed Tomographic Angiography and Computed Tomographic Perfusion
for Prediction of Tissue Fate in Ischemic Stroke. Stroke. 2017;48(4):939-45.
46. Grunwald IQ, Kulikovski J, Reith W, Gerry S, Namias R, Politi M, et al. Collateral Automation
for Triage in Stroke: Evaluating Automated Scoring of Collaterals in Acute Stroke on Computed
Tomography Scans. Cerebrovasc Dis. 2019;47(5-6):217-22.
47. Rodger M, Ramsay T, Fergusson D. Diagnostic randomized controlled trials: the final frontier.
Trials. 2012;13:137.
48. Berkhemer OA, Fransen PS, Beumer D, van den Berg LA, Lingsma HF, Yoo AJ, et al.
A randomized trial of intraarterial treatment for acute ischemic stroke. N Engl J Med.
2015;372(1):11-20.
122
49. Jovin TG, Chamorro A, Cobo E, de Miquel MA, Molina CA, Rovira A, et al.
Thrombectomy within 8 hours after symptom onset in ischemic stroke. N Engl J Med.
2015;372(24):2296-306.
50. Wahlgren N, Moreira T, Michel P, Steiner T, Jansen O, Cognard C, et al.
Mechanical thrombectomy in acute ischemic stroke: Consensus statement by ESO-Karolinska
Stroke Update 2014/2015, supported by ESO, ESMINT, ESNR and EAN. Int J Stroke.
2016;11(1):134-47.
51. Volny O, Bar M, Krajina A, Cimflova P, Kasickova L, Herzig R, et al. A Comprehensive
Nationwide Evaluation of Stroke Centres in the Czech Republic Performing Mechanical
Thrombectomy in Acute Stroke in 2016. Cesk Slov Neurol N. 2017;80/113((4)):1-6.
52. Mueller-Kronast NH, Zaidat OO, Froehler MT, Jahan R, Aziz-Sultan MA, Klucznik RP, et al.
Systematic Evaluation of Patients Treated With Neurothrombectomy Devices for Acute Ischemic
Stroke: Primary Results of the STRATIS Registry. Stroke. 2017.
53. Carvalho A, Cunha A, Rodrigues M, Figueiredo S, Paredes L, Gregorio T, et al.
Mechanical Thrombectomy in Acute Ischemic Stroke: Initial Single-Center Experience and
Comparison with Randomized Controlled Trials. J Stroke Cerebrovasc Dis. 2017;26(3):589-94.
54. Nikoubashman O, Jungbluth M, Schurmann K, Muller M, Falkenburger B, Tauber SC, et al.
Neurothrombectomy in acute ischaemic stroke: a prospective single-centre study and comparison
with randomized controlled trials. Eur J Neurol. 2016;23(4):807-16.
55. Sedova P, Brown RD, Zvolsky M, Kadlecova P, Bryndziar T, Kubelka T, et al.
Incidence of Hospitalized Stroke in the Czech Republic: The National Registry of Hospitalized
Patients. J Stroke Cerebrovasc Dis. 2017;26(5):979-86.
56. Urra X, Abilleira S, Dorado L, Ribo M, Cardona P, Millan M, et al. Mechanical Thrombectomy
in and Outside the REVASCAT Trial: Insights From a Concurrent Population-Based Stroke
Registry. Stroke. 2015;46(12):3437-42.
57. Tsivgoulis G, Goyal N, Mikulik R, Sharma VK, Katsanos AH, Zand R, et al.
Eligibility for mechanical thrombectomy in acute ischemic stroke: A phase IV multi-center
screening log registry. J Neurol Sci. 2016;371:96-9.
58. Chia NH, Leyden JM, Newbury J, Jannes J, Kleinig TJ. Determining the Number of Ischemic
Strokes Potentially Eligible for Endovascular Thrombectomy: A Population-Based Study. Stroke.
2016;47(5):1377-80.
123
59. Saver JL, Goyal M, van der Lugt A, Menon BK, Majoie CB, Dippel DW, et al.
Time to Treatment With Endovascular Thrombectomy and Outcomes From Ischemic Stroke: A
Meta-analysis. JAMA. 2016;316(12):1279-88.
60. Sheth SA, Jahan R, Gralla J, Pereira VM, Nogueira RG, Levy EI, et al. Time to endovascular
reperfusion and degree of disability in acute stroke. Ann Neurol. 2015;78(4):584-93.
61. Ribo M, Molina CA, Cobo E, Cerda N, Tomasello A, Quesada H, et al. Association
Between Time to Reperfusion and Outcome Is Primarily Driven by the Time From Imaging
to Reperfusion. Stroke. 2016;47(4):999-1004.
62. Bracard S, Ducrocq X, Mas JL, Soudant M, Oppenheim C, Moulin T, et al. Mechanical
thrombectomy after intravenous alteplase versus alteplase alone after stroke (THRACE):
a randomised controlled trial. Lancet Neurol. 2016;15(11):1138-47.
63. Sarraj A, Sangha N, Hussain MS, Wisco D, Vora N, Elijovich L, et al. Endovascular Therapy
for Acute Ischemic Stroke With Occlusion of the Middle Cerebral Artery M2 Segment. JAMA
Neurol. 2016;73(11):1291-6.
64. Heldner MR, Zubler C, Mattle HP, Schroth G, Weck A, Mono ML, et al. National Institutes
of Health stroke scale score and vessel occlusion in 2152 patients with acute ischemic stroke.
Stroke. 2013;44(4):1153-7.
65. Maas MB, Furie KL, Lev MH, Ay H, Singhal AB, Greer DM, et al. National Institutes of Health
Stroke Scale score is poorly predictive of proximal occlusion in acute cerebral ischemia. Stroke.
2009;40(9):2988-93.
66. Kim JT, Park MS, Chang J, Lee JS, Choi KH, Cho KH. Proximal arterial occlusion in acute
ischemic stroke with low NIHSS scores should not be considered as mild stroke. PLoS One.
2013;8(8):e70996.
67. Rajajee V, Kidwell C, Starkman S, Ovbiagele B, Alger JR, Villablanca P, et al. Early
MRI and outcomes of untreated patients with mild or improving ischemic stroke. Neurology.
2006;67(6):980-4.
68. Coutts SB, Modi J, Patel SK, Demchuk AM, Goyal M, Hill MD, et al. CT/CT angiography
and MRI findings predict recurrent stroke after transient ischemic attack and minor stroke: results
of the prospective CATCH study. Stroke. 2012;43(4):1013-7.
69. Asdaghi N, Yavagal DR, Wang K, Mueller-Kronast N, Bhatt N, Gardener HE, et al. Patterns
and Outcomes of Endovascular Therapy in Mild Stroke. Stroke. 2019;50(8):2101-7.
124
70. Sarraj A, Hassan A, Savitz SI, Grotta JC, Cai C, Parsha KN, et al. Endovascular Thrombectomy
for Mild Strokes: How Low Should We Go? Stroke. 2018;49(10):2398-405.
71. Nagel S, Bouslama M, Krause LU, Kupper C, Messer M, Petersen M, et al. Mechanical
Thrombectomy in Patients With Milder Strokes and Large Vessel Occlusions. Stroke.
2018;49(10):2391-7.
72. Dargazanli C, Arquizan C, Gory B, Consoli A, Labreuche J, Redjem H, et al. Mechanical
Thrombectomy for Minor and Mild Stroke Patients Harboring Large Vessel Occlusion
in the Anterior Circulation: A Multicenter Cohort Study. Stroke. 2017;48(12):3274-81.
73. Haussen DC, Lima FO, Bouslama M, Grossberg JA, Silva GS, Lev MH, et al. Thrombectomy
versus medical management for large vessel occlusion strokes with minimal symptoms: an analysis
from STOPStroke and GESTOR cohorts. J Neurointerv Surg. 2018;10(4):325-9.
74. Pfaff J, Herweh C, Pham M, Schonenberger S, Nagel S, Ringleb PA, et al. Mechanical
Thrombectomy in Patients with Acute Ischemic Stroke and Lower NIHSS Scores: Recanalization
Rates, Periprocedural Complications, and Clinical Outcome. AJNR Am J Neuroradiol.
2016;37(11):2066-71.
75. Wolman DN, Marcellus DG, Lansberg MG, Albers G, Guenego A, Marks MP, et al.
Endovascular versus medical therapy for large-vessel anterior occlusive stroke presenting
with mild symptoms. Int J Stroke. 2020;15(3):324-31.
76. Shang X, Lin M, Zhang S, Li S, Guo Y, Wang W, et al. Clinical Outcomes of Endovascular
Treatment within 24 Hours in Patients with Mild Ischemic Stroke and Perfusion Imaging Selection.
AJNR Am J Neuroradiol. 2018;39(6):1083-7.
77. Santos EMM, d'Esterre CD, Treurniet KM, Niessen WJ, Najm M, Goyal M, et al. Added value
of multiphase CTA imaging for thrombus perviousness assessment. Neuroradiology.
2018;60(1):71-9.
78. Menon BK, Al-Ajlan FS, Najm M, Puig J, Castellanos M, Dowlatshahi D, et al. Association
of Clinical, Imaging, and Thrombus Characteristics With Recanalization of Visible Intracranial
Occlusion in Patients With Acute Ischemic Stroke. JAMA. 2018;320(10):1017-26.
79. Siegler JE, Boehme AK, Kumar AD, Gillette MA, Albright KC, Martin-Schild S. What change
in the National Institutes of Health Stroke Scale should define neurologic deterioration in acute
ischemic stroke? J Stroke Cerebrovasc Dis. 2013;22(5):675-82.
125
80. Smith EE, Fonarow GC, Reeves MJ, Cox M, Olson DM, Hernandez AF, et al. Outcomes
in mild or rapidly improving stroke not treated with intravenous recombinant tissue-type
plasminogen activator: findings from Get With The Guidelines-Stroke. Stroke. 2011;42(11):3110-
5.
81. Barber PA, Zhang J, Demchuk AM, Hill MD, Buchan AM. Why are stroke patients excluded
from TPA therapy? An analysis of patient eligibility. Neurology. 2001;56(8):1015-20.
82. Nedeltchev K, Schwegler B, Haefeli T, Brekenfeld C, Gralla J, Fischer U, et al. Outcome
of stroke with mild or rapidly improving symptoms. Stroke. 2007;38(9):2531-5.
83. Smith EE, Abdullah AR, Petkovska I, Rosenthal E, Koroshetz WJ, Schwamm LH.
Poor outcomes in patients who do not receive intravenous tissue plasminogen activator because
of mild or improving ischemic stroke. Stroke. 2005;36(11):2497-9.
84. Mishra SM, Dykeman J, Sajobi TT, Trivedi A, Almekhlafi M, Sohn SI, et al. Early reperfusion
rates with IV tPA are determined by CTA clot characteristics. AJNR Am J Neuroradiol.
2014;35(12):2265-72.
85. Volny O, Krajina A, Belaskova S, Bar M, Cimflova P, Herzig R, et al. Mechanical
thrombectomy performs similarly in real world practice: a 2016 nationwide study
from the Czech Republic. J Neurointerv Surg. 2018;10(8):741-5.
86. Davis MJ, Menon BK, Baghirzada LB, Campos-Herrera CR, Goyal M, Hill MD, et al.
Anesthetic management and outcome in patients during endovascular therapy for acute stroke.
Anesthesiology. 2012;116(2):396-405.
87. Campbell BCV, van Zwam WH, Goyal M, Menon BK, Dippel DWJ, Demchuk AM, et al.
Effect of general anaesthesia on functional outcome in patients with anterior circulation ischaemic
stroke having endovascular thrombectomy versus standard care: a meta-analysis of individual
patient data. Lancet Neurol. 2018;17(1):47-53.