Please enable JavaScript.
Coggle requires JavaScript to display documents.
Stroke and 7T-MRI (Ultra high-resolution MRI) (FINER feasible: ->…
Stroke and 7T-MRI (Ultra high-resolution MRI)
FINER
feasible:
-> possible investigation/result: likelihood for a plaques to have caused a stroke identified on magnetic resonance imaging –> culprit, indeterminate, or nonculprit.
->
High-resolution 7T-MRI difficult to implement in in vivo imaging
. –> need to speak with neuroradiologist about 1) accurency of low resolution 7T-imaging (reduced time of acquisition) and how to improve the resolution with acceptable time of acquisition. Possible Use of
motion correction technique
or others programme to increase the imaging quality.
-> Acquisition
exclusively of multisequence carotid plaque imaging with dedicated carotid coils
->routine 3T-MRI: Brain and Neck-MRI followed by
7T Neck-MRI exclusively
to able 7T high-resolution investigation of ICA-plaques in vivo.
-> Use of a MD-candidate or a neuroradiologist as co-assessor.
interesting: Scientifically and technically interesting but not sure if clinically relevant so far 7T-MRI is not used routinely –> need of study design which adresses clinical issues.
novel: very few in vivo 7T-MRI with plaque imaging -> “Intracranial Vessel Wall Lesions on 7T MRI (Magnetic Resonance Imaging) Occurrence and Vascular Risk Factors: The SMART-MR Study“, Zwartbol et al.-> no detailed description of plaques characteristics -> resolution not high enough?
ethical: long time of acquisition for disabled patients-> how long lasts the acquisition of 7T-imaging.
relevant:
need to find a potential clinical application for our project:
–> plaques could be responsible for a part of ESUS -> Reclassification of Ischemic Stroke Etiological Subtypes on the Basis of High-Risk Nonstenosing Carotid Plaque, Kamel et al.,Stroke 2019
–> actual controversy about the optimal treatment to treat asymptomativ carotid stenosis (difficult to adress with a small group of patients.
Advantages:
Multiple study-designs: 3T vs. 7T MRI, stroke patients vs. healthy control, 7T MRI vs.Ultrasound (internal carotid artery), comparison intra- vs. extracranial plaques, investigation of small vs. large vessels plaques.
Carotid plaques generally easier to assess than intracranial plaques. Imaging carotid artery atherosclerosis has the advantage of easy access to ex vivo atherosclerotic plaque material for validation, using carotid endarterectomy specimens.
7T MRI –> High-resolution imaging difficulty feasible in clinical routine –> In clinical studies low resolution imaging performed. High-resolution 7T-imaging In vivo acquisitions are also challenged by motion artifacts and blood flow pulsation.
personal interest for MRI-technique -> Master-Thesis with O. Scheidegger about muscle MRI.
Part of financing through service of neuroradiology ?
Disadvantages:
neuroradiology hot topic–> lot of studies about plaques -> high-risk nonstenosing plaques in focus now -> subtyp momently classified as ESUS.
Few studies used 7T-MRI in vivo -> technical issues.
Expensive study-design? -> Financing unclear for me.
Intracranial stenosis less present in Europathan in Asia-> impact of enrollment
Per definition small patient-number.
Potential Selection bias also expected -> really disabled patients won’t be able to take part (informed consent and time of acquisition).
Further investigations:
1) Assessment of plaque vulnerability criterias (especially highrisk-nonstenosing plaques) and plaque stability with 7T-MRI after diagnosis (of stenosis with or without stroke-> to be define) with 3T-MRI -> according to actual litterature review not investigated directly in vivo with 7T-MRI so far -> certainly due to difficulties to perform high-resolution 7T imaging in vivo (see feasible from FINER) -> feasibility need to be evaluated.
2) sudivision in several subgroups: carotid plaques/ intracranial plaques / vertebralis or basilar plaques—> comparison of MRI-heterogenity in different locations.
3) comparison of cortical microinfarcts burden (3T vs. 7T) in patients with plaques (symptomatic only or also non-symptomatic -> to be define.
4) Artery positive remodeling is associated with unstable clinical presentations -> Analyse of plaque remodeling in 7T-MRI.
6) retrospectively find patients with plaques and mild/moderate or intensiv statin treatment –> Follow-up with 7T-MRI or prospectively determine if mild/moderate or intensiv statin therapy needed with 7T-MRI and follow-up with 3T MRI (or 7 T -MRI).
8) 7T-MRI comparison of symptomatic patients (stroke due to plaques/stenosis) and asymptomatic patients (plaques/stenosis without stroke).
9) identification
of fibrous cap
(= plaque stability?) and distinction to others cap components –> according to Zhao et al., 2019
difficult with 3T MRI.
10)
Assesment of plaque inflammation and neovascularisation with 7T MRI
(usually assesed only for research purpose and with PET). Inflammation and intraplaque neovascularisation might be also associated with stroke, but
evidence is inconclusive
source: Imaging biomarkers of vulnerable carotid plaques for stroke risk prediction and their potential clinical implications, Saba. et al., Lancet Neurol 2019.
11) comparison with other modalities: CT, Ultrasound, Histopathological preparations.
12) Try new sequences for representing intraplaque-hemorrhage, fibrous cap rupture or calcification. -> Need histology for correlation...
13) Development of a deep learning program for increase utilisability of 7T-MRI (especially for plaques imaging) in clinical use -> (CAS — Artificial Intelligence in medical imaging).
14) Comparison of leukoariasis burden (3T vs. 7T) -> more related to SVD than Plaques.
“
Embolic stroke of undetermined source: The role of the nonstenotic carotid plaque
” Bulwa et al., Journal of the Neurological Sciences, 2017
The importance of carotid artery plaques which do not cause significant stenosis as a source of emboli to the brain has generally been ignored given the long-standing focus of using percent stenosis measurements as the primary criterion for defining high-risk carotid atherosclerotic disease.
Advances in computed tomography angiography (CTA), magnetic resonance angiography (MRA), ultrasonography with and without contrast medium, microemboli signal detection, and 18F- fluorodeoxyglucose positron emission tomography (18F-FDG PET) have ushered in a new era of risk stratification based on the presence of highrisk plaque features including intraplaque hemorrhage, plaque ulceration, plaque neovascularization, fibrous cap thickness, the presence of a lipid-rich necrotic core, and evaluation of plaque inflammatory activity.
Freilinger et al. used
black-blood carotid magnetic resonance imaging (MRI)
to assess the prevalence of complicated plaques in cryptogenic stroke.
37.5% of carotid arteries ipsilateral to the ischemic stroke had American Heart Associated (AHA) lesion type VI plaques
compared to zero AHA type VI plaques contralateral to the stroke. The most common feature of AHA type VI plaques was
intraplaque hemorrhage (75%), followed by fibrous plaque rupture (50%), and luminal thrombus (33%)
. This study's imaging technique required the use of a dedicated surface carotid coil and an ~ 18 min multi-sequence protocol, thereby allowing for the detailed detection of various high-risk carotid plaque elements.
In 2015, Gupta et al. used
non-contrast 3-dimensional time-of-flight MRA
to enhance current stroke risk stratification. 22.2% of patients had
intraplaque high-intensity signal (IHIS)
, a marker for intraplaque hemorrhage, in nonstenosing carotid plaques ipsilateral to ischemic stroke. Unlike the work of Freilinger et al., this study determined the presence of high-risk plaque from a standard 5 min angiographic sequence.
In 2016, Hyafil et al. combined
18F-FDG PET imaging with MRI
to investigate morphological and biological aspects of nonstenotic carotid artery plaques in cryptogenic stroke.
The existing uncertainty in the literature about the causative role of nonstenotic carotid plaque in ischemic stroke likely arises from the lack of carefully controlled prospective studies in which the broad range of potential causes of cryptogenic stroke are tested in a systematic fashion in patients who are then observed for stroke recurrence.
It remains unclear whether intraplaque hemorrhage on MRA, plaque thickness measures on CTA, or plaque inflammatory activity on PET is the strongest marker of risk of stroke recurrence
and whether risk stratification could be improved using a composite of multiple imaging features in combination.
MRI-sequences
black blood carotid (MR vessel wall imaging):
Technique: Vessel wall imaging requires high spatial and contrast resolution to depict thin arterial walls apart from its surroundings. Several techniques to achieve this resolution are applied to sequences weighted towards various tissue contrasts (T1-weighted images before and after contrast most commonly, T2-weighted images, or proton density-weighted images).
Spatial resolution: First, due to the small size of the target vessels, high spatial resolution is imperative. Higher field strength (3T rather than 1.5T) is preferred due to the improved signal-to-noise ratio and acquisition times when selecting smaller voxel dimensions. For intracranial imaging at 3T, a 3D acquisition with isotropic voxel sizes in the 0.4-0.7 mm range is commonly used. If 2D sequences are used, because of the poor spatial resolution in the slice-select direction (at least 2 mm slice thickness), multiplanar imaging is required to optimally depict vessels in both short- and long-axes
Contrast resolution: Second, the signal from blood must be kept low to distinguish the vessel wall from the lumen.
These so-called black/dark blood techniques usually exploit the fact that blood flows while vessel walls remain stationary.
Turbo/fast spin-echo sequences intrinsically have low arterial blood signal (flow voids) due to time of flight effects (spins move out of the imaged slice before the 180° refocusing pulse) and/or intravoxel dephasing (spins moving with different velocities due to turbulent or laminar flow acquire different phases). Both mechanisms contribute when 2D techniques are employed but only the latter is important in 3D techniques in which a large slab is imaged at once. The most commonly used 3D sequences for intracranial vessel wall imaging use turbo/fast spin-echo with variable low refocusing flip angles, which maintain high spatial resolution during long echo trains. These sequences are known by brand names such as SPACE, Cube, and VISTA.
Other methods of blood suppression are available:
-> spatial presaturation (saturation band): in 2D techniques, the spins that start in a slab adjacent to the imaged slice can be selectively dephased with a spoiling gradient and cannot regain signal when they flow into the imaged slice
-> double inversion recovery: a nonselective inversion pulse with a TI near the null point of blood followed by a slice selective inversion pulse recovers signal in stationary tissue but not inflowing blood
-> delay alternating with nutation for tailored excitation (DANTE)
Finally, to depict the vessel wall discretely, signal must also be suppressed from the surrounding tissue. For vessel wall imaging of the extracranial head and neck, for instance, this means suppressing fat signal. For intracranial vessel wall imaging, this means suppressing CSF signal. Fortunately, because CSF flows, many techniques used to suppress signal from flowing blood also suppress CSF signal. Depending on the contrast weighting (CSF signal is low on T1-weighted images but not T2- or proton density-weighted images), dedicated methods to suppress CSF may be needed, such as inversion recovery, antidriven equilibrium, or DANTE.
Correlative luminal imaging: Vessel wall imaging is usually performed in addition to a bright blood MRA sequence, such as the gradient echo-based time of flight MRA, to assess the lumen caliber and localize abnormalities for further vessel wall assessment.
Intraplaques hemorrhages: T1-weighting sequences –> TOF, FSE (fast spin echo T1)
Whole brain vessel-wall-imaging
enables the combination of vessel-wall and Lenticulostriate artery (LSA) imaging in one image setting, which can provide information about plaque burden and LSA distribution.
Plaques morphology
A lesion is defined as an eccentric plaque if the minimum wall thickness is smaller than half of the maximum wall thickness at the maximal stenotic site, and as a concentric plaque if the minimum wall thickness is larger than half of the maximum wall thickness at the maximal stenotic site.
Plaque burden can be measured by normalized wall index –> NWI = wall area/total vessel area × 100%. Source (abstract only): Cao et al., „Atheroscleroticplaqueburden of middle cerebral artery and extracranial carotid artery characterized by MRI in patients with acute ischemic stroke in China: association and clinical relevance.“
The arterial remodeling ratio (RR) was calculated as outer wall area at the lesion site divided by outer wall area at the reference site after adjusting for vessel tapering. Arterial remodeling was categorized as positive if RR >l 1.05, intermediate if 0.95≤RR≤1.05, and negative if RR < 0.95. Source (abstract only): Qiao et al., Patterns and implications of intracranial arterial remodeling in stroke patients.
Stabilized plaque: thin lipid pool, thick fibrous cap, preserved lumen
Vulnerable plaque: thin fibrous cap, thick lipid pool, many inflammatory cells.
Healed ruptured plaque: Narrow lumen, fibrous interna
High risk plaques features/ plaque vulnerability
Ulcerations
Silent infarcts
High intensity transient signal
Intraplaque hemorrhage
Plaque echolucency
Stenosis progression
“Quantitative score of the vessel morphology in middle cerebral artery atherosclerosis
” Meng et al. Journal of the Neurological Sciences, 2019.
-> The parameters of plaque components such as intra-plaque hemorrhage and vessel wall enhancement have limited translational clinical value.
The occurrence rate of intra-plaque hemorrhage is too low to sensitively stratify stroke risk, while vessel wall enhancement is recently reported in healthy subjects, making its specificity questionable.
Comparatively, quantitative measurements of plaque morphology are available for each plaque and are promising to discriminate plaque types. Comparatively, quantitative measurements of plaque morphology are available for each plaque and are promising to discriminate plaque types.
Methods: Patients were selected from a high-resolution magnetic resonance imaging (HRMRI) study from January 2007 to December 2015. One hundred and three patients with acute cerebral infarcts due to MCA stenosis (>50%) and eighty-nine patients with asymptomatic MCA stenosis (>50%) were included. Quantitative measurements of MCA morphology, including lumen area, outer-wall and wall area at stenotic site and reference site, stenotic degree, plaque length, remodeling index and plaque eccentricity, were performed on HRMRI.
Results: Total plaque length (p < 0.001) was greater in symptomatic plaques than in asymptomatic plaques, combined with a greater stenotic wall area (p < 0.001) and stenotic outer-wall area (p < 0.001). Furthermore, symptomatic plaques exhibited a greater remodeling index (p < 0.001).
Discussion: Four plaque morphological parameters of symptomatic plaque score (SPS), including stenotic lumen area, stenotic wall area, plaque eccentricity and plaque length, have been shown to mainly represent plaque burden and vulnerability in coronary artery, carotid artery and MCA studies. In a HRMRI study of MCA atherosclerosis,
plaque burden
, represented by
normalized wall index (a ratio of stenotic wall area to stenotic outer wall area –> NWI = wall area/total vessel area × 100%)
is closely associated with stroke severity. Moreover, both normalized wall index and minimal lumen area are proposed as better indicators than stenosis degree in differentiating culprit and non-culprit MCA lesions.
Atherosclerotic luminal narrowing is caused by the
combined effects of plaque burden (mesured with normalized wall index) and arterial remodeling
.
Artery positive remodeling is associated with unstable clinical presentations,
whereas negative remodeling is more common in patients with stable clinical presentations.Therefore, symptomatic MCA stenosis can present with relatively large lumen area with positive remodeling.
It is reasonable to assume that the morphological transformation from symptomatic plaques to asymptomatic plaques may represent an improvement of plaque stability.
„7T TOF-MRA shows modulated orifices of lenticulostriate arteries associated with atherosclerotic plaques in patients with lacunar infarcts“
, Kong et al., European Journal of Radiology, 2019.
Introduction: Because of small caliber and relatively slow blood velocity,
the in vivo imaging method of Lenticulostriate arteries (LSAs) remains challenging in clinical settings.
Digital subtraction angiography (DSA) is a feasible way of imaging LSA, but it cannot provide information of vessel wall lesions and infarcts on brain parenchyma. As a result, LSA has not been extensively studied in Lacunar Infarct (LI), especially about its associations with atherosclerotic plaques on MCA.
The most prominent MRI techniques for imaging intracranial arterial pathologies are time-of-flight MR angiography (TOF-MRA) and vessel wall imaging (VWI)
. At ultra-high field (7 T), TOF-MRA was proved to be an efficient method of imaging LSA, without the need of exogenous contrast agent.
On the other hand,
3D high-resolution VWI has been developed to characterize intracranial vessel walls
, enabling reliable evaluation of atherosclerotic plaques in the CoW. Several studies have demonstrated morphological changes of LSA in vascular diseases using TOF-MRA at 7 T], while others have studied the distribution of atherosclerotic plaques on MCA using VWI.
According to the evolution of intracranial atherosclerosis, the orifices of LSAs is vulnerable to nearby atherosclerotic plaques on its parent MCA.
However, no study has ever incorporated TOF-MRA and VWI at 7 T together to investigate the associations between the impairment of LSA and the plaques on MCA.
Method: From January 2017 to December 2017,
eighteen patients
from Xuanwu Hospital were prospectively collected in the study. -> Among them 3 patients with motion artifact in final images were excluded. Seventeen healthy volunteers were also recruited in the study.
–> Analysis of plaque distribution: Atherosclerotic plaque was defined as eccentric wall thickening with luminal stenosis (>50%) identified on the reconstructed VWI images.
The plaque was not counted if narrowing was only identified on the VWI images but not on the TOF-MRA images.
The locations of plaques on vessel walls were classified into four quadrants (the superior, inferior, dorsal, or ventral side of MCA).
Results/Discussion: They demonstrate for the first time that the orientation distribution of LSA orifices on MCA trunks can be visualized and analyzed noninvasively with high-resolution TOF-MRA at 7T. This study reports a consistent result that most orifices were located on the superior and the dorsal sides both in patients and in volunteers.
The results showed that TOF-MRA at 7 T was a feasible tool for investigating intracranial perforating arteries, especially LSAs
. For LI patients, the orifices of LSA are found affected mainly on the ipsilateral side, especially on the ventral and inferior parts of MCA wall. Due to the influenced pattern of orifices, the blood flow of LSA is mostly supplied through the superior branches in LI patients. Among the patients included in the study, plaques are observed predominantly in the ventral and inferior wall.
–> The findings may be meaningful in the etiological study of LI. The forceful displacement of neighboring atheromatous material into the ostia of arterial branches was a major concern of intracranial angioplasty]. Before this study, there was no report studying the orifices of LSA. In a previous study, as the LSA could not be evaluated with TOF- MRA at 3 T, the displacement of neighboring atheromatous material into branch vessel ostia was speculated to be uncommon.
„
Intracranial Plaque Characterization in Patients with Acute Ischemic Stroke Using Pre- and Post-Contrast Three- Dimensional Magnetic Resonance Vessel Wall Imaging
“, Natori et al., 2015, journal of stroke and CVD.
Introduction: MR vessel wall imaging (VWI) has been applied to directly assess plaques in the intracranial major arteries of stroke patients by using novel techniques such as the high-resolution two-dimensional (2D) fast spin-echo (FSE) method and three-dimensional (3D) FSE method. These studies can detect plaques in the proximal middle cerebral artery (MCA) (M1) and vertebrobasilar arteries with/without substantial stenosis. The high
signal intensity
of cervical carotid plaques on
T1W images
reflects
unstable plaques,
which consist of
hemorrhagic and/or lipid/necrotic components
, as determined from specimens excised by carotid endarterectomy. High signal intensity, which suggests vulnerable plaques on T1- weighted (T1W) VWI,
is associated with the artery that is responsible for the stroke event
.
Purpose of study: Correlations between the intraplaque imaging
characteristics of intracranial arteries and the occurrence and extension of the infarcts
. –> Attempt to characterize atherosclerotic plaques of the proximal MCA in terms of
vulnerable components and inflammatory changes using a pre- and postcontrast 3D T1W-VWI techniqu
e to determine whether unstable plaques in the M1 can cause stroke.
Method: Prospectively examination of 30 consecutive patients with acute ischemic stroke in the proximal MCA territory, ie, the basal ganglia and/or corona radiata, due to noncardioembolic mechanisms. The patients underwent MR examinations using a 3-Tesla scanner with an 8-channel head coil at 3-27 days after stroke onset.
T1W 3D-VWI was performed before and after intravenous administration of .2 mL/kg gadopentetate dimeglumine
. The following sequence parameters were used: sagittal flow-sensitized T1W 3D-FSE. 3D-TOF before administrating of contrast agent. And finally T2-weighted and T2*- weighted images.
–> Qualitative and quantitative evaluations were performed according to methods published in a previous study (Evaluating Middle Cerebral Artery Atherosclerotic Lesions in Acute Ischemic Stroke Using Magnetic Resonance T1-weighted 3-Dimensional Vessel Wall Imaging, Natori et al., 2014.) 1) Visually assessed the presence of wall thickening, which indi- cates atherosclerotic plaque, at the ipsilateral and contralateral M1s on the reformatted images. 2)
Measure of the signal intensity of the plaque (Sp) three times with a manually placed polygon-shaped region of interest (ROI).
Measurements were made at the location where the signal intensity was the highest, and the values obtained were averaged. The signal intensity of the corpus callosum (Sc) was also measured on the midsagittal section using a polygon-shaped ROI while excluding the areas showing abnormal signal intensity due to ischemia or axonal degeneration. 3) The contrast ratio (CR) of the plaque against the Sc was calculated using the following equation: CR = (Sp/Sc) × 100. The presence of contrast enhancement in the plaques was evaluated on the postcontrast VWI. A CR increase of more than 1.2 times when compared with the precontrast VWI was defined as substantial contrast enhancement. 4) Measurement of the degree of stenosis in the bilateral M1 on MRA images
three times
using the same workstation according to the methods described in the Trial of Cilostazol in Symptomatic Intracranial Arterial Stenosis (or TOSS) study. In addition, the size of the acute infarcts on diffusion-weighted images was evaluated three times.
Signal to noise: In MRI,
the signal to noise ratio (SNR) increased proportionally with the magnetic field strength
, and is the most important factor deciding the resolution.
Another advantage of ultra-high field MRA is that T1 relaxation becomes longer under higher field, which improves the flow related enhancement in TOF-MRA and produces the longer visible length in small arteries like LSA.
In cervical carotid arteries,
plaque contrast enhancement has been confirmed to reflect inflammation and/or neovascularization
(Inflammation in Carotid Atherosclerotic Plaque: A Dynamic Contrast-enhanced MR Imaging Study, Kerwin et al., 2006).
it is difficult to
differentiate between fibrous, lipid, and intraplaque haemorrhage because of overlap in Hounsfield units
, which characterise radiation attenuation in different tissues. (Saba et al. CT attenuation analysis of carotid intraplaque hemorrhage. AJNR Am J Neuroradiol 2018)
Results: Lacunar and nonlacunar type infarcts were found in 17 and 13 pa- tients, respectively. The degree of M1 stenosis ranged from 0 to 33.3% (median, 0%) in the ipsilateral side and from 0 to 21.3% (median, 0%) in the contralateral side, indi- cating that none of the patients showed substantial stenosis. The precontrast CRs of the plaques in the ipsilateral and contralateral M1 ranged from 46.7 to 67.9% (median, 54.8%) and from 34.3 to 69.4% (50.0%), respectively, with the former being significantly higher than the latter. The differences were evident in the patients with lacunar infarctions (P = .002), but were not significant in patients with nonlacunar infarctions (P = .48). On postcontrast VWI, substantial contrast enhancement of the plaques was observed in six (20.0%) of the ipsilateral M1s and in five (16.7%) of the M1s on the contralateral side,
which were not statistically significant
. However, the
CRs of the enhanced plaques were significantly higher than the CRs of the nonenhanced plaques.
Discussion: Signal intensity of the plaques was significantly higher in the ipsilateral side than it was in the contralateral side, indicating that hyperintense plaques that may consist of unstable components are apparently relevant to the infarcts. Contradictory to our expectation, no significant differences in plaque signals were observed between the ipsilateral and contralateral sides in patients with nonlacunar large infarcts.
The signal intensity of intracranial plaques cannot predict the occurrence of large (non lacunar) infarcts.
In contrast, in patients with lacunar infarcts, the signal intensity of the ipsilateral plaques was significantly higher than the signal intensity of the plaques on the contralateral side.
–> No differences between ipsilateral and contralateral M1s regarding contrast enhancement of plaques were observed, although the enhanced plaques tended to be more frequent in the nonlacunar group than they were in the lacunar group and showed significantly higher signals.
Thus the clinical roles of enhanced intracranial plaques remain unclear.
–> Limitations:
No comparison the findings symptomatic and asymptomatic patients —> unable to determine whether hyperintense and/or enhanced intracranial plaques are a risk factor for stroke events.
„
Reproducibility of 3.0T High-Resolution Magnetic Resonance Imaging for the Identification and Quantification of Middle Cerebral Arterial Atherosclerotic Plaques
“, Zhao et al. Journal of Stroke and Cerebrovascular Diseases, Vol. 28, No. 7 (July), 2019
Several small cohort retro- spective studies have demonstrated the feasibility of using HR MRI to display the artery wall and lumen of the MCA. Using HR MRI, certain artery wall characteristics or plaque features, such as
positive remodeling, intraplaque hemorrhage and large plaque burden
, could potentially be used to identify high-risk patients requiring early prevention or needing more intensive treatments. In the present study, they aimed to determine the
intraobserver and interobserver agreement
not only by
quantifying MCA wall and lumen areas
but additionally to
identify plaque surface morphology, plaque location and plaque components.
HR-MRI scanning: A 3.0T MR scanner with an 8-channel phased-array head coil was used. First,
time-of-flight-MRA (TOF-MRA)
was obtained in the axial plane and data were reconstructed using a dedicated online postprocessing tool to determine blood-vessel architecture. HR MRI, including
black blood T1-weighted, T2- weighted, and proton density (PD)-weighted imaging
was obtained for the stenotic site, as determined using the 3D TOF-MRA (Fig 1). Next, the 3D-MRA images and TOF source images were used as the localizer tomogram to ensure that the cross-sectional images were perpendicular to the M1 segment of the MCA. The above imaging sequences were applied with the following parameters.
Image Analysis: Morphology and signal intensity of the MCA plaques were analyzed
against the signal intensity of the carotid artery and coronary artery plaque
Fibrous cap was defined as a
high signal band adjacent to the lumen on T2-weighted images
. Recent and fresh intraplaque hemorrhage was characterized by an area of
high signal intensity within the plaque, the intensity of which was more than 150% of the signal in adjacent muscles on T1- weighted images
. Plaque surface irregularity was defined as a
discontinuity of the plaque juxtaluminal surface
and regularity defined as smooth inner wall. Plaque distribution at the narrowest part of the lumen was evaluated by dividing the cross-section into 4 equal arcs, namely, the superior, inferior, dorsal, and ventral arcs, on the short axial T2-weighted images.
All images were first reviewed by 2 experienced readers who were blinded to the patients’ clinical data. The wall characteristics, including plaque surface morphology, plaque location, plaque components, and burden were identified and
measured twice, 4 weeks apart
, by 1 single reader.
Discussion: Recently, Yang et al investigated the reader agreement for component characterization and plaque area measurements in MCA using HR MRI at 3.0T. In Yang’s study,
substantial intraobserver and interobserver agreement was demonstrated for intraplaque hemorrhage identification. We observed similar results in the present study.
-> The lower agreement for intraplaque hemorrhage identification was observed with MCA compared to the carotid artery –> Compared to carotid artery plaques,
the lower incidence of MCA intraplaque hemorrhage, the relatively smaller hemorrhagic volume in the small MCA plaque, and very complicated and muddled hemorrhagic signal intensities (which usually overlap with other components within the limited MCA plaque space)
are likely explanations for the lower agreement.
-> Substantial intraobserver and interobserver agreement for identifying plaque surface irregularity found.
-> Based on previous carotid artery studies, fibrous caps are observed as high signal bands adjacent to the lumen on T2-weighted images. However, hyperintense regions of plaques on T2-weighted images represent not only fibrous cap but also intraplaque hemorrhage.
The intensity and the shape of the fibrous cap will be more variable and complicated in vulnerable or ruptured plaques. Hence, it is not easy to distinguish fibrous caps from other plaque components on T2 weighted images. The results highlight
the difficulties in distinguishing a fibrous cap from other plaque components using 3.0T
HR MRI. Much remains to be improved for HR MRI to confidently identify fibrous cap.
Multicenter large cohort studies are needed to minimize the impact on the agreement.
Review:
Imaging biomarkers of vulnerable carotid plaques for stroke risk prediction and their potential clinical implications
, Saba. et al., Lancet Neurol 2019
Introduction: European and US guidelines for prevention of stroke in patients with carotid plaques are based on quantification of
the percentage reduction in luminal diameter due to the atherosclerotic process
to select the best therapeutic approach. However, better strategies for prevention of stroke are needed because
some subtypes of carotid plaques (eg, vulnerable plaques) can predict the occurrence of stroke independent of the degree of stenosis.
Intraplaque haemorrhage is accepted by neurologists and radiologists as one of the features of vulnerable plaques, but other characteristicseg,
plaque volume, neovascularisation, and inflammation
are promising as biomarkers of carotid plaque vulnerability.
Intraplaque haemorrhage is one of the key features of vulnerable carotid plaques and
contributes to enlargement of the lipidrich necrotic core (LRNC) and rapid plaque progression.
In a metaanalysis of nine studies,
detection of carotid intraplaque haemorrhage by MRI was associated with increased risk for future ischaemic stroke in patients with symptomatic and asymptomatic carotid stenosis
. MRI is the best imaging technique for the de tection of intraplaque haemorrhage because its
appearance depends on the oxidative state of haemoglobin
and can be easily detected using common imaging sequences, such as
T1weighted fatsaturated turbo spin echo, in version recovery turbo field echo, or inversion recovery fast spoiled gradient recalled acquisition in the steady state.
MRI allows categorisation of intra plaque haemorrhage into
fresh (type 1), recent (type 2), and old (type 3) subtypes
, but no evidence is available to link the subtype of intraplaque haemorrhage with increased or reduced occurrence of future ischaemic strokes.
–> It is difficult to
differentiate between fibrous, lipid, and intraplaque haemorrhage because of overlap in Hounsfield unit
s, which characterise radiation attenuation in different tissues.
Lipid-rich necrotic core (LRNC) and fibrous cap: LRNC is a heterogeneous tissue composed of cholesterol crystal, debris of apoptotic cells, and particles of calcium. The fibrous cap is a layer of fibrous connective tissue that separates the core of the plaque from the arterial lumen.
The LRNC is predictive of increased risk for stroke.
Alterations to the fibrous cap (thin or ruptured cap) are an important feature of plaque vulnerability. MRI is the preferred technique to image this feature,
particularly with use of gadoliniumbased contrast agents.
Plaque inflammation and intraplaque neovascularisation:
Inflammation is often associated with angiogenesis and referred to as plaque activity
. Imaging of inflammation is not routinely used in clinical practic. It is only used for research. –> normaly investigated with PET. Detection of intraplaque inflammation by MRI showed an association with histological markers of inflammation, suggesting that
MRI could be a quantitative and noninvasive marker of plaque inflammation. Further confirmatory studies are needed.
Intraplaque neovascularisation, which is associated with plaque activity in terms of
increased risk for neovessel rupture, haemorrhage, and inflammation
. Inflammation and intraplaque neovascularisation might be also associated with stroke, but
evidence is inconclusive
.
Carotid plaque thickness: The maximum plaque thickness (quantified by MRI) was more strongly associated with cerebral ischaemic symptoms than was the degree of stenosis, showing that plaque size represents a parameter associated with the occurrence of stroke.
Surface morphology: The surface of carotid plaques can be categorised as
smooth, irregular (surface fluctuates from 0·3 mm to 0·9 mm), or ulcerated (cavities measuring at least 1 mm).
Carotid plaque volume: The volume of the carotid artery plaque was found to be associated with plaque vulnerability and stroke. MRI is highly useful for quantification of carotid plaque component volume. Although the spatial resolution of MRI is lower than that of CT, soft tissue contrast with MRI is superior to CT.
Prevention of stroke: Findings of a metaanalysis of five randomised controlled trials (3019 patients) published in 2017 showed
a modest but significant benefit for carotid intervention
in asymptomatic patients with severe carotid stenosis. It seems crucial to identify patients with asymptomatic carotid stenosis
with stable and with unstable plaques
and to select those patients who might benefit from a carotid intervention.
–> Some carotid plaque features are associated with a low risk for recurrent stroke in patients with severe stenosis—eg, the heavily calcified plaque. However, the effect of calcium in carotid artery plaques could be more complex than thought. Two types of calcium salts were identified in atheromatous plaques—
hydroxyapatite and calcium oxalate
. An association was noted between hydroxyapatite calcification and vulnerable carotid plaques, whereas calcium oxalate calcifications were mainly detected in nonvulnerable carotid plaques. This finding could further increase the
use of multienergy CT scanners
because of their potential to do spectral analysis and distinguish between hydroxyapatite and calcium oxalate calcifications.
„
Imaging the Intracranial Atherosclerotic Vessel Wall Using 7T MRI: Initial Comparison with Histopathology“
, van der Kolk et al., Am J Neuroradio, 2015.
Introduction: Intracranial arteries are smaller than carotid (or other major peripheral) arteries, necessitating a
high spatial resolution, and therefore high SNR, for plaque visualization
. Because the
SNR increases approximately linearly with field strength
, 7T MR imaging might provide the spatial resolution necessary to image small atherosclerotic plaques. In this feasibility study, ultra-high-resolution 7T MR imaging sequences with different image contrast weightings were developed and used in
an ex vivo setting
, to assess the ability (image contrast) of 7T MR imaging to image different intracranial atherosclerotic plaque components. For validation of our findings, results were compared with histology.
Method: Five specimens of the circle of Willis (CoW) were selected from 100 postmortem cases that were performed in our institution.
–> MR Imaging Protocol: For imaging, sequences with 4 different contrast weightings were used: After a
3D T1-weighted turbo field echo sequence
with full-specimen coverage was applied,
single-section proton-attenuation (PD)–weighted spin-echo-, single-section T2-weighted TSE, and single-section T2*-weighted turbo field echo sequences
with identical geometric parameters were performed for each of the 13 marked locations (resulting in 1 section per image contrast per sample location), by using the T1- weighted images for planning. The total scan duration for each CoW specimen was approximately 40.5 hours.
Results: MR Image Contrast Heterogeneity –> First, MR images were assessed by 2 investigators for the presence of image contrast heterogeneity within the arterial vessel wall for each sample; then, the corresponding areas on the histologic sections were assessed by another investigator for possible atherosclerotic changes that could explain the image contrast heterogeneity seen on the MR images.
Twenty-three of 44 samples (52.3%) showed
no image-contrast heterogeneity (–> understanding: vessel wall changes)
on any of the 4 MR image contrast weightings (T1-, PD-, T2-, or T2*-weightings). In 21 of 44 samples (47.7%), various patterns of image contrast heterogeneity were found.
–>
Eleven samples showed good correlation between the spatial organization of vessel wall MR imaging heterogeneities
(areas of decreased or increased signal intensity)
and the spatial organization of plaque components in histologic preparations
(eg, collagen-rich rim, areas of foamy macrophages).
Discussion: Areas of focal arterial vessel wall thickening on ultra-high-resolution MR images corresponded with histologically determined advanced atherosclerotic lesions. In all of these
more advanced lesions
, signal heterogeneities on 7T MR imaging enabled the spatial differentiation of different plaque components, like foamy macrophages and collagen.
In early lesions, no signal-intensity heterogeneity could be observed.
Of the 44 arterial samples that were assessed in this study,
correlation between MR imaging and histology was shown to be best in the samples with more advanced lesions
. None of the 5 samples with healthy arterial vessel walls showed areas of signal hypo- or hyperintensity on MR images. This finding was also true for 18 of 24 samples (75%) with intimal thickening, suggesting that
these early atherosclerotic changes are beyond the contrast-to-noise ratio
obtained with the ultra-high-resolution sequences used.
Of the more advanced lesions, all 8 samples (7 with fibrous plaque, 1 with fibrolipid plaque) showed
at least partial correlation
between the spatial organization of the MR signal heterogeneities and the spatial organization of plaque components of the corresponding histologic sections. In this small subset, a
hypointense signal on all sequences generally corresponded to the presence of foamy macrophages, increased proteoglycans, or a lipidrich core
(with or without additional intima-media artifacts). Areas of increased collagen content showed more ambiguous sig- nal intensities ranging from hypo- to hyperintense on the same image contrast weightings. In comparison, previous studies on plaque characterization in the carotid artery showed a lipidrich core to be hyperintense on T1-weighted imaging and iso- to hypointense on PD- and T2-weighted imaging; a fibrous (collagen-rich) area was shown to be isointense on T1-weighted imaging, iso- to hyperintense on T2-weighted imaging, and hyperin- tense on PD-weighted imaging.
Discrepancy may be due to the prolonged formalin fixation or to the less advanced atherosclerotic status of most of these samples.
–>
Only 1 sample with a lipid-rich core, and no samples with intraplaque hemorrhage or plaque rupture
.
Overall, the clearest histology-corresponding image contrast heterogeneity was seen on the T2- and T2*-weighted MR images, followed by the T1-weighted images. The PD-weighted images showed less clear image contrast differences among different plaques components.
The very long scan duration (approximately 40 hours) of the ultra-high-resolution sequences used in this study prohibits their use in vivo in clinical practice –> Need of lower spacial resolution to reduce scan duration.
Follow-up study –>
“Quantitative Intracranial Atherosclerotic Plaque Characterization at 7T MRI: An Ex Vivo Study with Histologic Validation“
. Harteveld et al. 2016.
Our results showed that the
T1 relaxation times gave the most differences between individual tissue components
of advanced intracranial plaques. Tissue components identified in the histologic sections as lipid accumulation, fibrous tissue, fibrous cap, calcifications, and vessel wall showed significant differences in T1 relaxation times, indicating that they can be distinguished from each other in a T1-weighted sequence.
–> Van der Kolk et al. showed the lipid-rich core to be hypointense on T1-weighted imaging of ex vivo CoW specimens at 7T, while Turan et al. demonstrated in vivo lipid and loose matrix to have an isointense signal intensity compared with the surrounding tissue on a T1-weighted image at 3T.
–> On the basis of this study, the T1 relaxation time seems to be the most promising parameter. Now that it is known which plaque components can be identified with ex vivo MR imaging sequences, a translation may be made to in vivo intracranial vessel wall MR imaging to obtain optimal image contrast between the different plaque components
„
Vessel-Wall Magnetic Resonance Imaging of Intracranial Atherosclerotic Plaque and Ischemic Stroke: A Systematic Review and Meta-Analysis“
, Han Na Lee, Front. Neurol., 2018
Intracranial plaques were assessed by high-resolution, multi-contrast vessel-wall MRI in 13 studies, of which seven studies used
three different sequences of T1-,T2- and proton density-weighted images
and six studies included T1- and T2-weighted images. All studies included T1-weighted MRI with
14 studies using different block-blood techniques
; six studies used 2-dimensional turbo spin echo or fast spin echo images; eight studies used 3-dimensional volume isotropic turbo spin echo acquisition or magnetization prepared rapid gradient echo or sampling perfection with application optimized contrasts using different flip angle evolution.
All studies except one involved more than one reader evaluating the vessel-wall MRI for ICAS
.
Stroke events were present in 11 studies (of 20 studies) evaluating the contrast enhancement of plaques.
Six of 11 studies classified the degree of contrast enhancement as a
two-level grading system (non-enhancement and enhancement)
and five studies classified it as a
three-level grading system (0: enhancement was less than or equal to that of intracranial arterial walls without plaque, 1: less enhancement than the pituitary stalk, 2: enhancement greater than or equal to that of the pituitary stalk).
Data analysis: Pooled estimates showed a consistent
strong positive correlation between ischemic stroke and plaque contrast enhancement
regardless of subgrouping. Five of 10 studies presented higher prevalence of culprit lesions for positive Intraplaques hemorrhages (IPH) than for negative IPH, the other five did not find a significant difference in culprit lesions related to IPH.
-The meta-analysis did not show any significant differences in ischemic events between the eccentricity and concentricity of stenosis (OR, 1.22). However a significant association between
positive remodeling and plaques irregularity and stroke events
within the corresponding vascular territory, with a random effect OR of 6.19 and 3.94, respectively was found.
Discussion: Intracranial plaques with contrast enhancement, positive remodeling, and wall irregularity are significantly more likely to be associated with ischemic stroke at the corresponding territory. These findings were
significantly different from the known vulnerability markers (including IPH, large lipid core, and thin fibrous cap) in the extracranial carotid plaque
.
„
Intracranial Vessel Wall Lesions on 7T MRI (Magnetic Resonance Imaging) Occurrence and Vascular Risk Factors: The SMART-MR Study
“, Zwartbol et al., 2018, stroke
Aim: 1) Study the frequency, distribution, and risk factors of intracranial vessel wall lesions on 7T magnetic resonance imaging in patients with a history of vascular disease.
2) Explore possible risk factors of this novel and direct marker of intracranial atherosclerosis (ICAS) 7T-MRI and relate these to known ICAS risk factors.
Method: For MRI, a 7T whole-body system (Philips Healthcare, Cleveland, OH) was used with a volume/transmit coil for transmission and a 32-channel receive head coil (Nova Medical, Wilmington, MA). Intracranial vessel wall imaging was performed with a T-weighted Magnetization-Prepared Inversion Recovery Turbo Spin Echo sequence. A maximum intensity projection of the first echo of a dual echo susceptibility-weighted imaging sequence was used as a faux MR angiography. For image assessment, transverse multiplanar reconstructions were made from the T1-weighted Magnetization-Prepared Inversion Recovery Turbo Spin Echo sequence, all angulated according to the nasion-foramen magnum line. All images were assessed, blinded to patient characteristics,
by 1 observer who was trained by a senior observer
. Vessel wall lesion was defined as either a clear focal or more diffuse thickening of the vessel wall, compared with the healthy contralateral or neighboring vessel wall. Inconspicuous lesions were evaluated in multiple planes for verification.
Vessel wall lesions were rated per arterial segment.
Lesions in bifur- cations that stretched into multiple segments (eg, from C7 segment into M1 segment) were counted as separate lesions for each affected segment.
Results: n: 130 –> Ninety-six percent of patients had evidence of ≥1 vessel wall lesion.
–> A mean number of 8.5±5.7 vessel wall lesions were identified in the total cerebral circulation (5.3 anterior circulation and 3.8 posterior circulation). On the arterial level, lesions were most frequently identified in the distal ICAs (70% [left]–62% [right] of patients) and basilar artery (59%). Increasing age was associated with a higher ICAS burden. Also, a significant association was observed between increasing systolic blood pressure and ICAS burden. No significant association was found between sex, smoking, alcohol intake, diastolic blood pressure or hypertension, and ICAS burden.
Several metabolic risk factors were found to be significantly associated with a higher ICAS burden, namely diabetes mellitus, hemoglobin A1c level and apoB. A significant association was found between increased hs-CRP and ICAS burden.
–> Eccentric Versus Concentric Vessel Wall Lesions: Eccentric lesions were found in 95% of patients, whereas concentric lesions were found in 26% of patients. Furthermore, 75% of all lesions were eccentric compared with 25% for concentric lesions.
Discussion: Recently, the ARIC study (Atherosclerosis Risk in Communities) was the first study to use high-resolution vessel wall–MRI at 3T to assess
vessel wall lesions, as a measure of ICAS, in a large population-based cohort
and reported a prevalence of ≥1 vessel wall lesion in 36% of participants. The reconstructed spatial resolution of this study is similar to the ARIC study,
the increased MR signal at 7T was used to optimize the contrast-to-noise ratio
, which may have enabled us to grade lesions with more certainty. A relatively regular distribution of vessel wall lesions was found, with the highest frequency in both distal ICAs and the basilar artery. Several factors may explain this phenomenon. First, it is thought that the proximal intracranial arteries— as opposed to the remaining intracranial vasculature still possess vasa vasorum, which facilitates arterial wall functioning and plays a critical role in
atherogenesis via initiation of an inflammatory cascade
.
Second, the distal ICA is a segment exposed to low shear stress, which is a hemodynamic risk factor for plaque formation.
Third, the distal ICA and the basilar artery have the largest radius of all intracranial arteries, which has been associated with the degree of atherosclerosis.
To date, the ARIC study is the only other study that has investigated intracranial vessel wall lesions
in nonstroke patients
in vivo using vessel wall–MRI.
Future studies on ultra-high field vessel wall imaging could focus on updating current MRI sequences, to keep up with advancements at lower field strengths and more extensive histopathologic validation studies. In addition, more complex multiparametric scores, such as the Gensini-score for coronary heart disease, are an interesting direction of research which should be further explored –> multiparametric scores like these will provide a more versatile way to study the relationship between treatment of risk factors, ICAS, and outcomes. Need for longitudinal studies investigating the clinical relevance of the visualized vessel wall lesions.
Basic science research report: „
Intracranial Vessel Wall Magnetic Resonance Imaging Does Not Allow for Accurate and Precise Wall Thickness Measurements: An Ex Vivo Study
“, van Hespen et al. 2019, Stroke
The measured vessel wall thickness potentially distinguishes between patient and control groups –> Unfortunately, the spatial resolution of the used MR acquisitions thwarts reliable thickness measurements because for reliable measurements, walls need to span at least 2 voxels. This is not feasible because acquisition schemes commonly have voxel sizes between 0.5 and 0.8 mm, whereas vessel wall thicknesses of the larger arteries of the circle of Willis range between 0.3 and 0.5 mm.
This results in an overestimation of the true vessel wall thickness.
The vessel wall thickness measured on
clinically feasible (low resolution) images
was compared with validated thickness measurements acquired on ultra high-resolution images –>
postmortem high-resolution acquisition
. Both images were acquired on a 7T MR imaging scanner with an
isotropic voxel size of 0.8/0.11 mm for the low/ultra high-resolution images, respectively
. –> Results show that normal or slightly thickened walls cannot be measured reliably with low resolution.
The measured vessel wall thickness deviates considerably from the true thickness for walls thinner than 1.0 mm.
From 0.2 to 1.0 mm, measurements on the low-resolution images were indistinguishable from each other. Only walls thicker than 1mm, for example, advanced plaques, could be measured accu-rately from
the clinically feasible low-resolution
images. The clinical application of the vessel wall thickness as a biomarker seems unfeasible given these results, as early wall thickening remains undetected. The results of this study, in accordance to the measurements by Antiga et al. pose a
significant challenge to the MR community to further improve measurement accuracy.
Currently, the
signal-to-noise ratio and acquisition duration are major limitations for high-resolution acquisitions
. Ongoing technological developments, including higher field strengths and com- pressed sensing, may allow for higher spatial resolutions in the future. Nonetheless, accurate measurements of the thinnest walls theoretically require a
resolution of 0.15 mm isotropic
. This resolution will likely remain unfeasible for a long period of time as the voxel volume and signal-to-noise ratio are an order of magnitude smaller than the currently highest resolution acquisitions. Alternatively, the use of image processing methods such as deep learning, which can learn relevant image features related to the vessel wall thickness, can be explored.
„Atherosclerotic Carotid Plaque Composition: A 3T and 7T MRI-Histology Correlation Study
“, Gonzalez et al., 2016, J Neuroimaging
The aim analyses of this study was to: (1) perform in vivo and then ex vivo imaging on the same atheromatous plaque, with comparison to histopathology; (2) define an algorithm allowing differentiation of plaque components at 7T; and (3) assess whether 7T offers potential additional utility in defining plaque composition over lower field strengths.
Method: Subjects were drawn from a group of 30 patients with minor acute (within prior month) cerebral ischemic symptoms, with corresponding >50% carotid stenosis or plaque ulceration on CT-Angiography. 13 subjects were included in this study having completed
CEA, 3T in vivo and 7T ex vivo MRI
of the CEA specimen. Following ex vivo MR –> Histological assessment of CEA specimen.
Results:
-> 3T MRI In Vivo versus Histology
The mean total plaque area was greater on MRI than histology (mean difference 8.59 cm2). This is an expected finding due to plaque shrinkage during fixation and histological processing. The areas of plaque subcomponents were corrected for this, by defining relative to total plaque area. -> Bland-Altman analysis suggests that
areas of fibrous tissue are overestimated by in vivo 3T MRI compared with histology.
-> 7T MRI Ex Vivo versus Histology
The mean total plaque area was again greater on ex vivo 7T MRI than histology (2.75 cm2), an expected finding due to plaque shrinkage during fixation and histological processing. The Bland-Altman plots show good agreement for area measurements of lipid-rich/necrotic core (LR/NC) with or without hemorrhage and calcification. MRI slightly overestimated regions of fibrous tissue (19%).
-> Correlation Analysis: There was fairly strong correlation between MRI and histology area measurements of LR/NC with Hemorrhage (w Hem) and fibrous tissue. A fair correlation was for hemorrhage. The correlation between MRI and histologic area measurements for calcification and LR/NC without Hem were less good.
In Vivo versus Ex Vivo:
An average difference of around 44.5 ± 18.7% in area was found when comparing in vivo 3T MRI measurements versus histopathology and 29.6 ± 5% when comparing in vivo 3T MRI and ex vivo 7T MRI modalities. The average difference when comparing 7T MRI and histopathology was 16.1 ± 16%.
Discussion: This is the first study to report a quantitative comparison between 3T MRI with histopathology in human atherosclerotic carotid plaques. In addition, this is the first study to simultaneously compare quantified high-resolution ex vivo 7T MRI with histopathology.
This study confirmed that in vivo 3T MRI can determine the relative areas of components of carotid atherosclerotic plaque compared to subsequent histopathology analysis.
-> MRI carotid plaque composition may aid risk stratification and treatment selection in acute stroke and TIA. For example, the presence of intraplaque hemorrhage has been reported as strong predictor of cerebral events. Fibrous cap rupture has been reported more prevalent in symptomatic patients compared with asymptomatic. Mixed results have been published regarding the association of intraplaque calcification and the stabilization of carotid plaques.
-> MRI in vivo and ex vivo overestimation has been reported earlier in a number of references. The main cause of overestimation of plaque components is specimen shrinkage due to the fixation procedure. Shrinkage of arterial rings specimens have been reported as 19-25%. Additionally, in vivo measurements have larger voxel size than ex vivo measurements which results in an overestimation of the wall and boundaries between plaque components, that is, partial volume effects.
-> comparison ex-Vivo-MRI and histology: Hemorrhage was overestimated and LC/NC wo Hem was underestimated. This discrepancy between histology and 7T MR in hemorrhage quantification
could be explained by the MRI techniques used (T1-w, T2-w, and DWI)
. -> Boekhorst and colleagues used a different protocol consisting of T1-w GE, T2-w FSE, PD-w FSE, and IR-SE MRI techniques at 9.4T, and reported that the combination of these sequences improved the sensitivity to intraplaque hemorrhage compared to the protocol of this study; T1-w, T2-w, and DWI. Hemorrhage as identified on histology may have distinct 3T and 7T properties that have been overlooked in this study, such as
susceptibility effects
.
-> In vivo 3T MRI results show good agreement with most histological components in absolute areas and relative to total plaque area except for hemorrhage and LR/NC wo Hem.
-> An additional source of error may arise from partial volume effects, where an individual voxel contains 2 or more tissue types. In such cases, the signal from the voxel is a weighted average of the component tissue types, which may lead to misclassification of plaque components from the averaging of the MRI signal over the whole voxel. This will be a bigger affect in the case of 3T MRI where the spatial resolution is lower compared with 7T and histology.
Conclusion: Development of clinical 7T MRI of artheroclerotic carotid arteries may improve identification of the morphological feature and composition of vulnerable plaques. On the other hand, there are still limitations such as the increase in specific absorption rates and the inhomogeneity in transmit fields. Therefore, the overall benefits of in vivo 7T MRI still need to be demonstrated and comparison made with lower field strengths, considering not only scanning length, signal-to-noise, and contrast-to-noise rations, but also the extensive contraindications for patient scanning.
“
Subacute vessel wall imaging at 7-T MRI in post-thrombectomy stroke patients
”
Truong et al., DIAGNOSTIC NEURORADIOLOGY, 2019
Method: The MRI protocol included a high-resolution black blood sequence with
prospective motion correction (iMOCO)
, acquired before and after contrast injection.
Using 7-T instead of 3-T field strength results in an
increased signal-to-noise ratio (SNR) and allows for im- age acquisitions with higher resolution
. This is also true when using low-SNR imaging techniques such as black blood vessel wall imaging. The higher resolution makes it possible to visualize non-diseased vessel walls without the use of a contrast medium. However, to be able to see this level of detail,
minimal patient motion is required during the scan
. This can be challenging for patients recovering from severe stroke, especially considering the relatively long scan times required.
To counter patient motion, there is a newly developed prospective motion correction technique that can be applied to the vessel wall sequence
, but this has not yet been performed in many studies.
Second goal of the study was to explore the use of motion correction for high-field vessel wall MRI to compensate for patient movements during the examination.
Patients n=7.
-> MRI was performed
within 2 days
of the thrombectomy using an actively shielded 7-T MRI unit with a dual channel transmit, 32 channels receive, head coil.
-> For quantifying motion,
fat-selective 3D gradient echo navigator volumes were inserted in the time gaps after each TSE readout in the 3D MPIR-TSE sequence.
Each reconstructed navigator volume was compared with the first volume in real-time, and the position and angulation of the MPIR-TSE volume was correspondingly updated before the next repetition of the MPIR-TSE. Furthermore, a
motion score combining translation and rotation was calculated, and if the detected motion was larger than a certain threshold (1 mm in this protocol), the last shot of the MPIR-TSE was reacquired using the updated geometry settings
. Reacquisition prolongs the scan duration but reduces motion artifacts. The iMOCO technique displays the detected motion on the operator console during acquisition, allowing the operator to monitor the amount of motion and reacquisitions throughout the scan.
-Results: Overall image quality was graded highest on a 3-grade scale (1 = non-diagnostic, 2 = acceptable, 3 = excellent) for all patients in all contrast phases, by both reviewers.
In total, 14 MPIR-TSE sequences were acquired. In eight of these, the motion never exceeded the reacquisition threshold, so the scan time was unaffected. In the remaining six MPIR-TSE scans, the percentage of reacquired data was between 1.6 and 43% (median 11%), which corresponds to a scan prolongation of between 10 s and 3:39 min, respectively.
-> Motion larger than the threshold triggered reacquisition of the last shot.
Remember
: The CREST (Carotid Revascularization Endarterectomy Versus Stenting Trial) demonstrated equivalent composite outcomes between carotid endarterectomy (CEA) and carotid artery stenting (CAS) for treating carotid stenosis in standard-risk patients in the largest randomized controlled trial to date (2522 patients), with outcomes reported through 10 years of follow-up. Overall, no significant difference was observed between CEA and CAS in the composite primary outcomes of stroke, myocardial infarction, or death. Subgroup analysis of individual end points suggested an increased risk of periprocedural stroke with CAS and an increased risk of periprocedural myocardial infarction with CEA.
„
A High-Resolution MRI Study of the Relationship Between Plaque Enhancement and Ischemic Stroke Events in Patients With Intracranial Atherosclerotic Stenosis
“, Wang et al., 2019, frontiers in neurology
Purpose: To investigate the relationships among the degree of intracranial atherosclerotic stenosis (ICAS), plaque enhancement (PE), and ischemic stroke events (ISEs) using 3. 0 T high-resolution magnetic resonance imaging (HR-MRI).
Fifty-two ICAS patients who underwent HR-MRI were retrospectively analyzed. The patients were divided into two groups according to the results of whole-brain digital subtraction angiography (DSA): the mild-moderate stenosis group (group MID) and the severe stenosis group (group SEV). Two neuroradiologists independently measured
the signal intensity of PE and pituitary enhancement
in the HR-MRI and calculated
the ratio of the two indices
. According to the ratio, the patients were divided into three groups:
the marked enhancement group (group MA), the mild enhancement group (group ME), and the no enhancement plaque group (group NO).
Discussion: In this study, there were 8 non-enhanced plaques, all of which were non-culprit plaques. The culprit plaques (acute/subacute plaques and chronic plaques) have different degrees of PE. –> The authors speculated that this phenomenon indicated that non-enhanced plaque might be stable.
As time progressed, the degree of PE began to gradually decline, and the duration of PE lasted longer and generally exceeded the acute phase. In this study, it was observed that some of the patients in chronic-phase culprit plaque group (1 month after ISE) also showed marked enhancement.
–> Studies have shown that extracranial PE was associated with increased neovascularization and endothelial permeability, both of which could promote the penetration of contrast agents into plaques, and the extent of carotid PE has a strong correlation with ischemic stroke.
However, the pathological mechanism of intracranial PE remains unclear.
–> In this study, the univariate and multivariate analysis showed a high correlation between marked PE and ISEs
after adjustment for the effect of arterial stenosis
, and there was a strong association between severe intracranial arteries stenosis and ISEs
after adjustment for the effect of enhancement.
–> This study has several highlights. First, most of the previous similar studies only analyzed the relationship between PE and ISEs
without the consideration of the effect of arterial stenosis degree
on ISEs.
Review: „
7 Tesla MRI in cerebral small vessel disease
“, Benjamin et al. 2015 World Stroke Organization
Opportunities and challenges at 7 Tesla:
The best known argument for using higher field strengths is the approximately linear increase in SiGnal-noise-ratio (SNR) with the main magnetic field (B0). This improvement in SNR may be traded to provide better spatial resolution (since SNR is also proportional to the voxel volume). SNR depends also on multiple other factors (relaxation constants, spin density and RF coil characteristics). Nevertheless, as a rule of thumb an isotropic voxel of 1 mm at 1·5T or 3T can be reduced at 7T to ≈ 0·6 mm and ≈ 0·75 mm, respectively, without SNR penalty.
This assumes that the acquisition time is kept constant by reducing the overall imaging volume.
A significant technical challenge at 7T is inhomogeneity in the B1 (radio-frequency) field, which is responsible for sample excitation. This results in spatially varying flip angles which
degrade image quality and affect image contrast.
The increasing B0 also leads to changes in the relaxation constants T1, T2 and T2*. The longitudinal relaxation constant (T1) increases significantly from 3T to 7T -> A disadvantage of the longer T1 at 7T is the need for a prolonged TR in T2 weighted imaging, such as in T2-FLAIR and T2 turbo-spin echo (TSE), which increases the acquisition time.
At 7T the transverse relaxation constant (T2) is slightly reduced and the same holds for T2
. The shortened T2
and the increased magnetic susceptibility effect (which is linearly proportional to B0) allow for
easier detection of paramagnetic substances like hemosiderin, calcium, deoxyhaemoglobin and iron.
A final issue relevant to high field is the MR compatibility of the subject. A device or implant being categorized MR safe at 1·5T or 3T does not guarantee MR safety at 7T. Metallic implants that have been tested safe at lower field strengths may experience reso- nance effects and local tissue heating at 7T. Various studies that have tested temperature and position changes of metallic implants and other objects already demonstrated safe at lower fields (25–29), have shown 7T to be safe
Vascular imaging:
MR angiography: 7T MRI allows non-invasive imaging of the cerebral perforating arteries, i.e. the pathological arteries themselves
Vessel wall imaging: 3T MRI is being increas- ingly used to visualize characteristics of the atherosclerotic plaque in the extracranial carotid arteries but the smaller size of the middle cerebral artery makes visualization of atheroma in this vessel much more challenging. Studies at 3T have shown that the vessel wall was visible only in the presence of large atherosclerotic lesions or vessel wall inflammation. At 7T, however, it is possible to visualize even healthy intracranial vessel walls using a 3D inversion recovery turbo spin-echo (MPIR-TSE) sequence. The inversion pulse was used to null the CSF and black blood was obtained due to flow between excitation and refocusing in the TSE train. As such, 7T MRI can be used to visualize basal intracranial vessel wall disease and therefore may be useful in determining the role of intracranial atheroma in the pathogenesis of lacunar infarction.
“
Cortical Cerebral Microinfarcts (CMI) on 3T Magnetic Resonance Imaging in Patients With Carotid Artery Stenosis
”, Takasugi et al., 2019, Stroke.
Method: Eighty-nine patients with >30% carotid stenosis on ultrasound were prospectively enrolled, and underwent brain and carotid artery magnetic resonance imaging.
Results: CMIs were identified in 26 patients and were associated with intraplaque hemorrhage (rate ratio, 1.95), lacunar infarcts (rate ratio, 1.54), and cortical infarcts (rate ratio, 3.22). These associations were also observed in asymptomatic patients (n=64). The presence of CMIs was associated with poor cognitive function.
“Reclassification of Ischemic Stroke Etiological Subtypes on the Basis of High-Risk Nonstenosing Carotid Plaque”
, Kamel et al, Stroke, 02/2020..
Under current stroke classification systems, an atherosclerotic plaque must cause ≥50% luminal stenosis for a stroke to be attributed to large-artery atherosclerosis. However,
several studies have found a higher prevalence of nonstenosing carotid plaque ipsilateral to brain infarction compared with the contralateral, noninfarcted side suggesting that vulnerable carotid plaques may rupture and cause downstream embolism even in the absence of significant luminal stenosis.
Approximately 20% of ischemic stroke patients appear to harbor nonstenosing plaque ipsilateral to their brain infarction. These findings suggest that
some proportion of ischemic strokes might be reclassified if high-risk nonstenosing plaque features were taken into account.
Method -> measurements: They measured (retrospectively in 1.5 T-MRI and 3 T-MRI performed between 2011 and 2015) 2 plaque biomarkers—
luminal stenosis and intraplaque hemorrhage
—that can be reliably ascertained from
noncontrast TOF using neurovascular coils with a large field-of-view
. They did not attempt to ascertain
plaque ulceration
because to detect it reliably requires
a multisequence technique and dedicated surface carotid coils, which is difficult in the routine clinical care of acute stroke.
Brain MRI studies were reviewed after the neck MRA studies
so that analysis of carotid vessels occurred without knowledge of the side of the diffusion-weighted imaging lesion(s). They used 2-dimensional- TOF images to determine the diameter of the distal ICA to use as a denominator for stenosis calculations based on the method of the NASCET (North American Symptomatic Carotid Endarterectomy Trial). They also used the 3-dimensional-TOF maximum intensity projection images to ensure that the NASCET-derived stenosis measurements were consistent with the qualitative assessment of luminal narrowing determined by the neuroradiologist making the original clinical interpretation. The presence of high-risk carotid plaque was determined by using a previously published technique in which
intraplaque hemorrhage is defined by signal hyperintensity in carotid plaque using region-of-interest analysis
. To be classified as intraplaque hemorrhage, a lesion’s signal must be
at least 50% greater than the background signal intensity of the sterno-cleidomastoid muscle and must be distinguishable from the normal flow-related enhancement visible in the patent ICA lumen.
As the T1-shortening effects of fat can result in high-signal intensity,
they distinguished intraplaque hemorrhage from perivascular fat near the ICA
by classifying an artery as harboring intraplaque hemorrhage
only when high signal was visible at the site of presumed atherosclerotic plaque, for which we required evidence of focal eccentric luminal narrowing
seen on source images. To compare intraplaque hemorrhage across ischemic stroke subtypes and assess stroke subtype reclassification, they used the TOAST classification system. For greater precision, they split the undetermined cause category into 3 separate categories: (1) strokes with an incomplete evaluation, which we excluded; (2) truly cryptogenic strokes, which we defined per the consensus ESUS definition and thus refer to as cryptogenic/ESUS; and (3) strokes with multiple potential causes.
Results: Among the 579 included patients, 175 had high-risk ICA plaque either ipsilateral, contralateral, or bilateral to their brain infarct. Among the 175 patients with any high-risk ICA plaque, 106 had high-risk plaque only ipsilateral to their brain infarct, 24 had high-risk plaque only contralateral to their brain infarction, and 45 had either bilateral high-risk plaque or bilateral brain infarcts.
After taking into consideration the presence of ipsilateral high-risk nonstenosing carotid plaque, 88 of the 579 patients were reclassified: 38 from cardioembolic to multiple potential etiologies, 6 from small-vessel occlusion to multiple, 3 from other determined cause to multiple, and 41from cryptogenic/ESUS to large-artery atherosclerosis.
Discussion: High-risk carotid plaque was more prevalent ipsilateral to recent brain infarction compared with the contralateral side. This side-to-side asymmetry was greatest among strokes already attributed to large-artery atherosclerotic disease with ≥50% luminal stenosis, as expected, but was also significant among strokes initially classified as ESUS or cardioembolic, who by traditional definitions had <50% luminal stenosis. This study suggest that accounting for high-risk nonstenosing carotid plaque may reclassify the causes of
up to 15% of ischemic stroke
cases in the study sample.
More causes may be reclassified among patients with 20% to 49% ipsilateral stenosis than among patients with 0% to 19% ipsilateral stenosis
. This study adds novel findings regarding 3 points: (1) a marker of high-risk non-stenosing plaque was significantly associated with ipsilateral brain infarction even in patients adjudicated to non-atherosclerotic stroke mechanisms, (2) accounting for high-risk non-stenosing plaque may reclassify ischemic stroke subtype in a substantial proportion of patients, and (3) such reclassification may be more common in patients with 20% to 49% ipsilateral stenosis than in patients with 0% to 19% ipsilateral stenosis.
selected Limitations: They did not assess other markers of high-risk plaque such as a lipid-rich necrotic core or thinning/rupture of a fibrous cap. They did not perform more detailed
multisequence carotid plaque imaging with dedicated carotid coils
, so we may have misclassified other high-risk plaque elements, such as plaque ulceration, as intraplaque hemorrhage. Last,
contrast-enhanced MRA is superior to TOF MRA for detecting luminal stenosis
. However, TOF MRA is 90% as accurate as contrast-enhanced MRA for the assessment of luminal stenosis and allows more rapid assessment of stenosis and high-risk plaque in a protocol that does not require gadolinium.