Keywords
near-infrared spectroscopy - intravascular ultrasound - carotid artery stenting
Ischemic stroke is a major public health issue because it is the most common cause
of disability and the third leading cause of death in Western countries.[1]
[2]
[3] It is estimated that atherosclerotic plaques in the internal carotid artery cause
approximately 15 to 20% of all ischemic strokes.[4] Because the disease is extremely debilitating, primary and secondary prevention
is a deeply important topic. In the vast majority of cases, stenosis is caused by
an atherosclerotic process.[5]
[6] Our options in the management of the disease include a conservative approach, with
optimal medication and carotid revascularization by either carotid endarterectomy
(CEA) or carotid artery stenting (CAS). The guidelines for the management of carotid
artery disease have mostly been based on clinical symptoms and the degree of stenosis.[1]
[2] Many points regarding this issue remain controversial, partly because we currently
do not have sufficient information about the composition of the plaque that causes
the disease. With the introduction of modern imaging modalities, which are capable
of providing such information, we might be able to optimize the management and tailor
it more suitably to each patient according to the composition of the atherosclerotic
plaque. Although noninvasive imaging methods will likely provide better solutions
once they are capable of providing a powerful enough resolution, the new invasive
imaging modality near-infrared spectroscopy (NIRS) seems to have great potential.
As we can assume from the research conducted on coronary arteries, NIRS could have
many useful applications in the management of carotid artery disease. It could help
with the optimization of CAS by providing background for decisions about the embolic
protection or stent used. It could also aid in the risk stratification of patients
to CAS or CEA and with decision making in the treatment of asymptomatic patients.
Herein, we provide background regarding this topic.
Current Evidence on Carotid Artery Disease
Current Evidence on Carotid Artery Disease
Invasive treatment of carotid artery disease has been a well-recognized form of stroke
prevention for decades. The first CEA, performed in 1953, was followed by the first
carotid angioplasty approximately three decades later.[7]
[8]
[9]
[10] Much scientific evidence for the effectiveness of the respective approaches followed
these milestones.
In the early 1990s, the benefit of CEA over the contemporary standard medical therapy
in symptomatic patients with greater than 50% carotid stenosis was confirmed by the
first large, randomized, controlled trials: The North American Symptomatic Carotid
Endarterectomy Trial (NASCET) and the European Carotid Surgery Trial (ECST).[11]
[12] The Asymptomatic Carotid Atherosclerosis Study (ACAS) and the Asymptomatic Carotid
Surgery Trial (ACST) subsequently provided background for the treatment of asymptomatic
patients.[13]
[14] The main limitation of all of these studies, despite highly recognized data, was
that they were performed before the introduction of modern pharmacotherapy, such as
statins or modern antihypertensives.[15]
CAS challenged CEA in the first randomized, controlled trials more than a decade later.
At first, the Stenting and Angioplasty with Protection in Patients at High Risk for
Endarterectomy (SAPHIRE) trial confirmed the noninferiority of CAS over CEA for symptomatic
patients at increased risk for surgical complications.[16]
In the first trials, which included patients with an average surgical risk, CAS fared
considerably worse. The Endarterectomy Versus Angioplasty in Patients with Severe
Symptomatic Carotid Stenosis (EVA-3S) study had to be terminated prematurely because
of a significantly higher incidence of 30-day complications with CAS.[17] In addition, the Stent-protected Percutaneous Angioplasty of the Carotid versus
Endarterectomy (SPACE) trial failed to prove noninferiority of CAS compared with CEA
for the periprocedural complication rate.[18] The International Carotid Stenting Study (ICSS), which enrolled a large cohort of
patients, documented statistically higher incidences of stroke, death, and periprocedural
myocardial infarction (MI) in the CAS group than in the CEA group.[19] However, all of the aforementioned trials were heavily criticized because of their
flawed designs (i.e., the trials included physicians with only limited experience
with CAS, and no embolic protection devices were required).[20]
The largest trial that compared CAS to CEA in patients at average surgical risk was
the Stenting Versus Endarterectomy for Treatment of Carotid Artery Stenosis (CREST)
study. This study, which was arguably the best-designed study, also yielded the most
promising results with CAS thus far. The study included a total of 2,502 patients
who were randomized to either CAS or CEA. At 2.5 years of follow-up, there was no
significant difference (7.2% vs. 6.8%, p = 0.51) in the estimated 4-year rates of the primary end point (a composite of stroke,
MI or death from any causes during the periprocedural period [30 days] or any ipsilateral
stroke). Importantly, although the incidence of the composite outcome was similar
between the groups, there was a statistically significant difference between the incidence
of minor stroke and MI. Stroke was more frequent with CAS (4.1 vs. 2.3%, p = 0.01), and MI was more likely to appear after CEA (2.3vs. 1.1%, p = 0.03).[21]
Despite such a long experience, the treatment of carotid artery disease remains a
challenge and the prevention of complications is crucial, given the large population
of patients affected. The new diagnostic method NIRS seems to have potential in the
risk stratification of patients and in the improvement of complication rates, particularly
in the reduction of distal embolization.
Near-Infrared Spectroscopy
Near-Infrared Spectroscopy
Near-infrared spectroscopy is a modern intravascular diagnostic modality that was
developed for the detection of the lipid content of atherosclerotic plaques. The technique
has been used for decades in science and industry to determine the chemical composition
of substances. The principle of the method is the different absorbances of electromagnetic
radiation in the near-infrared spectrum (wavelengths from 800 to 2,500 nm) based on
the presence of various chemical bonds in substances. When the backscattered signal
is analyzed, a characteristic map of absorbance is obtained. The map is typical of
every substance, only as a “fingerprint.”[22]
The idea that NIRS could be useful for the detection of cholesterol within atherosclerotic
plaques is more than 20 years old,[23] but it took many years before a catheter suitable for intravascular diagnostics
was developed. In 2008, Gardner et al published a validation study that proved the
efficacy of NIRS compared with histology as the gold standard.[22] A receiver operating characteristic analysis of the results yielded an area under
the curve value of 0.89 (95% confidence interval, 0.76–0.85), suggesting that the
method was accurate in the detection of lipid cores.[22] One year later, the safety and efficacy of the method were proved in vivo.[24] Subsequently, the method received Food and Drug Administration (FDA) and Conformité
Européenne (CE) approval for use in coronary arteries, triggering a boom of research
in the field.
Today, a commercially produced catheter is available. The imaging device consists
of the following three main parts: a dual modality 3.2 F imaging catheter that incorporates
both NIRS and intravascular ultrasound (IVUS), a device for mechanical pullback and
rotation of the catheter through a vessel and a console with which one can observe
the results of both imaging modalities in real time in the catheterization laboratory
([Fig. 1]).
Fig. 1 The near infrared spectroscopy imaging device consisted of the following three main
parts: a dual-modality 3.2 F imaging catheter that incorporates both near-infrared
spectroscopy and intravascular ultrasound; a device for the mechanical pullback and
rotation of the catheter through a vessel; and finally, a console with which one can
observe the results of both imaging modalities in real time in the catheterization
laboratory.
The IVUS results of the examination are depicted on both a tomographic ultrasound
image of the vessel and a longitudinal IVUS cross-section ([Fig. 2]). Together with the NIRS results, the IVUS results provide us with a so-called true
vessel characterization. From the IVUS image, we can learn about the anatomical structure
of the vessel, including the dimensions of lumen and external elastic membrane, the
degree of stenosis, and the plaque burden ([Fig. 2]).
Fig. 2 The results of the intravascular ultrasound examination has two outputs. From the
tomographic ultrasound image (panel A) of the vessel, we can learn about the anatomical
structure of the vessel, including the dimensions of the lumen (red line) and the
external elastic membrane (blue line). We can additionally calculate the lesion plaque
burden as a fraction of the two parameters. The longitudinal intravascular ultrasound
cross-section provides additional information about the lesion extent and severity
(panel B).
The NIRS data are presented as a chemogram. Every pixel of this rectangular color-coded
probability map represents the probability of lipid presence at the given location.
Low probabilities of lipids are depicted as red, while the other extreme of the scale
is represented as yellow. The X-axis of the chemogram indicates the pullback position
in millimeters, while the Y-axis represents the circumferential position in degrees,
as though the coronary vessel has been incised along its longitudinal axis ([Fig. 3]).[25]
[26] The lipid-core burden index (LCBI) was established to satisfy the need for the quantification
of the presence of lipids. The LCBI is defined as the proportion of yellow and red
pixels on the chemogram multiplied by 1000. The maximal LCBI per 4 mm describes the
region with the highest lipid burden.[22]
[25]
[26]
[27]
Fig. 3 The near-infrared spectroscopy data are presented as a chemogram. Every pixel of
this rectangular color-coded probability map represents the probability of the presence
of lipids at a given location. Low probabilities of lipids are depicted as red, while
the other extreme of the scale is represented by yellow. The X-axis of the chemogram
indicates the pullback position in millimeters, while the Y-axis represents the circumferential
position in degrees as though the coronary vessel has been incised along its longitudinal
axis. The lipid-core burden index was established to satisfy the need for the quantification
of the presence of lipids, defined as the fraction of yellow and red pixels on the
chemogram multiplied by one thousand. The maximal lipid core burden index per 4 mm
describes the region with the highest lipid burden.
NIRS–IVUS for Optimizing CAS
NIRS–IVUS for Optimizing CAS
Despite more than three decades of experience with carotid interventions, there still
seems to be much room for improving and fine tuning the procedure to prevent adverse
events.[9]
[16]
[17]
[20]
[21] Large prospective trials that compared CAS and CEA suggested that the higher incidence
of stroke due to distal embolization is an important limitation of CAS, leading to
worse outcomes of patients.[17]
[18]
[19]
[21] It was demonstrated that new ischemic lesions could be detected by diffusion-weighted
magnetic resonance imaging (MRI) more often after CAS than after CEA.[28] Although the CREST trial did not find a significant difference in its primary end
point, the incidence of minor stroke was higher in the CAS group,[21] indicating a great area of improvement in CAS, which could theoretically drive its
results even beyond those for CEA in a well-selected population of patients. This
fact is why it is crucial to distinguish patients with a higher risk of periprocedural
stroke and either to schedule them for CEA as a safer option or to tailor the CAS
procedure accordingly, using different types of embolic protection devices and stents.
Interestingly, the CREST study taught us that while younger patients tended to benefit
from CAS, older patients profited from CEA.[21] A possible explanation of this phenomenon might be the more advanced atherosclerosis
in older patients, leading to a higher risk of periprocedural stroke during CAS. Many
diagnostic methods have been tested for the detection of such high-risk carotid plaques.
Some results were obtained with conventional ultrasound.[29] Promising results were obtained with MRI, where some plaque features such as intraplaque
hemorrhage or lipid core were associated with plaque vulnerability.[30]
[31]
[32]
[33] A promising method, given its high spatial resolution and correlation with the histopathology
of carotid plaques, is multidetector computed tomography.[34]
[35] Because NIRS is an invasive imaging modality, it can be assumed that it will preferably
be used in patients already indicated for carotid angiography and/or stenting. In
such cases, the risks associated with the procedure will not be dramatically increased.
Current knowledge suggests that NIRS could play an important role in the detection
of lipid-rich high-risk plaques. Knowledge about the quantity of lipids could play
an important role in decisions about the treatment strategy. We hypothesize that CEA
might be the preferable treatment strategy in patients with a high lipid burden. Another
possible outcome is a stricter approach to embolic protection. Herein, we provide
hypotheses for the use of NIRS in carotid interventions based on the current findings
from coronary research. It is important to emphasize that these coronary data are
mostly from small pioneering studies, and most of the hypotheses must be verified
by larger trials. The later mentioned assumptions must therefore be considered cautiously.
Embolic Protection
As aforementioned, distal embolization of atherosclerotic debris is a major limitation
of CAS, and different embolic protection devices have been developed to reduce this
risk. Currently, either distal protection with a filter or proximal protection devices
are routinely used.[36]
[37]
[38]
[39]
[40]
As demonstrated in the coronary arteries, NIRS could be very helpful for predicting
the distal embolization of the lipid content. Studies have proved that high LCBI is
associated with periprocedural MI.[41]
[42]
[43]
[44]
[45] The CANARY trial proved that this association could still be observed when a distal
protection filter was used. It can be assumed that this knowledge could also be transferred
to carotid interventions (CANARY trial, Coronary Assessment by Near-infrared of Atherosclerotic
Rupture-prone Yellow, NCT01268319). We recently published the first case report of
the use of NIRS in CAS, demonstrating the distal embolization of lipid cores during
a filter-protected CAS that was associated with transient neurologic symptoms.[46] Although we currently do not have any supporting evidence, it can be hypothesized
that high LCBI is associated with a greater risk of distal embolization and periprocedural
stroke. A weak correlation between plaque composition and new ipsilateral MRI lesions
was observed in a small IVUS virtual histology study.[47] In such cases, NIRS might be used for the risk stratification of patients, either
to decide between the types of protection devices used or, in high-risk patients,
to switch the strategy to CEA. Obviously, these hypotheses are only speculation and
clinical trials are needed to prove them.
Optimization of Stenting
Spectroscopy might also be used for the optimization of stent length to cover the
whole lesion based on a more precise determination of the extent of the plaque or
to decide between the types of stent used. Coronary case studies have documented acute
in-stent thrombosis following percutaneous coronary interventions (PCIs) in which
the lipid core, imaged by NIRS, was not entirely covered by a stent.[48]
[49] A small study by Dixon et al suggested the potential of NIRS–IVUS in the optimization
of stent length to cover the whole LCP.[50] Furthermore, a small study by Papayannis et al observed an association between the
amount of LCBI and the formation of thrombi after PCI, as determined by optical coherence
spectroscopy.[51]
Some data have also suggested that the embolic load during a carotid intervention
might vary according to the type of stent used.[52] The idea here is that whereas the open cell type of stent is less likely to cause
kinking of a tortuous vessel, the closed cell type of stent affords fuller coverage
of the lesion and thus more effectively prevents distal embolization of the plaque's
content. Although other studies have disproved this hypothesis,[53] closed cell types of stents are indeed used in lesions with suspected high emboligenicity
in common practice. If the theory proves to be true, decision making about the type
of stent used might be another important use of NIRS.
The Role of NIRS–IVUS in the Risk Stratification of Patients
The Role of NIRS–IVUS in the Risk Stratification of Patients
Although the research into the management of carotid artery disease has already passed
several important milestones, the indications for revascularization in asymptomatic
patients remain uncertain. The relative benefit of invasive revascularization over
modern complex medical therapy, paired with smoking cessation programs, remains unclear.[20]
[54] Most of the evidences on which we base our decisions were obtained before the introduction
of modern drugs. It is essential to identify a subgroup of patients at higher risk
for stroke that will benefit from revascularization and subsequently to choose the
most suitable and safest form of revascularization.
The intention behind the development of NIRS was to identify the so-called vulnerable
atherosclerotic plaques. These lesions supposedly bear a greater risk of causing a
carotid event. Thus, the identification of such plaques in vivo might be essential
for the risk stratification of patients to either medical therapy or one of the types
of revascularization.
Current research in the field of NIRS is focused in this direction. From coronary
research, we have learned that lesions responsible for acute coronary syndrome more
often harbor lipid core plaques[55] and that the lesions responsible for ST-elevation MI can be characterized by a certain
threshold of LCBI.[56] A small, pioneering prospective trial recently found that the amount of LCBI was
correlated with the incidence of major cardiovascular events (the composite of all-cause
mortality, nonfatal ACS, stroke, and unplanned coronary revascularization).[57] Much larger prospective coronary studies are already in the process of patient recruitment,
and they are expected to yield results in a few years. Obviously, it might be difficult
to transfer the knowledge learned in the coronary studies to the carotid arteries
if we consider the different pathophysiologies of acute events in the different vascular
beds. Most carotid events are caused by embolisms, while acute coronary events are
usually caused by the rupture of lipid-rich vulnerable plaques. The information might
thus be untransferable.
Once we decide that revascularization is the more beneficial option, a decision between
CAS and CEA remains to be made. Obviously, patients with many internal comorbidities
who are at high risk for surgical complications or who have different attitudes about
surgery will probably benefit from the less invasive CAS, as we also can assume from
the results of the SAPHIRE trial.[16] In contrast, patients with lipid-laden plaques have a higher risk of periprocedural
stroke.[20]
[21] These patients will probably benefit from CEA. In such cases, NIRS could provide
important insights.
Conclusion
Carotid artery disease is a major cause of morbidity and mortality, and its treatment
plays an important role in the prevention of stroke. Current guidelines for the management
of the disease contain many possible flaws because they are partly based on outdated
evidence and do not emphasize the pathophysiology of the disease. Near-infrared spectroscopy
paired with IVUS might provide important information about the anatomy and composition
of an atherosclerotic plaque, and this information could play a crucial role in the
management of carotid artery disease in the near future. The authors of this review
have outlined a few possible applications of NIRS in the coronary arteries, ranging
from the more conceivable utilization in the optimization of CAS to prevent periprocedural
stroke to a less probable role in the risk stratification of asymptomatic patients.
However, it is important to emphasize that all of these speculations are based on
the first pieces of information that were obtained from coronary artery research.
No evidence is currently available for the utility of NIRS in carotid arteries.
Summary
Carotid artery disease is a major cause of morbidity and mortality, and its treatment
serves an important role in the prevention of stroke. Current guidelines on the management
of the disease contain many flaws because they are partly based on outdated evidence.
Near-infrared spectroscopy is a novel diagnostic method that could be useful for the
improvement of the management of carotid atherosclerosis by providing more thorough
information about the underlying carotid plaque. Currently, we are only at the beginning
of research in this field, and solid evidence is lacking. It seems plausible that
better risk stratification of patients based on the information provided by NIRS and
IVUS might decrease the incidence of adverse events during the invasive treatment
of carotid artery disease.
