doi: 10.1161/hq0701.092091
© 2001 American Heart Association, Inc.
Atherosclerosis and Lipoproteins |
Discrimination of Human Coronary Artery Atherosclerotic Lipid-Rich Lesions by Time-Resolved Laser-Induced Fluorescence Spectroscopy
From Laser Research and Technology Development, Cedars-Sinai Medical Center (L.M., W.S.G.); the Department of Biomedical Engineering, University of Southern California (L.M., J.-M.I.M.); and the Department of Pathology and Laboratory Medicine, UCLA School of Medicine (M.C.F.), Los Angeles, Calif.
Correspondence to Laura Marcu, PhD, Laser Research and Technology Development, Cedars-Sinai Medical Center, 650 S San Vicente Blvd, Los Angeles, CA 90048. E-mail lmarcu{at}bmsrs.usc.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Abstract—Lesion composition plays a significant role in atherosclerotic lesion instability and rupture. Current clinical techniques cannot fully characterize lesion composition or accurately identify unstable lesions. This study investigates the use of time-resolved fluorescence spectroscopy for unstable atherosclerotic lesion diagnosis. The fluorescence of human coronary artery samples was induced with nitrogen laser and detected in the 360- to 510-nm wavelength range. The samples were sorted into 7 groups according to the AHA classification: normal wall and types I, IIa (fatty streaks), III (preatheroma), IV (atheroma), Va (fibrous), and Vb (calcified) lesions. Spectral intensities and time-dependent parameters [average lifetime
f;
decay constants: 1 (fast-term),
2 (slow-term), A1
(fast-term amplitude contribution)] derived from the time-resolved
spectra of coronary samples were used for tissue
characterization. We determined that a few intensity values at longer
wavelengths (>430 nm) and time-dependent parameters at
peak emission region (390 nm) discriminate between all types of
arterial samples except between normal wall and type I
lesions. The lipid-rich lesions (more unstable) can be discriminated
from fibrous lesions (more stable) on the basis of time-dependent
parameters (lifetime and fast-term decay). We inferred that
features of lipid fluorescence are reflected on lipid-rich
lesion emission. Our results demonstrate that analysis of the
time-resolved spectra may be used to enhance the discrimination between
different grades of atherosclerotic lesions and provide a means of
discrimination between lipid-rich and fibrous lesions.
Key Words: atherosclerosis • lesion instability • time-resolved laser-induced fluorescence • spectroscopy
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Rupture of coronary atherosclerotic lesions leads to the acute coronary syndromes of unstable angina, acute myocardial infarction, and ischemic sudden death.1 2 3 Evidence suggests that lesion composition plays a crucial role in lesion instability and that lipid-rich lesions are more prone to rupture than fibrous lesions.1 2 3 Current clinical techniques (angiography, angioscopy, ultrasound) are limited in their ability to characterize lesion composition and identify lipid-rich lesions.3 Various techniques are currently under study as potential new tools for identification of lipid-rich lesions (MRI,4 near-infrared spectroscopy,5 Raman spectroscopy6 ) or markers of instability, such as macrophage activation in fibrous cap (local thermography7 ).
Several groups have investigated laser-induced fluorescence spectroscopy (LIFS) as a tool for analyzing plaque composition in the attempt to guide laser angioplasty and to evaluate the likelihood of restenosis.8 9 10 11 12 13 The research was carried out for both ex vivo8 9 10 11 12 and in vivo12 13 conditions. These early studies have demonstrated the potential of LIFS to characterize a few types of atherosclerotic lesions (fibrous, atheromatous, calcified) and to discriminate those from the normal arterial wall. These studies, however, have neither fully explored nor demonstrated the potential use of LIFS as a clinical tool for identification and characterization of unstable or lipid-rich atherosclerotic lesions. Previous work, including research from our group, suggested that time-resolved LIFS (TR-LIFS) improves the specificity of fluorescence measurements in tissue and enhances the ability of LIFS to characterize atherosclerotic lesions and to evaluate lesion composition.10 14 15 16 17 18 19 The use of TR-LIFS for arterial tissue characterization is suitable for several reasons. Time-resolved fluorescence measurement (1) can resolve the spectral overlap of endogenous fluorophores in tissue; (2) is independent of fluorescence emission intensity as long as the signal-to-noise ratio is commensurable, and thus independent of the presence of the endogenous chromophores in tissue (hemoglobin) or of excitation-collection geometry (optical assembly); and (3) is sensitive to microenvironmental parameters in tissue (pH, enzymatic activity, temperature), and thus may reflect inflammatory activity at the arterial wall level. These are important features for in vivo diagnostics.
The purpose of this research was to investigate the use of TR-LIFS as a diagnostic tool for assessment of unstable atherosclerotic lesions. Using a time-domain time-resolved technique, this work attempts to (1) determine the spectrotemporal fluorescence emission characteristics of excised human coronary tissue (normal and atherosclerotic vascular wall); (2) identify the main spectrotemporal fluorescence features that provide a means of discrimination between arterial tissue types, in particular the fluorescence parameters able to distinguish the lesions with lipid-rich core and thin collagenous cap (more unstable) from those with thick collagenous cap (more stable); and (3) infer correlations between fluorescence features and morphological and compositional changes that occur during the atherosclerotic process.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Instrumentation
The experimental setup has been described by our group.16 For details, please see Figure
I and the instrumentation description at
http://atvb.ahajournals.org.
|
Samples
The coronary artery samples (total of
58 coronary segments from 11 subjects; subject age 11 to 85
years; median age 48 years) were obtained at autopsy (24 to 48 hours
postmortem). The histological sections were examined by
a pathologist and classified according to the American Heart
Association (AHA)
classification3 20
as follows: normal, type I (early lesion), type II (progression-prone
type II corresponding to type IIa in Stary’s
classification), type III (preatheroma), type IV
(atheroma), type Va
(fibroatheroma), and type Vb
(complicated lesions with advanced calcification close to the intimal
surface but not exposed calcification). For tissue and
histological section processing and assessment, please
see http://atvb.ahajournals.org.
Experimental Procedures
The fluorescence response pulses derived from
each excitation laser pulse were measured at 5-nm intervals for the
360- to 510-nm spectral range. For details, please see
http://atvb.ahajournals.org.
Data Analysis
The time-integrated fluorescence spectrum
(conventional spectral emission) was computed from the measured
fluorescence response pulses. The time-resolved
fluorescence spectrum [fluorescence impulse response
function If() (FIRF)] was constructed by
deconvolving the measured laser pulse from the measured
fluorescence pulse at each wavelength across the spectrum.
Spectral intensities and time-dependent parameters
[average lifetime f; decay constants:
1 (fast-term), 2
(slow-term), A1 (fast-term amplitude
contribution)] derived from the time-resolved spectra of
coronary samples were used for tissue characterization. For
details, please see
http://atvb.ahajournals.org.17 21
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Histology
Of 58 histological samples that underwent spectroscopic examination, 46 sections were histopathologically classified as normal, n=13; type I, n=4; type II, n=5; type III, n=4; type IV, n=9; type Va, n=7; and type Vb, n=4. Representative histological sections are depicted in Figure 1
. For descriptions of sample composition and
morphology, please see http://atvb.ahajournals.org. The median
thickness of intima for each tissue type is shown in Figure II
(http://atvb.ahajournals.org). Although the median thickness gradually
increased as a function of lesion type, we recorded similar ranges
of intimal thicknesses for normal and type I samples and for types IV,
Va, and Vb samples. The
thickness of the intimal cap overlying the lipid pool ranged between
150 and 350 µm for type IV and 450 and 600 µm for type
Va lesions. For type Vb,
the cap above the calcium deposit varied in the 50- to 150-µm
range.
Time-Integrated Fluorescence
Spectra
The fluorescence spectra of coronary
specimens are shown in
Figure 2a. For a detailed description, please see
http://atvb.ahajournals.org.22
The variation of emission intensity with increased atherosclerotic
level was observed primarily at the longer wavelengths.
Figure 2b depicts the changes of emission intensity for 430
and 470 nm (I430, I470).
The intensity decreased considerably from normal coronary wall
(I470 
60%) to type Va
lesions (23%), except for type II lesions
(70%).
|
Time-Resolved Fluorescence
Spectra
Representative FIRFs of
coronary samples are shown in
Figure 3. The emission of normal coronary wall
(Figure 3a) lasted for 14 ns (5% of initial intensity) at
the wavelength region of the main peak emission (390 nm). Similar
values were observed across the spectrum. The emission of type I
lesions (not shown) presented characteristics similar to those
of normal wall. The emission of type II lesions
(Figure 3b) lasted for a shorter time (10 ns at 390 nm,
8 ns at 470 nm) than that of normal wall. For type III (not shown),
the emission at the main peak region was comparable to that of normal
wall, but it diminished as a function of wavelength (12 ns at 470
nm). The emission of type IV
(Figure 3c) lasted 13 ns at 390 nm and decreased as a
function of wavelength (10 ns at 470 nm). The emission of type
Va
(Figure 3d) lasted for 20 ns at the region of the main
peak and for 13 ns at longer wavelengths. For type
Vb lesions, the fluorescence decayed
faster than that of type Va. Their emission (not
shown) lasted 15 ns at peak range and 12 ns at a longer spectral
range. Note that for deconvolution, we used a weighting factor
proportional to the inverse of experimental variance. This factor is
probably not optimal for some experimental data, thus resulting in
artifacts (oscillations) in time in their
FIRFs.
|
Fluorescence Decay Dynamics
Characteristics
The average lifetime varied as a function of both
wavelength and lesion type. The lifetime of normal wall and all lesion
types except type II was increased at the peak emission region (390 nm:
1.9- to 3.25-ns range) compared with longer wavelengths (470 nm: 1.4
ns). For type II, the lifetime averaged 0.9 ns and appeared to be
constant along the emission spectrum. At the peak region, the lifetime
varied significantly with lesion progression.
Figure 4 describes the variation of lifetime for 390-nm
emission. Except for type II lesions, the lifetime values measured at
430 and 470 nm did not change with lesion type. The average lifetimes
along the emission spectrum for each lesion type are depicted by Figure
III (please see http://atvb.ahajournals.org).
|
The decay constant (fast- and slow-term time
constants; fast-term amplitude constant) variation as a function of
lesion type for 390-nm emission is depicted in
Figure 5. Both 1 and
2 values increased significantly from normal
(1=1.05 ns, 2=5.75
ns) to type Va lesions
(1=1.95 ns, 2=7.70
ns), whereas A1 did not vary significantly with
lesion progression. Type II lesions were different from these trends.
They were associated with a reduced 1 and
increased 2 and A1
values compared with normal and type I. Also, type
Vb lesions were characterized by lower
time-decay constant values than those observed for type
Va, suggesting a faster emission decay for
calcified lesions than that of collagenous lesions. The time constants
at 430 and 470 nm (not shown) did not vary significantly with lesion
progression.
Table
I (http://atvb.ahajournals.org) summarizes the decay
characteristics (f,
1, 2, and
A1) for 390-nm emission and fluorescence
intensity at 430 nm (I430) and 470 nm
(I470) values able to discriminate (post hoc
test) between various types of arterial tissue. The
variation of decay constants as a function of wavelength is
represented in Figure IV and described at
http://atvb.ahajournals.org.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This study examined the time-resolved spectra of ex vivo human coronary artery wall (normal and graded levels of atherosclerotic lesions types I, II, III, IV, Va, and Vb) to characterize them and to determine whether features of their fluorescence provide a means for lesion classification. We determined that 6 parameters derived from spectral emission at longer wavelengths (>430 nm) and time-resolved emission at the peak emission region (390 nm) discriminate between all types of arterial samples except between normal wall and type I lesions. Furthermore, the lipid-rich lesions can be discriminated from fibrous lesions primarily on the basis of time-dependent parameters, the lifetime and the fast-term decay component. The lipid-rich lesions exhibit faster emission decay than fibrous lesions.
Spectral and Temporal Fluorescence
Characteristics
The spectral emission of coronary artery has
been reported for both ex
vivo9 10 23 24
and in vivo12 13
investigation and for various excitation wavelengths: 308
nm,12 325
nm,23 24 337
nm,10 and 476
nm.9 The time-integrated
spectra of normal coronary artery described by our study are in
agreement with the emission spectra reported by the research groups
that used 337 nm and 325 nm for
excitation.10 24
Their emission was characterized by a broad spectrum (360- to 550-nm
range; peak 380 to 390 nm), modulated by a valley at 415 nm.
Furthermore, Andersson-Engels et
al10 and Gindi et
al24 showed that the
emission of atherosclerotic lesions was reduced at longer wavelengths
compared with the emission of normal coronary
arterial wall. Our results confirm these early findings by
displaying a gradual decrease of the intensity at wavelengths >420 nm
from normal tissue to type V lesions. In the early
studies,3 20 the
atherosclerotic specimens were differentiated from the normal samples
on the basis of their intimal thickness (<200 µm normal, >200
µm atherosclerotic). In our study, the lesions were classified
into 6 groups on the basis of their composition and morphology (AHA
classification).3 20
The discrimination between various lesion types provides a means for
interpretation of spectral data in correlation with lesions in diverse
clinical situations. For instance, the spectral emission data from the
red range allow discrimination
(Figure 2, Table
I) between early lesions from progression-prone areas
(type IIa) and normal coronary artery
wall, collagenous (type Va), and calcified (type
Vb) lesions but not between lipid-rich (type IV)
and collagenous (type Va) or calcified (type
Vb) lesions.
The time-resolved fluorescence emission of
coronary artery has been described by only a few studies, in
particular that by Andersson-Engels et
al.10 This research group
investigated the fluorescence of human aorta and
coronary artery samples generated by 337-nm picosecond laser
pulses and recorded with a photon counting detection technique.
Three exponentials were used for characterization of
fluorescence decay at 3 wavelengths: 380, 437, and 480 nm. The
samples were grouped on the basis of their intimal thickness: normal,
thin plaque (<500 µm), thick plaque (>500 µm), and
calcified. Our results agree with the previous
findings,10 although no
extensive comparison can be made between the present study and the
early observations, because different experimental methods, data
analysis techniques, and criteria for sample classification
were used. For instance, we observed an increase of fast- and slow-term
time constant values (390 nm) from normal tissue to type
Va lesions
(Figure 5) that resemble the trends reported
earlier,10 such as the
increase of time decay constants from normal (380 nm:
1=6.0 ns, 2=1.7 ns,
3=0.2 ns) to thick plaque
(1=7.3 ns, 2=2.4
ns, 3=0.3 ns). The present study
complements the previous work by reporting the characteristics of
fluorescence decay for graded levels of atherosclerotic
lesions, thus providing information regarding changes of time-resolved
fluorescence in relationship to the morphogenetic sequence of
lesion progression. We determined that progression-prone lesions (type
IIa) can be distinguished from normal wall and
advanced lesions on the basis of their fast decay, reflected by short
lifetime (<0.9 ns), short fast-term time constants
(1<0.7 ns), and large values of
1 contribution to the overall decay
(A1 >0.8). Furthermore, the lifetime and the
fast-term time constants derived from the peak spectral range can
provide discrimination between lipid-rich (type IV) and collagenous
(type Va) lesions. Because the intimal
thicknesses of type IV and Va lesions were not
resolved (Figure II), our results suggest that compositional
differences between the 2 types are reflected in their time-dependent
parameters. Overall, the time-dependent
parameters provide means of discrimination between the
majority of tissue groups investigated in this study
(Table
I) and complement the parameters inferred
from spectral information for lesion evaluation. In addition, this
study describes the variation of time-dependent parameters
as a function of wavelength.
Fluorescence of Coronary
Artery: Interpretation in Terms of Intrinsic Fluorescent
Components
Fluorescence of human arteries on UV
irradiation has been attributed to several intrinsic constituents. The
main fluorescent components of normal and diseased
arterial wall, together with their emission
characteristics, are depicted in
Table
II
(http://atvb.ahajournals.org).25 26 27 28 29 30 31 32
The fluorescence of the structural proteins elastin and
collagen was related primarily to artery fluorescence. Previous
research8 25
demonstrated that the fluorescence of elastic fibers within the
internal lamina or media predominates over the emission of normal
coronary artery or aorta, respectively, whereas the
fluorescence of collagen (30% to 60% dry
weight3 ) prevails over the
emission of advanced collagenous lesions (type
Va or fibrous plaques). The spectrotemporal
fluorescence of both proteins is well documented for 337-nm
excitation
(Table
II).
The atherosclerotic stages between normal and advanced
collagenous lesions are characterized by accumulation within
arterial intima of various types and amounts of lipids
(20% to 25% in type II, 35% to 30% in type III, 60% in type IV)
that alter the intimal
morphology.33 Although
reported by several studies, the contribution of lipids to
arterial wall fluorescence is still not well
understood. Blood-derived particles that undergo
transendothelial diffusion, such as LDL or VLDL and
their oxidative products, fluoresce, and are thus likely to
modulate the fluorescence emission of matrix structural
proteins in arterial
tissue.28 30
Cholesteryl esters (cholesteryl oleate and linoleate) account for
>60% of the lipid composition of types II and III
lesions.20 Cholesteryl
oleate is the major lipid component of type II lesions, and the
oleate/total ester (oleate and linoleate) ratio decreases with lesion
progression,20 34 35
whereas free cholesterol accumulates largely within the
necrotic core of type IV lesions, and the free/total
cholesterol ratio increases with lesion progression.
Previous LIFS studies reported characteristics of
cholesterol fluorescence emission
(Table
II). Both free and esterified
cholesterols exhibit emission trends that were distinct
from those of elastin and collagen. Thus, the cholesterol
emission features are likely to affect the fluorescence
dynamics of lesions with high cholesterol content. Also, 2
lipopigments, ceroid and carotenoids, were related to atherosclerotic
lesion fluorescence
emission.15 31 32
Ceroid, an end product of lipoprotein oxidation, is found in
macrophages and in lesions with lipid-rich necrotic cores.
Carotenoids are present in lipid-rich lesion cores. Their peak
emission, however, is >510 nm, a spectral range not explored by this
study.
Other components, such as
glycosaminoglycans, tryptophan, and calcium, are
also reported to fluoresce. Glycosaminoglycans
represent <2% of the organic matrix within normal
arterial wall and 0.4% of the organic matrix within
atherosclerotic wall; thus, it has been
suggested25 that these
components may not contribute significantly to the fluorescence
of arterial wall. Tryptophan fluorescence induced
by excitation wavelengths <310 nm was shown to strongly influence the
emission of arterial wall at wavelengths <360
nm,8 but it had minimal
influence on the emission at >360
nm.8 15 Calcium
exhibits sharp fluorescence peaks in the 350- to 650-nm range
on 308-nm excitation,28 but
its emission on longer excitation wavelengths (325 nm) was not reliably
detected.25 Consequently,
these components are less likely to significantly influence the
fluorescence characteristics reported in our study (excitation
337 nm).
Early work reported a penetration depth of irradiation at
337 nm of 150 to 200
µm.8 Therefore, the
fluorescence of normal-wall and type I specimens in our study
is likely to originate not only from intima but also from media
(Figure 1), whereas the fluorescence of advanced
lesions originates entirely from diseased intima. Furthermore, the
fluorescence of type IV lesions is likely to be generated by
both the thin cap and the top layers of the lipid-rich core, in
contrast to type Va lesions, in which the
emission may be yielded only by the thick collagenous cap.
Like previous
studies,10 24
ours found spectral trends for normal coronary wall and type I
lesions that reflect the emission of elastin (broad spectrum: 70 nm
full width at half maximum, peak 410 nm) modulated by that of
collagen (blue-shifted peak: 385 nm). In addition, we determined
that time-resolved data complement the spectral information by
displaying a slightly slower emission decay for arterial
specimens (14 ns) along the wavelengths than for elastin (11 ns),
but faster than for type I collagen (20 ns), thus suggesting the
contribution of both elastin (more) and collagen (less) to the overall
fluorescence of the samples. These results are in agreement
with the histopathological analysis of arterial
samples as well as with the early interpretation of the origin of
arterial wall
fluorescence.24
Also, it is known that both types I and III collagen are the major
collagen types in the arterial wall. Immunological
studies36 have shown that
type III collagen is localized in the subendothelial
space of the normal intima, but no type I collagen was found at that
location. Consequently, for our study, another possible fluorophore is
type III collagen.
Previously,27 we found an
emission decay of 15 ns for type III collagen. These early results
also support our findings for normal and near-normal coronary
wall.
The time-resolved spectra of type Va lesions, characterized by a narrow-band emission focused in the blue spectral range, closely resembled the emission of type I collagen (Table II). These results are in agreement with our histopathological analysis, which identified thick layers of collagen fibers as a dominant feature of the type Va lesion caps. Also, they agree with the chromatographic analysis of the composition of arterial wall collagen types, which identified type I collagen as prevalent (70%) in the fibrous cap of advanced lesions.3
To the best of our knowledge, this is the first study that
describes the time-resolved spectra of early lipid-laden lesions (fatty
streaks or type II lesions), early atheroma (type III), and
atheroma (type IV) in coronary artery. The emission
features of type II lesions, characterized by a faster
fluorescence dynamics and red shift of secondary peak emission
compared with that of normal specimens, suggest the contribution of a
component with fast and enhanced fluorescence emission at
longer wavelengths. These characteristics correlate well with the
emission of free and esterified cholesterols that exhibited
a fast dynamics and a secondary peak in the 470- to 490-nm
range.29 Consequently, the
emission of type II samples could be the result of the lipid emissions
superimposed on the emission of elastin and collagen from the
underlying structures. This hypothesis correlates with the
histopathological analysis of our samples. The analysis
reveals accumulation in the intima of a large number of
macrophages known to contain considerable amounts of
cholesteryl oleate. The discrimination of macrophage deposits
is particularly important, because macrophages (foam cells) are
markers of plaque
instability.1 2 The
emission of type III and IV lesions probably originates from the
competing fluorescence emission of the large amount of lipid
components (small f and
1 values) located deeper in the intima and
the collagen fibers (large 2 values) observed
in the proximity of the endothelial surface by
histopathological analysis. The larger
A1 values and decreased emission intensity at
longer wavelengths for type III than type IV lesions correlates with
the increased amount of collagen in type IV compared with type III
lesions. Although very heterogeneous intimal structure
characterizes the transition from normal to advanced atherosclerotic
wall, our results demonstrate that TR-LIFS data reflect the
compositional changes induced by lipid accumulation and offer the
possibility of interpretation of coronary artery spectra as a
function of lipid composition.
The emission of type Vb lesions (calcified but not exposed calcification) reflects the fluorescence of the thin intimal layers that covered the calcium deposits. Consequently, collagen, elastin, and lipids are likely to contribute to their fluorescence.
Comparison Between Aortic and Coronary
Artery Wall Fluorescence
The morphology and composition of the
arterial wall varies with the type of blood vessel and
arterial wall layers. For instance, collagen-elastin ratio
and distribution in normal coronary wall (more collagen and
elastin fibers localized mainly in the internal elastic lamina) are
different from those of normal aortic wall (less collagen and elastin
distributed in the media). These morphocompositional differences have
been reported to be reflected on the spectral emission trends of the 2
types of arterial
tissue.13 24
Higher values of the emission intensity were
reported18 37 for
normal aortic wall (intensity in the red range: 80% of main peak
value) relative to those found for normal coronary artery
(65%) in this study. Also, the time-dependent emission of normal
aorta was shorter (11 ns at the main peak region) than that of
normal coronary artery (14 ns). These results suggest that
the greater collagen/elastin content in coronary wall relative
to that in aortic wall is reflected not only by the decreased
fluorescence emission intensity at longer wavelengths but also
by the long-lasting fluorescence at the region of main peak
emission.
Unlike normal arterial wall, the
histopathological analysis of the collagenous cap of advanced
lesions reveals similar morphology and composition for both
arterial beds, although the cap is much thinner in
coronary artery than aortic wall. The spectroscopic results of
this study show that the emission characteristics for type
Va lesions in coronary artery are
similar to those
previously18 37
described for aorta (inten- sity in the red-range spectrum 20%,
1=2.1±0.2,
2=7.45±0.4). These findings suggest that
the collagenous matrix of type Va lesions is
characterized by unique spectroscopic features that resemble the
emission of type I collagen. The fluorescence emission
characteristics of the lipid-rich lesions in coronary artery
show trends comparable to those determined for aorta in our early
studies,18 37
albeit different thicknesses and heterogeneous composition
and morphology of the 2 different types of arterial walls.
For both tissue types, the emission was characterized by a faster
fast-time decay component than the emission of type
Va lesions. These observations suggest that
lipid accumulation in the intima of distinct arterial beds
generates particular fluorescence emission characteristics that
provide a means of discrimination between lipid-rich (unstable) and
collagenous (stable) lesions. It remains to be determined, however, how
in vivo environmental conditions and tissue structure and geometry
influence the time-resolved spectra measurements of
arterial tissue.
Conclusions
Our results demonstrate that analysis of the
time-resolved fluorescence spectra can be used to enhance the
discrimination between different grades of atherosclerotic lesions. The
lipid-rich lesions can be differentiated from the other lesion types
(in particular, fibrous lesions) and normal arterial wall.
On the basis of indirect evidence, an increased lipid/collagen content
ratio has been related to decreased mechanical strength of the
atherosclerotic lesions and thus to increased risk of lesion rupture.
This research shows that spectroscopic features derived for lipid
components are reflected in the emission of lipid-rich lesions, whereas
characteristics of type I collagen are identified in the emission of
fibrous lesions. By inferring such a relation, this study promotes the
development of clinical instrumentation based on spectroscopic
characterization of lipid/collagen content to predict and monitor the
clinical evolution of individual lesions in vivo. Furthermore, the
analysis of the time-resolved spectra provides a wealth of
spectroscopic parameters. By use of suitable algorithms,
the optimal parameters that provide a means of
discrimination between lesion types could be identified. Our results
suggest that a few parameters that combine spectral
features at longer wavelengths and time-resolved characteristics from
the peak emission region are the best selections for coronary
artery lesion discrimination. Parameters derived from
time-resolved spectra, such as the lifetime and the fast-time decay
constants, are most likely to differentiate between lipid-rich and
fibrous lesions and be used for diagnosis. In addition, by showing that
parameters derived from time-resolved spectra can enhance
the discrimination of atherosclerotic lesions, this study demonstrates
that the TR-LIFS technique can be used for characterization of
arterial tissue and advances a paradigm that may be
applicable for other disease states, such as
neoplasms.
|
Acknowledgments |
|---|
This work was supported by the American Heart Association, Greater Los Angeles Affiliate (1082-GI), and the Biomedical Simulations Resource (NIH P41-RR01861), University of Southern California. The authors are grateful to Dr Prediman K. Shah for helpful discussions, to Thanassis Papaioannou for technical assistance, and to Dr Stephanie Erlich (Los Angeles County, Department of Coroner) for providing the arterial samples.
|
Footnotes |
|---|
Submitted April 26, 2000; revision submitted March 29, 2001; revision accepted April 4, 2001.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Felton CV, Crook D, Davies MJ, Oliver MF. Relation of plaque lipid composition and morphology to the stability of human aortic plaque. Arterioscler Thromb Vasc Biol. 1997;17:1337–1345.
2. Shah PK. Pathophysiology of plaque rupture and the concept of plaque stabilization. Cardiol Clin. 1996;14:17–29.[Medline] [Order article via Infotrieve]
3.
Stary HC, Chandler
AB, Dinsmore RE, Fuster V, Glagov S, Insull W Jr, Rosenfeld ME,
Schwartz CJ, Wagner WD, Wissler RW. A definition of advanced types of
atherosclerotic lesions and a histological
classification of atherosclerosis: a report from the
Committee on Vascular Lesions of the Council on
Arteriosclerosis, American Heart Association.
Arterioscler Thromb Vasc Biol. 1995;15:1512–1531.
4.
Shinnar M, Fallon
JT, Wehrli S, Levin M, Dalmacy D, Fayad ZA, Badimon JJ, Harrington M,
Harrington E, Fuster V. The diagnostic accuracy of ex-vivo
MRI for human atherosclerotic plaque characterization.
Arterioscler Thromb Vasc Biol. 1999;19:2756–2761.
5. Moreno PR, Lodder RA, O’Connor WN, Vyalkov VA, Purushothaman KR, Muller JE. Characterization of vulnerable plaques by near infrared spectroscopy in an atherosclerotic rabbit model. J Am Coll Cardiol. 1999;33:66A.
6.
Romer TJ, Brennan
JF III, Fitzmaurice M, Feldstein ML, Deinum G, Myles JL, Kramer JR,
Lees RS, Feld MS. Histopathology of human coronary
atherosclerosis by quantifying its chemical composition
with Raman spectroscopy.
Circulation. 1998;97:878–885.
7.
Stefanadis C,
Diamantopoulos L, Vlachopoulos C, Tsiamis E, Dernellis J, Toutouzas K,
Stefanadi E, Toutouzas P. Thermal heterogeneity within
human atherosclerotic coronary arteries detected in vivo: a new
method of detection by application of a special thermography catheter.
Circulation. 1999;99:1965–1971.
8. Baraga JJ, Rava RP, Taroni P, Kittrell C, Fitzmaurice M, Feld MS. Laser induced fluorescence spectroscopy of normal and atherosclerotic human aorta using 306–310 nm excitation. Laser Surg Med. 1990;10:245–261.
9. Richards-Kortum R, Rava RP, Fitzmaurice M, Kramer JR, Feld MS. 476 nm excited laser-induced fluorescence spectroscopy of human coronary arteries: applications in cardiology. Am Heart J. 1991;122:1141–1150.[Medline] [Order article via Infotrieve]
10. Andersson-Engels S, Johansson J, Svanberg S. The use of time-resolved fluorescence for diagnosis of atherosclerotic plaque and malignant tumours. Spectrochim Acta. 1990;40A:1203–1210.
11. Deckelbaum LI, Stetz ML, O’Brien KM, Cutruzzola FW, Gmitro AF, Laifer LI, Gindi GR. Fluorescence spectroscopy guidance of laser ablation of atherosclerotic plaque. Lasers Surg Med. 1989;9:205–214.[Medline] [Order article via Infotrieve]
12. Morguet AJ, Gabriel RE, Buchwald AB, Werner GS, Nyga R, Kreuzer H. Single-laser approach for fluorescence guidance of excimer laser angioplasty at 308 nm: evaluation in vitro and during coronary angioplasty. Lasers Surg Med. 1997;20:382–393.[Medline] [Order article via Infotrieve]
13. Bartorelli AL, Leon MB, Almagor Y, Prevosti L, Swain JA, McIntoch CL, Neville RF, House MD, Bonner RF. In vivo human atherosclerotic plaque recognition by laser-excited fluorescence spectroscopy. J Am Coll Cardiol. 1991;17:160B–168B.
14. Baraga JJ, Taroni P, Park YD, An K, Maestri A, Tong LL, Rava RP, Kittrell C, Dasari RR, Feld MS. Ultraviolet laser induced fluorescence of human aorta. Spectrochim Acta. 1989;45A:95–99.
15. Andersson-Engels S, Baert L, Berg R, D’Hallewin MA, Johansson J, Stenram U, Svanberg K, Svanberg S. Fluorescence characteristics of atherosclerotic plaque and malignant tumors. SPIE Proc. 1991;1426:31–40.
16. Marcu L, Grundfest WS, Maarek JM. Photobleaching of arterial fluorescent compounds: characterization of elastin, collagen and cholesterol time-resolved spectra during prolonged ultraviolet irradiation. Photochem Photobiol. 1999;69:713–721.[Medline] [Order article via Infotrieve]
17. Maarek JM, Marcu L, Snyder WJ, Grundfest WS. Time-resolved fluorescence spectra of arterial fluorescent compounds reconstruction with Laguerre expansion technique. Photochem Photobiol. 2000;71:178–187.[Medline] [Order article via Infotrieve]
18. Marcu L, Maarek JM, Fishbein M, Grundfest W. Atherosclerotic lesions classification by time-resolved laser-induced fluorescence spectroscopy: clinical identification of lipid-rich lesions. J Am Coll Cardiol 1999;33(suppl A):66A.
19. Lakowicz JR. Principles of Fluorescence Spectroscopy. New York, NY: Plenum Press; 1985.
20.
Stary HC,
Chandler AB, Glagov S, Guyton JR, Insull W Jr, Rosenfeld ME, Schaffer
SA, Schwartz CJ, Wagner WD, Wissler RW. A definition of initial, fatty
streak, and intermediate lesions of atherosclerosis: a
report from the Committee on Vascular Lesions of the Council on
Arteriosclerosis, American Heart Association.
Arterioscler Thromb. 1994;14:840–856.
21. Marmarelis VZ. Identification of nonlinear biological systems using Laguerre expansions of kernel. Ann Biomed Eng. 1993;21:573–589.[Medline] [Order article via Infotrieve]
22. Richards-Kortum R, Rava RP, Cothren R, Metha A, Fitzmaurice M, Ratliff NB, Kramer JR, Kittrell C, Feld MS. A model for extracting of diagnostic information from laser induced fluorescence spectra of human artery wall. Spectrochim Acta. 1989;45A:87–93.
23. Lucas A, Radosavljevic MJ, Lu E, Gaffney EJ. Characterization of human coronary artery atherosclerotic plaque fluorescence emission. Can J Cardiol. 1990;6:219–228.[Medline] [Order article via Infotrieve]
24. Gindi GR, Darken CJ, O’Brien KM, Stetz ML, Deckelbaum LI. Neural network and conventional classifiers for fluorescence-guided laser angioplasty. IEEE Trans Biomed Eng. 1991;38:246–252.[Medline] [Order article via Infotrieve]
25.
Laifer LI,
O’Brien KM, Stetz ML, Gindi GR, Garrand TJ, Deckelbaum LI. Biochemical
basis for the difference between normal and atherosclerotic
arterial fluorescence.
Circulation. 1989;80:1893–1901.
26. Yan W, Perk M, Chagpar A, Wen Y, Stratoff S, Schneider WJ, Jugdutt BI, Tulip J, Lucas A. Laser-induced fluorescence, III: quantitative analysis of atherosclerotic plaque content. Lasers Surg Med. 1995;16:164–178.[Medline] [Order article via Infotrieve]
27. Marcu L, Cohen D, Maarek JMI, Grundfest WS. Characterization of type I, II, III, IV, and V collagens by time-resolved laser-induced fluorescence spectroscopy. SPIE Proc. 2000;3917:93–101.
28. Morguet AJ, Körber B, Abel B, Hippler H, Wiegand V, Kreuzer H. Autofluorescence spectroscopy using a simultaneous plaque ablation and fluorescence excitation. Lasers Surg Med. 1994;14:238–248.[Medline] [Order article via Infotrieve]
29. Marcu L, Maarek JM, Grundfest WS. Time-resolved laser-induced fluorescence of lipids involved in development of atherosclerotic lesion lipid-rich core. SPIE Proc. 1998;3250:158–167.
30.
Oraevsky AA,
Jacques SL, Pettit GH, Sauerbrey RA, Tittel FK, Nguy JH, Henry PD. XeCl
laser-induced fluorescence of atherosclerotic arteries:
spectral similarities between lipid-rich lesions and peroxidized
lipoproteins. Circ Res. 1993;72:84–90.
31. Verbunt RJM, Fitzmaurice MA, Kramer JR, Ratliff NB, Kittrell C, Taroni P, Cothren RM, Baraga J, Feld MS. Characterization of ultraviolet laser-induced autofluorescence of ceroid deposits and other structures in atherosclerotic plaques as a potential diagnostic for laser angiosurgery. Am Heart J. 1992;123:208–216.[Medline] [Order article via Infotrieve]
32. Hunt JV, Bottoms MA, Skamarauskas J, Carter NP, Mitchinson MJ. Measurement of ceroid accumulation in macrophages by flow cytometry. Cytometry. 1994;15:377–382.[Medline] [Order article via Infotrieve]
33. Fuster V. Syndromes of Atherosclerosis: Correlations of Clinical Imaging and Pathology. Armonk, New York: Futura; 1996.
34.
Guyton JR, Klemp
KF. Development of the atherosclerotic core region: chemical and
ultrastructural analysis of microdissected atherosclerotic
lesions from human aorta. Arterioscler
Thromb. 1994;14:1305–1314.
35. Yla-Herttuala S. Biochemistry of the arterial wall in developing atherosclerosis. Ann N Y Acad Sci. 1991;623:40–59.[Medline] [Order article via Infotrieve]
36.
Stary HC,
Blankenhorn DH, Chandler AB, Glagov S, Insull W Jr, Richardson M,
Rosenfeld ME, Schaffer SA, Schwartz CJ, Wagner WD, Wissler RW. A
definition of the intima of arteries and of its
atherosclerosis-prone regions: a report from the
Committee on Vascular Lesions of the Council on
Arteriosclerosis, American Heart Association.
Circulation. 1992;85:391–405.
37. Maarek JM, Marcu L, Fishbein MC, Grundfest WS. Time-resolved fluorescence of aortic wall: use for improved identification of atherosclerotic lesions. Lasers Surg Med. 2000;27:241–254.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  S W E van de Poll, K Kastelijn, T C B. Schut, C Strijder, G Pasterkamp, G J Puppels, and A van der Laarse On-line detection of cholesterol and calcification by catheter based Raman spectroscopy in human atherosclerotic plaque ex vivo Heart, September 1, 2003; 89(9): 1078 - 1082. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Chen, C.-H. Tung, U. Mahmood, V. Ntziachristos, R. Gyurko, M. C. Fishman, P. L. Huang, and R. Weissleder In Vivo Imaging of Proteolytic Activity in Atherosclerosis Circulation, June 11, 2002; 105(23): 2766 - 2771. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||