- Top of page
Although a wide variety of imaging modalities is applied for imaging of coronary plaques angiography, intravascular ultrasonography (IVUS),1,2 optical coherence tomography (OCT),3,4,5 virtual histology,6 near-infrared spectroscopy (NIRS),7,8 computed tomography (CT),9 magnetic resonance imaging (MRI),10 conventional angioscopy using white light,11–15 positron-emission computed tomography,16 intravascular radiation detectors,17 near-infrared Raman spectroscopy,18 radioisotope-labeled antibodies,19 and intravascular MRI20 direct and 2-dimensional imaging of individual lipid components such as cholesterol (C) and cholesteryl esters (CE) deposited inside the atherosclerotic coronary plaques is clinically not successful.
If the individual components of the lipids deposited inside the plaques become visible in vivo, the progression of plaques and the effects of medical and interventional therapies on them can be evaluated more objectively.
Near-infrared light penetrates deeper into the tissues than visual light and evokes near-infrared fluorescence (NIRF) of the targeted substances. Detection of C and CE by NIRS in atherosclerotic plaques has been attempted in vitro but not clinically.21–23 NIRS shows the substance as spectra, and therefore is not suitable for direct and 2-dimensional imaging of the substances.
Therefore, the present study was carried out to obtain direct and 2-dimensional images of C and CE within human coronary plaques by a near-infrared fluorescence angioscopy (NIRFA) system in human coronary plaques in vitro and in vivo.
- Top of page
In this study, C, CE, and Ca individually did not exhibit NIR autofluorescence. Any other substances composing atherosclerotic plaques21 did not exhibit autofluorescence. Ye et al found that β-carotene accumulates in atherosclerotic plaques and exhibits fluorescence in the range of visual light.27 Blankenhorn et al found that this substance exists in the human atherosclerotic plaques and exhibits fluorescence in the range of visual light28; however, they did not examine NIRF of this substance.
In the present study, β-carotene did not exhibit NIR autofluorescence. However, the mixture of β-carotene and C, CE, or Ca exhibited NIRF. Because β-carotene is lipotrophic, it may have conjugated with these substances, forming an adduct and evoking a NIRF at wavelengths of ≥780 nm.
In the in vitro study on human coronary plaques, the NIRFA image patterns were well correlated with those obtained by NIRFM scanning of cut wall. Moreover, the location and patterns of NIRF by NIRFM scanning coincided with those of lipids and Ca determined by histology, indicating that the NIRF images by NIRFA were caused by the substances that evoke NIRF, such as C, CE, and Ca.
It was concluded that lipids deposited within a depth of 700 μm from the plaque surface could be imaged by the present NIRFA system. Although its potency is limited, it has the advantage that lipids inside the plaque can be visualized. The NIRFA images of the plaques were classified into homogenous, doughnut-shaped, and spotty types in both in vitro and in vivo studies.
In the yellow plaques observed by conventional angioscopy, the homogenous type showed deposition of lipids in the entire plaque histologically, indicating that C and CE were deposited throughout the entire plaque.
In the doughnut-shaped type, the NIRF-absent portion was surrounded by strong NIRF regions. Histological studies showed that a lipid-laden fibrous cap and necrotic core underneath existed in all samples of this type. The NIRF-absent portion featured yellow color when observed by conventional angioscopy, indicating existence of the lipids. Therefore, it is probable that the fibrous cap and necrotic core were filled with other NIRF-absent substances, such as oxidized low-density lipoprotein, triglycerides, and others.
In the spotty type, NIRF spots were observed in the NIRF-absent fibrous cap by NIRFM scanning and disseminated small Ca particles in lipid-laden thin fibrous cap by histology, indicating that Ca was disseminated in the fibrous cap.
In this study, not only C and CE, but also Ca exhibited NIRF. Calcium exhibited strong and clearly demarcated spotty NIRF even in the presence of C or CE in the in vitro study, probably due to its hydro- and lipophobic nature. Therefore, differentiation of Ca from C and CE was easy.
Discrimination of C and CE by NIRS in vitro was reported by Weinmann et al.23 These substances were not discriminated by NIRFA in the present study. However, NIRFA has an advantage in that it can image 2-dimensional distribution of these substances.
These substances in the plaques can be discriminated by changing BPF and BAF of NIRFA in future. Thus, coronary plaques that were simply classified by their surface color by conventional angioscopy into white and yellow became classifiable into 4 image patterns by NIRFA. It was considered that these were determined by distribution patterns of C, CE, and Ca.
Compared with NIRS, quantitative measurement of C, CE, and Ca deposited in the plaques was beyond the scope of the present NIRFA system because NIRF intensity measured by NIRFA was inversely correlated to the depth from the plaque surface to these substances. Although lipids and Ca were discriminated because the latter showed a clearly demarcated spotty configuration, discrimination of C and CE was beyond the scope of the present NIRFA system. Thus, these substances located ≥700 μm in depth from the plaque surface were not imaged.