JACC: Cardiovascular Imaging
Integrated IVUS-OCT for Real-Time Imaging of Coronary Atherosclerosis
Author + information
- Jiawen Li, MS,
- Xiang Li, PhD,
- Dilbahar Mohar, MD,
- Aidan Raney, MD,
- Joseph Jing, MS,
- Jun Zhang, PhD,
- Abbey Johnston, MD,
- Shanshan Liang, MS,
- Teng Ma, BS,
- K. Kirk Shung, PhD,
- Sari Mahon, PhD,
- Matthew Brenner, MD,
- Jagat Narula, MD,
- Qifa Zhou, PhD∗ (qifazhou{at}usc.edu),
- Pranav M. Patel, MD† (pranavp{at}uci.edu) and
- Zhongping Chen, PhD‡ (z2chen{at}uci.edu)
- ↵∗1042 Downey Way, University Park, NIH Ultrasonic Transducer Resource Center, University of Southern California, Los Angeles, California 90089.
- ↵†101 City Drive South, City Tower, Suite 400, Orange, California 92868.
- ↵‡1002 Health Sciences Road East, Beckman Laser Institute, University of California, Irvine, Irvine, California 92612.
Accurate assessment of atherosclerotic plaque characteristics and the subsequent tailoring of optimal therapy holds great promise for preventing acute coronary syndromes (ACS) and life-threatening sequelae (1). Combined use of optical coherence tomography (OCT) and intravascular ultrasound (IVUS) was proposed as a potential method for accurate assessment of plaque characteristics and vulnerability (2,3). However, significant challenges remain in trying to adapt an integrated OCT-IVUS system for clinical applications. We report here a fully integrated intracoronary OCT-IVUS imaging technique to visualize atherosclerotic plaque in living animals and human coronary arteries from cadavers with high resolution and deep penetration capability simultaneously.
First, we created lesions, similar to human atherosclerotic plaques, in male New Zealand white rabbits by feeding them a high-cholesterol diet and subjecting them to de-endothelization procedures (4). We imaged plaques in rabbit aortas using the OCT-IVUS system with a 3.4-F integrated catheter (5) and a 10-ml Omnipaque (iohexol, 350 mgI/ml, GE Healthcare, Princeton, New Jersey) injection flushing into the artery at ∼3 ml/s for blood clearance. A total of 10 volumetric datasets were obtained from 5 rabbits. Ten 2- to 20-mm long aorta segments were imaged at 5 mm/s pull-back speed. A representative OCT-IVUS image pair and corresponding histology of a rabbit abdominal aorta with a thick-cap fibroatheroma is shown in Fig. 1 (row I). IVUS enables the visualization of the layer structure of the artery wall. Intimal thickening and a low-density acoustic signal region (denoted by the arrow in Fig. 1Ia) demonstrate plaque in the IVUS image. However, this image also illustrates the inability of IVUS to determinate the plaque type and the plaque cap boundary. At the same site in the OCT image (Fig. 1Ib), a homogenous boundary and weak signal region under a high signal region indicates that this plaque is a necrotic/lipid plaque with an overlying fibrous cap. In addition, the minimum thickness of the cap can be easily measured to be ∼200 μm by using OCT, which is indicative of a thick-cap fibroatheroma. The classification of plaque type is validated by the corresponding histology photo (Fig. 1Ic), which shows loose necrotic material. This area is covered by smooth muscle and fibrous proliferations at the luminal surface, which is consistent with a fibrous cap. All IVUS-OCT images of rabbit aortas were matched with histology for correlation of accuracy. Linear regression showed a high correlation between plaque circumference percent (PCP) (defined as the circumference of lumen in which there is plaque divided by the entire lumen circumference) determined from histological analysis and the estimated PCP of OCT and IVUS (R2 = 0.955, p < 0.001 between OCT and histology; R2 = 0.970, p < 0.001 between IVUS and histology).
Representative Images Acquired by the Fully Integrated Intracoronary OCT-IVUS System
(Top row) in vivo imaging of rabbit abdominal aorta with an optical coherence tomography (OCT)–intravascular ultrasound (IVUS) system. (Ia) IVUS and (Ib) OCT cross-sectional images of atherosclerosis microstructure in a rabbit; (Ic) corresponding hematoxylin and eosin (H&E) histology. The artifact circles in the IVUS images (* in Ia) are due to the ultrasound pulse ring-down effect and the reflection of the catheter sheath. The artifact circle in the OCT image (* in Ib) is caused by the high back-reflection from the interface between the prism and the gradient index lens. The shape of this artery changed between (Ia) in vivo imaging and (Ic) histology due to the reduced intralumen pressure after this artery was harvested. (Second row) an OCT-IVUS image pair obtained in a normal swine coronary artery in vivo. (IIa) IVUS image, (IIb) OCT image, and (IIc) corresponding H&E histology. Guidewire (G) artifact is denoted by * in IIb. Yellow boxes denote the left anterior descending branch. From the center of the OCT image, there is a high-signal, thin band corresponding to the intima, followed by a high-signal strip corresponding to the external elastic lamina, and finally a low-signal area corresponding to the adventitia. (Bottom 3 rows) imaging of a coronary artery with a fibrous plaque (third row), calcified plaque (fourth row), and lipid plaque (bottom row). (IIIa) (IVa) (Va) IVUS images, (IIIb) (IVb) (Vb) OCT images, and (IIIc) (IVc) (Vc) corresponding histology images. Insets in Vc are highly magnified images of the histology slides: left inset, stained with H&E; right inset, stained with cluster of differentiation 68. Arrows denote the location of plaques. Plaque types can be classified on the basis of optical scattering contrast of different tissue types: 1) signal-rich regions from 5 o' clock to 7 o'clock in IIIb indicate a fibrous plaque; 2) a sharp-boundary, signal-poor region from 11 o' clock to 4 o'clock in IVb indicates a calcified plaque; 3) a diffusive-boundary, signal-poor region from 5 o' clock to 7 o'clock in Vb indicate a lipid plaque. The histology results confirm the classification of plaque types by the OCT and IVUS images: the dense, compact eosinophilic fibers in IIIc represent increased amounts of collagen seen in a fibrous plaque. In IVc, the residual calcium can be seen as numerous irregular refractile purple crystals. In Vc, foam cells and dark brown staining in the cluster of differentiation 68 stain slide verify that this is a lipid plaque. Note for Vc: because this excised tissue is older (∼10 months’ post-mortem), the degraded nuclear material did not stain well with hematoxylin. Although the tissue is predominantly pink in color, the structure and architecture are preserved. Scale bar = 1 mm. A = adventitia; E = external elastic lamina; I = intima; T = tissue; V = vessel.
Second, a female Yorkshire white swine was imaged by conventional femoral access and angiography guidance under the same flushing procedure as in the rabbits. The goal was to test the feasibility of translating this technology into clinical applications (Figs. 1, row II). In the IVUS image (Fig. 1IIa), the 3-layer structure of the swine artery (wall thickness ∼0.4 mm) is barely visualized with an IVUS axial resolution of 60 μm. In Figure 1IIb, the OCT image differentiates the 3 structural layers of the artery wall.
Last, we collected 14 cadaver coronary arteries from 6 patients who died of complications from ACS or were diagnosed with atherosclerotic heart disease. Representative OCT-IVUS image pairs of a fibrous plaque, calcified plaque, and lipid plaque from different cadavers are shown in Figure 1/rows III, IV, and V, respectively. An acoustic shadow in Figure 1IVa shows the location of a calcified plaque. However, it is difficult to classify the plaque morphology in Figure 1IIIa and Figure 1Va by using IVUS imaging because of intrinsically limited resolution and low soft tissue contrast. The OCT imaging is able to classify plaque morphology by optical scattering contrast of different tissue types. However, with limited penetration depth, the OCT image cannot provide a clear visualization of the media and adventitia layer at this intima-thickening coronary segment. These results clearly demonstrate the complementary nature of OCT and IVUS imaging. A total of 28 OCT-IVUS image pairs, obtained from 14 plaque samples (2 pairs from each sample, pull-back and repull-back), were analyzed for quantitative validation of the technique's accuracy and reproducibility. Linear regression showed a high accuracy (R2 = 0.911, p < 0.001 for OCT histology; R2 = 0.923, p < 0.001 for IVUS histology) and high reproducibility (R2 = 0.937, p < 0.001 for OCT; R2 = 0.971, p < 0.001 for IVUS) of PCP measurements.
Our fully integrated in vivo imaging system has high resolution to identify the thin cap and deep penetration to visualize the necrotic core simultaneously. Such a device may lead to a more accurate assessment of vulnerable plaques and especially thin-cap fibroatheroma. Moreover, most of the current understanding about ACS has been achieved through static histopathology research. This novel, in vivo integrated OCT-IVUS imaging technique is anticipated to improve our understanding of the process of this disease through longitudinal in vivo studies.
Acknowledgments
The authors acknowledge Mr. E. Steward, Ms. T. Burney, Mr. D. Mukai, and Mr. D. Yoon for their assistance during surgical procedures, as well as Ms. L. Liaw and Ms. L. Li for their assistance in histological analyses. The authors also thank individuals who donated their bodies and tissues for the advancement of education and research.
Footnotes
Please note: This work was supported by the National Institutes of Health under grants R01EB-10090, R01HL-105215, R01EY-021519, P41EB015890 (Laser Microbeam and Medical Program), R01CA124967, and K25HL-102055. Dr. Chen has a financial interest (founder, chairman, and consultant) in OCT Medical Imaging Inc., which, however, did not support this work. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Jiawen Li and Xiang Li have equal contributions to this work and are treated as cofirst authors. Marc D. Feldman, MD, and Sherif Nagueh, MD, served as Guest Editor’s for this article.
- American College of Cardiology Foundation
