Intracoronary optical coherence tomography: a comprehensive review clinical and research applications

Abstract

Cardiovascular optical coherence tomography (OCT) is a catheter-based invasive imaging system. Using light rather than ultrasound, OCT produces high-resolution in vivo images of coronary arteries and deployed stents. This comprehensive review will assist practicing interventional cardiologists in understanding the technical aspects of OCT based upon the physics of light and will also highlight the emerging research and clinical applications of OCT. Semi-automated imaging analyses of OCT systems permit accurate measurements of luminal architecture and provide insights regarding stent apposition, overlap, neointimal thickening, and, in the case of bioabsorbable stents, information regarding the time course of stent dissolution. The advantages and limitations of this new imaging modality will be discussed with emphasis on key physical and technical aspects of intracoronary image acquisition, current applications, definitions, pitfalls, and future directions.

Figure 1. Scheme of TD-OCT and FD-OCT

A schematic representation of time-domain optical coherence tomography (TD-OCT, left panel) and frequency or Fourier-domain optical coherence tomography (FD-OCT, right panel depicts a subtype of FD-OCT called optical frequency domain imaging [OFDI OCT]). Both systems use a reference arm and an interferometer to detect echo time delays of light. The interferometer uses a beamsplitter, dividing the light into a measurement arm (tissue sample) and a reference arm. The reference arm in TD-OCT is mechanically scanned (by a moving mirror) in order to produce a time-varying time delay. In the FD-OCT, because the light source is frequency swept, the interference of the 2 light beams (tissue and reference) oscillates according to the frequency difference. In both systems the interference of the signal ultimately provides amplitude and frequency data. In the FD-OCT system, all echo delays are acquired simultaneously enabling significant increases in the speed of image acquisition.

Figure 2. Most Frequent Artifacts

Cross-sectional image of the human coronary artery. Most frequently observed artifacts: (A) incomplete blood displacement, resulting in light attenuation. (B) Eccentric image wire can distort stent reflection orientation, the struts align toward the imaging wire “sunflower effect” and are elongated “merry-go-round.” (C) Saturation artifact, some scan lines have a streaked appearance. (D) Sew-up artifact: result of rapid wire or vessel movement along 1 frame formation, resulting in misalignment of the image. (E) Air bubbles, formed inside the catheter, produce an attenuated image along the corresponding arc. Detail reveals the bubbles, bright structures, between 5 and 9 o’clock. (F) Fold over artifact (Fourier-domain optical coherence tomography system), the longitudinal view demonstrates that the cross section is located at the level of a side branch (blue line).

Figure 3. Malapposition Quantification

Cross-section image illustrating various levels of stent strut protrusion. Strut located at 7 o’clock is malapposed (detail). Malapposition is defined when the measured distance from the surface of the blooming to the lumen contour is higher than the total thickness of the stent strut + polymer + one-half of the blooming (the stent surface should be, theoretically, located at one-half the distance of the blooming thickness). In this particular case of a Taxus Express (Boston Scientific, Natick, Massachusetts) stent, the total, estimated, strut thickness = 164 μm (strut thickness = 132 μm + polymer = 16 μm + one-half of the blooming = 18 μm). The measured distance was 200 μm confirming a malapposed strut. Because of the eccentric wire position, the blooming component is elongated on the struts distant from the wire (5 and 7 o’clock), “merry-go-round” effect.

Figure 4. Fibrotic Cap Measurement

Representative cross section containing atherosclerotic plaque with different fibrotic cap thicknesses. (A and B) Thin fibrotic cap, measuring 40 μm. (C and D) Thick fibrotic cap, measuring 250 μm.

Figure 5. Optical Coherence Tomography: Histology Correlation

(A) Fibrotic plaque: characterized by high signal (high backscattering) and low attenuation (deep penetration). (B) Predominantly calcified plaque: calcified regions have a sharp border, low signal, and low attenuation permitting deeper penetration. (C) Lipid-rich plaque: the lipid core has a diffuse border. High light attenuation results in poor tissue penetration (in contrast to calcified regions). The overlying fibrotic cap can be readily measured; in this case a thick cap (>200 μm) is present. ‡Calcified region; *lipid core. Courtesy of LightLab Imaging (C. Y. Xu and J. M. Schmitt).

Figure 6. Coronary Imaging Acquired With an FD-OCT System

A longitudinal reconstruction (lower panel) and cross-sectional images (upper panels) acquired with a frequency or Fourier-domain optical coherence tomography (FD-OCT) system (Lightlab) at 20 mm/s immediately after stent implantation. Note that a 5-cm coronary segment was imaged with a 3-s contrast injection. (A) Distal edge dissection, with corresponding longitudinal view (arrows). (B) Well-expanded and well-apposed stent struts, with corresponding longitudinal view. (C) Malapposed struts between 11 and 1 o’clock, with corresponding longitudinal view (arrows). (D) Proximal calcified plaque with minimal fibrous coverage, with corresponding longitudinal view (arrows).

Figure 7. Quantitative Stent Analysis

(A) Representative frame depicting a Core-Lab area measurement and strut level analysis. Strut level analysis consists of a qualitative assessment for strut coverage and quantitative measurement from the surface of the blooming artifact to the lumen contour. A heterogeneity of strut coverage is observed within a single frame (C to N refer to labeled struts): covered struts (E to G), uncovered apposed struts (H to J), and malapposed struts (K to N, C and D). (B) Magnification of the automatic 360° chord system, applied between the stent and lumen contours, allowing a detailed measurement of the stent coverage. (C) Automatic lumen and stent detection: stent struts (green dots) and lumen contour (red line). (D) En face 3-dimensional view: the highly reflective stent surface allows easy discrimination of the struts from the surrounding tissue. (C and D) LightLab’s automated stent strut analysis software (R&D program, not released) courtesy of LightLab Imaging (C. Y. Xu and J. M. Schmitt), images obtained at Wakayama University (Prof. Akasaka). NIH = neointimal hyperplasia.

Figure 8. Comparison Between OCT Capabilities and Others’ Imaging Methods

Graphic representation of tissue penetration versus spatial resolution of optical coherence tomography (OCT) compared with other imaging methods. The solid boxes represent the maximal resolution and depth achieved with current technologies. The larger open boxes represent the range of resolution and tissue penetration for each of the technologies. The dashed box illustrates the hypothetical capabilities of future OCT systems. Resolution requirements to accurately detect neointimal hyperplasia observed after drug-eluting stent (NIH) usually exceeds the capabilities of intravascular ultrasound (IVUS). Similarly, fibrous cap thickness can only be assessed in vivo by OCT. However, current OCT systems are not suitable to assess tissue at depths beyond 2 mm. Currently, evaluation of single cell endothelial layers can only be assessed in vitro by microscopy or advance bench-top OCT systems. Atherosclerosis = early and advanced stages of the process; Endothelium = single intimal cell layer; Fibrous Cap = thickness range related to thin cap fibroatheroma.