Optical coherence tomography (OCT) is an emerging imaging modality analogous to intravascular ultrasound imaging but uses light instead of sound. The integration of a fiber-optic probe with frequency domain OCT enables video images that display the location and changes of coronary plaques and stent apposition in live patients. This chapter details the basic principles of intravascular optical coherence tomography (IV-OCT) in clinical practice. The system architecture and catheter structure consisting of an optical probe and a protective sheath are discussed in detail. Also, recent technology advances in IV-OCT are briefly introduced.

1 Introduction to OCT

Optical coherence tomography (OCT) is a high-resolution imaging modality that provides real-time cross-sectional images of tissue microstructures using the near-infrared light [1]. As OCT has common features with ultrasound imaging and microscopy in medical applications, it has been clinically adopted in ophthalmology, dermatology, and cardiology [2,3,4]. In most tissues, OCT imaging plays an important role in filling a gap between microscopy and ultrasound in comparison with resolution and imaging depth as shown in Fig. 10.1 [5]. Microscopy performs very high-resolution (~1 μm) imaging of en face tissue plane, but imaging depth in biological tissues is limited up to only a few hundred micrometers due to the signal attenuation from large optical scattering. The resolution of medical ultrasound imaging varies 0.1–1 mm depending on the sound wave frequency. It is, however, possible to see internal organs even if the imaging depth is limited to only millimeter ranges at high frequencies of ultrasound waves [6].

Fig. 10.1
figure 1

Image resolution and penetration depth for OCT, microscopy, and ultrasound. OCT plays a role in filling a gap between microscopy and ultrasound in medical imaging. The image resolution in OCT is about 1–15 μm and the imaging depth is limited to 2–3 mm. Microscopy performs very high-resolution (~1 μm) imaging of en face tissue plane, but the imaging depth in biological tissues is limited up to only a few hundred micrometers due to the signal attenuation from large optical scattering. The resolution of medical ultrasound imaging varies 0.1–1 mm depending on the sound wave frequency, but it is possible to see internal organs even if imaging depth is limited to only millimeter ranges at high frequencies of ultrasound waves

Full size image

Compared with ultrasound imaging, OCT has the same operation principle, echo signal detection, but utilizes infrared light instead of ultrasound. In general, imaging is performed by measuring the magnitude and time delay of backscattered or backreflected signal from internal biological tissues. As a sound wave travels at 340 m/s in air, the echo signal can be measured with a time resolution of ~100 ns, which is within the limits of the electronic detection process. However, it is impossible to electrically measure echoes of backscattered light due to the light speed of 3 × 108 m/s in air, and optical interferometric techniques were then proposed [7,8,9]. Optical interferometers are widely used in science and engineering to measure small displacements and spatial irregularities by measuring interference patterns. To achieve microscale resolutions of the optical sectioning ability, low-coherence interferometry, using the short coherence length of a broadband light source, is required. Low-coherence interferometry used in OCT is a unique solution to measure the echo signal of backscattered light with a very high signal-to-noise ratio, which is termed system sensitivity. The incident beam of light source is divided into a reference beam reflected from a mirror and a sample beam illuminated on biological tissues, and the sum of two beams is directed to a photodetector that measures the intensity of the combined signals [10].

1.1 Time Domain OCT and Frequency Domain OCT

OCT systems mainly consist of a light source, a coupler- or a circulator-based interferometer, and a photodetector. OCT is categorized into time domain OCT (TD-OCT) and frequency domain OCT (FD-OCT) as illustrated in Fig. 10.2 [11, 12]. TD-OCT employs a broadband light source such as a superluminescent diode and a scanning reference arm. The path length difference between the sample and reference arms of the interferometer is modulated by scanning a reference path length. The envelope of interference fringes is then extracted as a function of time, which means that image data is finally generated in a time domain. A single travel of the reference mirror creates a depth profile or an A-scan. Development of the high-speed scanning delay line of the reference arm allows OCT imaging speeds of several thousand axial profiles per second and video frame rate [13, 14]. FD-OCT additionally requires a spectrometer as a photodetector or wavelength-swept laser as light source without scanning a reference length. The former case is referred to as spectral domain OCT (SD-OCT), whereas the latter is termed swept-source OCT (SS-OCT) or alternatively optical frequency domain imaging (OFDI) [15,16,17,18,19]. SS-OCT has the advantages of easy implementation of polarization diverse detection as well as large depth range over SD-OCT. In the case of SS-OCT, a narrow instantaneous linewidth over a broad spectral range is tuned in wavelength as a function of time, and all echo signals from different depths are measured simultaneously. Thus, system sensitivity and imaging speeds can be dramatically improved [11, 19].

Fig. 10.2
figure 2

Comparison between TD-OCT and FD-OCT. (a) Time domain optical coherence tomography (TD-OCT) system configuration consisting of a low-coherent light source, an interferometer, and a photodetector. To generate axial profiles, the reference arm is scanned as a function of time. (b) Frequency domain OCT (FD-OCT) system configuration. A wavelength-swept source is utilized and the reference arm is stationary. Interferometric patterns are measured as a function of wavelength and time, rather than as a function of time alone. The delay of echo signals from different positions in tissue results in different frequency modulations which are measured by Fourier transform

Full size image

2 Intravascular OCT System

Intravascular OCT (IV-OCT) is a catheter-based imaging modality using a fiber-optic probe (Fig. 10.3). In general, in vivo intracoronary imaging is challenging because a suitable OCT catheter and contrast agent flushing protocols to remove blood need to be developed. IV-OCT additionally requires a catheter system that is composed of a rotary junction and a catheter as shown in Fig. 10.4 [20, 21]. A rotary junction plays the important role of pulling back and rotating a catheter as well as optically connecting an OCT system console with a catheter. A catheter consists of an optical probe and a protective sheath. An optical probe is composed of a single mode fiber, a small lens, and a hollow torque cable. The torque cable has multiple threads and layers to accurately transmit the proximal end rotation to the distal tip of a fiber probe in curved environments. As a rotating optical fiber is prone to fragility, it is inserted into a torque cable and fixed. To focus light on the vessel, an angle-polished ball lens or a gradient index (GRIN) lens is utilized. An inserted catheter is rotated to create two-dimensional cross-sectional images of coronary artery while it is pulled back to generate multiple frames (Fig. 10.5). IV-OCT creates original rectangular OCT images that are converted from polar to Cartesian coordinates for display as shown in Fig. 10.6. To examine the coronary artery by IV-OCT, blood in the artery must be removed to avoid massive optical scattering and attenuation by red blood cells. The first catheter-based imaging of a human artery ex vivo was conducted by Tearney et al. [22, 23]. This study reported that OCT images were capable of differentiating the intima, media, and adventitia of the artery. The first in vivo IV-OCT imaging in human patients was conducted by Jang et al. who demonstrated a comparison of OCT with IVUS images of tissue prolapse in a stent [24].

Fig. 10.3
figure 3

Configuration of intravascular OCT. Intravascular OCT (IV-OCT) additionally requires a catheter system, which includes a rotary junction and a catheter. The rotary junction connects an IV-OCT platform to a catheter

Full size image
Fig. 10.4
figure 4

Rotary junction and catheter structure . To create cross-sectional OCT images, a catheter needs to be rotated and pulled back by a rotary junction. The rotary junction also couples OCT light from the IV-OCT system to the fiber-optic probe in the protective sheath. To focus light beam, a GRIN lens or a polished ball lens is utilized. To provide the stable torque transmission from the proximal end to the distal tip of a fiber probe, the fiber probe is inserted into a hollow torque cable and fixed

Full size image
Fig. 10.5
figure 5

Image data acquisition by helical scanning. Image data in IV-OCT are acquired by helical scanning of a catheter. This helical scanning consists of transverse scanning and pullback motion

Full size image
Fig. 10.6
figure 6

Converting image in polar to Cartesian coordinate . IV-OCT images are generated by axial, lateral scannings and pullback motion. The original polar OCT images are transformed into Cartesian coordinates. (a) Rectangular OCT image in the polar domain naturally generated by IV-OCT. (b) Cross-sectional OCT image converted to Cartesian domain for visualization of coronary artery

Full size image

The first commercial IV-OCT product was the M2 OCT system (LightLab Imaging, Inc., Westford, MA, USA, now part of St. Jude Medical, Inc.) with regulatory approval in Europe and Japan in 2004, and the M3 system was launched 3 years later in Japan. Since both M2 and M3 systems were based on TD-OCT technology, the imaging speeds were limited (frame rate, 15.6/s for the M2 vs. 20/s for the M3, and pullback speed, 3.0 mm/s vs. 2.0 mm/s) [25, 26]. The balloon occlusion with lactated Ringer’s solution or normal flushing was performed from proximal location to the lesion during the OCT imaging. Dramatic advancements in the wavelength-swept source in FD-OCT enabled much faster frame rates (100 frames/s) and pullback speeds (5–20 mm/s), resulting in non-occlusive OCT imaging with flushing of viscous contrast [4, 27]. The commercial version of FD-OCT, a Dragonfly imaging catheter and C7-XR OCT system (St. Jude Medical/LightLab Imaging, Westford, MA, USA), was launched with the world’s first regulatory approval in 2010 [26]. The feasibility and safety of the Terumo OFDI system allowing 160 frames/s were evaluated in human patients in 2011 [28], and the first product of a LUNAWAVE OFDI system (Terumo Corporation, Tokyo, Japan) was then launched in more than 30 countries in Europe in 2013. The main specifications of TD-OCT and FD-OCT systems are summarized in Table 10.1 [4, 26, 28].

Table 10.1 Comparison of TD-OCT and FD-OCT
Full size table

3 Image Quality of OCT

Since image quality is determined by imaging modality resolution, developing a high-resolution optical imaging system has been one of main research topics, involving high-speed and penetration depth imaging modalities. The image resolution of OCT is divided into axial resolution and lateral or transverse resolution. The axial resolution is the smallest distance between two objects that can be resolved along the axis of the incident beam. It is independent of the lens design and proportionally dependent on the center wavelength of light source and inversely proportional to the source bandwidth [29]. The lateral resolution in OCT imaging is the minimum resolvable distance between two objects which lie perpendicular to the OCT beam at the same depth position. It is mainly dependent on the focusing lens in the imaging core [30]. Another important parameter to determine the image quality is the depth of focus (DOF). DOF is twice the Rayleigh range, defined as the axial distance from the position of the minimum spot size (d) to the position of \( \sqrt{2}d \). A trade-off exists between lateral resolution and DOF. Thus, increasing the lateral resolution to acquire better image quality results in decreasing the DOF. Since the relative position of a catheter from the vessel wall significantly varies during IV-OCT imaging, great variations of the lateral resolution induce distortion of images. The axial resolution for IV-OCT is typically ~10 μm at a center wavelength (~1300 nm) in the light source, and the lateral resolution is between 20 and 40 μm and the DOF is ~1.3 mm [31, 32].

3.1 Image Distortion in IV-OCT

Image artifacts in IV-OCT are mainly caused by catheter motion and cardiac dynamics. To create OCT cross-sectional images of vessels, a catheter is inserted into the coronary artery and rotated with an automatic pullback. Here, the torque applied at the proximal end of the catheter is not evenly transmitted to the distal imaging core, and thus the catheter is nonuniformly rotated since a catheter is placed through tortuous vessels or a crimped imaging sheath or a tight hemostatic value. This distortion is referred to as nonuniform rotational distortion, which also occurs in IVUS imaging [32]. Cardiac dynamics causes image distortion. During the cardiac cycle, heart motion directly affects the catheter motion in both the radial and longitudinal direction as shown in Fig. 10.7 [33, 34]. As a result, an axial discontinuity in the cross-sectional image of coronary artery may appear at the transition region between the first and the last A-scan, and the repeated images of cross-sectional coronary artery can be generated by the longitudinal motion of a catheter that is pulled back and forth. In addition, there is also image distortion by saturation due to high optical backscattering, shadowing effect caused by blood inside of a catheter or stent struts, and suboptimal flushing [32].

Fig. 10.7
figure 7

Representative motion artifacts caused by cardiac motion. A catheter is significantly fluctuated in the coronary artery due to cardiac dynamics. Cardiac motion mainly generates the probe motion in the radial and longitudinal directions. During the cardiac cycle, the probe oscillates in both directions. (a) Cross-sectional view of coronary artery, (b) longitudinal cutaway view of the stented vessel caused by the radial motion artifact and (c) unfolded longitudinal view of the stented vessel due to the longitudinal motion artifact

Full size image

4 Advances in IV-OCT Technology

Imaging speed, resolution, and penetration depth are most importantly considered when the performance of imaging modalities is evaluated. High-speed imaging may provide motion-artifact-free images and reduce the contrast flush volume. Recently, an ECG-triggered high-speed OCT system demonstrated a cardiac motion-free imaging at a rate of 500 frames/s and a pullback speed of 100 mm/s, and the in vivo imaging experiment was conducted in a beating swine heart [35]. It was also reported that micro-OCT (μOCT ) with enhanced lateral resolution of 1 μm proved an ability to observe cells, extracellular components, and endothelial coverage in situ although a μOCT catheter needs to be developed [36]. However, it may be difficult to improve penetration depth in tissue without increasing the center wavelength of light source, which decreases the axial resolution.

There has been a great deal of interest in identifying high-risk plaques by characterizing tissue components. Polarization-sensitive OCT (PS-OCT ) is a functional extension and provides the property of tissue birefringence which may be associated with collagen and smooth muscle cell content [37, 38]. Recently, an in vivo human pilot study demonstrated the feasibility and robustness of intravascular PS-OCT by achieving improved tissue characterization such as plaque rupture [39, 40]. For other light-based approaches to detecting high-risk plaques, multimodality OCT combined with near-infrared fluorescence (NIRF) or near-infrared autofluorescence (NIRAF) has been demonstrated. Multimodality IV-OCT and NIRF imaging system accurately identified lipid-rich inflamed plaques using a FDA-approved indocyanine green (ICG) in rabbit models [41]. A first-in-human IV-OCT and NIRAF study was conducted with a 2.6-F coronary catheter. This study showed that an elevated NIRAF signal was focally associated with a high-risk morphological phenotype as determined by IV-OCT [42]. It was recently noted that the most advanced IVUS-OCT system at a rate of 72 frames/s was successfully demonstrated in a rabbit artery in vivo. The accurate registration between IVUS and OCT data sets showed great potential to accelerate the clinical adoption for accurate identification of vulnerable plaques in humans [43].

5 Summary

IV-OCT has a light-based imaging modality using an ultrathin catheter. Development of the high-speed imaging technology based on FD-OCT has enabled real-time non-occlusive coronary artery imaging. As IV-OCT has a great potential for further understanding and treatment for atherosclerotic coronary artery disease, new technologies will be constantly developed and revolutionized by taking multidisciplinary approaches.