Bench-to-bedside success: intravascular optical coherence tomography
If one were to ask a physician to describe an ideal medical imaging technology, the resulting wish list might include high resolution, high sensitivity, and rapid acquisition speeds, all delivered at low cost with minimal risk to the patient in a platform that is simple and easy to use. Neglect any one of these requirements, and the technology in question might at best be relegated to a niche research role. Since its invention in 1991 by a research team led by James G. Fujimoto at the Massachusetts Institute of Technology (MIT), optical coherence tomography (OCT) has promised to meet the imaging requirements of a diverse range of medical specialties from gastroenterology to neurosurgery.1 OCT is an interferometric optical imaging modality, somewhat analogous to ultrasound, that provides cross-sectional and 3D images of biological tissue with extremely high resolutions reaching down to the micron scale. After finding initial widespread commercial success in ophthalmology, however, researchers and companies have struggled to develop the next viable market for the technology.
Interventional cardiology presents a unique set of opportunities and challenges for in vivo imaging. Drug-coated intravascular stents can be used to treat many forms of coronary artery disease without the need for painful and expensive surgery, but cardiologists are typically limited to low-resolution 2D fluoroscopy to plan, guide, and evaluate the procedure. This reliance on legacy x-ray imaging can lead to suboptimal therapeutic delivery in addition to exposing the patient to ionizing radiation. In contrast, OCT can be performed with a flexible fiber-optic catheter delivered through a minimally invasive percutaneous puncture and provides image data that is unobtainable with fluoroscopy or ultrasound. Consequently, it was recognized very early on by the Fujimoto group as a potential paradigm change for intravascular imaging.2 However, two formidable obstacles would block the road to clinical adoption: the need to temporarily displace blood from the arteries to prevent scattering of the imaging laser light by red blood cells, and the need for ultra-miniaturized delivery catheters to allow access to narrow coronary arteries.
Coherent Diagnostics Technology, later to become LightLab Imaging Inc., was spun out of MIT in 1998 to commercialize the nonophthalmic applications of OCT. Intravascular imaging soon became the focus of the company's efforts due to the pressing clinical need and large market opportunity. LightLab rapidly developed a lens system consisting entirely of fusion-spliced optical fiber segments, allowing construction of an intravascular catheter with a diameter less than 0.5mm. This breakthrough solved the fundamental problem of accessing tight stenoses in the coronary arteries, but the issue of blood clearance remained.
Providing the physician with sufficient data to guide clinical decision making requires analyzing a segment of an artery several centimeters long. A spiral scan protocol was developed for intravascular OCT catheters, where the fiber-optic imaging core rotates and simultaneously pulls back inside a stationary outer sheath to prevent damage to the vascular endothelium. Unfortunately, since OCT systems were at that time constrained to acquisition speeds of only a few thousand image lines per second, this ‘pullback’ method required 30–60 seconds to complete. Blood displacement was therefore only possible by totally occluding the vessel with a balloon, leading to high procedural complexity. Nevertheless, with regulatory approval in Europe and Japan in 2004, LightLab launched its first commercial product, the M2 system and ImageWire catheter. The M2 was followed in 2007 by the M3, which provided modestly higher imaging speeds but was built around the same core platform. These early products saw significant use by clinical researchers despite the challenges associated with the balloon occlusion procedure, underscoring the tremendous value of OCT image data for interventional cardiologists. By the end of their production runs, M2 and M3 systems had been installed in cardiac catheterization labs throughout the world, with adoption particularly high in Japan.
As LightLab was making the first forays into commercial intravascular OCT, multiple academic research labs were developing a new generation of OCT technology that would transform the field. The advent of frequency domain OCT (FD-OCT), which is based on spectral interferometry rather than time-domain interferometry, led to dramatic improvements in imaging speeds and sensitivities. LightLab also adopted this new technology and in 2009, just two years after the release of the M3 system, obtained regulatory approval for the world's first commercial intravascular FD-OCT platform, the C7XR system and DragonFly catheter. With an imaging speed more than 10 times higher than its predecessor, the C7XR could scan a 5.4cm vessel segment in less than three seconds. Reaching this threshold allowed blood clearance during imaging to be obtained with a simple intracoronary injection of conventional x-ray contrast fluid rather than a cumbersome occlusion balloon, dramatically reducing procedural complexity.
As a graduate student in the Laser Medicine and Medical Imaging Group at the Research Laboratory of Electronics at MIT, my early work on OCT focused on designing ultra-high-speed imaging systems and optical catheters.3 Later, our group worked with collaborators at the VA Boston Healthcare System to develop clinical applications in gastroenterology.4 After graduating from MIT in 2009, I joined LightLab Imaging and began work on next-generation FD-OCT systems to support the company's cardiovascular product line.
In addition to imaging speed, the LightLab team also improved the image quality and ranging depth on the new FD-OCT platform, while advances in rendering and visualization software made 3D display of OCT data sets possible. Figure 1 shows an example of an OCT data set acquired from a human coronary artery in vivo. The top portion of the screen displays a cross-sectional view at a single longitudinal position within the vessel, while the bottom portion displays a longitudinal cut through the length of the vessel. At this particular location, a large calcified plaque is visible at a branch in the artery as a hypointense inclusion at three o'clock in the cross-sectional view and near the 43mm mark in the longitudinal view. Figure 2 shows an example of a 3D rendering of the vessel created in real time from the same OCT data set. 3D visualization allows an intuitive appreciation of an enlargement of the lumen in this area of the data set. Note that this 3D functionality is not yet cleared for human clinical use by the US Food and Drug Administration.
As these technological advances were occurring in the research and development laboratory, clinicians were beginning to focus more on appropriate selection, optimal delivery, and careful post-interventional assessment of coronary therapies. The capabilities of FD-OCT aligned perfectly with their needs, and researchers and mainline clinicians soon adopted the technology.5, 6 LightLab Imaging was acquired by St. Jude Medical in 2010, which further increased the breadth of market adoption and accelerated the pace of technological progress. To date, LightLab/St. Jude Medical systems have been used to perform tens of thousands of coronary imaging procedures, making OCT a rare bench-to-bedside success in the biomedical optics community.
In the near-term future, aggressive research and development efforts aim to place even better technology in the hands of interventional cardiologists. Multi-modality platforms, such as the ILUMIEN combined OCT and fractional flow reserve system, released by St. Jude Medical in 2011, are intended to provide truly comprehensive lesion assessment from a morphological and physiological perspective. Finally, we are working on improved software tools to increase automation of image analysis, leaving the physician free to focus on clinical decision making rather than data interpretation.
C7XR, ILUMIEN, ImageWire, and DragonFly are trademarks of St. Jude Medical and its related companies.
Desmond Adler is a principal research and development engineer in the Advanced Development group at St. Jude Medical. His areas of expertise are OCT, intravascular imaging, and clinical application development. He received a PhD from MIT and a BS from the University of Alberta.