2020–2021 BCSC Basic and Clinical Science Course™

Clinical Education /

Book Excerpts /

Basic and Clinical Science Course - Excerpt

2020–2021 BCSC Basic and Clinical Science Course™

Go to Academy Store Learn more and Purchase.

12 Retina and Vitreous

Part I: Fundamentals and Diagnostic Approaches

Chapter 02: Diagnostic Approach to Retinal Disease

Imaging Technologies

Optical Coherence Tomography

Optical coherence tomography (OCT) is a noninvasive, noncontact imaging modality that produces micrometer-resolution images of tissue. Low-coherence light is directed into tissue and into a reference arm. An interferometer combines the light returning from the tissue with the light from the reference arm, producing an interferogram. The benefit of using low-coherence light is that its spectral makeup changes rapidly with time; thus, light produced at a particular instant will not interfere well with light produced at other times. This means the exact position from which the interfering light came can be determined by the resolution dictated by the coherence length of the light source, which is typically 5–7 μm. In any given A-scan, time-domain OCT is used to interrogate each point in the tissue sequentially. In more modern techniques, for example either spectral-domain (SD) or swept-source (SS) OCT, a more efficient approach is taken. In SD-OCT, a broad-spectrum light source is used, and the resulting interferogram produced varies with the reflectivity of the tissue. SS-OCT uses a more complicated light source that sequentially scans through successive wavelengths of light across a spectral range (see Optical Coherence Tomography Technologies: A Comparison of SD-OCT and SS-OCT).

OPTICAL COHERENCE TOMOGRAPHY TECHNOLOGIES: A COMPARISON OF SD-OCT AND SS-OCT

In SD-OCT, a broad-spectrum light source is used, and the resulting interferogram varies with the reflectivity of the tissue, which depends on how far that reflecting surface is from the same distance in the reference arm; this imaging focal point is known as the zero-delay line. Thus, in SD-OCT the interferogram has frequency encoding of depth; a Fourier transform can be used to calculate the amount of reflectivity and its depth in the tissue. However, the ability of SD-OCT to detect the interferogram decreases at higher frequencies; deeper structures can be visualized, but not as effectively. A variation on this technique, enhanced depth imaging (EDI) OCT, places the zero-delay line at the deepest part of the tissue being evaluated. Deeper structures may be better visualized, but at the expense of seeing the more superficial structures. In conventional SD-OCT, the peak sensitivity is placed in the vitreous, making it possible to visualize a structure with very low reflectivity. The choroid is not visualized well with this approach. EDI-OCT is able to image the choroid, but the vitreous is not visualized well. Therefore, the clinician must choose the most appropriate imaging modality according to the situation in order to obtain an optimal image from the desired level.

In swept-source (SS) OCT, a light source sequentially scans through successive wavelengths of light across a spectral range. Over time, each sweep of light builds an interferogram. The light sources currently available are based on microelectromechanical systems, which are small and fast. The disadvantages to using physical methods to obtain a swept laser output include phase jitter and mode hopping. Coming to market in the near future is a new light source, called an akinetic laser, which uses electronic means to sweep wavelengths, thereby avoiding the mechanical method of selecting wavelengths. The akinetic laser reduces phase jitter and is able to select narrow regions of light wavelengths; with these abilities, systems may soon be able to scan the entire depth of the eye, from cornea to choroid, in 1 scan.

SS-OCT systems display little change in sensitivity with increasing depth. Therefore, it is not necessary to choose whether to highlight the vitreous or choroid in an image, because both can be imaged well simultaneously. With improvements in spectrometers, SD-OCT systems might have lower amounts of sensitivity fall-off with increased depth and thus may remain competitive in the marketplace. SS-OCT is generally done at somewhat longer wavelengths because of the availability of light sources. The longer wavelengths help visualize deeper structures, with a slight trade-off in decreased lateral resolution (remember, the smallest spot size is related to wavelength).

Finally, SD-OCT systems use a line-charge coupled device to detect the interferogram after it has passed through a diffraction grating. The detector’s sensitivity and response time are the major determinants of the device’s scan speed. In SS-OCT, the detector is the relatively simple and high-speed photodiode, and the scan speeds can be quite fast. However, because safety standards limit the amount of light used in the eye, higher-speed imaging often comes at the expense of decreased signal-to-noise ratio. Faster scan speeds are good for area coverage and reduced imaging time, but they are a disadvantage in terms of noise.

Both SD-OCT and SS-OCT create an A-scan through tissue. B-scans consist of a collection of many A-scans conducted through a plane of tissue. A volume scan consists of an assembly of numerous B-scans; this volume scan is stored in computer memory as a block of data in which each memory location stores a value that corresponds to a specific small volume of tissue. The voxels (a portmanteau of volume and pixels) in the volume of data may be represented in many ways; 1 simple way is to make planar slices producing an image called a C-scan. C-scans are difficult to interpret because in a curved structure, many planes of tissue can be crossed. Another, more advanced method is to segment the data according to tissue planes; a thickness of voxels presented this way is called an en face scan. The tissue thicknesses can be measured in an en face scan; the retinal nerve fiber layer is commonly measured. Maps of the thickness of the retina or a specific retinal layer can also be produced. Actual correlation between OCT scans and histology of the retina has not yet been thoroughly explored, and the correlation of the identity of structures seen in OCT has changed numerous times. The International Nomenclature of OCT Panel has proposed nomenclature terminology, but these assignments are likely to change over time (Activity 2-1).

ACTIVITY 2-1 Optical coherence tomography (OCT) nomenclature terminology, based on the International Nomenclature for OCT Panel for Normal OCT Terminology.

From Staurenghi G, Sadda S, Chakravarthy U, Spaide RF; International Nomenclature for Optical Coherence Tomography (IN•OCT) Panel. Proposed lexicon for anatomic landmarks in normal posterior segment spectral-domain optical coherence tomography: the IN•OCT consensus. Ophthalmology. 2014;121(8):1572–1578.

Access all Section 12 activities at www.aao.org/bcscactivity_section12.

Some OCT scanners offer eye movement tracking. With the addition of this feature, ocular motion can be detected and corrected in the final image, improving the quality of the resulting scan. Tracking methods rely on recognizing fundus features and registering the scan pattern with the fundus image. This capability expands the utility of OCT; with it, scans interrupted by patient blinks still produce usable images. In addition, it is possible to perform repeated scans of the same fundus location over time, enabling assessment of disease progression (Fig 2-2).

In addition to B-scans and en face imaging, volume rendering shows the 3-dimensional character of the tissue. Compared to ordinary B-scans, volume rendering is computationally intensive. It is used in radiology, but is not yet widely used in ophthalmology (Fig 2-3).

Figure 2-2 Evolution of a macular hole, visualized with optical coherence tomography (OCT). A, OCT image of a patient with a perifoveal posterior vitreous detachment and no obvious traction on the macula. B, After 1 year, the patient experienced visual distortion; the image shows obvious traction with foveal tractional cavitations. C, Image taken 2 months later; note the full-thickness macular hole. D, Image taken 1 month after macular hole surgery; the hole is closed. Note the subtle area of increased reflectivity in the center. E, Image taken 3 months later shows the fovea with a nearly normal contour and laminar structure.

(Courtesy of Richard F. Spaide, MD.)

Figure 2-3 Imaging of a lamellar macular hole. A, Fundus photograph of a patient with prominent drusen and distorted vision. B, A B-scan section of the OCT imaging shows a lamellar hole with an unusually thick epiretinal membrane, which is sometimes seen in association with lamellar macular holes. C and D, Two different views from volume-rendered imaging of the lamellar macular hole taken in sections, showing the thick epiretinal membrane, as well as the absence of induced distortion of the retina. Note the cavities within the undermined retina and the attachment of the proliferation to the central foveal tissue.

(Courtesy of Richard F. Spaide, MD.)

Excerpted from BCSC 2020-2021 series: Section 10 - Glaucoma. For more information and to purchase the entire series, please visit https://www.aao.org/bcsc.

2020–2021 BCSC Basic and Clinical Science Course™

12 Retina and Vitreous

Part I: Fundamentals and Diagnostic Approaches

Chapter 02: Diagnostic Approach to Retinal Disease

Imaging Technologies

Optical Coherence Tomography

OPTICAL COHERENCE TOMOGRAPHY TECHNOLOGIES: A COMPARISON OF SD-OCT AND SS-OCT

OUR SITES

FOLLOW THE ACADEMY