Optical Coherence Tomography in Retinal Imaging






Definition


Optical coherence tomography (OCT) is a noninvasive imaging technique based on the principle of optical reflectometry light, which enables precise anatomical examination of ocular structures.




Key Features





  • High-resolution evaluation of tissue pathology at the cellular level, achieving axial resolution of up to 2–3 µm in tissue.



  • Direct correspondence to the histological appearance of the retina, cornea, and optic nerve in health and disease.



  • Critical tool in the diagnosis and monitoring of ocular disease involving the retina, choroid, optic nerve, and anterior segment.





Associated Features





  • Easy to use, noninvasive, reproducible, safe.



  • Obtainable through most media opacities, including vitreous hemorrhage, cataract, and silicone oil.



  • Recent advances allow for a dramatic improvement in the cross-sectional image resolution with improved acquisition speed.



  • Helpful in the interpretation of pathologies in all layers of the retina as well as the vitreous–retinal interface.



  • Also used for the detection and monitoring of optic nerve, glaucoma, and anterior chamber pathology.





Introduction


Optical coherence tomography (OCT) is a noninvasive imaging technique that allows for the examination of ocular structures. This technique utilizes light waves to create the image in a manner similar to ultrasonography, except that reflected light, rather than sound, is used to create the image. Low-coherence light is scanned across the tissue and focused with an internal lens on the ocular structure of interest. A second beam internal to the OCT unit is used as a reference, and a signal is formed by measuring the alteration of the reference beam and comparing this with the reflected beam. The interface between different ocular tissues can be determined by changes in reflective properties between the tissues. Detection of these beams is based on time-domain or spectral-domain protocols.


The use of light allows for high resolution and evaluation of tissue pathology at the cellular level, achieving resolution of 2–3 µm. Other advantages include ease of use, reproducibility, noninvasiveness, safety, and repeatability. In addition, OCT can image through most media opacities, including vitreous hemorrhage, cataract, and silicone oil.




OCT Technology Platforms


Time-Domain OCT


In time-domain (TD)-OCT, an individual A-scan is acquired by varying the length of the reference arm in an interferometer such that the scanned length of the reference arm corresponds to the A-scan length. The image is then constructed by using a false color scale that represents the quantified amount of backscattered light, with brighter colors representing high reflectivity and darker colors representing little or no reflectivity. The main limitations in the clinical use of TD-OCT are the limited resolution and slow acquisition.


Spectral-Domain OCT


In spectral-domain (SD) or Fourier-domain OCT, the light composing the interference spectrum of echo time delays is measured simultaneously by a spectrometer and a high-speed, charge-coupled device, which allows information of the full-depth scan to be acquired within a single exposure. The interference spectrum is made up of oscillations with frequencies that are proportional to the echo time delay. By calculating the Fourier transform, the machine calculates the axial scan measurements without adjusting the reference mirror. This results in improved sensitivity and image acquisition speed compared with TD-OCT. As a result, SD-OCT is several orders of magnitude more sensitive than TD-OCT. SD-OCT’s higher acquisition speeds allow for a shift from two-dimensional to three-dimensional images of ocular anatomy.


Multifunctional OCT


Functional extensions to OCT add to the clinical potential of this technology. Polarization-sensitive OCT (PS-OCT) provides intrinsic, tissue-specific contrast of birefringent (e.g., retinal nerve fiber layer [RNFL]) and depolarizing (e.g., retinal pigment epithelium [RPE]) tissue with the use of circular or otherwise polarized light. This allows PS-OCT to be useful in glaucoma diagnosis and for the diagnosis of RPE disturbances associated with some diseases, such as age-related macular degeneration (AMD).


Doppler tomography enables depth-resolved imaging of flow by observing differences in phase between successive depth scans. This technology provides valuable information about blood flow patterns in the retina and choroid, allowing absolute quantification of flow within retinal vessels. Ultimately, this adjunct of OCT could potentially reduce the need for fluorescein angiography.


Time-Encoded Frequency-Domain OCT (Swept-Source OCT)


Swept-source OCT, a variation on Fourier-domain OCT, sweeps the frequency of a narrow band continuous wave light source and collects the time-dependent interference signal. Here, the advantage lies in high signal to noise ratio detection technology, achieving very small instantaneous bandwidths at high frequencies (20–200 kHz). This dramatically increases acquisition speed and scan depth. Drawbacks include nonlinearities in the wavelength, especially at high scanning frequencies, and high sensitivity to movements of the scanning target.


High-Speed, Ultra-High-Resolution OCT


Another variation on Fourier-domain OCT, high-speed, ultra-high-resolution OCT (hsUHR-OCT) allows for a dramatic improvement in cross-sectional image resolution and acquisition speed. The axial resolution of hsUHR-OCT is approximately 3.5 µm, compared with the 10 µm resolution in standard OCT. Imaging speeds are approximately 75 times faster than that with standard SD-OCT. The ultra-high resolution enables superior visualization of retinal morphology in a number of retinal abnormalities. hsUHR-OCT further improves visualization by acquiring high-transverse-pixel density, high-definition images.


Adaptive Optics OCT


The resolution of OCT systems in the axial dimension is set by the coherence properties of the light source. Current light sources can provide axial resolution below 3 µm, which is more than sufficient to resolve the axial dimensions of most retinal cells. However, the lateral resolution is substantially degraded from the diffraction limit by optical aberrations present in the eye. Consequently, most ophthalmic OCT systems are designed to be operated with a lateral resolution in the range of 15–20 µm. Adaptive optics OCT (AO-OCT) measures aberrations by using a wavefront sensor and uses a wavefront corrector to compensate for the measured aberrations. The ability to correct for diffraction from ocular imperfections allows for very high resolution (2–3 µm), sufficient for resolution of individual cells ( Table 6.7.1 ).



TABLE 6.7.1

Commercially Available OCT Systems

















































System (Company) Axial Resolution (µm) A-Scans per Second Advanced Features
Cirrus HD-OCT5000 (Carl Zeiss Meditec) 5 68 000 Fixation-independent scan adjustment; multilayer en face C-scan visualization; high-resolution anterior segment imaging; drusen and GA mapping; OCT angiography (AngioPlex)
Spectralis HRAOCT (Heidelberg Engineering) 7 (digital 3.5) 42 000 Point-to-point registration with eye tracking; up to six diagnostic methods in one platform with pinpoint registration between imaging devices; wide-field imaging module for OCT and FA/ICGA; OCT angiography
Avanti RTVue XR (Optovue) 5 70 000 14-mm wide-field macular scans with wide-field analysis; drusen and GA mapping; OCT angiography (Angiovue); software for quantitative analysis of OCTA images
3D-OCT 2000 (Topcon) Capable of exportation to common multimedia devices; able to import time-domain Stratus OCT images
OCT-HS100 (Canon) 3 70 000 10-mm scan length; 10-layer segmentation analysis; multilingual interface; Doppler retinal blood flow analysis capable
SDOCT (Bioptigen) 4 20 000 Handheld head for pediatric patients or animal research; portability facilitates use in an operating room; Doppler retinal blood flow analysis capable
RS-3000 Advance (Nidek) 3 53 000 12-mm scan length option; segmentation analysis of six distinct retinal layers; OCT angiography capable (Angioscan)
DRI-OCT-Triton (Topcon) 5 100,000 Swept Source device; deeper penetration and 12-mm scan length; OCT angiography capable

FA, Fluorescein angiography; GA, geographic atrophy; ICGA, indocyanine green angiography; OCT, optical coherence tomography; SOCT, spectral OCT; SDOCT, spectral-domain OCT.




Time-Domain OCT


In time-domain (TD)-OCT, an individual A-scan is acquired by varying the length of the reference arm in an interferometer such that the scanned length of the reference arm corresponds to the A-scan length. The image is then constructed by using a false color scale that represents the quantified amount of backscattered light, with brighter colors representing high reflectivity and darker colors representing little or no reflectivity. The main limitations in the clinical use of TD-OCT are the limited resolution and slow acquisition.




Spectral-Domain OCT


In spectral-domain (SD) or Fourier-domain OCT, the light composing the interference spectrum of echo time delays is measured simultaneously by a spectrometer and a high-speed, charge-coupled device, which allows information of the full-depth scan to be acquired within a single exposure. The interference spectrum is made up of oscillations with frequencies that are proportional to the echo time delay. By calculating the Fourier transform, the machine calculates the axial scan measurements without adjusting the reference mirror. This results in improved sensitivity and image acquisition speed compared with TD-OCT. As a result, SD-OCT is several orders of magnitude more sensitive than TD-OCT. SD-OCT’s higher acquisition speeds allow for a shift from two-dimensional to three-dimensional images of ocular anatomy.




Multifunctional OCT


Functional extensions to OCT add to the clinical potential of this technology. Polarization-sensitive OCT (PS-OCT) provides intrinsic, tissue-specific contrast of birefringent (e.g., retinal nerve fiber layer [RNFL]) and depolarizing (e.g., retinal pigment epithelium [RPE]) tissue with the use of circular or otherwise polarized light. This allows PS-OCT to be useful in glaucoma diagnosis and for the diagnosis of RPE disturbances associated with some diseases, such as age-related macular degeneration (AMD).


Doppler tomography enables depth-resolved imaging of flow by observing differences in phase between successive depth scans. This technology provides valuable information about blood flow patterns in the retina and choroid, allowing absolute quantification of flow within retinal vessels. Ultimately, this adjunct of OCT could potentially reduce the need for fluorescein angiography.




Time-Encoded Frequency-Domain OCT (Swept-Source OCT)


Swept-source OCT, a variation on Fourier-domain OCT, sweeps the frequency of a narrow band continuous wave light source and collects the time-dependent interference signal. Here, the advantage lies in high signal to noise ratio detection technology, achieving very small instantaneous bandwidths at high frequencies (20–200 kHz). This dramatically increases acquisition speed and scan depth. Drawbacks include nonlinearities in the wavelength, especially at high scanning frequencies, and high sensitivity to movements of the scanning target.




High-Speed, Ultra-High-Resolution OCT


Another variation on Fourier-domain OCT, high-speed, ultra-high-resolution OCT (hsUHR-OCT) allows for a dramatic improvement in cross-sectional image resolution and acquisition speed. The axial resolution of hsUHR-OCT is approximately 3.5 µm, compared with the 10 µm resolution in standard OCT. Imaging speeds are approximately 75 times faster than that with standard SD-OCT. The ultra-high resolution enables superior visualization of retinal morphology in a number of retinal abnormalities. hsUHR-OCT further improves visualization by acquiring high-transverse-pixel density, high-definition images.




Adaptive Optics OCT


The resolution of OCT systems in the axial dimension is set by the coherence properties of the light source. Current light sources can provide axial resolution below 3 µm, which is more than sufficient to resolve the axial dimensions of most retinal cells. However, the lateral resolution is substantially degraded from the diffraction limit by optical aberrations present in the eye. Consequently, most ophthalmic OCT systems are designed to be operated with a lateral resolution in the range of 15–20 µm. Adaptive optics OCT (AO-OCT) measures aberrations by using a wavefront sensor and uses a wavefront corrector to compensate for the measured aberrations. The ability to correct for diffraction from ocular imperfections allows for very high resolution (2–3 µm), sufficient for resolution of individual cells ( Table 6.7.1 ).



TABLE 6.7.1

Commercially Available OCT Systems

















































System (Company) Axial Resolution (µm) A-Scans per Second Advanced Features
Cirrus HD-OCT5000 (Carl Zeiss Meditec) 5 68 000 Fixation-independent scan adjustment; multilayer en face C-scan visualization; high-resolution anterior segment imaging; drusen and GA mapping; OCT angiography (AngioPlex)
Spectralis HRAOCT (Heidelberg Engineering) 7 (digital 3.5) 42 000 Point-to-point registration with eye tracking; up to six diagnostic methods in one platform with pinpoint registration between imaging devices; wide-field imaging module for OCT and FA/ICGA; OCT angiography
Avanti RTVue XR (Optovue) 5 70 000 14-mm wide-field macular scans with wide-field analysis; drusen and GA mapping; OCT angiography (Angiovue); software for quantitative analysis of OCTA images
3D-OCT 2000 (Topcon) Capable of exportation to common multimedia devices; able to import time-domain Stratus OCT images
OCT-HS100 (Canon) 3 70 000 10-mm scan length; 10-layer segmentation analysis; multilingual interface; Doppler retinal blood flow analysis capable
SDOCT (Bioptigen) 4 20 000 Handheld head for pediatric patients or animal research; portability facilitates use in an operating room; Doppler retinal blood flow analysis capable
RS-3000 Advance (Nidek) 3 53 000 12-mm scan length option; segmentation analysis of six distinct retinal layers; OCT angiography capable (Angioscan)
DRI-OCT-Triton (Topcon) 5 100,000 Swept Source device; deeper penetration and 12-mm scan length; OCT angiography capable

FA, Fluorescein angiography; GA, geographic atrophy; ICGA, indocyanine green angiography; OCT, optical coherence tomography; SOCT, spectral OCT; SDOCT, spectral-domain OCT.




Anatomical Results


OCT images correspond to the histological appearance of the retina. The highly reflective nerve fiber layer (NFL) is represented by a red signal. Similarly, the RPE layer, Bruch’s membrane, and choriocapillaris are represented by a red signal because of their higher reflectivity. A third red line represents the junction of the inner and outer segments (called outer segment ellipsoid zone ). Inner cellular layers have lower reflectivity and are represented by yellow, green, and blue colors. The nonreflective vitreous cavity has a black signal, but the posterior hyaloid face can occasionally be seen as an additional reflective layer anterior to the NFL ( Fig. 6.7.1 ).




Fig. 6.7.1


Optical Coherence Tomography (OCT) Images of a Normal Retina.

(A) Conventional OCT. (B) Spectral-domain (SD) OCT. (C) Average of 12 SD-OCT images (SD-OCT with multiple B-scan averaging). Retinal structures visible include the hyperreflective retinal nerve fiber layer (NFL), hyporeflective inner nuclear layer (INL), hyperreflective outer plexiform layer (OPL), hyporeflective outer nuclear layer (ONL), hyperreflective lines that correspond to the junction of the inner and outer photoreceptor segment layers (IS/OS), retinal pigment epithelium (RPE)/Bruch’s membrane (BM) complex (RPE/BM), external limiting membrane (ELM), inner plexiform layer (IPL), ganglion cell layer (GCL), and choroidal vessels (Ch). Red arrows indicate structures delineated only by single-scan SD-OCT or SD-OCT with multiple B-scan averaging.

(From Sakamoto A, Hangai M, Yoshimura N. Spectral-domain optical coherence tomography with multiple B-scan averaging for enhanced imaging of retinal diseases. Ophthalmology 2008;115(6):1071–1078.)






The choroid is a highly vascular structure with blood flow and thickness varying in relation to the intraocular pressure, perfusion pressure, refractive error, disease state, and age. It is possible to image the choroid with conventional OCT imaging ( Fig. 6.7.2 ).




Fig. 6.7.2


Comparative Choroidal Imaging With Optical Coherence Tomography (OCT) Sections Through the Fovea Obtained With Various OCT Instruments.

(Top) Heidelberg Spectralis spectral-domain (SD)-OCT using enhanced depth imaging to obtain more choroidal details, including better visualization of the hyporeflective line indicating the edge of the choroid. (Bottom) Cirrus SD-OCT image in which choroidal details are visible with oversampling.




Image Optimization


OCT measures the intensity of a backscattered optical signal, which represents the optical properties or reflectivity of the target tissue. The tissue reflectivity varies among different structures, allowing for measurements that can be displayed as false or pseudo-color or gray-scale images. The gray scale runs continuously from high signal (white) to no signal (black), and images can contain up to 256 shades of gray corresponding to the optical reflectivity of the various tissue interfaces. The standard color scale uses a modified continuous rainbow spectrum in which darker colors, such as blue and black, represent regions of minimal or no optical reflectivity and lighter colors, such as red and white, represent a relatively high reflectivity, as described in the Anatomical Results section above.


Studies have shown that compared with the color images, the gray-scale images are easier to interpret and are more informative because of their improved ability to visualize subtle retinal structures, such as photoreceptor inner and outer segment junction (IS/OS), and subtle pathologies, such as thin epiretinal membranes (ERMs).


Another method to improve image quality is to average multiple OCT scans. Frames with the least amount of motion artifacts are chosen. These frames are then averaged. Each pixel value is calculated as an average intensity from all frames, to create one frame. On average, 50 frames are used to create one image.




OCT Image Interpretation


Preretinal


The use of OCT has facilitated the diagnosis and description of diseases involving the vitreoretinal interface, including vitreomacular traction syndrome, ERMs, macular holes, and schisis.


Posterior Vitreous Detachment


The vitreous in a healthy eye is optically clear. When the vitreous is completely attached, the vitreoretinal interfaces can be detected by the marked change in reflectivity between the vitreous and the internal limiting membrane ( Fig. 6.7.3 ).




Fig. 6.7.3


Stratus time-domain (TD)-OCT (A) and prototype, ultra-high-resolution optical coherence tomography (UHR-OCT) (B) scans of a normal subject. I, Macular linear scans, whose orientations are depicted in the upper image. II, Peripapillary circular scans. III, Optic nerve head (ONH) linear scans. Inf, inferior; INL, internal nuclear layer; IS/OS, inner segment/outer segment; NFL, nerve fiber layer; ONL, outer nuclear layer; OPL, outer plexiform layer; RPE, retinal pigment epithelium; Sup, superior; Temp, temporal.

(From Wollstein G, Paunescu LA, Ko TH, et al. Ultrahigh-resolution optical coherence tomography in glaucoma. Ophthalmology 2005;112(2):229–37.)


Vitreomacular Traction


Vitreomacular traction (VMT) is a complication of anomalous partial posterior vitreous detachment (PVD), where the vitreous is separated from the retina throughout the peripheral fundus but remains adherent in a broad region encompassing the macula and/or the optic nerve. A subtle variant demonstrates a localized perifoveal vitreous detachment with a small, focal vitreofoveolar adhesion resulting in an anterior–posterior tractional force that may lead to retinal distortion, cystoid macular edema (CME), or even a macular hole. Several OCT studies have documented that surgical or pharmacological separation of the vitreofoveal adhesion promotes the resolution of macular thickening, usually with improvement in visual acuity in patients with vision loss caused by vitreomacular traction ( Fig. 6.7.4 ).




Fig. 6.7.4


Vitreomacular Traction (VMT).

Vitreofoveolar traction resulting in an anterior–posterior tractional force leading to cystic macular edema.


Epiretinal Membrane


An ERM is a result of proliferation of abnormal tissue on the surface of the retina. It is semitranslucent and proliferates on the surface of the internal limiting membrane. ERM has been found to consist of glial cells, RPE cells, macrophages, fibrocytes, and collagen fibers.


On OCT, the ERM appears as a highly reflective thick membrane on the surface of the retina. The strength of the reflection can differentiate it from the posterior hyaloid, which appears as a minimally reflective signal ( Fig. 6.7.5 ).




Fig. 6.7.5


(A) Fundus photograph showing an epiretinal membrane (ERM) on the macula. (B) Reconstructed fundus image corresponding to black box on fundus photograph. (C) Spectral-domain optical coherence tomography image showing the ERM.

(From Legarreta JE, Gregori G, Knighton RW, Punjabi OS, Lalwani GA, Puliafito CA. Three-dimensional spectral-domain optical coherence tomography images of the retina in the presence of epiretinal membranes. Am J Ophthalmol 2008;145(6):1023–30.)






Macular Holes


OCT has become the gold standard in diagnosing and monitoring macular holes. OCT technology has been instrumental in the classification of macular hole development, following the sequence of events from anteroposterior vitreofoveal traction to full-thickness macular hole (FTMH) ( Fig. 6.7.6 ).


Oct 3, 2019 | Posted by in OPHTHALMOLOGY | Comments Off on Optical Coherence Tomography in Retinal Imaging

Full access? Get Clinical Tree

Get Clinical Tree app for offline access