Optical Coherence Tomography: A Primer


Device (manufacturer)

Axial resolution

Scanning rate

Special characteristics

Scans available

3D-OCT 2000 (Topcon, Tokyo, Japan)

5 μm

27 kHz

Fundus camera

Fast map, box scan

5- and 9-line raster, line scan

12-line radial scans

Bioptigen SD-OCT (Bioptigen, Research Triangle

Park, NC, USA)

4 μm

20 kHz

Designed for research

Rectangular volume, mixed volume

Linear scan

Radial volume

Cirrus HD-OCT (Carl Zeiss Meditec, Dublin,

CA, USA)

5 μm

27 kHz
 
Macular cube

5-line raster scan, 1-line raster scan

RTVue-100 (Optovue, Fremont, CA, USA)

5 μm

27 kHz
 
3D macular scan, MM5

Line scan, HD scan, cross line, HD cross line

Radial slicer, MM6

MM5 mesh scan

Spectral OCT SLO (Opko, Miami, FL, USA)

6 μm

27 kHz

Microperimetry

3D macular scan

Line scan

Spectralis OCT (Heidelberg Engineering, Heidelberg, Germany)

8 μm

40 kHz

Eye-tracking fluorescein

ICG angiography, autofluorescence

Volume scans

7-line raster scan






1.3 Scanning Protocol


The three scanning protocols are:


  1. (i)


    Macular cube scan

     

  2. (ii)


    Line scan

     

  3. (iii)


    C-scan

     


1.3.1 Macular Cube Scans


They are volume scans similar to computed tomography or magnetic resonance scans that acquire volumetric cubes of data. Rapid series B-scans are acquired, generally in a 6 mm × 6 mm square area centered on the fovea. Manual centering of the cube scans other areas of interest besides the fovea. Similarly, optic nerve topographic scans are cube scans centered on the optic nerve. The Cirrus HD-OCT (Carl Zeiss Meditec Inc., CA, USA) provides two macular cube scan protocols. The 512 × 128 protocol analyzes 128 horizontal scans at high resolution (512 A-scans per B-scan). The 200 × 200 protocol achieves faster imaging at the cost of resolution. It takes 1.6 s to acquire 200 horizontal scans (200 A-scans per B-scan). As the average foveal thickness is 160–180 microns, the resolution of A-scan is less than 4 microns theoretically. In Heidelberg Spectralis volume scan uses a scanning protocol with fast 25 B-scans, each of which consists of 512 A-scans or a dense 1024 × 49 default scanning protocol. The Topcon 3D OCT has a 512 × 128 and a 256 × 256 scanning protocol. In RTVue 3D macular scans have a 4 mm × 4 mm macular cube scan with 101 B-scans each consisting of 512 A-scans and a MM5 protocol. Raster and radial scans are also available to scan the macula.


1.3.2 Line Scans, Cross Lines, and Raster Scans


SD-OCT line scans are a single B-scan composed of a higher number of A-scans than the macular cube scans enabling higher resolution scans of the retinal tissue. In Cirrus 5-line raster, five horizontal 6 mm lines are scanned four times and averaged. Oversampling increases the signal-noise ratio, thereby increasing the resolution of the image. The five lines in the raster can be integrated to obtain a single line scan consisting of 20 averaged B-scans. The RTVue uses cross line scans consisting of horizontal and vertical lines scans. In Heidelberg 7-line raster scans are used.


1.3.3 C-Scans


The C-scans are also known as en face image or OCT fundus image. The literal meaning of the term en face is “on the face.” En face image is obtained by summating data from all B-scans and has the same appearance as a red-free image of fundus. These provide with a global overview of the morphology of the retinal surface.


1.4 Three-Dimensional Spectral Domain Optical Coherence Tomography


The high scanning speed of the SD-OCT allows multiple B-scans to be obtained and thus acquires a large dataset for the macular cube. The inbuilt software, 3D-volume rendering, in Cirrus HD-OCT (Carl Zeiss Meditec Inc., CA, USA) reconstructs a three-dimensional image of the whole retinal cube. The three planes of 3D SD-OCT are the X, Y, and Z planes. The X plane corresponds to the horizontal B-scan or cross section. The Y plane is the vertically reconstructed B-scan. The Z plane or the coronal plane (C-scan) corresponds to the reconstructed en face image of the retina.


1.5 Topographical Maps


A topographical map of the individual retinal layers, internal limiting membrane (ILM) and retinal pigment epithelium (RPE) are generated by 3D segmentation analysis. The ILM segmentation map is color-coded pale blue; the RPE/choriocapillaris complex is brown or cream. These maps provide an effortless pictorial representation of the morphological alterations as well as localization of the pathology. The topographical map of the retina represents the thickness of the retina, between ILM and RPE, with a cold/warm color scale. The color coding provides a global representation of the difference in thickness of the retina in comparison to a normative database.


1.6 Masking


Masking is another feature of the SD-OCT, useful when visualizing the three-dimensional retinal cube. Masking of the tissue below a desired layer from cutting enables peeling back of the tissues above the desired layer. By applying the internal limiting membrane (ILM) mask, the tissue below the ILM is preserved from being cut, and the ILM and the tissues above this layer can be peeled back. “Niche” allows for masking in two dimensions, the X and Y planes simultaneously. These facilitate imaging of the retinal pathology at various depths.


1.7 Retinal Pigment Epithelium Fit


The RPE fit feature of the Cirrus HD-OCT (Carl Zeiss Meditec Inc., CA, USA) is an important advancement to understand various retinal pathology (Lumbroso et al. 2009). It provides C-scan images adapted to retinal curvature. En face images that follow the curvature of posterior pole can be obtained. RPE fit software identifies RPE/choriocapillaris complex to display it as a curved 3D section plane. The software adapts to concavity of RPE/choriocapillaris (Wei et al. 2012). Sections with thickness varying from 2 to 20 μm can be obtained with thinner sections known as slices and thicker sections known as slabs. Blood vessels of the choroid can be seen by changing the thickness of sections. A slab of selective thickness, the slab thickness of interest, can be studied by selecting and adjusting the anterior and posterior boundaries represented by two separable same-color dashed lines. The slab image represents an average signal intensity value for each A-scan location through the selected depth of the slab.


1.8 Optical Coherence Tomography Interpretation



1.8.1 Qualitative Data Analysis


Qualitative interpretation reviews individual line scans of the imaged retina and assess the pathology on the basis of knowledge of normal anatomy. Comparison with the previous scans allows assessment of course of the underlying disease and response to treatment. There are three things to be taken care of: registration, sampling error, and subjective evaluation. Registration implies that future line scans must be registered to past scans so that precisely same anatomic areas are scanned. Sampling error is important as to rule out any pathology, so multiple line scans through the macula is examined. Line scans in quantitative assessment tends to be individualized as qualitative numbers are lacking.


1.8.2 Quantitative Data Analysis


The inbuilt software in commercially available OCT devices are capable of analyzing the thickness of the total retina, individual layers like retinal nerve fiber layer, and the macular volume. This is facilitated by the software’s ability to distinguish inner and outer boundaries of the retina. Unlike time domain OCT mapping, which measures the total retinal thickness as distance from the photoreceptor inner and outer segments (IS/OS) junction to the ILM, the SD-OCT thickness mapping strictly corresponds to anatomical retinal thickness, that is, the distance from the RPE to the ILM.


1.8.3 Mapping Retinal Thickness


In the context of a thickness map the early treatment of diabetic retinopathy study (ETDRS) grid is superimposed on the fovea. The grid has an innermost circle of 1 mm diameter centered at the fovea, an inner circle of 3 mm, and an outer circle of 6 mm diameter. The outer two circles are divided into superior, nasal, inferior, and temporal quadrants. Average thickness values for each sector are indicated numerically and also by a color scale. A color-coded representation demonstrates the difference in the retinal thickness with age-matched normative database.


1.9 Recent International Nomenclature for Optical Coherence Tomography


Recently, an international panel with expertise in retinal imaging reached a consensus for OCT imaging terminology (Staurenghi et al. 2014). They suggested the terms band, layer, and zone for the layers of the retina. The term band refers to the three-dimensional structure of the retinal layers anatomically. The term zone describes those regions on OCT whose anatomical correlation is not clearly delineated. The RPE/Bruch’s complex is one of the layers ascribed as zone as they are inseparable owing to interdigitation of cellular structure or tissue.


1.10 Interpretation of Normal Retinal Layers


Histologically, the retina consists of ten layers, four of them are cellular and two are neuronal junctions. The axonal layers, nerve fiber layer, and plexiform layers are capable of potent light scatter and hence are hyperreflective. The light scattering potential of nuclear layers is lower. The first layer visible on OCT images is the hyperreflective ILM line at the vitreoretinal interface. The hyperreflective retinal nerve fiber layer lies next to this layer. Outer to this is a hyporeflective ganglion cell layer. Subsequently, hyperreflective plexiform layers are imaged with an inner nuclear layer situated between them. The relatively thick hyporeflective outer nuclear layer is visible next. A thin hyperreflective line underneath the outer nuclear layer corresponds to the location of the external limiting membrane (ELM). The hyperreflective layer of photoreceptor junction of inner and outer segments, currently termed as the photoreceptor inner segment ellipsoid zone, lies beneath this layer. Due to the increased length of outer cone segments in the central fovea, this line is slightly elevated in the foveal region. The outermost layer imaged is the hyperreflective RPE-Bruch’s complex. The choriocapillaris and the choroid are visible outer to the RPE-Bruch’s zone. The fovea is recognized on a cross-sectional image by its characteristic depression, due to the thinning of the retina with the absence of the inner layers at the macula (Fig. 1.1) (Table 1.2).

A324343_1_En_1_Fig1_HTML.jpg


Fig. 1.1
Reflectivity of different layers on optical coherence tomography (Heidelberg Engineering)



Table 1.2
Reflectivity of different layers on optical coherence tomography.



















Layer

OCT finding

Nomenclature

1

Hyperreflective

Posterior cortical vitreous

2

Hyporeflective

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jul 12, 2017 | Posted by in OPHTHALMOLOGY | Comments Off on Optical Coherence Tomography: A Primer

Full access? Get Clinical Tree

Get Clinical Tree app for offline access