OCT-A: another piece to complete the clinical picture puzzle.
Optometrists [OCULAR COMMUNICATION] Therapeutic optometrists [OPTIONS] Dispensing opticians [OCULAR EXAMINATION]
1 CET POINT
Optical coherence tomography (OCT) first became commercially available in 1991. Since then the technology has undergone numerous hardware and software developments in order to obtain high resolution structural retinal scans within seconds. A natural progression was the development and implementation of anterior OCT and subsequently that of OCT angiography (OCT-A). With these new developments both OCT and OCT-A have been likened to be the 'Swiss army knife of retinal imaging.'
OCT-A allows for fast, non-invasive (dye-free), high contrast (and high resolution) imaging of the retinal vasculature providing an alternative to fluorescein angiography (FA) imaging.
How it works
OCT-A is the 'functional extension' to structural OCT scans. A structural retinal OCT image is generated by interferometrically measured amplitude and delay of the reflected / backscattered light. Each B-scan (cross-sectional scan) image is generated by sequential acquired multiple A-scans as the light beam is scanned in a transverse direction across the area of observation. Subsequently, the volumetric data is generated by sequentially obtained B-scans covering a region of the retina or the anterior eye, depending on the area of interest.
As the retina is stationary (while the patient is viewing a non-moving fixation target), successive B-scans acquired at the same position are similar, except for locations where blood is moving within the tissue. This means that at sites where blood flows, the reflectivity or scattering is altered from one scan to another, producing a change in signal amplitude and/or phase between successive B-scans. OCT-A generates a high-resolution map of the microcirculation and assesses blood flow by analysing these variations in the signal amplitude and / or phase properties between the repeated B-scans.
It is important to appreciate that the way in which scans are acquired, how the differences are determined, and the changes defined to be significant, are not the same across different devices. Different algorithms to compute blood flow measurements from repeated B-scans have been developed and are based on OCT signal phase, amplitude, or both. Those using phase signal-based technology are referred to as Doppler OCT, whereas amplitude signal-based technology includes split-spectrum amplitude decorrelation, speckle variance and correlation mapping. Other more complex signal-based OCT-A technology (those using phase and amplitude data) are referred to as optical microangiography (OMAG) and multiple signal classification OMAG.
Currently, commercially available OCT-A devices include split-spectrum amplitude decorrelation angiography based algorithms as implemented in the AngioVue system (Oprovue, Inc., Fremont, CA, USA), (1) OMAG used in Angioplex (Carl Zeiss Meditech Inc., Dublin, CA, USA), (2) amplitude decorrelation algorithm in swept source OCT (Angio, Topcon Corp., Tokyo, Japan), (3) and the so-called OCT-A ratio analysis algorithm of the Heidelberg Spectralis OCT-A (Heidelberg Engineering, Heidelberg, Germany). (4) Besides using different algorithms, wavelength and scan rate (between 68-100 kHz) varies between devices. Each algorithm, scan rate and wavelength have been chosen to improve certain functions such as reducing noise and motion artefacts, compensate for eye movements, penetrate to deeper layers, and improve scan speed. However, these implementations each have a trade-off in terms of performance, potential errors and/or cost. Munk and colleagues have qualitatively and quantitatively compared the aforementioned commercially available OCT-A devices. (5) They have measured vessel density, vessel bifurcations and graded artefacts and used each parameter to create a ranking of the devices. They concluded that while each device used a different approach to generate optimal high-resolution images with minimum possible artefacts, overall the Zeiss Cirrus 5000-HD OCT was superior in most evaluated features. (5) There are other devices available from Nidek and Optopol that were not included in the study and it was noted that each device had their strengths and weaknesses but vessel density, for example, was comparable across all devices. (5)
As the retina is one of the body's most metabolically active tissues, it requires a rich blood supply. In order to meet its demands, the retina has a dual supply: the choroidal circulation for the outer part of the retina and choroidal tissues; and the retinal circulation for the inner retina. The inner retinal arterioles supply blood to the superficial capillary network which is located in the nerve fibre layer as well as to the deeper capillary plexus which is located at the level of the inner nuclear layer. OCT-A has great potential as a clinical tool in the evaluation, management and follow up of a number of ocular diseases which compromise the vascular system, such as age-related macular degeneration (AMD), diabetic retinopathy (DR), glaucoma, uveitis and ocular inflammatory conditions. While FA (and indocyanine green) can visualise leakage, pooling and staining, the angiogram obtained is based on the fluorescence of a plasma protein bound dye molecule, which does not offer any information on red blood cell (RBC) flow and/or oxygenation. (6,7) However, plasma and RBC flow are not necessarily the same and their relative relationship depends largely on local haematocrit. Additionally, blood flow in the microvasculature differs considerably from what is referred to as laminar flow in a larger vessel. (6,7)
Signal strength, noise, artefacts, and other factors that can affect OCT-A quality
Limitations of this technology are similar to those of other imaging tools and can be broadly split in two categories: a) acquisition-based limitations; b) technology-based limitations. Acquisition-based limitations include: poor patient compliance, fixation, media opacities and severely dry eyes, whereas technology-based limitations are: motion (eye, body and software motion), projection artefacts, and small field of view. However, the latter is almost always resolved with the latest updates allowing for easy-to-use montage imaging (see Figures 1-3). While these new montage images can provide a much larger view, they do not yet necessarily provide all quantitative output as seen with central, optic nerve head (ONH) and macular angiography analyses.
Some factors can be managed by the clinician and/ or the software, but not all can be tackled. Hence, knowledge of what constitutes an artefact and what has caused it is necessary in order to be able to interpret the data correctly. Firstly, good signal strength is paramount to obtain a good quality image. Media opacities such as cataract can lead to signal attenuation resulting in a low contrast image. However, other factors such as patient (eye, pupil) alignment and dry eye can also lead to low signal strength, but these can be overcome by manually improving alignment, providing artificial lubricants as well as some device specific support with alignment (inbuilt contact sensors and software guided alignment). Aforementioned factors mostly lead to the whole image lacking contrast and clarity whereas vitreous floaters, for example, can lead to a local loss of signal strength which will cause only a localised lack of contrast/clarity of the scan obtained. Depending on the size, type and location of the floater, repeated imaging following head and/or eye movement, or a short rest might help shift the floater out of the scan area. The importance of low signal strength lies in the fact that smaller vessels will produce smaller signals which means in situations where the signal is attenuated due to opacities or artefacts these small vessels might not be visualised at all and could lead to falsely concluding a vessel dropout or low density is present. All manufacturers have different approaches to reduce noise and improve the signal-to-noise ratio. However, there is limited published information as to how they work. The movement of the eye, head or body can also lead to poor quality images due to a decorrelation over the entire B-scan referred to as motion artefacts, (8) these can produce shearing distortions and/or gaps in the enface image. To tackle motion artefacts, most manufacturers have developed eye tracking software that can help to correct/ reduce some motion artefacts and gaps in the data. Other artefacts that can occur are projection and segmentation artefacts. Projection artefacts can lead to vessels/ flow falsely visualised at an incorrect level, that is to say, in an area without vasculature. Segmentation artefacts can, for example, occur where the enface image shows vasculature of different depths as if they were located at the same level; this can, for instance, occur in high myopes. It is, therefore, important to not only view the enface image but also the cross-sectional view to ensure the segmentation has captured the correct vascular level.
OCT-A is another piece to complete the clinical picture in today's multimodal imaging approach encompassing colour fundus and wide-held imaging auto-fluorescence capture and OCT scanning to better understand diseases such as AMD, DR, glaucoma and other vascular pathology affecting the eye. Ocular vascular pathology is multifactorial and perfusion abnormalities appear to play an important role in disease pathogenesis and progression, hence a multimodal imaging approach allowing for structural and functional vascular assessment will hopefully provide not only a better understanding of these diseases but can be used efficiently in screening diagnosing and monitoring in the near future.
Similar to utilising structural OCT data, OCT-A images need to be interpreted by qualitative and quantitative methods. The vasculature can be assessed by visual inspection to establish its relative location within the retinal tissue and proximity to retinal landmarks and/or lesions and can be measured quantitatively using inbuilt analysis software to establish vessel density/perfusion and area of the foveal avascular zone (FAZ). Vessel density is the total length of perfused vasculature per unit area in the region of measurement. In addition, size of the neovascular area, distance to lesions, depth and density can be quantified using semi-automated methods. The following section provides a glance at what we know to date and how OCT-A could fit into clinical practice.
Age-related macular degeneration
AMD is multifactorial and ocular vascular parameters have been widely explored with not always conclusive or consensual outcome other than that circulatory abnormalities of the choroidal vasculature may contribute to the pathological changes. Research predating commercially available OCT-A showed reduced subfoveal choroidal blood flow in patients with dry AMD compared to controls which was thought to be linked to the amount of drusen present. (9,10) Toto and colleagues were able to demonstrate a link between structural OCT signs and retinal blood supply in patients with intermediate AMD. (11) They showed that patients with signs predicting development of geographic atrophy, such as thinner foveal thickness, have a reduced flow in the superficial retinal vascular plexus. Another study examining AMD patients using FA and OCT-A found that the overall sensitivity to detect CNV was 64.4% (75.7% for classic CNV and 48% for occult CNV, respectively). (12)
For the purpose of routine clinical practice, the use of OCT-A to follow CNV progression depends on its capability to provide supplemental information to traditional structural only OCT but also on its reproducibility. Recent findings from a study by Amoroso and colleagues suggests that OCT-A provides reproducible images for evaluation of the neovascular size in AMD patients. (13) They found excellent inter-grader agreement and good agreement for inter-visit measurements of the mean choroidal neovascularisation area. However, there are areas where more research is necessary. For example, where contradictory findings from the structural OCT and OCT-A show that subretinal and intraretinal fluid do not necessarily go hand-in-hand with neovascularisation. While these findings appear to be contradictive, they provide a new insight into disease pathogenesis and progression which will improve our understanding and most likely will lead to new treatment pathways and diagnostic endpoints. (14)
DR is characterised by marked alterations to the retinal vasculature and is becoming increasingly prevalent worldwide. The current gold standard in screening for DR is retinal photography which is, depending on the disease stage, often followed up with structural OCT and FA imaging. Many features of DR could be imaged accurately with OCT-A and, therefore, help improve disease classification; this is because due to light scattering in the retina, capturing the deep capillary plexus with FA is not possible but can be visualised with OCT-A. It also allows for simultaneous observation of structural and vascular features while at the same time minimising acquisition time, increasing patient comfort and lowering the risk of side effects due to dye injection in FA. While there are still improvements needed to reliably capture and visualise microaneurysms, which are a hallmark of DR, OCT-A has already shown its potential in a study by Scarinci and colleagues, which reported that capillary dropout in the deep capillary plexus was associated with macular photoreceptor disruption in patients with DR. (15) Similar findings have been reported by histological studies and are now confirmed by OCT-A, namely vascular changes such as the development of aneurysms occurring much earlier and more severe in the deep capillary plexus than in the superficial plexus. (16) Other findings using OCT-A in DR show an increase of the FAZ with severity of disease and deep capillary plexus vessel pattern changes correlating with disorganisation of the inner retinal layers. (17,21) Also, cystoid oedema appears to preferentially form in regions with absent deep plexus flow. (22,23) Changes in the deep vascular plexus can begin very early and might be present prior to alterations seen on gold standard photography screening and FA observations.
Optic nerve disorders
The radial peripapillary capillary network supplies the nerve fibre layer around the ONH. Early studies of this peripapillary vasculature have shown a reduction in vessel density and flow index in preperimetric glaucomatous eyes. (24) More recent studies have shown a reduction in vessel density in glaucoma patients compared to normals, (25) where vessel density reduction correlated with visual field parameters and structural OCT, (25) significantly reduced ONH perfusion in glaucoma but similar in OHT and NTG patients, (26,27) and a reduction of peripapillary density in glaucomatous eyes without high myopia was spatially co-localised with the location of the visual field defect. (28) These findings all highlight the potential of OCT-A, but also illustrate the need for longitudinal studies which employ multimodal imaging protocols to gain further insight into the disease pathogenesis and progression while obtaining experience and knowledge on how the different techniques compare and complement each other best for clinical use. A note of caution in regard to assessing the peripapillary vasculature is warranted here especially when considering the technical aspects of OCT-A: currently, capturing and visualising capillaries with slow flow is limited, which can lead to potentially false low capillary density measurements and capillary dropout noted. This is particularly true in patients with slow flow as found in those suffering from glaucoma or other systemic vascular conditions characterised by poor circulation. While this appears to be a limitation of the technique, it also opens up the possibility to utilise this feature as a way of monitoring patients undergoing treatment--for example, those who have slow flow being treated to restore normal flow which would initially present with lower vessel density but following treatment show a higher vessel density, this could serve as a surrogate marker for flow.
The prevalence of myopia has risen rapidly in the past 50 years. (29) This is of particular concern as eyes with increased axial length and deformation of the posterior segment can lead to several lesions including myopic choroidal neovascularisation, lacquer cracks, chorioretinal atrophy, posterior staphyloma and macular retinoschisis. (29,30) Macular neovascularisation has poor prognosis and often leads to severe loss of central vision. Current diagnostic tests include a combination of FA and OCT. Indocyanine green angiography (ICGA) can visualise the choroidal vasculature, but as with FA, due to its invasive nature it might be complemented in the future by OCT-A.
OCT-A is still in its infancy as there is no agreed standard protocol (meaning there are no normative datasets available to date) as well as the absence of clinical guidelines. Nevertheless, the results published to date are very promising as they show a rethinking in regard to disease staging and treatment decisions may be warranted given the fact that structural and functional alterations appear not always to go hand-in-hand. This fact isn't new but that we can capture both at the same time in high resolution is, and this is what allows us to study both at the same time. In the absence of standard protocols and clinical guidance, it is paramount to still apply gold standard examination protocol but to supplement with OCT-A in order to build up the necessary long-term experience. It is also important to note that as in structural OCT imaging, numerical output alone is not the Holy Grail. Qualitative output such as the presence or absence of neovascularisation in itself is equally important to form a final clinical decision.
Under the enhanced CET rules of the GOC, MCQs for this exam appear online at www.optometry.co.uk. Please complete online by midnight on 19 April 2019. You will be unable to submit exams after this date. Please note that when taking an exam, the MCQs may require practitioners to apply additional knowledge that has not been covered in the related CET article.
CET points will be uploaded to the GOC within 10 working days. You will then need to log into your CET portfolio by clicking on 'MyGOC' on the GOC website (www.optical.org) to confirm your points.
Course code: C-70255 Deadline: 19 April 2019
* Understand the use of OCT-A for retinal assessment (Group 3.1.3)
* Be aware of the use of OCT as a tool for retinal assessment (Group 3.1.3)
* Be able to interpret retinal abnormalities identified using OCT-A (Group 2.1.2)
Dr Rebekka Heitmar is a German-trained dispensing optician and optometrist. After completing her PhD, she became a lecturer at Aston University. Her research includes a number of ocular imaging modalities (such as OCT, OCT-A, fundus photography, retinal oximetry and vessel dynamics) to explore how ocular features and systemic vascular diseases impact on the ocular circulation and visual function.
Visit www.optometry.co.uk, and click on the 'Related CET article' title to view the article and accompanying 'references' in full.
Caption: Figure 1 Left eye montage OCT-angio image 6 x
Caption: Figure 2 Structural OCT and OCT-A of the optic nerve head
Caption: Figure 3 Example of a macular 3 x 3mm OCT-A image on the left, followed by another 3 x 3mm OCT-A image taken with the macula off centre to examine an area of RPE dropout as visible from the structural OCT scan below. The image furthest to the right is of the same individual but shows a 6 x 6mm scan. The smaller images help navigation through the different depths whereas the structural OCT scans below the angio images show the segmentation lines illustrating the vascular bed shown in the en face image
|Printer friendly Cite/link Email Feedback|
|Date:||Mar 1, 2019|
|Previous Article:||Residual visual function in cortical vision loss.|
|Next Article:||Orthokeratology lens design for myopia management.|