Dr.Dheepak Sundar Malaravannan, M19066, Dr.Nasiq Hasan, Dr.Surbhi Agarwal, Dr.Rohan Chawla
ABSTRACT:
Purpose
To correlate the imaging characteristics of hypointense areas of certain retino-choroidal pathologies on Optical coherence tomography angiography (OCT-A) and Indocyanine green angiography (ICG-A)
Methods
Retino-choroidal pathologies were divided into 5 groups namely geographic atrophy (GA), serous pigment epithelial detachment (PED) and neurosensory detachment (NSD) secondary to central serous chorioretinopathy (CSC), drusenoid PED, choroidal neovascularization (CNV) and acutechoroiditis (AC). 10 cases in each of the 5 groups were studied individually using Swept source OCT-A and Spectralis ICG-A. The imaging characteristics were critically analyzed. The hypocyanescent areas on ICG-A were correlated with the corresponding hypo intense areas on OCT-A.
Results:
All 10 cases in each of the 5 groups uniformly showed the following features. GA showed hypocyanescent/hypo areas on ICG-A/OCT-A with unearthing of underlying prominent larger choroidal vessels. CSC (serous NSD/PED) and drusenoid PED showed corresponding hypointense areas on both ICG-A and OCT-A, but the larger underlying choroidal vessels were not seen through these areas. A similar hypo zone was seen around CNV. The larger underlying choroidal vessels were not visualized through the hypointense zone surrounding a CNV. AC initially showed hypo areas (without the visibility ofchoroidal vessels) which disappeared after oral steroid therapy.
Conclusion:
Corresponding hypo areas at the level of choriocapillaris on OCT-A and ICG-A should be carefully analysed and they do not necessarily imply a loss of choriocapillaris or choroidalhypoperfusion. A definite conclusion of hypoflow/loss of choriocapillaris can be made if the underlying larger choroidal vessels get unearthed through this window (hypo area). Else the hypointense defect may be due to shadowing or inability of the imaging modality to detect flow in presence of altered anatomy.
Keywords: ICG-A, OCT-A, low flow, dark areas, hypocyanescence
INTRODUCTION
Fluorescein angiography has provided us valuable information regarding retinal vascular flow. However, due to certain limitations it does not provide us with accurate information about anomalous flow in the sub-retinal area and the choroid. Indocyanine green angiography (ICGA) and now Optical coherence tomography Angiography (OCT-A) are better techniques for analysis of flow in these structures (1,2).
OCT-A provides flow information by analysing the decorrelation of the OCT signal reflected by moving red blood cells. Areas that show higher decorrelation (greater than a preset cutoff value) are represented as areas with flow and are seen as white lines (presumably vessels) on the scan. Additionally we sometimes seehypointense (dark) areas on the scan, which could be due to an actual lack of flow or a shadowing artifact. OCT-A scans segmented below the normal retinal pigment epithelium (0-15 microns) show a granular uniform hyperintense-reflective pattern. This is presumably from the flow in the choriocapillaries. Since these are fine fenestrated vessels, we are unable to see the exact morphology of the capillaries but only appreciate an area of diffuse flow. OCT-A is limited by the fact that it fails to detect flow below a certain flow rate and also the amount of decorrelation saturates above a certain flow (3).
Indocyanine green angiography (ICGA) is another investigation, which is able to provide us with information regarding the choroidal circulation. The larger and middle layer choroidal vessels can be better appreciated on ICGA. The individual choriocapillary vessels are not evident even on ICGA. However, in the central fundus a diffuse background hypercyanescence from the choriocapillaris is seen which obscures some of the deeper vessels. Hypocyanescenct areas seen on ICGA depend on actual lack of blood flow as well as blocked cyanescence due to overlying structures. Other lesser -understood and lesser – acknowledged causes of hypocyanescence on ICGA are due to the selective binding affinity of ICGA and diffusion properties of the dye (4).
Focal areas of hypocyanescence on ICGA with corresponding hypointense/dark areas on OCT-A are seen in certain retino-choroidal pathologies. The exact cause of these dark areas is controversial. Certain authors believe that the hypointense/dark areas seen on OCT-A, which correspond tohypocyanescent areas on ICG, represent focal areas of choroidalischaemia (5,6). We did a study to correlate the imaging characteristics of few pathologies on ICGA and OCT-A to try and understand their significance andcorrecting interpretation.
MATERIALS AND METHODS
This index observational study was performed at our tertiary care ophthalmic centre. The institutional review board approval was obtained. The study adhered to the tenets of the Declaration of Helsinki. The pathologies were divided into 5 groups namely geographic atrophy (GA), central serous chorio-retinopathy (CSC), drusenoid PED, choroidal neovascularization (CNV) and acute choroidits (AC). The CSC group included serous neurosensory detachment (NSD) and/or pigment epithelial detachment (PED). 10 patients in each group were selected from our outpatient services. Informed consent was obtained from all the patients.The study patients underwent a detailed anterior segment and fundus examination
All patients underwent fundus imaging, optical coherence tomography (OCT), OCTA and ICGA. Fundus photography, OCT and OCT-A were performed using Swept- source OCT/ OCT-A (DRI OCT Triton, Topcon Inc, Japan). The device uses OCTARA (OCT-A Ration analysis) algorithm, which benefits from being paired with the swept-source OCT.
Concurrent Fluorescein angiography (FFA) and ICGA images were obtained using confocal scanning laser ophthalmoscope – SPECTRALIS® High-resolution angiography (HRA), Heidelberg Engineering with the Heidelberg Eye Explorer software platform (Heidelberg Engineering Inc. United States).
A correlation of all these imaging modalities was done to put forward a probable hypothesis regarding the presence of hypocyanescent areas on ICGA which correspond to apparent hypoflow (dark) areas seen on OCT-A.
RESULTS
The first pathology we studied was GA. Both on ICGA and OCT-A lack of the diffuse hypercyanescent/hyperflow signal respectively was evident from the atrophic patches. Additionally on both investigations there was unearthing of underlying prominent larger choroidal vessels in the sharply demarcated atrophic patch in all the 10 cases. (Figure 1a&b)
The second pathology studied was CSC (Figure 2). ICGA again revealed an area of hypocyanescence in the area of the PEDin early phase (Figure 2c). However, in this case the underlying choroidal vessels were not visible unlike the case of geographic atrophy. OCT-A revealed dark areas (Figure 2d). A similar dark impression on OCT-A was also seen in the case of NSD (Figure 3). Drusenoid PED(Figure 4 a and b) also revealed corresponding hypo areas on both OCT-A and ICGA. (Figure 4c&d)
Next we analysed CNV with associated minimal bleed and sub-retinal fluid (Figure 5a). ICGA revealed hypercyanescence in the area of the neovascular membrane surrounded by a zone of hypocyanescence (Figure 5b). The exact pattern seemed to be replicated on OCT-A (Figure 5c). In the surrounding hypointense/dark zone, there was no overlying hemorrhage or significant fluid/exudation to cause a shadowing effect on ICG or prevent proper detection of the OCT signal on OCT-A.
Acute choroiditis (AC) group included Vogt koyanagiharada (VKH) disease (4 cases) and multifocal choroidits (4 cases) and serpigineouschoroiditis (2 cases). ICGA showed well definedhypocyanescent spots persisting till late phases (Figure 6b). OCT-A showed dark areas on scans segmented below the RPE (Figure 6c). However, after around 10 to 30 days of oralsteroid therapy, as the disease started resolving, these dark areas started reducing with restoration of visibility of the near normal choriocapillaris flow pattern on both ICG and OCT-A (Figure 6e&f).
Appropriate sample images in each group, has been shown in the figures 1 to 6.
DISCUSSION
Interpretation of choroidal vascular flow has become possible with ICGA and OCT-A. Corresponding hypointense/dark areas seen on these imaging modalities have been interpreted as areas of focal choroidal ischemia or lack of flow through the choriocapillaris (5,6). We suggest caution in making such conclusions based on our observations described above.
The hypointense/dark areas at the choroidal level on OCT-A/ICGA can be due to various reasons which include: Shadowing caused by the overlying pathology such as blood /exudates/fluid on ICGA/ OCT-A, inability of ICG to stain certain pathologies or actual lack of flow through these regions on both imaging modalities. Hence it is necessary to differentiate the causes to finalize the diagnosis in every individual case.
Analyzing the ICGA and OCT-A images of GA we infer that focal areas of loss of choriocapillaries (lack of flow) act as a window and enhance visibility of deeper choroidal vessels on OCT-A and ICG. ICG images of areas of GA permit visualization of the underlying choroidal vessels through the atrophic areas due to the absence of diffuse hypercyanescence of the lacking overlying choriocapillaries. Similarly on OCT-A the lack of interference of a deccorelation signal from the choriocapillaris enhances the capability of the OCT-A machine to interpret the deccorelation from deeper choroidal vessels in the atrophic patches. Thus, focal areas of actual loss of choriocapillaries and thus focal choroidalischaemia are seen as window defects unearthing the deeperchoroidal vessels and not simply as hypointense/dark areas. According to various reports, retinal pigment epithileal (RPE) loss and choriocapillary flow impairment areprominent findings in a case of geographic atrophy (7,8,9,10). Additionally, ICG as a dye and the higher wavelength of Swept source OCT both can penetrate through intactretinal pigment epithelium as well. Hence, in cases of GA true choriocapillary loss rather than just RPE atrophy is the major factor responsible for the increased visibility of larger choroidal vessels(11).
A serous PED is a sharply demarcated orangish red elevation of the RPE due to collection of fluid between the RPE and bruchs’ membrane. Serous PED’s are most commonly associated with age related macular degeneration, polypoidalchoroidalvasculopathy and CSC. The fluid in the PED is able to mask the underlying the choridalhypercynasescence on confocal scaning laser ophthalmoscope based ICG angiography, both in early and late phases (12). The OCTA features in CSC have been recently well elaborated in a study by Kanika et al (5), where they have described two distinct features: mottled dark areas corresponding to NSD and dense dark areas conforming to PED. The enface OCT over the PED also showed signal attenuation, hence these dense dark areas were thought to be due to signal loss rather than true flow void (choroidal ischemia). Our observations in PED and NSD are akin to the findings described above. Similarly drusenoid PEDs seen as hypointense areas on OCT-A and ICG, are likely due to inability of the imaging modalities to visualize the choroidal circulation due to the material deposited between the RPE and the Bruchs, rather than lack of choroidal flow.
OCT-A and ICGA of a CNV revealed a central bright zone corresponding to the neovascular flow surrounded by a dark rim. Few authors have suggested that the dark halo is due to focal choroidal ischemia (13). A mere shadowing caused by the surroundingproliferative RPE or the less vascular (fibrotic) component of the CNVcanalso be the cause of the dark rim.When the CNV invades the RPE, the outer layer ofthe complex is lined by an inverted component of RPE (14). This dark halo may hence correspond to the proliferative RPE surrounding the margin of the CNV.
Hence from the above findings, we comprehend that those areas that show the underlying larger choroidal vessels (like GA) may invariably represent anatomical loss of choriocapillaries or a defect in the flow through these vessels. Whereas isolated dark areas may correlate to the shadowing caused by an overlying structure or inability of current generation OCT-A machines to detect flow due to alterted anatomy or slow flow.
Dark areas on OCT-A are also seen in the case of acute choroiditis described by us.. The dilemma in interpretation of this finding is that are these dark areas are actually due to focal choroidalischaemia or something else?If we analyse them based on our conclusions drawn from cases of GA, then these donot represent true areas of choriocapillaris loss. This is because they are isolated dark areas and the underlying larger choroidal vessels are not visible through them.The dark spots further disappeared very early following institution of steroid therapy (within a week). It is difficult to accept that the choriocapillaris would regenerate so quickly within a week of steroid therapy. Rather a thin film of inflammatory cells at the RPE- Choriocapillary level might have caused shadowing or slowed the flow leading tothe appearance of the dark spots. The steroid therapy lead to early resolution of inflammation restoring the visibility of the choriocapillaryflowduring the followup. Similar views have been expressed by kanika et al (5).
Hence, it is difficult to conclude presence of choroidal ischemia or focal loss of choriocapillaris based on the correlation of hypointense areas on ICG and dark areas on OCTA. These dark areas rather reflect the inability of ICG and OCT-A to comment on the status of choroidal perfusion in presence of the above pathologies. A small sample size and lack of histo-pathological evidence are some of the limitations of this study.
REFERENCES:
(1) Yannuzzi LA. Indocyanine green angiography: a perspective on use in the clinical setting. American journal of ophthalmology. 2011;151(5):745-51.
(2)Sambhav K, Grover S, Chalam KV. The application of optical coherence tomography angiography in retinal diseases.Survey of Ophthalmology. 2017;62(6):838-66.
(3)Bandello F, Souied EH, Querques G (eds): OCT Angiography in Retinal and Macular Diseases. Dev Ophthalmol. Basel, Karger, 2016, vol 56, pp 57-61
(4)Desmettre T, Devoisselle JM, Mordon S. Fluorescence properties and metabolic features of indocyanine green (ICG) as related to angiography. Survey of Ophthalmology. 2000;45(1):15-27.
(5) Aggarwal K, Agarwal A, Deokar A et al. Distinguishing features of acute Vogt-Koyanagi-Harada disease and acute central serous chorioretinopathy on optical coherence tomography angiography and en face optical coherence tomography imaging. Journal of ophthalmic inflammation and infection. 2017;7(1):3.
(6) Howe LJ, Woon H, Graham EM, Fitzke F, Bhandari A, Marshall J. Choroidalhypoperfusion in acute posterior multifocal placoid pigment epitheliopathy: an indocyanine green angiography study. Ophthalmology. 1995;102(5):790-8.
(7) Bressler SB, Bressler NM. Age-Related Macular Degeneration: Non-neovascular Early AMD, Intermediate AMD, and Geographic Atrophy. In Retina Fifth Edition. Vol. 2. Elsevier Inc.2012. p. 1150-1182.
(8)McLeod DS, Grebe R, Bhutto I, Merges C, Baba T, Lutty GA. Relationship between RPE and choriocapillaris in age-related macular degeneration. Investigative ophthalmology & visual science. 2009;50(10):4982-91.
(9) McLeod DS, Taomoto M, Otsuji T, Green WR, Sunness JS, Lutty GA. Quantifying changes in RPE and choroidal vasculature in eyes with age-related macular degeneration. Investigative ophthalmology & visual science. 2002;43(6):1986-93.
(10) Biesemeier A, Taubitz T, Julien S, Yoeruek E, Schraermeyer U. Choriocapillaris breakdown precedes retinal degeneration in age-related macular degeneration. Neurobiology of aging. 2014;35(11):2562-73.
(11) Moult EM, Waheed NK, Novais EA et al. Swept-source optical coherence tomography angiography reveals choriocapillaris alterations in eyes with nascent geographic atrophy and drusen-associated geographic atrophy. Retina. 2016;36:S2-11.
(12) Mrejen S. Multimodal imaging of pigment epithelial detachment: a guide to evaluation. Retina. 2013;33(9):1735-62.
(13) El Ameen A, Cohen SY, Semoun O, Miere A, Srour M, Quaranta-El Maftouhi M, Oubraham H, Blanco-Garavito R, Querques G, Souied EH. Type 2 neovascularization secondary to age-related macular degeneration imaged by optical coherence tomography angiography. Retina;35(11):2212-8.
(14) Grossniklaus H, Gass JD. Clinicopathologic correlations of surgically excised type 1 and type 2 submacularchoroidalneovascular membranes. American journal of Ophthalmology. 1998;126(1):59-69.
Legends
Figure 1 (a) :Indocyanine angiography (ICG-A) of a typical case of geographic atrophy showing well defined patch of hypocyanescence with enhanced visibility of the underlying larger choroidal vessels. (b) OCT angiography of the same patient showing a well demarcated area within the choriocapillaries network through which deeper prominent larger choroidal vessels are seen.
Figure 2 (a) Fundus photograph of a patient who presented with serous pigment epithelial detachment (PED). (b) Swept source OCT of the same patient showing a localized serous PED. (c) Indocyanine angiography (ICGA) showing hypocyanescence corresponding to the serous PED (d) OCT angiography exhibiting a dark zone similar to ICGA
Figure 3 (a) Fundus photograph of a case of neurosensory detachment (NSD) (b) OCT-A showing a dull dark area corresponding to the NSD indicating the shadowing effect caused by the overlying fluid.
Figure 4(a) Fundus photograph and (b) swept source OCT of a case of drusenoid PED (c) Indocyanine angiography (ICGA) of the same patient shows central hypocyanescence and (d) OCTA reveals dense dark areas at the choriocapillarylevel due to the masking effect of the drusenoid PED
Figure 5(a) Fundus picture of a representative case of Choroidal neovascularization (CNV) (b) ICGA reveals a dull central hypercyanescence with a surrounding hypocyanescent border. This image is better discerned with the help of OCTA (c) an irregular hyperflow central zone with a dark margin
Figure 6A Case of acute choroiditis showing multiple focal areas of activity on (a) fundus photography; (b) ICGA reveals numerous dark spots corresponding to the lesion (c) OCTA also shows multiple darks patches at the choriocapillary level The same patient significantly improved following oral steroid therapy and follow up scans revealed (d) resolution of the inflammatory lesions with reduction in number, density and size of these spots (e) OCTA also revealed a reduction in the dark patches.
Figures have not been added as each figure has a size of around 5mb.
Leave a Comment