Choroidal and Choriocapillaris OCT-A Analysis in Patients Affected by Active Central Serous Chorioretinopathy

Abstract

Purpose: To analyze choroidal and choriocapillaris changes in eyes affected by active unilateral central serous chorioretinopathy (CSC).

Methods: A total of 17 eyes suffering from naïve CSC were enrolled. In addition, 17 healthy fellow eyes were analyzed, and 10 eyes were enrolled as controls. Main outcome measures were choroidal vascularity index (CVI), best-corrected visual acuity (BCVA), central macular thickness (CMT), subfoveal choroidal thickness (SFCT), and pigment epithelial detachment (PED) presence and maximum height (PED-MH). In addition, choriocapillaris and choroidal flow analysis in the two concentric areas surrounding the fovea were analyzed.

Results: CCT was higher in affected eyes than healthy ones (P = 0.007). CVI was significantly higher in affected eyes (P = 0.027) and in fellow eyes (P = 0.027) compared to healthy. The choriocapillaris analysis showed interesting results in the first ring, with statistically significant differences between diseased eyes and fellow eyes and in diseased eyes compared to healthy ones. Besides, in the second ring analysis a lower flow in choriocapillaris was found in diseased eyes compared with healthy (P = 0.019). The choroidal flow analysis showed lower flow in affected eyes in the first and second ring when comparing diseased eyes with healthy controls (P = 0.006).

Conclusions: Choroidal and choriocapillaris flow abnormalities occur in both eyes affected by CSC and fellow eyes with different trends depending on the area of study reinforcing the key role of choroid and choriocapillaris in the pathogenesis of disease.

Translational Relevance: Understanding choroidal and choriocapillaris flow abnormalities in CSC eyes could give us new biomarkers able to monitor disease.

Introduction

Central serous chorioretinopathy (CSC) is a common ophthalmic disorder characterized by low visual acuity and metamorphopsia and typically affects men aged 20 to 60.¹ Typical signs of the condition are the presence of idiopathic serous focal retinal detachments associated with choroidal hyperpermeability and retinal pigment epithelium (RPE) outer blood-retina barrier defects.² To date, the pathological process leading to CSC has not been completely cleared, although the central role of corticosteroids as a contributing cause of the disease has universally been assessed. Besides, recent studies have included CSC in the pachychoroid spectrum diseases because of the choroidal hyperpermeability, which characterizes the condition, frequent bilaterality and a thicker choroid in both affected and fellow eyes.³ Thus all evidence supports the key role of choroid and choriocapillaris in the CSC pathological process. Several authors have ascertained choroidal abnormalities as factors promoting the disease.^4–8 Choroidal thickening and hyperpermeability, together with increased hydrostatic pressure, are believed to induce RPE detachments with fluid leakage into subretinal space, resulting in active CSC.² Spaide et al.,⁹ in their recent review, have proposed a hypothetical framework for central serous chorioretinopathy and allied disorders by observing that all the underlying etiologies that are supposed to cause the disease seem to share the same pathologic process of underlining venous overload chorioretinopathy. Recent advancements in optical coherence tomography (OCT) and OCT angiography (OCT-a) have made studying choroid and choriocapillaris more accessible for quantitative and qualitative analysis.^10,11 The role of imaging has permitted us to comprehend the disease better and analyze typical patterns central to disease diagnosis and management. Different authors have tried to investigate CSC nature by using various imaging tools and techniques. Toto et al. have reported in their study that changes in the choroidal vascularity index (CVI) occur after eplerenone treatment in eyes with chronic CSC. After treatment, they found that the choroidal parameter is reduced in both affected and fellow eyes.¹² Cohen et al., in a recent study, considered the imaging identification of mid-phase hyperfluorescent plaques on enhanced-depth-imaging OCT (EDI-OCT), blue-light fundus autofluorescence (BAF), fluorescein angiography (FA) and indocyanine green angiography ICG-A as clinical signs of both vascular permeability and RPE dysfunction, both central processes in CSC.¹³ Consequently, a few authors have been interested in investigating whether choroidal and choriocapillaris circulation behave differently depending on the area of study.^14,15 In the present study, we aimed to analyze changes in CC and choroidal flow surrounding the macular area in patients affected by active CSC. 

Methods

Seventeen eyes of 17 patients with unilateral active CSC with evidence of subfoveal subretinal fluid (SFSRF) were enrolled in our observational study between February 2023 and February 2024 at Ophthalmological Clinic, University “G. d'Annunzio” of Chieti-Pescara, Italy. Inclusion criteria were idiopathic CSC with symptoms lasting less than one month, no previous or active local treatments (including laser or intravitreal injections), and any presence of fluids in studied or fellow eyes at presentation other than CSC. Exclusion criteria were patients under 18 years old with local or general diseases other than CSC and previous CSC episodes. In addition, 17 healthy fellow eyes (FE) were analyzed, and 10 patients matched in age and gender were enrolled as controls (HE). All subjects underwent complete ophthalmic evaluation, including best-corrected visual acuity (BCVA) measurement using ETDRS chart by assessing the logarithm of the minimal angle resolution (LogMAR), slit lamp examination, and intraocular pressure measurement using Goldmann applanation tonometry. Moreover, every patient underwent a complete multimodal imaging study with spectral domain OCT (SD-OCT), fluorescein angiography, indocyanine green angiography (ICGA), and OCT-angiography. Main outcome measures were BCVA, central macular thickness (CMT), central choroidal thickness (CCT), presence of pigment epithelium detachment (PED) and its maximum height (PED-MH), choroidal vascularity index (CVI) and choriocapillaris (CCF) and choroidal flow (CF) in the first and second ring surrounding the macular area on OCT-a slab analysis. 

OCT Protocol

OCT was performed using Spectralis HRA+ OCT (Heidelberg Engineering, Heidelberg, Germany) (Fig. 1). The acquisition protocol included 49 horizontal raster dense linear B- scans on the fovea. Moreover, enhanced depth imaging (EDI) mode horizontal and vertical B-scans centered on the fovea were performed in all subjects. CMT was measured with the central 1 mm diameter circle of the ETDRS thickness map. SCT was assessed using the inbuilt manual caliper on EDI horizontal OCT scans by measuring the vertical distance from the outer border of the RPE to the inner border of the sclera. PED was defined as the separation between the RPE and Bruch's membrane, and its maximum height (PED-MH) was manually measured using the caliper on the scan, revealing the most prominent lesion site within the central 1 mm-diameter circle of the ETDRS grid. CVI was calculated through a validated algorithm.¹⁶ EDI-OCT horizontal and vertical scans centered on the fovea were acquired and thus exported on ImageJ software version 1.52° (National Institutes of Health, Bethesda, MD, USA; available at http://rsb.info.nih.gov/jj/). First, choroidal boundaries were identified, which delimited the choroid. Thus images were automatically binarized, and the choroid-EPR junction and the sclero-choroidal junction were considered the ROI's limits. CVI was calculated as the ratio between the luminal choroidal area (LCA) and total choroidal area (TCA), identified through binarization using Niblack's auto-local threshold.¹⁷ All measurements were performed by two independent, experienced readers (MLR and LT), and images with poor signal strength (<25) were excluded and thus repeated. In disagreements, a third specialist expert in the retinal field was consulted (RM). 

Figure 1.

 

Imaging examinations of a CSC case with foveal subretinal fluid (A), healthy fellow eye (B), and aged-matched control (C). Central macular thickness (1) and subfoveal choroidal thickness measured on an OCT macular line B scan (2) and CVI (3) are shown in A, B, and C images.

OCT-a Protocol

OCT-a scans were obtained using a PLEX Elite 9000 device (Carl Zeiss Meditec Inc., Dublin, CA, USA). Scans of 3 × 3 volume were acquired, and FastTrack motion correction software was applied. Low-quality scans (affected by motion artefact, incorrect segmentation, or signal strength <8) were excluded and thus repeated until good-quality scans (signal strength ≥8) were achieved. The instrument segmentation automated algorithm was used to identify the choroidal and choriocapillaris en face OCT-a slabs. Two retina specialists (AQ and LT) analyzed images for segmentation accuracy. Identified scans were exported as JPEG files to be analyzed with ImageJ software version 1.52° (National Institutes of Health, Bethesda, MD, USA; available at http://rsb.info.nih.gov/jj/). Two concentric areas surrounding the fovea were identified using a described methodology¹⁴ in each 3 × 3 en face choriocapillaris and choroidal OCTA en face scan. A first ring was identified in the foveal area as previously defined on en face OCT-a images as a 1 mm-diameter circle centered on the fovea.¹⁸ As previously described,¹⁶ three progressive 40 pixels concentric rings were generated from the foveal area using ImageJ's “Distance Map” function. Each ring (R1, R2, and R3) was added to the region of interest (ROI) manager for flow analysis. Images were binarized using the Phansalkar method as previously described,^19–21 and the “analyze particles” function software was applied to proceed with flow calculation in the two concentric rings in both choroidal (CH) and choriocapillaris (CC) slab¹⁶ (Fig. 2). 

Figure 2.

 

The sequential process (A–G) shows the binarization and flow calculation in a DE in the rings surrounding the foveal area. The three progressive 40 pixels concentric rings were created using the “Distance Map” function in image J from the foveal area. Two regions were thus identified as a first (yellow) ring and a second external ring (blue) The flow was calculated in the two areas.

Statistical Analysis

Descriptive statistics are reported as median and quartiles [first; third] for continuous variables, whereas categorical data were summarized as absolute frequency and percentage. Paired t-test was used to demonstrate the primary outcome differences between the treated and control eyes. Assuming a pooled standard deviation of 0.01 for first-ring CC, the study would require a sample size of 17 eyes for each group to achieve a power of 90% and a significance level of 5% (two-sided), for detecting a true difference in means between groups. The association was evaluated using the Fisher exact test (frequency ≤5). The Mann U Whitney test assessed differences in unpaired samples (diseased eyes [DE] and FE vs. HE), and the Wilcoxon rank sum test for paired samples (DE vs. FE). All statistical tests were two-tailed, with a significance level set at P ≤ 0.05. Analyses were performed using the R software environment for statistical computing and graphics (version 4.2; http://www.r-project.org/). 

Results

A total of 17 eyes of 17 patients with active unilateral treatment naïve CSC were included in our observational study. Of the patients, 17.6% were female, whereas 82.4% were male. The median age was 53.0 [48.0; 57.0]. The median BCVA was 0.10 [0.00; 0.20]. The median CMT in affected eyes was 292 [212; 373], and PED was present in all DE (100%) with a medium PED maximum height of 104 [85.0; 162]. Patient characteristics are extensively reported in the Table. CCT was found to be significantly higher in DE when compared to HE (P = 0.002) and in the comparison between FE and HE (P = 0.042). Besides, no statistically significant difference between DE and FE (P = 0.535) was found. Interestingly, CVI resulted to be significantly higher in DE compared to HE (P = 0.018). Similarly, FE showed a significantly higher CVI when compared to HE (P = 0.013). The choriocapillaris analysis showed interesting results in the first ring, with statistically significant differences between DE and FE (lower values in DE compared to FE P = 0.026), as well as in DE compared to HE (with reported lower choriocapillaris flow in the first ring compared to HE p = 0.002). Besides, in the second ring analysis, a lower flow in choriocapillaris was found in DE compared with HE (P = 0.006). The choroidal flow analysis showed statistically significant results when comparing DE with HE, with a lower flow in DE in the first ring (P = 0.001). Similarly, the flow analysis in the second ring showed a significant reduction in flow when comparing DE with healthy controls (P = 0.002). 

Table.

 

Summary Descriptive Statistics by Groups

 

Discussion

The key role of the choroid in CSC disease has already been universally assessed. Spaide et al.,⁸ by investigating contributing factors in CSC pathophysiology, identified the presence of venous outflow abnormalities in eyes suffering from CSC, with similar traits encountered in eyes affected by pachychoroid syndrome, thus reinforcing the concept of including CSC in the category above. Although different factors have already been identified in the pathogenesis of the disease, including the central role of endogenous or exogenous corticosteroids, abnormal venous outflow from the choroid has thus been appointed as a unique underlying phenomenon explaining the dysregulation occurring in affected eyes, according to previous authors.⁹ Choroidal hyperpermeability has been appointed as a central factor in disease development in previous years.^6,8 The condition is easily detectable by ICG leakage, which has led to the development of treatments directed to hyperpermeability areas, such as PDT.¹³ Previous authors have tried to explain venous overload with different theories. Prünte et al.⁷ in 1995 proposed choroidal congestion caused by delayed filling of choroidal vessels, followed by capillary or venous congestion and leakage. Later, Hiroe et al.²² and Kishi et al.²³ proposed that CSC disease is characterized by vortex vein congestion developing in eyes with asymmetric vortex veins. Recently, an interesting “multi-hit theory” has been proposed by Cheung et al.²⁴ identifying a sequence of events that may be responsible for CSC development. In their theory, the mechanism underlying CSC comprises a sequence of events in which in case of (1) anatomical predisposition, the occurrence of an inciting event may result (2) in venous overload, which triggers compensatory attempts (3) that may cause (4) saturation, decompensation, and vicious cycle, and finally lead to (5) visual impairment. In the author's theory, the progression through stages relies on different factors, including an appropriate compensatory mechanism to venous overloading.²⁴ Undoubtedly, it must be acknowledged that both choroid and choriocapillaris play a central role in disease pathogenesis. The occurrence of vascular changes in eyes affected by CSC has already been recently analyzed by Kuroda et al.,²⁵ who reported the microvasculature of the inner choroid at the large choroidal vessels to be larger in eyes with CSC when compared to age-matched normal eyes. After this finding, Ma et al.,²⁶ in their research article, evaluated changes in choroidal vascularity after half fluence PDT in patients with CSC by finding that choriocapillaris had the earliest decrease in the vascular proportion of en face images on swept-source optical coherence tomography (SS-OCT) analysis. In contrast, Sattler's and Haller's layers showed later decreases, explaining this temporal difference due to the CSC pathophysiological mechanism and HF-PDT therapeutical effect. Indeed, recent findings have permitted the inclusion of CSC in the pachychoroid spectrum disease. A recent article by Meng et al.²⁷ analyzed widefield OCT angiography assessment of choroidal thickness and choriocapillaris in eyes affected by CSC and found CSC to be bilateral disease characterized by asymmetrical manifestations. Moreover, the authors observed local factors in the affected eyes to have a key role in CSC development in the dynamic and regional changes that arise in choriocapillaris, thus proposing widefield OCTA as a useful tool to study CSC pathogenesis. Consistent with these results, Chan et al.²⁸ revealed greater choriocapillaris width in eyes affected by CSC compared to healthy fellow eyes. In a recent paper, Viggiano et al.¹⁶ performed a topographical analysis of the choriocapillaris reperfusion after loading anti-VEGF therapy in neovascular AMD by analyzing the percentage of the choriocapillaris (CC), flow deficit percentage (FD%), FD average area (FDa), and FD number in five progressive concentric rings surrounding the dark halo the MNV finding CC flow deficits to be greater around the associated dark halo before treatment, followed by a progressive recovery in CC flow after intravitreal therapy. In the present study, it was our interest to analyze whether the occurrence of vascular changes in choroid and choriocapillaris in a setting of vascular impairment because of the underlying CSC was different in two concentric areas surrounding the foveal area in eyes affected by active naïve CSC with the presence of foveal subretinal fluid by means oct swept-source OCTa in both choroidal and choriocapillaris slabs. Moreover, we aimed to assess whether the occurring changes also involved healthy fellow eyes. Interestingly, choriocapillaris flow appeared to be significantly lower in the first ring surrounding the foveal area in diseased eyes compared to healthy fellow eyes and between diseased eyes and healthy controls. Besides, in the second ring analysis, distancing from the foveal area, only the comparison between diseased and healthy eyes showed significance with lower flow in affected eyes than in healthy eyes. Previous studies have demonstrated that alterations in choriocapillaris flow patterns may be possible in both eyes of CSC patients, which can be interpreted as an index of focal choriocapillaris ischemia.^29,30 Yun et al.,³¹ in their article investigating the characteristics of choriocapillaris flow in fellow eyes of patients with central serous chorioretinopathy, found the underlining choroidal vessels to affect choriocapillaris perfusion in pachychoroid eyes, resulting in significant choriocapillaris hypoperfusion. Interestingly, the presence of areas of detectable choriocapillaris flow surrounded by an area of undetectable or diminished flow in patients diagnosed for active CSCR was reported by Estawro and colleagues,³² thus defining the presence of “choriocapillaris island.” In our study, no choriocapillaris islands were evidenced within the first ring area, despite the presence of lower choriocapillaris flow value when compared to the second external ring area. However, the presence of defined topographical differences in choriocapillaris flow highlights the necessity of better characterizing occurring changes, with future studies aimed at focusing on vascular changes occurring within the area of neurosensory detachment. Nevertheless, it must be considered that in our analysis, choriocapillaris flow was affected in the first ring closer to the foveal area in both comparisons with healthy patients and fellow eyes. Besides, the flow was affected only in the second ring compared to healthy patients. This finding can be explained as a greater flow depletion occurring in affected eyes than in fellow eyes, thus meaning that although pachychoroid healthy eyes may present choriocapillaris flow irregularities, CSC eyes are affected by deeper flow changes. Nevertheless, it must be assessed that evaluating flow changes in conditions with overlying fluid may be difficult due to the occurrence of signal interference. The choice of using a full-spectrum swept-source OCT aims at reducing this possible bias, because of its ability in representing CC signals despite the presence or absence of fluid as previously demonstrated.³³ However, although reduced by the methodology applied, the issue must be considered, thus representing a limitation in our study. Previous studies have reported that eyes with CSC are affected by choroidal blood flow alterations.^34–36 Similarly, in our article, the choroidal flow analysis showed reduced flow values in affected eyes compared to healthy fellow eyes in both the first and second rings. On the contrary, the comparison with healthy fellow eyes did not show significance. It has been demonstrated that eyes affected by pachychoroid spectrum disease report the presence of dilated choroidal vessels with increased choroidal thickness and thinning of both choriocapillaris and inner choroid. The so-called “pachyvessels” have been found to have a key role in affecting the overlying RPE, having a mechanical effect on the RPE-Bruch membrane.^15,37,38 Hwang et al.³⁹ in their article found choriocapillaris flow on OCT angiography to be reduced in acute CSC eyes, finding the dysregulated choroidal flow to affect the choriocapillaris flow. Undoubtedly, choroid appears to be dysregulated in CSC eyes. This blood flow dysregulation appears to be a causative factor in increased choroidal thickness, which has been demonstrated to be reversible in treated eyes, as demonstrated by our group in a previous article, where changes in choroidal structural and functional parameters were found in eyes treated by oral eplerenone using CVI reduction.^12,39 In our analysis, CVI appeared to be higher in affected eyes than healthy eyes, thus enhancing the role of CVI as representative of vascular dysregulation occurring in CSC eyes. Moreover, CVI was found to be altered in fellow eyes compared to healthy patients, thus confirming the pachychoroid nature of the disease, which frequently presents bilaterally.⁴⁰ Similarly, CCT was higher in diseased eyes compared to healthy patients, thus enhancing the agreement of structural and vascular parameters representing the choroid. In conclusion, choroidal and choriocapillaris flow abnormalities occur in both eyes affected by CSC and fellow eyes; however, they have different trends depending on the study area and the underlying condition. These results reinforce the role of choroid and choriocapillaris in disease pathogenesis, proposing new parameters to deepen inside vascular changes occurring in CSC eyes. However, future studies should be conducted to study variation in choroidal and choriocapillaris flow in pachychoroid diseases, to reinforce the role of functional and structural parameters for disease evaluation, response to treatment, and course assessment. 

Acknowledgments

Disclosure: M.L. Ruggeri, None; M. Passamonti, None; A. Quarta, None; O. Koci, None; A. Porreca, None; M. Di Nicola, None; L. Zeppa, None; R. Mastropasqua, None; L. Toto, None