Evaluation of acute central serous chorioretinopathy using enhanced depth imaging OCT and multifocal electroretinography

Background chorioretinopathy multifocal electroretinography optical coherence (EDI-OCT) Methods This prospective observational study included 57 patients with unilateral CSC. Both eyes underwent mfERG and EDI-OCT. Peak amplitudes and implicit times of the ﬁrst kernel responses were analyzed and compared with those in 25 age-matched normal controls. Correlational analyses were performed between the mfERG results and EDI-OCT parameters. The thicknesses of the central retina, subretinal fluid, and choroid was measured at baseline and 3 months later. Results Compared with the normal controls, the amplitude and implicit time on mfERG were significantly impaired in the area with serous retinal detachment (SRD). The P1 amplitude and implicit time of the areas beyond the SRD were also found to be significantly impaired in the affected eyes. Eyes with a greater reduction in SRD had a less impaired mfERG response in fellow eyes than those whose retinal detachments were not spontaneously decreased by more than 90% after 3 months. Correlational analysis did not reveal any significant correlations between mfERG values and OCT parameters except for central choroidal thickness. The subfoveal choroidal thickness was negatively correlated with the mfERG parameters. The findings this study indicate diffuse functional impairment in acute CSC which involves both eyes and areas beyond the SRD. The retinal of the unaffected eye associated with SRD functional abnormality correlated changes the choroid.

space between the photoreceptor and retinal pigment epithelium (RPE) layers, leading to neurosensory detachment of the retina. [1] Although the cause of CSC has not been fully elucidated, individuals who have recently experienced a stressful life event and those with a type A personality are known to be at higher risk and young-to-middle-aged men are most frequently affected. [ [11,12] Evaluation of the correlations between EDI-OCT and mfERG parameters might aid understanding of the relationships between the structural and functional changes in CSC. Therefore, the aim of this study was to evaluate the functional and structural abnormalities in patients with acute CSC using these assessment techniques. An attempt was also made to identify whether or not any OCT or mfERG parameters in affected and unaffected eyes with CSC were associated with spontaneous regression of SRD caused by CSC.

Methods
This observational study was performed in the Department of Ophthalmology at Dongtan Sacred Heart Hospital between December 2018 and December 2019. Patients with CSC were recruited prospectively for EDI-OCT and mfERG. Patients were included in the study if they had a diagnosis of unilateral CSC with a symptom duration of less than 1 month and had SRF involving the macula on OCT examination associated with leakage on fluorescein angiography. The following exclusion criteria were applied: (1) age older than 55 years or younger than 19 years; (2) myopia of more than -6 dioptres; (3) a history of ocular treatment or systemic corticosteroid therapy; (4) presence of a retinal disorder associated with SRF; (5) media opacity; (6) an abnormality in the fellow eye; (7) bestcorrected visual acuity (BCVA) <20/20 in the fellow eye; and (8) underlying disease or history of drug abuse. Twenty-five eyes of 25 age-matched individuals without any ophthalmic disease were enrolled as controls. The study was approved by the institutional review board and was conducted in accordance with the tenets of the Declaration of Helsinki. Informed consent was obtained from all patients.
At the time of presentation, all subjects underwent a complete ophthalmic examination, including measurement of BCVA with the Snellen chart, intraocular pressure, slit-lamp biomicroscopic examination, colour fundus photography, fluorescein angiography, OCT examination, and mfERG recording. Patients were observed prospectively for 3 months without treatment (including laser photocoagulation, intravitreal injection, or photodynamic therapy to determine whether or not the CSC would regress spontaneously. Regression of CSC was defined as complete disappearance of SRF on fundus photographs and OCT scans. OCT imaging EDI-OCT images were obtained using a Spectralis OCT instrument (Spectralis; Heidelberg Engineering, Heidelberg, Germany) with 7 horizontal lines of 20 × 20 degrees through the centre of the fovea. The eye tracking system in this instrument was used to obtain each image and 100 scans were averaged to improve the signal-to-noise ratio. Using the built-in automated software, the distance between the inner surface of the retina and the inner border of the RPE at the central fovea was calculated manually as the central macular thickness (CMT). The central retina, SRF and choroid height were measured using the built-in measurement software. The central retinal thickness (CRT) was measured from the anterior surface of the retinal nerve fibre layer to the outer border of the detached retina.
The central SRF thickness (SRFT) was measured from the anterior surface of the RPE layer to the outer border of the detached retina and the central choroid thickness (CCT) was measured as the space between the outer portion of the hyperreflective line corresponding to the RPE layer and the chorioscleral junction. (Fig. 1). Given the significant circadian fluctuations in CCT, all images were acquired on the same day within one hour between 10 am and 2 pm. cd/m 2 ) stimuli changed independently in each hexagon at a rate of 75 Hz. Before recordings were contained, the pupils were fully dilated by 1% tropicamide and 2.5% phenylephrine hydrochloride. mfERG recordings were obtained after 15 minutes of adaption to ordinary room light and before insertion of an ERG Jet corneal contact lens electrode. Proparacaine hydrochloride 0.5% drops were instilled for topical anaesthesia. During mfERG testing, the N1, P1, and N2 peaks for each response were automatically identified in real time over a group of up to five rings from 0° to 25° of eccentricity. We analyzed the average amplitudes and implicit times of the first-order kernel mfERG responses (N1 and P1) from maps of N1, P1 and N2 wave peaks on each individual ring.

Statistical analysis
The Mann-Whitney U test was used to compare the mfERG values between the patients with CSC and the controls. Serial comparisons of the OCT and mfERG values were performed using the paired t-test.
The correlations between the logMAR BCVA, OCT, and mfERG parameters were examined using Pearson's correlation coefficient. Multiple linear regression models were used to assess potential predictors for a decrease in SRF. Statistical modelling was used to identify factors that were independently associated with the change in SRF between baseline and 3 months. Only predictors that were statistically significant in exploratory analysis were included. A post hoc analysis was also performed to identify differences between patients who did and did not achieve a ≥90% reduction in SRF. All statistical analyses were performed using SPSS version 19.0 (IBM Corp., Armonk, NY, USA). A P-value <0.05 was considered statistically significant.

Results
Sixty-one patients were enrolled in the study. Four of these patients were excluded because of loss to follow-up, leaving 57 patients with acute unilateral CSC and no history of other ocular disease for inclusion in the study. Both eyes of each patient were analyzed. The mean age at the time of the initial episode of CSC was 47.75 ± 7.98 (range, 36-55) years and the mean age of the control subjects was 45.88 ± 5.28 (range, 35-52) years. Fifty-three of the patients with CSC were male and four were female while 23 of 25 control subjects were male and two were female ( Table 1).
The changes in BCVA and OCT parameters after 3 months of observation are shown in Table 1. The mean BCVA in affected eyes was 0.17 ± 0.21 logMAR at baseline, and improved significantly to 0.08 ± 0.11 logMAR after 3 months without treatment (P<0.0001). The mean CMT and SRFT values were significantly decreased after 3 months of observation (P<0.0001) but there was no significant change in CRT (P = 0.383). The mean CCT in the affected eyes decreased significantly from 385.79 ± 53.32 μm at baseline to 327.77 ± 46.5 μm at 3 months (P<0.0001). The mean CCT in the normal fellow eyes also decreased significantly after 3 months of observation (from 295.08 ± 54.97 μm to 281.17 ± 31.49 μm, P = 0.045). Tables 2 and 3 show the results of mfERG recordings from rings 1-5 of affected eyes and fellow eyes.
Compared with normal controls, the mean amplitude and implicit times of retinal responses were significantly impaired in affected eyes, especially in the N1 and P1 response amplitudes (Table 2).
There was a significant difference in the N1 amplitudes of rings 2-5 between control eyes and fellow normal eyes, but not in the other mfERG parameters.
Correlational analyses between logMAR BCVA, OCT, and mfERG parameters for eyes with CSC revealed no significant correlations between mfERG values and visual acuity (VA) or with OCT parameters except for CCT (Tables 4 and 5). The correlation analyses found significant negative correlations between CCT and the N1 amplitudes for rings 1-5 and the P1 amplitudes for rings 1,2, 4, and 5 (Table 4). Furthermore, a greater CCT was significantly associated with a delayed implicit time of N1 in rings 1-5 and P1 in ring 1 (Table 5).
Multiple and simple linear regression models were used to identify potential predictors of a decrease in SRF at 3 months. In addition to the demographic data (sex and age) and baseline VA, baseline OCT parameters (SRFT, CMT, CRT, CCT) were included in the multiple linear regression analysis. Statistical modelling identified the baseline VA to be the only significant predictor of SRF reduction. A better VA at baseline predicted a better reduction in SRF at 3 months (coefficient -1.716, 95% CI -0.306, -3.126, P = 0.045). The correlations between SRF reductions and mfERG values in affected and unaffected eyes were calculated, and only the P1 implicit times at rings 3-5 of unaffected fellow eyes were negatively correlated with SRF reductions (P<0.0001, P = 0.019, and P = 0.002, respectively).
Thirty-six of the 57 patients showed a more than 90% reduction in SRF between baseline and 3 months and 21 patients achieved an SRF reduction of less than 90% in the same period. Patients who achieved a more than 90% reduction in SRF were defined as a group with regression and the patients who did not were defined as a group with less regression. Except for the baseline VA, there was no significant between-group difference ( Table 6). The group with regression had a significantly better VA at baseline than the group with less regression. The mfERG results in the affected and unaffected eyes were compared between these two groups (Tables 7 and 8). There was no significant betweengroup difference in most of the mfERG parameters between the two groups of affected eyes (Table 7).
However, the mean amplitude and implicit times in normal fellow eyes showed significant differences within rings 1-4 ( Table 8). The retinal response in normal fellow eyes was better in the group with regression than in the group with less regression.

Discussion
This study evaluated a group of patients with acute CSC using mfERG and EDI-OCT. mfERG is an objective examination that provides a regional map of retinal function across the posterior pole. Unfortunately, these studies were not able to resolve this debate because of the relatively small numbers of patients enrolled. Timothy et al. performed a cross-sectional study in 45 eyes with acute CSC using mfERG and found a reduction in the second-order mfERG response for the more peripheral macular area. [17] In the present study, mfERG examination in 57 patients with acute CSC showed that retinal responses and the implicit time of mfERG were significantly impaired in the area with SRD, which is consistent with previous reports. The P1 amplitude and implicit time of rings 4-5 was found to be significantly impaired in affected eyes when compared with the controls. This finding suggested that the area of retinal dysfunction in CSC is larger than the SRD observed clinically. In contrast with the mfERG results in affected eyes, the retinal response in fellow eyes was less severely impaired over the entire area examined. However, a post hoc analysis revealed that the patients with a better reduction in SRD showed a less impaired mfERG response in fellow eyes than the patients whose retinal detachments were not spontaneously decreased by more than 90% after 3 months. This finding means that some of the patients with unilateral CSC had fellow eyes with good retinal responses and others did not. For this reason, there has been debate regarding whether normal fellow eyes in patients with CSC have abnormal retinal responses on mfERG examination. This study clearly indicates that the pathogenesis of CSC involves areas beyond the SRD in affected eyes and areas throughout the posterior pole of non-affected eyes; this finding support the view that CSC is a disease affected by systemic humoral factors or by diffuse underlying choroidal vascular disease. [18,19] The advent of EDI-OCT has provided a non-invasive method for examination of the choroid and has contributed greatly to current understanding of the pathogenesis underlying CSC.
[2] EDI-OCT imaging has shown that the choroid is abnormally thickened in both the affected and non-affected eye during the acute phase of CSC and that the choroidal thickness gradually decreases with regression of the disease. The EDI-OCT imaging measurements in this study also showed that the choroid thickness was increased in both the affected and normal fellow eyes but decreased significantly as SRD regressed. In addition to measuring choroidal thickness, various parameters from the OCT images, including SRFT, CRT, and CMT were measured manually and correlations were sought between the mfERG and OCT parameters. Previous studies that investigated patients with CSC using mfERG and OCT performed similar analyses to identify associations between the functional and structural Therefore, a thickened choroid implies that a pathological change in the choroid, compromised photoreceptor metabolism, and an ischemic outer retinal layer causes the retinal dysfunction. In this study, retinal dysfunction become more prominent as the choroid thickened.
The major limitation of this study is the lack of follow-up mfERG results. There has been debate as to whether retinal dysfunction is fully recovered after resolution of SRD.[10, 11] This study was not able to answer this question. A further prospective study that includes follow-up mfERG evaluation in a large number of patients with CSC is needed for a better understanding of retinal dysfunction in CSC.
The second limitation is that only the thickness of the choroid was measured, in spite of EDI-OCT allowing for in-depth observation of the choroid and the analysis was unable to compare the exact location between OCT images and mfERG measurements. In spite of these limitations, this study is the first to examine patients with acute CSC using a combination of EDI-OCT and mfERG and found bilateral diffuse impairment of the retinal response.
The retinal response of the unaffected eye was associated with regression of SRD, suggesting that the mfERG results for the unaffected eye may serve as an indicator of spontaneous regression. By applying the mfERG in routine clinical examination for CSC, patients will be able to obtain more precise information regarding their course of disease. Furthermore, pathological changes in the choroid identified by EDI-OCT had a negative correlation with retinal function measured with mfERG.
These findings imply that structural and functional impairment of the choroid plays an important role in the pathogenesis of CSC. Further evaluation using EDI-OCT and mfERG will improve current understanding of the choroid in patients with CSC and explain the subtle variation in disease progression from patient to patient.

Ethical approval and consent to participate
The study was approved by the ethical committee of Dongtan Sacred Heart Hospital. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the study.

Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Conflicts of interest
None.

Consent for publication
Not applicable.

Author contributions
In Hwan Hong and Jae Ryong Han designed and conducted the study.
In Boem Chang collected the data.
In Hwan Hong and In Boem Chang analyzed and interpreted the data.
In Hwan Hong, In Boem Chang, and Jae Ryong Han prepared, reviewed, and approved the manuscript.