Highlights
-
•
Inner choroidal fibrosis (ICF) is a marker of severity in complex chronic central serous chorioretinopathy (CSC) eyes. It has distinct imaging characteristics and can be of 2 morphological subtypes.
-
•
ICF lesions persist and enlarge over long-term follow-up. None of the cases show resorption. Eyes with CSC with ICF have poor visual outcomes, persistent fluid and limited therapeutic response.
-
•
Long term analysis with indocyanine green shows that remodeling of the inner choroid due to hypoxia is a dynamic continuous process which leads to ICF. Our study sheds light on this yet unexplored distinct entity in chronic choroidal diseases.
Purpose
To study imaging characteristics of inner choroidal fibrosis (ICF) in eyes with chronic central serous chorioretinopathy (CSC) and report long term changes.
Design
Retrospective interventional case series.
Methods
Records of chronic CSC patients between January 2004 and September 2024 were reviewed to identify and study longitudinal changes in eyes with ICF. ICF was divided into morphological types: type 1 and type 2. Type 1 ICF represented disorganized inner choroid without well-defined accumulations, while type 2 ICF represented accumulations with a demarcation line.
Results
Thirty-eight eyes of 23 patients were identified. All the eyes had complex CSC, with a mean of 1.7 lesions per eye. Patients were followed up for a median period of 36.5 months. The mean BCVA reduced from 0.67 ± 0.46 logMAR (20/95) to 0.78 ± 0.44 logMAR (20/121). Mean BCVA at time of presentation (0.83 logMAR, 20/135 vs 0.51 logMAR, 20/65; P =.1), as well as at last visit (0.88 logMAR, 20/152 vs 0.66 logMAR, 20/91; P =.7) was worse in eyes with Type 2 as compared to Type 1 ICF. Resolution of CSC was seen in 5 eyes (12.8%) at the last visit. There was a non-significant decrease in SFCT (456 vs 444 microns, P =.8) and significant increase in hypo-fluorescence area on indocyanine green angiography of 4.25 sq. mm ( P <.00001). Higher duration of follow-up and persistent fluid were associated with increase of size in ICF on multivariate regression analysis.
Conclusion
ICF in complex CSC was associated with persistent fluid and poor response to therapy. It was seen to increase in size with increasing chronicity.
INTRODUCTION
C entral serous chorioretinopathy (CSC) is a posterior segment disease characterized by localized and limited serous detachments of the neurosensory retina often associated with focal detachments of an altered retinal pigment epithelium (RPE). ,, While the disease was initially described in the 19th and 20th centuries. ,,, Donald Gass was the first to classify CSC as a choroidopathy, and described its clinical and diagnostic features along with its natural history and treatment options.
CSC eyes undergo several changes over time. and there are well defined imaging biomarkers which help us identify the chronic form of the disease. We recently identified a potentially novel marker of chronicity, termed “inner choroidal fibrosis” (ICF). which manifests in eyes with complex CSC as a subretinal grey-white lesion. On swept-source or enhanced depth imaging optical coherence tomography (OCT), the lesion exhibits a characteristic region of hyper-reflectivity below the RPE, which distorts and may push away the surrounding choroidal vessels. The lesion appears hypo-autofluorescent on autofluorescence (AF) imaging and hypo-fluorescent during the early phases of fluorescein angiography (FA), though it may stain in the recirculation phases. Additionally, it remains as a persistent hypo-fluorescent lesion on indocyanine green angiography (ICGA). ICF is always located below the RPE, which helps differentiate it from any subretinal pathology. Other mimickers of ICF include choroidal osteoma and serous maculopathy due to atypical choroidopathy (SMACH) which can be differentiated with multimodal imaging. Interestingly, ICF appeared to have 2 morphological types upon analysis of a greater number of cases. While some cases presented with deposits that had well-defined boundaries, others did not have any clear separation from the surrounding choroidal vessels, on OCT. Also, it was unclear whether these 2 entities had different clinical significance. Cases of both types of ICF had the same features on autofluorescence, dye-based angiography and OCT angiography (OCTA). Our current study extends our previous report, involving a much larger number of cases followed over many years. We seek to address some unanswered queries regarding the origins, behavior, associations and follow-up of these inner choroidal changes.
METHODS
This was a retrospective, interventional, longitudinal case series done in eyes with a diagnosis of chronic CSC, presenting to a multitier ophthalmology hospital network. Records of patients presenting between January 2004 and September 2024 were reviewed. All files were screened to look for the presence of ICF. Eyes with poor documentation of follow-up or poor OCT image quality that prevented visualization of the choroid were excluded. CSC was classified as either simple or complex according to the latest multimodal imaging-based classification by the Central Serous Chorioretinopathy International Group. ICF lesions were confirmed when they met all the criteria described in our initial report. Pertinent images were reviewed by the graders (SH and NKS). In case of any discrepancy, it was resolved by the senior adjudicator (JC). All included cases had a subretinal grey lesion, which was hypoautofluorescent and showed persistent hypofluorescence on ICGA. The OCT through the lesion showed a region of hyper-reflectivity below the RPE, helping differentiate it from any subretinal lesion. It can also be distinguished from 2 of its main mimickers—SMACH and choroidal osteoma. SMACH typically has a characteristic stellate appearance, while choroidal osteoma is usually unilateral, unifocal, larger, and appears at an earlier age. Furthermore, all cases of ICF were identified in eyes with a history and/or imaging features of CSCR, which is uncommon among the other 2 entities.
ICF was classified into 2 types:
-
1.
Type 1 inner choroidal fibrosis (Type 1 ICF) ( Figure 1 ): Clinically appears as a subretinal grey zone without sharp boundaries. OCT shows disorganization in the inner choroid without a well-defined border separating it from the mid and outer choroidal layers. The outer larger choroidal vessels do not get pushed outward from this lesion. On OCTA, these areas show patchy loss of de-correlation signal. FA and ICGA show corresponding persistent areas of hypo-fluorescence.
FIGURE 1 A. Color fundus photograph of the right eye shows the macula to have a yellow grey sheen due to a subretinal lesion (dashed blue circle). The foveal reflex is blunted and subretinal pigmentary changes can be appreciated (blue circle). B. Confocal autofluorescence shows mottling in the macula. C. The recirculation pahse of the fluorescein angiogram and D. The mid phase of indocyanine green angiogram shows persistence hypofluorescence in the macula (blue circles respectively). This corresponds to the area of pigmentary changes seen in panel A. E. The choriocapillaris slab of the OCT-angiogram shows flow voids in the macula corresponding to the area of persistent hypofluorescence seen in panel C and D. F. The vertical line scan of the swept source optical coherence tomogram passing through the area of hypofluorescence on the indocyanine green angiogram shows an area of hyper-reflectivity in the inner choroid (blue box). This area is distinct from the surrounding choroidal tissue and the choriocapillaries cannot be seen within it. Vessels of the choriocapillaris and Sattler’s layers can be seen terminating at the edges of the hyper-reflective disorganized tissue. Cystoid macular degeneration, loss of outer retinal layers, subretinal fluid and pachyvessels can also be seen. The multimodal imaging findings are characteristic of type 1 inner choroidal fibrosis. In this case the patient had a follow up of 7 years, during which time despite treatment his disease did not resolve, leading to a final visual acuity of 20/250.
-
2.
Type 2 inner choroidal fibrosis (Type 2 ICF) ( Figure 2 ): Clinically appears as subretinal grey-white macular scar with a well-defined sharp boundary. OCT shows homogeneous hyperreflectivity of the inner choroid, pushing the outer choroidal vessels outward, and is associated with a well-defined border of greater reflectivity than the lesion itself. OCTA shows the absence of a decorrelation signal. FA and ICGA show a corresponding area of well-defined hypo-fluorescence.
FIGURE 2 D, F. Color fundus photograph of the right eye and left eye shows well demarcated areas of subretinal scarring (yellow and blue circles respectively). A, G. These areas are absolutely hypoautofluorescent (yellow and blue circle respectively). B, H. Corresponding lesions show persistent hypofluorescence up to the mid phase of the indocyanine green angiogram (yellow and blue circle respectively). E, J. The line scans of the swept source optical coherence tomogram passing through the lesions shows homogenous areas of hyper-reflective accumulation in the inner choroid with a well-defined boundary, with the surrounding pachyvessels pushed away (yellow and blue box respectively). These lesions are of type 2 inner choroidal fibrosis. I. The optical coherence tomogram line scan passing through the fovea in the left eye shows disorganization in the inner choroid (red box). Subretinal fibrosis can also be seen (red arrow) which corresponds to the subretinal lesion seen temporal to the optic disc in panel F. C. The optical coherence tomogram line scan passing through the fovea in the right eye does not show any changes in the inner choroid.
Macular neovascularization (MNV) was identified based on OCT, FA, ICA, and OCTA findings as well as retrospective clinical records.
IMAGING PROTOCOL
The ICF lesions (nonoverlapping discrete lesions based on ICGA images) were counted and categorized as per the location, using clinical photograph and confirmed using the mid-phase of the ICGA as follows: foveal (when the lesion involves the foveal avascular zone [FAZ]), parafoveal (lesion border within 500 microns of the FAZ border), peripapillary (lesion border within 500 microns of the optic disc border), macular (lesions in the macula which could not be classified as foveal, parafoveal or peripapillary), or extramacular. SS-OCT was performed using the Topcon DRI Triton SS-OCT (Topcon, Tokyo, Japan), while spectral-domain OCT was carried out using the enhanced depth imaging (EDI) mode of Zeiss Cirrus HD OCT (Carl Zeiss Meditec, Dublin, CA) and Heidelberg Spectralis OCT (Heidelberg Engineering, Heidelberg, Germany). The subfoveal choroidal thickness (SFCT) was measured using the in-built caliper tools from individual consoles, by determining the distance between the inner border of Bruch’s membrane and the sclera-choroidal junction. The mean of the measurements obtained from vertical and horizontal line scans was calculated. ICGA was performed using the HRA 2 or Heidelberg Spectralis HRA + OCT (Heidelberg Engineering).
ICGA images at baseline and final visit were used to measure the area of hypo-fluorescence corresponding to the ICF lesion on the fundus image. The freehand measuring tool (Heyex software viewing module) was used to calculate the lesion size at each visit during the mid-phase of the angiogram. The mid-phase of ICGA was defined as frames seen between 5 and 7 minutes of dye injection. In cases with ill-defined borders, the minimum area encompassing the entire hypo-fluorescent lesion was included. Similarly, for multiple lesions, the area of individual lesions was measured separately and added. The measurements were performed by a single grader (SH), after establishing good inter-observer agreement (intraclass correlation coefficient > 0.8) between 2 graders (SH and NKS) in a sample of 20 eyes. Eyes were also evaluated for the presence of any macular choroidal venous anastomosis on ICGA. The degree of choroidal vascular hyperpermeability (CVH) in the 44-degree posterior pole centered on the fovea was classified as: (1) Uni FISH (Unifocal Indistinct Signs of Hyperpermeability), (2) Multi FISH (Multifocal Indistinct Signs of Hyperpermeability), or DISH (Diffuse Indistinct Signs of Hyperpermeability). Any degree of confluent CVH ≥ 1 disc diameter was classified as DISH.
STATISTICAL ANALYSIS
In patients with bilateral involvement, both eyes were taken into analysis. Continuous variables were expressed as mean ± SD. Before and after comparisons were done using Wilcoxon signed rank test and Paired T test. Differences between groups were assessed with Unpaired T test and Mann-Whitney U test. Difference of proportion of findings between groups were assessed with the Chi Square test. Linear function of generalized estimated equations was used for analyzing factors predicting final visual acuity and change in ICF area. Statistical analysis was performed using R studio (v2024.4.2). For analysis, Snellen visual acuity values were converted into logMAR units. A P -value of less than.05 was considered statistically significant.
RESULTS
Out of the 255 eyes with chronic CSC screened, 38 eyes from 23 patients with ICF were included in the study (prevalence of 14.3%). The mean duration of symptoms was 50 months (±56 months) (Range 3-835 weeks). The mean age of diagnosis of CSC was 52 ± 7.1 years, while the mean age at which ICF was diagnosed was 52.9 ± 7.3 years. A total of 9 patients (36.5%) had unilateral ICF, while 14 patients (61.5%) had bilateral ICF. All patients were male and had complex CSC. Various risk factors for CSC, and treatment administered in these eyes have been enumerated in Table 1 . The SS-OCT features and ICGA findings have been described in Table 2 .
TABLE 1
Risk Factors and Treatment Administered for Central Serous Chorioretinopathy.
| Risk Factors for CSC Noted Amongst 23 Patients | |
|---|---|
| Hypertension | 15 (65.2%) |
| Type A personality | 4 (17.3%) |
| History of corticosteroid use | 4 (17.3%) |
| History of tobacco consumption | 4 (17.3%) |
| Treatment undertaken for CSC over median period of follow up of 36.5 months (range 1.5-192 months) (n = 38 eyes) | |
| Oral MRcA | 24 (63.1%) |
| Anti-VEGF injection | 21 (55.2%); Mean of 1.9 injections per eye |
| MPL | 16 (42.1%); Mean 1.8 sessions per eye |
| Focal laser | 16 (42.1%); Mean 1.4 sessions per eye |
| PDT | 7 (18.4%) |
| External SRF drainage | 1 (2.6%) |
| History of MNV over median period of follow up of 36.5 months (n = 38 eyes) | |
| MNV detected | 8 (21.1%) |
| Mean anti-VEGF injections over total period of follow up per eye | 2 |
| Mean anti-VEGF injections after detecting ICF per eye | 1.1 |
| MNV noted in non ICR eye (n = 9) | 3 (33.3%) |
| Mean anti-VEGF injections per eye | 3 |
CSC = central serous chorioretinopathy; ICF = inner choroidal fibrosis; MNV = macular neovascularization; MPL = micro pulse laser; MRcA = mineralocorticoid receptor antagonist; PDT = photodynamic therapy; SRF = subretinal fluid; VEGF = vascular endothelial growth factor.
TABLE 2
Swept Source Optical Coherence Tomography and Indocyanine Green Angiography Features of Eyes With Inner Choroidal Fibrosis.
| SS-OCT Features at Baseline (n = 38 Eyes) | |
|---|---|
| Mean SFCT (Microns) | 456 (SD ± 112) |
| Mean sublesional CT (Microns) | 443 (SD ± 112) |
| SRF at fovea | 20 (52.6%) |
| SRF at lesion | 21 (55.2%) |
| IRF | 23 (60.5%) |
| DLS | 31 (81.5%) |
| FCE | 14 (36.8%) |
| Outer retinal disruption at fovea | 15 (39.4%) |
| Outer retinal disruption at lesion | 22 (57.8%) |
| SS-OCT features at last follow up (Median period of follow up of 36.5 months) (n = 38 eyes) | |
| Mean SFCT (Microns) | 444 (SD ± 90) |
| Mean sublesional CT (Microns) | 418 (SD ± 101) |
| SRF at fovea | 20 (52.6%) |
| SRF at lesion | 13 (34.2%) |
| IRF | 25 (65.7%) |
| DLS | 26 (68.4%) |
| FCE | 15 (39.4%) |
| Outer retinal disruption at fovea | 26 (68.4%) |
| Outer retinal disruption at lesion | 28 (73.6%) |
| Features at first instance of ICGA (n = 33 eyes) | |
| Uni FISH | 0 |
| Multi FISH | 11 (33.3%) |
| DISH | 22 (66.7%) |
| Mean area of persistent hypofluorescence (sq.mm) | 6.07 (SD ± 5.4) |
| Vortex vein anastomosis | 4 (12.1%) |
| Features at last instance of ICGA (n = 22 eyes) | |
| Uni FISH | 0 |
| Multi FISH | 5 (22.7%) |
| DISH | 17 (77.2%) |
| Mean area of persistent hypofluorescence (sq.mm) | 9.9 (SD ± 8.5) |
| Vortex vein anastomosis | 5 (22.7%) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
