Retinal atrophy in idiopathic central serous chorioretinopathy1

Retinal Atrophy in Idiopathic Central Serous Chorioretinopathy MARIA S. M. WANG, MD, BIRGIT SANDER, PHD, AND MICHAEL LARSEN, MD, DMSC

● PURPOSE:

To study retinal atrophy in idiopathic central serous chorioretinopathy (ICSC). ● DESIGN: Observational case series. ● METHODS: Twenty-four eyes in 23 consecutive patients aged 33 to 67 years and 50 eyes in 25 healthy volunteers aged 24 to 70 years were studied. Observational procedures included: biomicroscopy, fundus photography, fluorescein angiography, optical coherence tomography, photocoagulation. Retinal thickness after resolution of detachment, definite attenuation being defined as foveal thickness less than the mean ⴚ2 SD in healthy volunteers; duration of symptoms; best-corrected visual acuity (BCVA). ● RESULTS: Definite retinal attenuation, retinal thickness ranging from 51% to 74% of normal, was found in nine eyes of nine patients with idiopathic central serous chorioretinopathy (ICSC). Their duration of symptoms was longer (P ⴝ .0014) and their BCVA was lower (P ⴝ .015) than eyes in patients with normal-range foveal thickness who, nevertheless, had thinner foveae than healthy subjects (P ⴝ .10). The most severe reduction of foveal thickness, to half the normal (three eyes of three patients), was associated with BCVA 0.5 or less, multifocal retinal pigment epithelial abnormalities, and more than 10 years’ duration of symptoms. Retinal attenuation was seen only after a duration of symptoms of more than 4 months, being most pronounced in the photoreceptor layer, particularly in the fovea. ● CONCLUSION: Foveal attenuation in ICSC is associated with more than 4 months’ duration of symptoms and persistent BCVA reduction despite resolution of the serous detachment. We found no other likely cause of atrophy than the prolonged absence of contact between photoreceptors and retinal pigment epithelial cells. (Am J Ophthalmol 2002;133:787–793. © 2002 by Elsevier Science Inc. All rights reserved.) Accepted for publication Feb 12, 2002. InternetAdvance publication at ajo.com Feb 28, 2002. From the Department of Ophthalmology, Herlev Hospital, University of Copenhagen, Copenhagen, Denmark. Reprint requests to Maria S. M. Wang, MD, Department of Ophthalmology, Herlev Hospital, University of Copenhagen, Herlev Ringvej 75, DK-2730 Herlev, Denmark; fax: (⫹45) 4 494-1041; e-mail: loja@ herlevhosp.kbhamt.dk 0002-9394/02/$22.00 PII S0002-9394(02)01438-1

©

2002 BY

I

DIOPATHIC CENTRAL SEROUS CHORIORETINOPATHY

(ICSC) is characterized by retinal pigment epitheliopathy, often highly localized, and a secondary involvement of the neurosensory retina, which is passively detached from the retinal pigment epithelium (RPE) by the accumulation of subretinal fluid. The notion that the RPE is the focus of the disease process is corroborated by the prompt resolution of the detachment following focal photocoagulation treatment of the RPE at the source of leakage. Best-corrected visual acuity (BCVA) may be normal even in the detached state and, if not, it will often return to normal after resolution of the detachment, as do the associated dyschromatopsia and relative scotoma. This constellation of findings is typical of cases of recent onset,1,2 whereas long-standing or recurrent cases often are characterized by mottling of larger areas of the RPE and a more severely and irreversibly depressed visual dysfunction.3 The mechanisms underlying the transition from early focal disease to the more complex and widespread changes of the RPE in chronic ICSC with functional involvement of the neurosensory retina, and its structural correlate2,4 – 6 have not been elucidated. We have attempted to study this process by examining the structure of the retina after resolution of ICSC using optical coherence tomography.

METHODS ALL PATIENTS DIAGNOSED WITH ICSC AT THE HERLEV

Hospital between February 1, 1998, and October 1, 2000, were included in the study. The diagnosis required a visible or clinically suspected serous neuroretinal detachment of any part of the macula, documented by optical coherence tomography, accompanied by one or more focal pigment epithelial changes, visible biomicroscopically or by fluorescein angiography, in the absence of evidence of subretinal new vessels, postinflammatory chorioretinal scarring, posterior segment tumors, vitreoretinal traction, epiretinal membranes, subretinal hemorrhage, drusen, and other causes of serous detachment unrelated to ICSC. Only cases with a follow-up of at least 4 months were included. Intra-study exclusion criteria included visible subneuro-

ELSEVIER SCIENCE INC. ALL

RIGHTS RESERVED.

787

retinal deposits of amorphous material after resolution of the detachment (eight eyes) and persistent neuroretinal detachment at the end of the study (five eyes). All patients except one had monocular serous detachment at presentation, but most patients had fluorescein angiographic evidence of bilateral pigment epithelial changes. The one patient with bilateral serous detachment had a 25-year history of poor vision in the right eye and recent recurrence of complaints in the left eye. The clinical examination included best-corrected Snellen visual acuity determination, slit-lamp biomicroscopy using a 90-diopter precorneal lens, and a Goldmann three-mirror contact lens, Amsler grid testing, applanation tonometry, and manual perimetry. Photocoagulation treatment using a 532-nm continuous-mode laser was offered principally to patients with more than 3 months’ duration of symptoms. Fifteen of 18 patients accepted. Digital black- and -white fundus photography in red-free illumination and intravenous fluorescein angiography with late-phase frames up to 40 minutes after injection were performed using a Topcon TRC-50X (Topcon Corp., Tokyo, Japan) retina camera equipped with a Kodak Megaplus model 1.4 (San Diego, California, USA) digital backpiece and a PC-based image handling system (Ophthalmic Imaging Systems, Inc., Sacramento, California, USA). Optical coherence tomography was performed using a Humphrey–Zeiss instrument (Zeiss AG, Oberkochen, Germany) with a nominal axial resolution of 10 ␮m.1 The thickness of the attached neuroretina was defined as the distance between its inner and outer surfaces as determined by the manufacturer’s software version CA A4.1. The algorithm automatically traces the vitreoretinal interface and the interface between the minimally reflecting photoreceptor layer and the highly reflective RPE. The height of any neurosensory retinal detachment, defined as the distance between the outer boundary of the photoreceptor layer and the inner boundary of the pigment epithelium, was measured by the same software pattern recognition utility on aligned and normalized scans. Horizontal and sagittal transfoveal scans were made using nominal scan widths of 2.83 mm. The three best out of 5 to 10 scans from each session were used for further analysis. Data analysis was based on retinal thickness measurements at the center of the fovea and on either side of the fovea at horizontal distances of 0.5 mm and 1.0 mm from the foveal center. The criterion for assigning a patient into the atrophic or the normal range group was the retinal thickness at the center of the fovea. Reference data were obtained in 50 eyes of 25 healthy volunteers aged 44 ⫾ 15 years (mean ⫾ SD; range, 24 –70 years) with no history of eye disease or chronic systemic disease and refractive errors less than ⫾5 diopters. The scan-to-scan reproducibility of the retinal thickness measurement was 7%. Over a range from 20 to 60 years of age there was no significant change in foveal thickness (P ⫽ 788

AMERICAN JOURNAL

FIGURE 1. Foveal thickness (mean ⴞ 2 SD) in 50 eyes in 25 healthy subjects (left) compared with 15 eyes in 15 patients with idiopathic central serous chorioretinopathy (middle) in whom foveal thickness after resolution of the detachment was within the normal range and nine eyes in nine patients with the same disease (right) in whom foveal thickness was outside the normal range, defined as the mean ⴚ 2 SD in healthy subjects.

.14). The normal retinal thickness range was defined as the mean ⫾ 2 SD. For the center of the fovea the mean value ⫾ 2 SD in healthy subjects was 157 ⫾ 38 ␮m, that is, ranging from 76% to 124% of the mean value. The study was approved by the medical ethics committee of Copenhagen County. A dichotomous analysis was made of all data from ICSC patients using data from healthy subjects to define eyes with ICSC as being within the normal range or, if the mean of three foveal thickness measurements was found to be lower than the mean—2 SD of healthy subjects (that is, 119 ␮m), as being atrophic. All other variables were then compared between the normal range and atrophic groups using the t test on log-transformed data and the Pearson correlation test. Regression analysis was made on data from the entire ICSC population. The level of statistical significance was set at P values lower than .05.

RESULTS FOVEAL ATROPHY WAS FOUND IN 9 (37.5%) OUT OF 24 EYES

(Figure 1) after complete resolution of the serous detachment following extrafoveal photocoagulation treatment (eight patients) or occurring spontaneously (one patient). The central foveal thickness after resolution in these 9 eyes was 96 ⫾ 13 ␮m (mean ⫾ SD; range, 80 to 116 ␮m, P ⫽ .001 compared with healthy subjects; Figure 2). The most pronounced attenuation of the fovea, to nearly 50% of normal central foveal thickness, was found in three patients with an ICSC-compatible history of more than 10 years’ duration and a chronic type of multifocal retinal pigment epithelial changes (Table 1, Figure 3). Although the resolution of the serous detachment was accompanied by subjective reduction of the central scoOF

OPHTHALMOLOGY

JUNE 2002

FIGURE 2. Retinal thickness in relation to horizontal distance from the center of the fovea in the nasal (left) and temporal (right) direction. The mean value (diamonds) and the ⴞ 2 SD range (bars) in 50 eyes in 25 healthy subjects are shown together with the individual values (other markers) of nine eyes in nine patients with idiopathic central serous chorioretinopathy and definite retinal atrophy defined as a central foveal thickness below the mean ⴚ2 SD range of healthy subjects. The attenuation of the retina was generally most pronounced in the center of the fovea.

toma and improved dark adaptation in all three of these patients, metamorphopsia persisted in two of the three eyes and BCVA remained 0.5 or less in all (Table 1). The remaining six of the nine patients with definite foveal attenuation had a central retinal thickness of 60% to 74% (Figure 4) of the normal and an ICSC-compatible history of 0.3 to 4.0 years. Resolution of the serous detachment followed to photocoagulation (five patients) or spontaneously (one patient). The pretreatment height of the serous detachment in the nine atrophic eyes was 125 ⫾ 56 ␮m (range, 57–211 ␮m). Normal-range foveal thickness after resolution of the serous detachment in ICSC was found in 15 (62.5%) eyes with a foveal thickness of 147 ⫾ 23 ␮m (mean ⫾ SD; range, 120 –181 ␮m, P ⫽ .10 compared with healthy subjects). These 15 eyes had post-reattachment BCVAs of 20/32 or better, a duration of symptoms of 1 week to 1.5 years at presentation, and a pretreatment serous detachment height of 145 ⫾ 102 ␮m (range, 51–333 ␮m). Resolution of the detachment followed photocoagulation treatment (eight eyes) or occurred spontaneously (seven eyes). As in the atrophic group of eyes, the attenuation affected primarily the photoreceptor layer and was most pronounced in the center of the fovea. The post-reattachment thickness of the retina did not differ between photocoagulation-treated and spontaneously resolved cases (P ⫽ .26). In patients with post-reattachment foveal attenuation the mean duration of symptoms was 6 years. This was markedly longer than the 5 months mean duration of symptoms found in patients without foveal attenuation (P ⫽ .0014; Table 1). Patients with foveal attenuation also had lower mean BCVA than those without foveal attenuation (P ⫽ .015). The shortest duration of symptoms at VOL. 133, NO. 6

presentation associated with foveal attenuation was 4 months. Analysis of continuous variables after resolution of the serous detachment in the total population of 24 ICSC eyes demonstrated significant negative correlations between foveal thickness and duration of symptoms (r ⫽ .56; Figure 5, left) and between BCVA and duration of symptoms (r ⫽ .68; Figure 5, right).

DISCUSSION WE HAVE DOCUMENTED THE PRESENCE OF RETINAL ATRO-

phy after resolution of the serous detachment in a subgroup of eyes with ICSC. In eyes with foveal atrophy, the mean duration of symptoms was significantly longer (6 years) than in eyes without evidence of foveal atrophy (5 months). Additionally, foveal atrophy was associated with lower BCVA at the final visit (20/32) than was a normal range foveal thickness (20/20). Overall, both foveal thickness and BCVA were proportional to the duration of symptoms. Retinal atrophy was found only in relation to presumed characteristics of chronic or recurrent ICSC, such as multifocal irregular pigment epithelial changes and absence of visible point source leakage on fluorescein angiography. It is unknown whether the transition from monofocal or paucifocal ICSC to the irregular multifocal mottling of the RPE seen in chronic ICSC is a cause or a consequence of the long-standing detachment of the neurosensory retina. In the present study, however, patients with a duration of symptoms of less than 4 months and monofocal retinal pigment epithelial lesions invariably have normal foveal thickness after the resolution of the

RETINAL ATROPHY

IN

ICSC

789

790

TABLE 1. Clinical and Optical Coherence Tomographic Characteristics of Patients With Idiopathic Central Serous Chorioretinopathy (ICSC) at Presentation and After Resolution of Serous Detachment (Retinal Atrophy Was Defined as Central Foveal Thickness Less Than the Mean ⫺ 2 SD of Healthy Subjects)

At Presentation

Patient No., Age [yrs], Gender (M/F)

AMERICAN JOURNAL OF

OPHTHALMOLOGY

ICSC, normal range 1 (OS) 67 M 11 43 M 12 41 M 13 46 M 14 55 F 15 44 F 16 45 M 17 45 M 18 44 F 19 41 M 20 55 M 21 56 M 22 36 M 23 36 F 24 35 M Mean 46 SD 9 ICSC, atrophic 1 (OD) 67 M 2 52 M 3 53 M 4 66 M 5 41 M 6 35 M 7 42 M 8 57 M 9 33 M Mean 50 SD 13

Duration of Symptoms (yrs)

ICSC Episode (no.)

1.5 0.5 0.2 0.2 0.25 0.2 0.02 0.2 1 0.5 0.2 1.5 0.02 0.03 0.5 0.45 0.49

2 2 1 1 1 1 1 1 2 1 1 1 1 1 1

25 10 10 0.5 1.5 0.3 1 4 0.5 5.9§ 8.2

5 1 2 4 2 1 1 3 2

†

RPE Lesion Type on IVFA* Monofocal

Multifocal

x x x x x x x x x x x x x x x

x x x x x x x x x

Sidedness B ⫽ bilateral, U ⫽ unilateral

Height of Detachment (␮m)

B U U U U U U U B U U U B U U

B U B B B U B U B

Final Examination Foveal Visual Thickness Acuity (␮m)

Visual Acuity

Photocoagulation Treatment

Followup (mos)

57 138 66 295 51 333 182 226 182 58 66 66 51 299 98 145 102

20/25 20/25 20/25 20/63 20/25 20/25 20/20 20/20 20/40 20/25 20/25 20/100 20/25 20/63 20/20 20/32

Y N N N Y N N Y Y Y Y Y N Y N

32 6 6 9 10 6 6 12 9 9 6 18 6 9 6 10.0 6.9

20/20 20/16 20/20 20/16 20/20 20/20 20/16 20/20 20/25 20/16 20/20 20/32 20/20 20/25 20/20 20/20

124 126 120 126 181 178 134 137 141 165 149 178 153 122 173 147 23

72 126 57 N/A 131 196 124 211 124 125 56

20/200 20/32 20/100 20/32 20/50 20/25 20/25 20/40 20/20 20/40

32 6 12 6 15 6 27 6 6 12.9 10.0

20/100 20/125 20/40 20/20 20/20 20/20 20/25 20/32 20/25 20/32§

80 82 80 94 116 106 98 110 102 96§ 13

Y Y Y N Y Y Y Y Y

JUNE 2002

ICSC ⫽ idiopathic central serous chorioretinopathy; IVFA ⫽ intravenous fluorescin angiography; RPE ⫽ retinal pigment epithelium; SD ⫽ standard deviation. *Type of retinal pigment epithelium lesion on intravenous fluorescein angiography. † Unilateral or bilateral retinal pigment epithelium lesions on fluorescein angiography. ‡ Central foveal thickness as measured by optical coherence tomography, percentage of mean of healthy subjects. § Significantly different from normal range group (P ⬍ .05).

Foveal Thickness, Relative to Mean of Healthy Controls (%)‡

79 80 76 80 115 113 85 87 90 105 95 113 97 78 110 94 14 51 52 51 60 74 68 62 70 65 61§ 9

FIGURE 3. Fundus fluorescein angiograms (mid-phase) from the right eye (upper left) and from the left eye (upper right) of a man aged 67 years (patient 1) with a documented history of idiopathic central serous chorioretinopathy in the right eye of 25 years and similar complaints of 1.5 years’ duration in the left eye. Best-corrected visual acuities (BCVAs) were 20/200 in the right eye and 20/25 in the left eye. Circumscribed areas of irregular pigment epithelial translucency indicate presumed chronic-type ICSC changes in both eyes. Transverse sections of the right macula by optical coherence tomography demonstrated foveal detachment and an overlying cystoid macula edema (second row left) before treatment and attenuation of the neuroretina after photocoagulationinduced resolution of the serous detachment (third row left). The left macula demonstrated detachment and splitting of the neuroretina before treatment (second row right) and attenuation of the neuroretina following photocoagulation-induced resolution of the serous detachment (third row right). The retina in both eyes is visibly thinner than in a healthy eye (bottom) after treatment; the foveal thickness in the right eye being reduced to 80 ␮m and the left eye to 124 ␮m (51% and 79% of the mean in healthy subjects). Also evident is a near-complete absence of the low-reflectance photoreceptor layer of the outer retina. Following treatment, BCVA in the right eye improved from 20/200 to 20/100 without any detectable visual field change. After treatment of the left eye, BCVA improved from 20/25 to 20/20 and a bothersome upper temporal paracentral scotoma spanning 10 ⴛ 15 degrees of the visual field and reaching within 2 degrees of the center of the visual field was reduced in extent to 5 ⴛ 8 degrees, the retraction being most pronounced in the direction away from the center of the visual field.

VOL. 133, NO. 6

RETINAL ATROPHY

IN

ICSC

791

in relating the duration of symptoms to the duration of disease in a predominantly monocular disease. Consequently, we propose that both retinal neurosensory and pigment epithelial atrophy are consequences of the prolonged absence of contact between the two layers of the retina in long-standing ICSC. Retinal atrophy was found to be most pronounced in the fovea, where early retinal pigment epithelial lesions are rare, but where a strong predilection for serous detachment is found, even when the source of leakage is extrafoveal.7 This supports that the atrophy is caused by the detachment and not by the primary disease process in ICSC. Furthermore, retinal attenuation was seen to be most pronounced in the photoreceptor layer of the retina by which is the layer directly affected by the interruption of intercellular contact. Permanent superior visual field loss associated with chronic and recurrent serous detachment of the inferior retina has previously been demonstrated.4 The duration of detachment required to induce retinal atrophy can only be crudely estimated because the reported duration of symptoms is an unreliable measure of disease duration, notably in patients with a healthy fellow eye, as was found in the majority of our patients. It seems safe, however, to conclude that clinically significant irreversible functional loss does not occur within 4 months of the onset of symptoms. The presence of multifocal pigment epithelial changes on fluorescein angiography should raise caution in evaluating the reported duration of symptoms because they appear to be associated with long-standing detachment. It remains to be demonstrated that intervention by photocoagulation treatment in the late stage of ICSC improves the long-term visual prognosis. The association between chronic serous detachment and retinal atrophy supports intervention, but caution should be advised in making the differential diagnosis between occult subretinal neovascularization and chronic ICSC. Photocoagulation treatment, notably if incompletely covering a neovascular membrane, is likely to accelerate the growth of subretinal new vessels. It may be difficult to verify or reject the suspicion of a serous foveal detachment by biomicroscopy in patients with symptoms of a central scotoma, especially when the detachment is lower than 100 ␮m.7 In such cases, optical coherence tomography may be particularly useful. Furthermore, foveal attenuation demonstrated by optical coherence tomography appears to be a strong indication that loss of central vision is attributable to foveal atrophy. In conclusion, we have found that in eyes with ICSC attenuation of the fovea after reattachment of the retina is associated with more than 4 months’ duration of symptoms. In such eyes, a persistent reduction of BCVA may be present despite resolution of the serous detachment. The proper treatment of chronic ICSC remains to be determined. In the present study population, we were successful in inducing resolution of the serous detachment by pho-

FIGURE 4. Fundus fluorescein angiogram (top) from the left eye of a 42-year-old man (patient 7) with central serous chorioretinopathy and a duration of symptoms of 1 year. The presence of a serous neuroretinal detachment with a height of 124 ␮m was confirmed by optical coherence tomography (middle). Three months after photocoagulation the detachment had resolved (bottom), the foveal thickness being 98 ␮m, which is only 62% of the mean thickness of 157 ␮m found in healthy subjects. When compared with a scan from a healthy subject (see Figure 3, bottom) the reduction in retinal thickness was seen to be attributable mainly to a pronounced attenuation of the low-reflectance photoreceptor layer.

serous detachment. Thus, all functional and morphologic parameters of foveal integrity within the study population demonstrate regression toward normal values with decreasing duration of symptoms, despite the inherent uncertainty 792

AMERICAN JOURNAL

OF

OPHTHALMOLOGY

JUNE 2002

FIGURE 5. Relationship between foveal thickness and duration of symptoms (left) and best-corrected visual acuity (BCVA) and duration of symptoms (right) in 24 eyes in 23 patients with idiopathic central serous chorioretinopathy examined after resolution of the serous foveal detachment occurring spontaneously (8 patients) or following photocoagulation treatment (16 patients). Regression analysis demonstrated a significant decrease in relation to the duration of symptoms of both retinal thickness (r ⴝ .68) and BCVA (r ⴝ .56).

tocoagulation treatment without evoking secondary subfoveal neovascularization. There is reason to be aware, however, that the risk of inducing choroidal neovascularization may be heightened by the diffuse pigment epitheliopathy seen in chronic ICSC.8 The study identified no other likely cause of atrophy than the prolonged absence of contact between the photoreceptors and the retinal pigment epithelial cells. ACKNOWLEDGMENTS

Maria S. M. Wang was supported by the Danish Eye Health Society (Vaern om Synet) and the Danish Association of the Blind (Øjenfonden).

REFERENCES 1. Hee MR, Puliafito CA, Wong C, et al. Optical coherence tomography of central serous chorioretinopathy. Am J Ophthalmol 1995;120:65–74.

VOL. 133, NO. 6

2. Gilbert CM, Owens SL,. Smith PD, Fine SL. Long-term follow-up of central serous chorioretinopathy. Br J Ophthalmol 1984;68:815– 820. 3. Yannuzzi LA. Type A behavior and central serous chorioretinopathy. Retina 1987;7:111–130. 4. Gass JDM, Little H. Bilateral bullous exudative retinal detachment complicating idiopathic central serous chorioretinopathy during systemic corticosteroid therapy. Ophthalmology 1995;102:737–747. 5. Bennet G. Central serous chorioretinopathy. Br J Ophthalmol 1955;39:605– 618. 6. Go´ mez-Ulla F, Seoane I, Labella F, Torreiro J, Ruiz C. An image analyzer study of central serous chorioretinopathy. Optom Vis Sci 1993;70:118 –122. 7. Wang M, Larsen M, Sander B, Lund-Andersen H. Central serous choroidopathy with foveal detachment demonstrated by optical coherence tomography. Acta Ophthalmol Scand 1999;77:402– 405. 8. Schatz H, Yannuzzi LA, Gitter KA. Subretinal neovascularization following argon laser photocoagulation treatment for central serous chorioretinopathy: complication or misdiagnosis? Trans Am Acad Ophthalmol Otolaryngol 1997;83:893–906.

RETINAL ATROPHY

IN

ICSC

793