Introduction: a short historical review

Central serous chorioretinopathy (CSC) was first described by von Graefe in 1866, who named it as ‘relapsing central luetic retinitis’.1 A variety of names have since been used to describe this idiopathic detachment of the neurosensory retina.2 In 1965 and 1967, the current terms of CSC3 and idiopathic central serous choroidopathy4 (ICSC) were first introduced to describe the same disease. The different terms adopted between 1866 and 19845 reflect the lack of understanding of the pathophysiology of CSC.

Initially investigators tried to explain CSC aetiology and pathophysiology based on the psychogenic-related hypothesis proposed by Horniker in 1927.6 He suggested that angioneurotic patients were susceptible to retinal angiospasm and exudation in the macula.

Maumenee7 was the first to describe a leak at the level of the retinal pigment epithelium (RPE) in 1965 using fluorescein angioscopy (FA). Subsequently, Gass4, 8, 9, 10, 11 provided detailed descriptions of the fluorescein angiographic characteristics of CSC.

Knowledge of the anatomy of the choriocapillaris-Bruch's membrane-RPE layer allowed ophthalmologists to surmise that there is a diffuse dysfunction of the RPE cells, the choroids, or both,12 regardless of the primary cause or the initiating event.

Infections,1, 13, 14, 15, 16, 17, 18 toxins,19, 20, 21, 22, 23 an immunological reaction,24, 25, 26 neuronal,2, 27, 28, 29, 30, 31 circulatory,32, 33, 34 and hormonal regulatory factors35, 36, 37, 38, 39 have all been implicated in the initiation of the damaging process that leads to CSC.

As a result of the multiple theories and despite CSC having a favourable natural course, various treatments have been proposed. Psychotherapy was suggested in 1948,40 followed by drug therapies such as α-adrenergic blockade,41 β-adrenergic inhibitors,42 and acetazolamide.43 Laser treatments have included retinal photocoagulation,44, 45, 46, 47 transpupillary thermotherapy (TTT),48 and currently photodynamic therapy (PDT).49, 50, 51, 52, 53 More recently, anti-VEGF agents54, 55, 56, 57 and corticosteroid antagonists58, 59 have also been investigated.

Pathophysiology of CSC

Choroid dysfunction theory

Gass60 suggested that a focal increase in the permeability of the choriocapillaris was the primary cause of damage to the overlying RPE in patients with ICSC. He suggested that this could create detachment of the RPE, serous retinal detachment and, in 10–15% of patients, serofibrinous subretinal exudation.

Guyer et al32 suggested a potential model for the pathogenesis of CSC based on ICG-videoangiography (ICG-V). They noted diffuse hyperpermeability around active leakage sites seen with ICG-V but not with FA. Therefore they concluded that hyperpermeability was at the level of the choroid rather than the RPE. They proposed that choroidal hyperpermeability causes serous detachments of the RPE, which can induce a rip or decompensation of the RPE. This subsequently causes RPE leakage, that is, diffusion of water, electrolytes, and proteins that leads to a neurosensory retinal detachment.

Alterations in choroidal circulation may also cause choroidal ischaemia. This was first noted by Hayashi et al61 who used similar diagnostic equipment and found areas of choroidal ischaemia as well as leakage of ICG dye from the choriocapillaris. Fluorescein angiography (FA) and indocyanine green angiography (ICG-A) with a scanning laser ophthalmoscope and a digital imaging system were performed to evaluate choroidal circulation changes in CSC by Prunte and Flammer.33 In their study, dilated capillaries and dilated draining venules in one or more choroidal lobules, following a localised delay in arterial filling, might explain choroidal hyperpermeability in the area of the damaged RPE. These observations are suggestive of a localised lobular inflammatory or ischaemic choroiditis.

However, the cause of the choroidal abnormality is still unknown. The answer may lie in changes of the autoregulation in the choroidal blood flow.12

Tittl et al62 suggested a persistent abnormal regulatory response was involved in the pathogenesis of chronic CSC. They found that subfoveal choroidal blood flow regulation in patients with chronic CSC was impaired. This dysregulation occurred after substantial functional restoration at least 6 months after the last episode. Measurements of ocular fundus pulsation in patients with newly diagnosed active CSC have previously provided evidence that choroidal perfusion in the macular region might be abnormal.63

RPE dysfunction theory

An alternative theory suggests that CSC results from dysfunction of the RPE (Figures 1a and b). This occurs following an undefined insult. It results in either a few impaired RPE cells or even a single RPE cell, which causes a reverse in fluid movement in a chorioretinal direction. This, in turn, leads to leakage of fluid in the subretinal space and finally to the development of a neurosensory retinal detachment.12 Spitznas64 suggested that focal damage to the RPE can reverse the direction of ion secretion and thus lead to greater fluid movement towards the retina than to the choroid.

Figure 1
figure 1

(a) Fundus and red-free photo of the left eye of a patient with recurrent idiopathic CSC associated with depigmentation of the RPE in the macular area as well as with small PEDs. (b) FFA of the above patient showing window defect and a focal RPE leak inferiorly to it.

Full size image

A limitation of many theories of the aetiology of CSC is the lack of a suitable animal model to test hypotheses. In a rare example of the creation of an animal model of CSC, Negi and Marmor65 made small non-rhegmatogenous retinal detachments (blebs) in rabbits over regions of RPE. The RPE was damaged mechanically or by laser photocoagulation. Focal RPE damage appeared to facilitate water movement from, rather than into, the subretinal space.

Marmor66 also postulated that there is a more diffuse RPE metabolic dysfunction and that a focal RPE ‘leak’ can overload the system so that the serous fluid accumulates and persists (Figure 2).

Figure 2
figure 2

Multiple focal areas of staining (multiple zones of hyperfluorescence), more prominent in the juxtapapillary region, indicative of chronic ICSC, corresponding with areas of RPE atrophy, caused by recurrent serous retinal detachment.

Full size image

Combined choroid and RPE dysfunction theory

Alternatively there could be a combination of increased fluid leakage from the choriocapillaris and impaired RPE function.12 A persistent choriocapillaris abnormality could lead to prolonged stress of the RPE cells, which would not be able to pump in a retinochoroidal direction and therefore fluid would accumulate and cause a neurosensory detachment.

Serous retinal detachment and pigment epithelial detachment associated with CSC

Decompensation of the RPE in CSC results in neurosensory retinal detachment, serous pigment epithelial detachment (PED), and RPE atrophy. PED may result from any number of choroidal disorders that disrupt the normal junction between the basement membrane of the RPE and the inner collagenous layer of Bruch's membrane. Thus, serous fluid from the underlying choriocapillaris gains access into the sub-RPE space. Fluorescein and ICG-A show early hyperfluorescence of the PED, which persists throughout the angiogram, demonstrating late pooling. Leakage into the sensory retina occurs only in cases of concurrent serous retinal detachment. Previously idiopathic serous detachment of the RPE was characterised as a ‘benign’ variant of idiopathic CSC, in terms of the final visual outcome,67, 68, 69 whereas in others the presence of PED was associated with poor visual recovery.69, 70, 71 The combination of PED and serous retinal detachment increases the probability of CSC diagnosis. Therefore, the definition of terms ‘serous retinal detachment’ and ‘pigment epithelial detachment’ is helpful in the understanding of CSC pathogenesis as well as of great importance for clinicians.

CSC: a bilateral disease with systemic associations

More recently, the use of three-dimensional optical coherence tomography (Fourier/spectral domain OCT) has helped investigators to study morphological alterations of the RPE both in symptomatic and asymptomatic eyes of the CSC patients.72, 73, 74, 75, 76 Not surprisingly, RPE irregularities, at the site of leakage, were noted in symptomatic eyes.76 However, Gupta et al77 studied three-dimensional single-layer RPE scans and also found that the majority of asymptomatic eyes of CSC patients also showed an uneven RPE surface. This finding was not present in control eyes of healthy volunteers. The authors postulated that accumulation of sub-RPE fluid along with RPE dysfunction, which results in the formation of RPE bumps, visible on spectral-domain OCT, may represent a preclinical or subclinical stage of the disease.

The above findings agree with previous observations, noted with multifocal electroretinogram (mERG) and ICG-V, respectively, that mERG changes and choroidal permeability changes are present in both affected and fellow eyes of CSC patients.78, 79

The first systematic investigation of the relationship between a type A behaviour pattern (quickness to anger, competitiveness, and need to be in control) and macular disease was conducted by Yannuzzi.2 This was the first cross-sectional study that employed strict clinical definitions and matched controls to assess CSC patients to classify them as a type A behaviour pattern. The latter was statistically more frequent in CSC patients than in both the control groups used in this study.

Experimental evidence to support the relationship of type A behaviour and CSC was provided by Yoshioka et al a few years before Yannuzzi's investigation. The authors observed that intravenous epinephrine produced experimental CSC. They also suggested that the serous detachment of the neurosensory retina in CSC was biochemically mediated via stimulation of adrenergic receptors; this resulted in choriocapillaris hyperpermeability and degeneration of the RPE cells above the damaged endothelial cells.30, 31, 80

The above clinical and experimental evidence demonstrated an association of the disease with a sympathetic response. Unfortunately treatment of the disorder based on these observations has not proven helpful. However, these papers do illustrate the systemic associations of CSC.

Role of corticosteroids in the development of CSC

Jampol et al81 stated that corticosteroids might sensitise the choroidal blood vessels or RPE to the effects of endogenous catecholamines.

There is clinical and experimental evidence that corticosteroid intake contributes to the development of CSC.35, 36, 37, 38, 39 Central serous chorioretinopathy induced by the systemic use of steroids was first reported in 1984.35 Almost 20 years later, Carvalho-Recchia et al82 published the first report of a consecutive series of patients with acute CSC studied prospectively for an association with corticosteroids. They found a statistically significant difference in corticosteroid exposure between study patients and controls. In addition, CSC has been described in patients with endogenously high levels of corticosteroids (Cushing's syndrome, pregnancy, and stress)39 as well as in patients with hypercortisolism due to the treatment of ocular (optic neuritis, ischaemic optic neuropathy, solar retinopathy, scleritis, anterior uveitis, and chorioretinitis)83, 84, 85, 86, 87, 88 or systemic diseases.89, 90 Bouzas et al91 suggested a category of systemic diseases that have been associated with CSC only in cases where glucocorticoids were used (such as asthma, allergic rhinitis, sinusitis, myasthenia, back pain, hepatitis, and breast cancer). They postulated that glucocorticoid intake in these cases was more important for the development of CSC than the underlying disease itself.

Many components of the hypothalamus–pituitary–adrenal axis and the autonomic (sympathetic) nervous system have been implicated in the pathogenesis of CSC.92, 93 In an attempt to identify endocrine and metabolic abnormalities in patients with CSC, Haimovici et al94 prospectively evaluated serum and urinary catecholamines, glucocorticoids, mineralocorticoids, serum testosterone, and thyroid-stimulating hormone function of 24 CSC patients. They found that 50% of patients with active acute CSC had an elevated 24-h urine cortisol or tetrahydroaldosterone levels.

At the cellular level, glucocorticoids have been shown to increase the expression of β-adrenergic receptors.95 At the molecular level, the same authors provided evidence linking glucocorticoid effects of β-adrenergic receptors to an increase in receptor mRNA. Sakaue and Hoffman96 found that the expression of the α-1B-adrenergic receptor gene is induced by glucocorticoids and results in an increase in the number of α-1B-adrenergic receptors.

There is evidence of the influence of corticosteroids on adrenergic receptor gene transcription and expression, which represents the genomic effects of corticosteroids mediated by intracellular receptors.

Important neural effects of glucocorticoids are also exerted by non-genomic mechanisms. This may explain the delay in the expression of genomic effects.97 Some of the glucocorticoid receptors that mediate the non-genomic effects are localised in the cell membrane. These include ion channels and neurotransmitter receptors.97 Jampol et al81 linked the rapid onset of the non-genomic effects of corticosteroids to the occurrence of serous retinal detachments after the use of high-dose systemic corticosteroids.

Evidence of familial clustering of CSC

Weenink et al reported 27 patients with characteristic, mostly bilateral, fundus lesions of chronic CSC. Out of 80 investigated relatives, 35 (44%) had fundus lesions: 22 had chronic CSC in one eye, 20 of them had chronic CSC or RPE atrophy in the fellow eye; 13 relatives had RPE atrophy in one or both eyes.98 The mode of inheritance could not be established. Fawzi et al99 described the demographic and clinical characteristics of CSC after solid organ transplantation. One of their patients with the severe variant of CSC with bullous retinal detachment and a cardiac transplant had two second-degree cousins who carried the diagnosis of Vogt–Koyanagi–Harada syndrome. This syndrome has similarities to the severe variant of CSC with bullous retinal detachment. According to the investigators, this observation raised the possibility of an underlying predisposition in these patients that might have been genetically determined.

Pathogenesis of visual loss in CSC patients

Foveal attenuation, cystoid macular degeneration, and damage of the foveal photoreceptor layer cause visual loss in CSC.48, 100, 101 Cystoid macular degeneration, which is generally known as chronic macular oedema, was defined by Iida et al100 as cystoid spaces without intraretinal fluorescein leakage in the fovea.

Posterior cystoid retinal degeneration, generally known as RPE atrophy or decompensation or depigmentation, is defined as cystoid retinal degeneration located in the posterior pole, sparing the fovea.102 Duration of symptoms more than 5 years and subretinal fibrosis have been identified as risk factors for the development of posterior cystoid retinal degeneration in CSC, but this is not necessarily associated with marked reduction in visual acuity.103 On the contrary, there are cases of foveal atrophy associated with severe visual loss in long-standing serous macular detachment without any cystoid changes in the retina.101

One intriguing feature of CSC is the ability of photoreceptors to continue to function above a serous retinal detachement. This compares to the profound visual loss associated with rhegmatogenous retinal detachment. Experimental studies of retinal detachment have suggested that retinal detachment has a greater impact on cones than on rods.104, 105 The OCT studies have confirmed that photoreceptor changes are more prominent at the fovea.101 However, the same authors observed that there was a much longer survival of photoreceptors in CSC. This, they explained on the basis of differences between the pathophysiological conditions underlying the neuroretinal damage in CSC patients and the behaviour of experimental animal models.

Conversely, Chuang et al106 suggested that rod dysfunction was more pronounced than cone dysfunction in CSC. Mori et al107 declared that photopigment degeneration appears to take place in the rod photoreceptors of the detached retina. Wang et al108 postulated that visual dysfunction should be more evident under scotopic conditions because the RPE visual cycle subserves predominantly the rods whereas a separate visual cycle in the neurosensory retina supports cone function.109

CSC impact on vision in untreated eyes

Visual acuity usually recovers well in untreated eyes affected with CSC.110 However, quality of vision may be affected.111 Patients with resolved CSC may complain of metamorphopsia, decrease in brightness, and alteration in colour vision in the affected eye for several months.4 Wong et al112concluded that ophthalmologists used to ‘trivialize the situation as patients sometimes suffer the consequences following presumed resolution of the disease’. They also found a strong correlation, although statistically insignificant, of visual acuity and contrast sensitivity in both normal and ICSC-affected eyes in their long-term follow-up of resolved ICSC. Koskela et al113 found a statistically significant correlation between visual acuity and contrast sensitivity after resolution of ICSC.

Therefore, clinicians need to be aware that patients may still be visually symptomatic despite visual acutiy returning after an episode of ICSC.

Treatment of CSC

Types of CSC and treatment criteria

The literature does not really distinguish well between what one might call ‘acute CSC’ and ‘chronic CSC’, although there is a general belief that such a distinction would correlate with the long-term visual prognosis as well as the decision of whether to treat or wait and when to start treatment.

The term acute CSC usually refers to the self-limiting CSC that resolves spontaneously over a few months without any treatment and minimal residual changes on imaging. Most clinicians would agree with the definition that Spaide et al114 gave to chronic CSC, meaning a serous macular elevation, visible biomicroscopically or detected by OCT, that is associated with RPE atrophic areas and subtle leaks or ill-defined staining by FA. Polak et al36 noted that the major distinction between chronic and acute disease is the fact that chronic disease has widespread pigment epithelial changes without overt detachment in most cases, whereas in acute disease there is focal pigment epithelial abnormality and marked detachment.

Diffuse RPE or sick RPE syndrome is a term that has been defined in literature as chronic CSC by some investigators36 or as a form of chronic CSC by others.50 It has been reported as an idiopathic condition50 or as a complication of systemic corticosteroid treatment.36 Its natural course is favourable despite the remaining problems in colour vision and some degree of metamorphopsia. The use of any of the treatment forms available today, which we describe further in this article, is questionable, especially in corticosteroid induced disease that recovers well when steroid daily dosage is diminished.

Bullous serofibrinous exudative retinal detachment occurs in some patients with ICSC. Owing to its peculiar clinical findings, it can present a diagnostic dilemma and may lead to inappropriate diagnosis of rhegmatogenous RD or serous RD because of other causes.115 Corticosteroid therapy, organ transplantation, haemodialysis, and pregnancy have been reported to relate to this form of CSC. It has been described however as an idiopathic CSC form as well.116, 117, 118 In the largest case series of ICSC with spontaneous bullous exudative RD, visual prognosis was good without any treatment.115

Photoreceptor atrophy in the fovea, despite successful reattachment, occurs after duration of symptoms of approximately 4 months.47 Attenuation of the foveal photoreceptor layer is associated with permanent visual loss as mentioned above. Imminent damage of foveal photoreceptors or foveal atrophy defined by the combination of chronic CSC signs with current activity involving or immediately threatening the fovea could be considered as the high-risk group for which treatment should be applied.

Therefore, treatment should be considered in recurrent chronic CSC or a single CSC episode, of greater than 3 months duration, with some signs of chronic CSC. Previous permanent visual loss in the fellow eye caused by a similar procedure would also indicate that treatment should be instituted even in the absence of chronic CSC signs or even if foveal photoreceptors were not immediately threatened.

Previous pharmacological and other treatment suggestions

Before the development of other treatments, psychotherapy was suggested as a therapy.40 It was abandoned though as soon as there was progress in the understanding of the pathogenic nature of the disease.

Corticosteroids were presented as the only treatment choice in CSC several years ago. They were administered subconjuctivally or systemically, but are no longer recommended as knowledge of CSC pathogenesis has evolved.119, 120, 121

Adrenocorticotropic hormone,119 anti-inflammatory drugs,122 retrobulbar tolazoline injections,123 subconjuctival injections of milk, albumin and salt solutions, anti-syphilitic and anti-tubercular drugs, insulin-free pancreatic extract, and thyroid extract have all also been suggested in the past.124 The use of the above agents was not proven to be effective by any clinical trials.2

The role of stimulation of adrenergic receptors in the pathogenesis of CSC led some investigators to suggest that β- or α-adrenergic blockade could be utilised in the treatment of CSC.41, 42 Their suggestions were based on small case series of patients and experimental models. However, there is no significant proof to support such a therapeutic approach.

Acetazolamide has also been tried as a means of treatment of the chronic macular oedema caused by CSC or other chorioretinal diseases with short-term encouraging results but no evidence of long-term benefit.43

Current and future treatment options for CSC

Natural course of CSC and the use of argon laser photocoagulation

Focal photocoagulation of the leakage sites does not have a significant effect on visual acuity.124, 125, 126, 127 Robertson and Ilstrup127 suggested a reduction in CSC recurrences and shortening of the duration of detachment with direct laser photocoagulation compared with sham or indirect (away from the site of leakage) photocoagulation within a follow-up period of 18 months. Dellaporta128 concluded that untreated eyes were 3.3 times more likely to develop a recurrence than treated eyes. Gilbert et al129 found no difference either in final visual acuity or in recurrence rate between eyes treated with argon laser photocoagulation and untreated eyes in their retrospective long-term follow-up study of CSC patients.They also explained discrepancies in recurrence rates in various studies by the different follow-up durations and different treatment techniques (laser spot size).

Wang et al130 recommended early laser photocoagulation treatment in detachments that have lasted longer than duration of symptoms, suggested by the presence of subretinal granular deposits. Robertson131 suggested early laser photocoagulation in cases with visual acuity of 20/40 or less and multiple recurrences.

In the study, with the longest follow-up (6–12 years), argon laser photocoagulation treatment was not shown to reduce the incidence of recurrent disease or of chronic CSC. This study also showed that the role of argon laser photocoagulation in CSC with good visual acuity is limited to hastening relief of symptoms by achieving speedier resolution of serous detachment.132

Overall, a review of literature on argon laser photocoagulation for CSC suggests that this treatment is mainly effective in acute CSC with obvious focal leakage observed on FA. In these cases an earlier resolution of the serous detachment is achieved compared with natural history. However, if the area of leakage is subfoveal or juxtafoveal, photocoagulation may induce secondary choroidal neovascularisation (CNV) and/or of damage foveal photoreceptors.108 Therefore other treatment options appear safer.

Discontinuation of corticosteroids

Sharma et al133 reported an observational case series of atypical severe CSC treated with corticosteroids for their ocular condition. Discontinuation of corticosteroids resulted in reattachment of the retina in 88% of affected eyes. They concluded that discontinuation of corticosteroids in atypical CSC could lead to obliteration of RPE leaks and retinal reattachment without laser treatment.

Transpupillary thermotherapy

Shukla et al48 performed TTT in long-standing CSC, which resulted in the resolution of CSC with subfoveal angiographic leaks and significant improvement in visual outcome, in comparison with the natural history of persistent CSC. Long-term results are unknown.

Photodynamic treatment with verteporfin

Photodynamic therapy with verteporfin has recently been utilised to treat ICSC. Photodynamic treatment using verteporfin has both an occlusive effect on CNV and also affects normal choroidal perfusion.134 Chan et al53 postulated that PDT could be beneficial for the treatment of CSC by its effect on the structure of choroidal vasculature, causing alterations in choroidal permeability.

Both Fluorescein and ICG-A-guided PDT have been recommended for treatment of ICSC. Choroidal abnormalities, demonstrated on ICG-A, imply that treatment can be guided by ICG-A.53

It has also been also suggested that PDT acts by both decreasing choroidal hyperpermeability and by tightening the blood–retinal barrier at the level of the RPE. Therefore, other investigators believe that treatment directed at the area of RPE decompensation on FA can be adequate to allow subretinal fluid resolution.135

Choroidal hypoperfusion, which is the main mechanism of action of PDT in CSC, can also lead to complications, especially if conventional PDT is performed, according to the Treatment of age-related macular degeneration with Photodynamic Therapy Study guidelines.136 Lee et al137 described three cases of abrupt visual loss due to severe choroidal ischaemia after using standard PDT in CSC patients. RPE atrophy, juxtafoveal CNV, and transient reduction in macular function demonstrated by mERG led investigators to reconsider PDT parameters for the treatment of CSC.51, 53, 138, 139

Lai et al140 reduced the dosage of verteporfin and shortened the interval between infusion and laser application to induce choroidal vascular remodelling and minimise any collateral damage to adjacent retinal structures. The safe and beneficial effect of ‘safety enhanced’ (half-dose verteporfin)-PDT was demonstrated in both acute and chronic CSC by the same investigators.141, 142 Zhao et al143 suggested that 30% of verteporfin full dose was the effective lowest dose in the treatment of their acute CSC patients.

Reibaldi et al144 described two cases of long-standing CSC treated with ICG-A-guided low-fluence PDT. The anatomic and functional outcomes were encouraging. They concluded that ICG-A-guided low-fluence PDT seemed effective and safe for treating long-standing chronic CSC. The same investigators studied the efficacy of ICG-guided low-fluence PDT compared with standard PDT in a prospective non-randomised clinical trial; both standard-fluence PDT and low-fluence PDT resulted in complete subretinal fluid reabsorption with visual acuity improvement. They postulated that choroidal hypoperfusion related to PDT could be reduced by low-fluence PDT.145

Recently, Inoue et al146 shortened the irradiation time and reduced the total energy using the same light intensity and the same verteporfin dosage as the standard protocol. They reported that the success rate of PDT for CSC depends on the degree of hyperfluorescence seen on ICG-A. PDT is not effective or the recurrence rate is predicted to be high in eyes without intense hyperfluorescence.

Long-term efficacy from this form of treatment is unknown. In addition, the number of patients in most of the relevant studies is limited and often there is no dramatic improvement in terms of the functional outcome of the treatment despite impressive anatomic outcomes observed with OCT. Additionally, using conventional PDT, visual improvement may be limited in patients with prolonged symptom duration, in those who have baseline confluent RPE atrophy, disintegrity of the junction between foveal outer and inner photoreceptor layer or progression of RPE atrophy after PDT. The risk of PDT-induced foveal injury in these patients should also be considered.147

As far as direct comparison of laser photocoagulation with PDT treatment is concerned, we now have some results coming from study by Maruko et al,148 in which authors showed that the choroidal thickness and hyperpermeability seen during ICG-A was reduced after PDT. They suggested that PDT reduces the choroidal vascular hyperpermeability seen in CSC and it may act by a different mechanism than laser photocoagulation.

Most published studies suggest PDT with verteporfin is a safe and efficacious treatment even in chronic CSC and that complications are rare. This is especially true when PDT parameters are changed to minimise potential damage. Unlike argon laser photocoagulation, it can be performed for subfoveal leakage (Figure 3a and b).

Figure 3
figure 3

(a) OCT image of a CSC patient prior to photodynamic treatment showing a large amount of subretinal fluid involving the fovea. (b) OCT image of the same patient 3 months after half-dose photodynamic treatment showing complete resolution of subretinal fluid.

Full size image

Intravitreal bevacizumab

A hypothesis that VEGF antibodies could reduce choroidal hyperpermeability and choriocapillaris ischaemia associated with CSC55 has resulted in treatment of acute and chronic forms of CSC with intravitreal injections of bevacizumab (Avastin).54, 55, 56, 57

However, all of the related reports are small, uncontrolled case series with a short duration of follow-up. Larger, controlled trials are still needed to evaluate the efficacy and safety of anti-VEGF agents for this indication.

Micropulse diode laser photocoagulation

Subthreshold micropulse diode laser (810 nm) has recently been assessed for the treatment of chronic CSC.149, 150, 151, 152 To avoid retinal damage, caused by conventional laser photocoagulation, it has been used in chronic CSC, in eyes with either well-defined leaking sites or diffuse leakage.

In the largest of these case series (26 eyes), a gain of visual acuity of three lines or more was achieved in 58% and a gain of between one and three lines was achieved in 23% of the treated eyes.149 Ricci et al150 assessed the efficacy of indocyanine green-enhanced subthreshold micropulse diode laser photocoagulation and found no worsening of the serous detachment or of visual acuity in patients with incomplete recovery at 12 months.

Subthreshold micropulse diode laser photocoagulation therefore appears to be a safe form of treatment in chronic CSC. Results however are not superior to PDT. Furthermore, micropulse diode laser photocoagulation seems unsuitable for diffuse leakage and diffuse RPE decompensation. These are frequent findings in chronic CSC cases. Larger, controlled, randomised clinical trials are needed to establish the role of diode laser photocoagulation in the treatment of CSC. Currently this treatment may be considered as an alternative to PDT for eyes with focal and well-defined areas of leakage.

Corticosteroid antagonists

Jampol et al81 first suggested that glucocorticosteroid antagonist activity may be of value in preventing or treating episodes of CSC. This was based on the association of endogenous hypercortisolism with the development of CSC.39 The potential treatment of CSC episodes using antiglucocorticoid agents includes RU486 (mifepristone) and ketoconazole.

RU486 is an active anti-glucocorticosteroid and anti-progesterone agent. This dual action results from similarities between receptors involved.153 Its use in voluntary early pregnancy termination has delayed the initiation of ophthalmic clinical trials in the United States.

Ketoconazole is also an adrenocorticoid agent. It was first tested as a potential treatment for CSC by Golshahi et al,58 in a prospective, case-controlled study. Patients received 200 mg of the drug per day for 4 weeks. The clinical benefit of this trial was not statistically significant. After 3 years, an increase in dosage of ketoconazole to 600 mg daily for 4 weeks was tried by Meyerle et al59 who found a delayed therapeutic response at 8 weeks after initiation of treatment. They postulated that their inconclusive results were because of short duration of treatment or/and normal baseline cortisol levels of the patients involved and they suggested larger, controlled trials to test the efficiency of ketoconazole in CSC patients.

Aspirin

Caccavale et al154 evaluated low-dose acetyl salicylic acid (aspirin) in 107 CSC patients with a mean follow-up time of 20 months. They found a rapid recovery of visual acuity and a reduced number of recurrences in their patients. They surmised that in all the diseases, associated with CSC, plasminogen activator inhibitor-1 (PAI-1) was increased and that aspirin is effective in lowering PAI-1 levels and platelet aggregation.

Search strategy

We searched the MEDLINE/PubMed database for articles from March 1969 to January 2010 after following MeSH suggestions for articles including the terms: CSC, chorioretinopathies, and central serous retinopathies. The headline used to locate related articles in PubMed was ‘central serous chorioretinopathy’ and to restrict search we used the headlines ‘pathophysiology of central serous chorioretinopathy’, ‘treatment of central serous chorioretinopathy’, and ‘photodynamic treatment in central serous chorioretinopathy’. A manual search was also based on references from these articles as well as review articles.

Comments

After reviewing the voluminous literature on the aetiology and pathogenesis of CSC it certainly seems that CSC is a multifactorial disease. It appears to result from a complex interaction of known and unknown environmental and genetic factors. This ultimately leads to a bilateral disease with systemic associations.

In 1986, Yannuzzi2 stated there was a lack of a definitive, universally accepted treatment for CSC. This could also be stated today. The multifactorial aetiology and complex pathophysiology of the disease and its generally favourable natural history provide no clear proof of the necessity and long-term efficacy of any of the treatment choices that have been reviewed in this article. Further large, prospective or even retrospective long-term follow-up studies are required to decide on one or more safe and effective forms of treatment, which will be generally accepted by clinicians.

Until then, it seems reasonable to suggest reduced dose/fluence/irradiation time verteporfin PDT in recurrent chronic CSC or in single CSC episodes, not resolving for a period of at least 3 months, accompanied by signs of chronic CSC. In both of which there is active leakage involving the fovea or a juxtafoveal area. Micropulse diode laser treatment, applied on well-defined leaking sites, can be considered as an alternative. The use of corticosteroid antagonists, possibly after evaluation of patients’ cortisol profile (for example, urine cortisol or tetrahydroaldosterone levels), is an interesting future option that merits further investigation. In addition, counselling about discontinuation of steroid treatment for systemic or ocular conditions and explanation of the relation of the disease to stress is helpful in the management of CSC patients.