Background To evaluate the long-term efficacy and complications of half-dose photodynamic therapy (PDT) for treating chronic central serous chorioretinopathy (CSC).
Methods A retrospective, interventional case series of patients with chronic CSC (symptoms ≧3 months) receiving half-dose PDT (3 mg/m2 verteporfin). Optical coherence tomography, fundus short-wave autofluorescence and near-infrared autofluorescence (SW-AF and NIR-AF) were taken at each visit. Central foveal thickness (CFT) and relative CFT (RCFT) were measured manually. Outcome measures included best-corrected visual acuity (BCVA), resolution of subretinal fluid (SRF), CFT and RCFT, and changes of SW-AF and NIR-AF.
Results 56 eyes of 56 patients (45 male and 11 female patients; mean age 45 years) were included in this study. The anatomic resolution was obtained in 100% at 1 year and at the last follow-up, respectively. Four cases developed recurrence of SRF after one session of PDT. The mean session of PDT was 1.00. The mean follow-up was 55.5 months. Final CFT and RCFT showed a positive correlation with final BCVA. Complications included enlargement of retinal pigment epithelial atrophy in one case and choroidal neovascularisation in another two cases at 12 and 14 months after PDT.
Conclusions PDT with half-dose verteporfin has a long-term efficacy and safety profile in chronic CSC. However, complications may still happen despite the reduced dosage.
Trial registration number The CCH IRB No. 131007.
- Treatment Lasers
Statistics from Altmetric.com
Central serous chorioretinopathy (CSC) is characterised by serous detachment of the neurosensory retina at the posterior pole. Indocyanine green angiography (ICGA) studies have demonstrated the hyperpermeability of the choroidal vasculature in CSC, supporting the theory that CSC primarily affects the choroidal circulation.1 Though cases of CSC are self-limiting and have favourable clinical outcomes most of the time, recurrence may take place. A minority of cases progress to chronic CSC with persistent serous retinal detachment and decompensation of retinal pigment epithelium (RPE).
The optimal management of CSC remains controversial. Photodynamic therapy (PDT) with standard dose of verteporfin (6 mg/m2) (Visudyne; QLT, Vancouver, British Columbia, Canada) has been used to treat chronic CSC and has been demonstrated to have beneficial visual outcomes in the majority of patients.1–3 Using a half-dose instead of standard dose PDT has been popular in recent years for treating chronic CSC because of its similar efficacy and reduced risk of complications.4–6 However, most of the published literature has only reported short- or mid-term efficacy of half-dose PDT. We thus report the long-term outcomes of a retrospective study in patients with chronic CSC managed with half-dose PDT.
Materials and methods
A retrospective, interventional case series analysis was conducted in 56 eyes of 56 consecutive patients with chronic CSC undergoing PDT using half the standard dose of verteporfin at Changhua Christian Hospital, Taiwan, between March 2006 and March 2009. This study enrolled patients with symptomatic chronic CSC with subretinal fluid (SRF) persisting three or more months with or without a recurrent history of CSC. The inclusion and exclusion criteria are listed in box 1. Patients who had previously undergone focal thermal laser photocoagulation 3 months or more prior to the PDT for treating CSC, but without resolution of SRF, or had recurrence of SRF were also included. This study was approved by the Institutional Review Board of Changhua Christian Hospital and the procedures conformed to the tenets of the Declaration of Helsinki.
The inclusion and exclusion criteria of the patients recruited in the study
Patients older than 18 years
Presence of subfoveal fluid persisting for 3 months or more on OCT
Lack of either spontaneous improvement or improvement induced by empirical treatment
Presence of active angiographic leakage in FA caused by CSC but not CNV or other diseases
Choroidal vascular hyperpermeability and abnormal dilation of choroidal vasculature on consistent with the diagnosis of CSC25
Any evidence of maculopathy, including previous macula involved CSC, in the fellow eye
Evidence of CNV
Any other ocular diseases that could affect visual acuity
Systemic steroid treatment in the previous 12 months
Previous laser within 3 months before PDT
Previous treatment by PDT
Media opacity such as cataract that could interfere with adequate acquisition of OCT, fundus, autofluorescence, FA and ICGA images
▸ Formal contraindication for PDT or FA
CNV, choroidal neovascularisation; CSC, central serous chorioretinopathy; FA, fluorescein angiography; ICGA, indocyanine green angiography, OCT, optical coherence tomography; PDT, photodynamic therapy.
All the patients received half the standard dose of verteporfin (3 mg/m2) infusion over 10 min followed by delivery of diode laser at 689 nm (Visulas690S; Carl Zeiss Meditec, Dublin, California, USA) 15 min after the start of the infusion with a total energy of 50 J/cm2, with over 83 s delivered to the area of choroidal hyperpermeability observed on ICGA. The PDT target was determined to ensure inclusion of only the area of choroidal abnormalities of dilated and congested choroidal vessels, and the area with sub-RPE extravascular leakage in the macula according to the findings by ICGA. In the presence of multifocal lesions responsible for the subfoveal fluid, consecutive non-confluent laser spots were used with different laser spot size according to the lesion size on ICGA and 83 s duration for each spot. Generally speaking, the lesion occupying the foveal area was first treated, followed by the lesions in the extrafoveal area. The spot size of PDT ranged from 1200 to 4200 µm. After treatment, patients were given protective spectacles and instructed to avoid strong light for 3 days.
Patients were assessed at baseline and followed up at 1, 3, 6, 9 and 12 months, and then every 6 months, after PDT. Patients were recorded regarding age, sex, first attack or recurrence, and duration of the latest onset at initial visit. Complete ophthalmic examinations including best-corrected visual acuity (BCVA), slit-lamp examination, intraocular pressure, dilated fundus examination, fundus short-wave autofluorescence and near-infrared autofluorescence (SW-AF and NIR-AF) and optical coherence tomography (OCT) imaging were performed at baseline and post-PDT visits. BCVA was measured with the decimal visual acuity (VA) chart at 6 m and was converted to logarithm of the minimum angle of resolution (logMAR) for analysis. To evaluate the status of RPE, SW-AF and NIR-AF were performed with a confocal scanning laser imaging system (HRA-II, Heidelberg Engineering, Heidelberg, Germany). OCT was performed with Stratus OCT (Carl Zeiss Meditec) using the 6 mm cross hair mode to evaluate the presence of SRF and pigment epithelial detachment (PED). Assessment of central foveal thickness (CFT) was determined by using the ‘calliper’ option measuring the distance between the internal limiting membrane and the outer border of neuroretina manually in good quality scans centred on fovea. SRF thickness was measured in the same fashion from the outer border of the neurosensory retina to the inner border of RPE. Considering that there is an inherent difference of retinal thickness between individuals, all patients had both eyes measured for CFT before and after PDT treatment. The value of relative CFT (RCFT) was obtained by using the CFT of the treated eye divided by the CFT of the fellow eye. Measurement was repeated by a masked reader using the same scan line on both measures and the mean values from both readers were used for the statistics.
Fluorescein angiography (FA) and ICGA (HRA-II) were performed in all patients at baseline and 1 year after PDT, and then as needed in patients with persistence or recurrence of SRF as well as other complications during the follow-up period. Retreatment was considered if the neurosensory detachment had failed to resolve or had recurred.
A total of 56 eyes of 56 patients diagnosed with chronic CSC treated with half-dose PDT were recruited in this study. In all, 45 patients were men and 11 were women. The mean age was 45.0±8.0 years (range 29–68 years). The duration of symptoms before treatment was 10.7±11.0 months (range 3–56 months) and the mean length of follow-up was 55.5±11.5 months (range 41–90 months). At baseline, the mean logMAR BCVA of all patients was 0.36±0.33 (range 0.0–1.3) and the mean CFT was 148.4±39.1 µm (range 73–261 µm). SRF was present in all eyes at the macula and 11 (20%) eyes had combined macular PED. Fifteen (26.8%) eyes had received focal thermal laser photocoagulation prior to this study.
The mean number of PDT treatment was 1.00. One single session of half-dose PDT achieved complete resolution of SRF in 56 of 56 eyes (100%). In 56 eyes, 17 eyes had multifocal lesions and four of them with initial good response were noted to have recurrent SRF during the follow-up. Case 1 was a 60-year-old male patient with 5-month duration of symptoms without previous history of laser. Recurrent macula involved SRF was noted 10 months later. The repeated ICGA showed reduced but persisted choroidal hyperfluorescence at one of the three previous treated areas (the second spot of treatment). He received another session of PDT targeting on this area and had SRF resolved 1 month later, which was kept stable till the end of follow-up. Case 2 had the recurrent SRF noted 65 months after PDT. He also denied any history of previous laser treatment. Repeated FA and ICGA showed that the choroidal hyperfluorescent area was located at the second of the three previous treated spots. The SRF resolved spontaneously 1 month later without any treatment. Case 3 was a 50-year-old male patient without previous laser treatment who had the recurrent SRF noted 27 months after PDT treatment. Repeated ICGA showed the leakage area was in one of the three previous treated areas (the third spot). The patient chose observation and the SRF resolved spontaneously 4 months later without further recurrence during the following 2 years of follow-up. The last case was a 29-year-old male patient with 15 months of symptoms before PDT treatment. He had a previous laser treatment for CSC 4 years before the PDT. Recurrent macula involved SRF was noted 19 months after PDT. Repeated ICGA showed choroidal hyperfluorescence at one of the two previous treated areas (the second spot). Though the patient refused any further treatment, the SRF resolved spontaneously 5 months later and was kept stable during the following 30 months of follow-up.
VA changes after PDT
In our study, older age and thicker SRF height were found to be correlated with worse baseline BCVA (p=0.01 and 0.039, respectively) after linear regression and it had no relationship with duration of symptoms, previous laser treatment, presence of PED, preoperative CFT or RCFT.
At 1 month after PDT, the mean logMAR BCVA improved significantly from 0.36 to 0.24 (p<0.001). At 6 months, the mean logMAR BCVA further improved to 0.13 (p<0.001) and remained stable thereafter. The visual improvement at different time points for all patients is shown in figure 1. In all, 46 (82%) eyes had improved vision (more than two lines improvement) at the end of follow-up and 8 (14%) eyes had unchanged VA and 2 (4%) eyes had worse VA. The eyes with thicker initial RCFT (RCFT>1) had better BCVA at every follow-up visit (figure 2). However, after the linear regression, thicker initial RCFT was not found to be related to better final VA or to increased visual improvement. Patients with a younger age or thicker final RCFT were found to be more likely to have better final vision and more visual improvement (p=0.013 and <0.001, respectively). Worse baseline BCVA was another factor for greater visual improvement. Final BCVA was not correlated with duration of CSC, presence of PED or previous laser treatment nor was visual improvement correlated with these factors (table 1).
OCT and anatomic changes after PDT
Complete resolution of SRF at the macula was achieved in 46 (82%) eyes at 1 month, and further increased to 56 (100%) eyes at 3 months. In all, 56 eyes (100%) had complete resolution of the serous macular detachment at the end of follow-up. Four eyes had recurrent SRF at 10, 19, 27 and 65 months later, respectively, two of which had subfoveal SRF. Three of the four eyes had SRF spontaneously resolve without further treatment till the end of follow-up. The other eye with recurrent subfoveal SRF 10 months after PDT needed another session of half-dose PDT and had SRF reabsorbed till the end of follow-up.
Eleven eyes presented PED with SRF at baseline; all of them had SRF resolution and PED disappear after a single session of PDT.
The status of CFT changes are presented in figure 3. Eyes with thicker initial RCFT (RCFT>1) had CFT thinned rapidly during the first month and gradually reached a similar thickness to the eyes with thinner initial RCFT (RCFT≦1, figure 4). Both initial and final RCFT were not related to the duration of symptoms (p=0.41 and 0.35, respectively).
None of the patients in the study developed any systemic adverse event. None of the patients experienced any subjective or objective drop in vision after treatment, except one patient who mentioned a transient scotoma for 1 day after PDT, which resolved completely without any sequel. This was a 52-year-old male patient, with duration of symptom for 3 months, without PED or previous laser treatment. He was treated according to the protocol with a laser spot size of 3400 µm. For the serial SW-AF and IR-AF follow-up, enlarged RPE atrophy was noted in one eye (figure 5). Despite the enlarged RPE atrophy, this patient still had the VA maintained at 20/20 and no recurrent SRF was noted during the follow-up period. Two eyes developed secondary choroidal neovascularisation (CNV) 12 and 14 months after PDT. One of them is a 43-year-old man with laser treatment history. The symptom duration was 15 months, and there was no PED noted before the treatment. The PDT spot size was 2700 µm with duration of 83 s. The CNV appeared 1 year later at the previous laser and PDT treated area. The CNV was under control after intravitreal injection of bevacizumab. He received four as needed injections during the 4 years of follow-up. The other is a 54-year-old man without PED or previous laser treatment with 8 months duration of symptoms before PDT. He had two consecutive PDT spots with the size of 2100 and 1400 µm, and the duration of 83 s for each spot. The CNV appeared 14 months after PDT at the PDT treated area. He had intravitreal injection of bevacizumab for CNV and received totally seven injections in the 3 years of follow-up after having CNV noted.
Angiographic and AF outcome
At baseline, the finding on ICGA was mild to moderate focal or multifocal hyperfluorescence in all of the eyes. In the follow-up ICGA 1 year after PDT, narrowing of the dilated and congested choroidal vessels was observed in all cases. Complete resolution of the choroidal vascular hyperpermeability and extravascular leakage was observed in all of primary treated lesions, and complete or partial resolution observed in consecutive treated lesions. In all cases, FA showed disappearance of leakage either in cases with focal leaking spots or in cases with diffuse RPE oozing at 1 year of follow-up. For the autofluorescence (AF) images, the SW-AF examination showed hypo-AF changes corresponding to the leakage point on FA in most cases (figure 6). More obvious hyper-AF and hypo-AF changes were noted 6 months after at the area of fluid resorption (figures 6 and 7). The granular pattern of hyper-SW-AF might become more evident and extensive after fluid resorption and gradually faded away during the follow-up period (figure 8). NIR-AF examination showed hypofluorescent spots corresponding to the leakage area in FA (figure 6) in every case. The changes of NIR-AF generally were more extensive to that of SW-AF (figures 5 and 6). The NIR-AF images taken 6 months after PDT were similar to the preoperative changes, but usually showed more prominent hyper-AF and hypo-AF changes (figure 6). Generally speaking, there was only minimal or no changes of both SW-AF and NIR-AF images 6 months after PDT during the follow-up period (figures 6⇓–8). Only one eye showed obviously enlarged RPE atrophy during the long-term follow-up (figure 5). However, this patient had BCVA stable which is 20/20 without recurrence of SRF despite the changes of AF.
Although CSC is generally considered to be a self-limiting and benign condition,7 some patients may progress to chronic CSC which is characterised by diffuse pigment epithelial impairment with persistent SRF.8 However, the treatment for chronic CSC has not been well established. In past decades, thermal laser photocoagulation was the standard of treatment for extrafoveal leakage points.7 Unfortunately, whereas it may hasten the resorption of SRF, choroidal hyperpermeability remains unchanged.7 Recent studies have shown favourable outcomes of PDT with standard (6 mg/m2) verteporfin in the treatment of chronic CSC.1–3 The exact mechanism of PDT in treating chronic CSC is not conclusive but is postulated to be caused by the short-term choriocapillaris hypoperfusion and long-term choroidal vascular remodelling, leading to reduction in choroidal congestion, vascular hyperpermeability and extravascular leakage.9 Despite the demonstrated efficacy of PDT with full dose verteporfin, it is not completely harmless to ocular structures. The application of conventional PDT in chronic CSC may have complications as the development of RPE atrophy, choriocapillaris ischaemia and secondary CNV.2 Studies have shown that cytotoxicity and vascular damage associated with PDT are dosage dependent.10 ,11 Thus, to modify the drug dosage or duration of laser application may lessen the chance of associated complications. Some authors used half fluence protocol (from 50 to 25 J/cm2) to minimise the potential adverse effects and obtained good treatment efficacy in the treatment of chronic CSC.12 Others used half-dose (3 mg/m2), standard fluence PDT and had similar efficacy and reduced risk of complications.4–6 ,13 Recent reports comparing eyes receiving half-dose, standard fluence PDT versus eyes receiving half fluence, standard dose PDT showed either a comparable result14 or a better result for eyes receiving half-dose, standard fluence PDT.15
Despite the short-term success of PDT on persistent CSC, up to now, there have been few studies about the long-term efficacy and safety of PDT in treating CSC. In the report of Valmaggia et al16 on the initial response to standard dose PDT of 46 eyes with chronic CSC, two eyes had recurrence in the 2-year follow-up period, but no treatment-related complications were noted. In another retrospective study of standard dose PDT for chronic CSC by Silva et al,17 93.4% eyes were free of SRF at the end of the 4-year follow-up and the mean session of PDT was 1.08±0.3; there were also no treatment-related complications. Karakus et al5 reported the results of half-dose PDT with an average follow-up period of 25.3 months: 91.7% of the eyes were free of SRF at the end of the follow-up. In our series, the average follow-up period was 55.5 months, during which persistent SRF was present in none of the eyes (0%) and recurrent SRF in four eyes (7.14%). The mean session of PDT was 1.00±0.13 and 56 eyes (100%) had no SRF at the end of the follow-up. The mean VA improved gradually and stabilised 6–12 months after PDT till the end of the follow-up. While these results are similar to those of previous reports, this study further proves the efficacy of half-dose PDT on VA with anatomic outcome being maintained for at least 4 years.
For the prognostic factors, prolonged duration of symptoms was previously reported to be related to poor visual outcome.4 ,5 It is reasonable to suppose a prolonged duration of symptoms to be associated with more permanent damage of outer retinal layer, with a subsequent poor visual outcome. However, in this study, patients with longer duration of symptoms were not noted to be statistically related to initial or final poor BCVA. One of the possible reasons is that the recorded duration of symptoms was according to the patients’ subjective presentations which might be different from the real disease period. Besides, many of our patients were recurrent cases, and the duration of symptoms in our study was recorded as the duration of the latest onset disease, not including the previous attacks, which might not really reflect the chronicity of the disease. For the RCFT, we initially supposed that the disease chronicity should be associated with a thinner RCFT and a poorer final BCVA. Though this study showed that eyes with initial thicker RCFT had both better mean initial and final BCVA (figure 1), regression analysis failed to demonstrate its statistical significance. The possible explanation may be that there are factors other than disease chronicity affecting the initial RCFT. The height of SRF at initial presentation may affect the oedematous status of the overlying retina, which prevents the initial RCFT as a reliable indicator of the retina reserve and a prognostic factor of the final BCVA. Besides, the integrity of cone outer segment tips, which was recently reported to be associated with final retinal sensitivity,18 was also not evaluated by the measurement of initial RCFT. Preoperative BCVA is also not correlated to final BCVA in this study. This result is different from a recent report, in which better initial VA is correlated with better final VA.19 The possible explanation is that there are many factors affecting initial BCVA, including chronicity of disease, permanent damage from previous attacks, height of SRF and so on. Since our patients were heterogeneous in these factors, the baseline BCVA may not be predictive of the final BCVA. In this study, baseline BCVA is inversely correlated with the final improvement of VA. This result nicely corresponds with the previous report.19 This could be explained by the fact that eyes with initial better BCVA have less room for visual improvement. PED at baseline were reported previously to be related to poor visual outcome and less responsive to PDT treatment.4 Since in our study eyes with PED showed similar treatment response to eyes without PED, it is of no surprise that PED bears no relationship to the final visual outcome in the present study.
In this study, only age, final CFT and final RCFT correlated with final visual outcome and visual improvement. Younger age has been shown to be a good prognostic factor for final BCVA in a previous report.20 Our study also confirmed this result (p=0.013). The reasons for the better prognosis in younger patients includes possibly less previous attacks of CSC, better retinal reserve and better ability to recover. Final RCFT and CFT are also well correlated with the final VA (p<0.001). Because all of our patients had no SRF and the reactive retina oedema at final follow-up, the final RCFT and final CFT may represent the retinal reserve of the outer nuclear layer. This corresponds with the previous reports that outer nuclear thickness is correlated with final VA.21
We also used SW-AF and NIR- AF to check the RPE status at macular area during the follow-up. RPE changes are almost universal findings in CSC patients, and previous studies have shown that both SW-AF and NIR-AF are useful tools in studying the status of choroid, outer retina and RPE.22 Ayata et al23 found that hypo NIR-AF corresponded to the area of the serous retinal detachment and the leakage point. The authors hypothesised that the hypo NIR-AF areas that were not leaking on the angiogram were also pathological sites of the chorioretinal disturbance. Thus, the findings from AF images may be another sensitive method to detect chorioretinal disturbance and whether there is still subclinical activity of CSC. In our serial follow-up, most patients had AF changes becoming more prominent after PDT. This is probably due to the fact that the presence of SRF preoperatively would interfere with the visibility of AF, and thus the changes in the early months after PDT may be related to the disappearance of SRF, not necessarily the atrophic changes of RPE. In our series, all but one eye showed relatively stable status of AF during the long-term follow-up since 6 months after PDT, which may indicate that the RPE status is stationary in most cases after the treatment. In the eye with more obvious changes of SW-AF and NIR-AF (figure 5), the shape of the AF changes did not correspond to the shape of the PDT spot. This may indicate that it was not a PDT-related complication. Instead, it may represent a deteriorating chorioretinal condition due to the subclinical activity of CSC, even though it was still well compensated without recurrent SRF and the BCVA was still maintained at 20/20. Further follow-up is necessary to investigate the clinical correlation with the changes of AF.
CNV was noted in two of our cases at the treated area 12 and 14 months after PDT. One of the patients had been treated previously with photocoagulation for chronic CSC. Development of secondary CNV has been reported in the full dose PDT treatment for CSC.2 Since CNV may also be a complication of chronic CSC24 or induced by laser photocoagulation, the secondary CNV in our cases may not be attributed to the complication of half-dose PDT, but rather to a complication of the previous laser photocoagulation or the natural history of chronic CSC. However, it is still worthwhile to pay attention to these possible complications.
In our study, four of the 17 eyes with multifocal lesions developed recurrence of serous detachment at 10, 19, 28 and 65 months after PDT, and none of the eyes with monofocal lesion had recurrence. The common characteristic in these four cases was that they all had more than one PDT laser spots applied in one session. All the recurrent lesions responsible for the leakage were the second or third spots. The delay in timing of PDT probably compromised its efficacy since the drug concentration in the choroidal circulation at the application time is lower than optimal concentration for PDT. To reinforce the efficacy of PDT on the second or third laser spot, a different treatment protocol for multifocal lesions may be necessary. A longer treatment time for the second or third spot may be considered to compensate for the lower concentration of verteporfin in the choroid at the laser application time. Further studies are necessary to clarify the optimal management of these cases.
The power of this study is the long duration of follow-up. The weak points are the retrospective nature and the lack of control group. Besides, some of our patients had previous laser treatment, though all the patients had laser treatment more than 3 months before the PDT treatment. In conclusion, the current study extends the observations of previous short-term studies of half-dose PDT for persistent CSC and demonstrates the efficacy of this treatment regime. The treatment effects were sustained more than 4 years. These findings suggest that half-dose PDT can be an effective treatment option for chronic CSC with long-term efficacy. However, recurrence may still happen in the long run in cases with multifocal lesions, as one of our cases had recurrence noted at 65 months after PDT. Furthermore, half-dose PDT may still have the risk of complications such as secondary CNV and RPE atrophy despite the reduced dosage, and thus long-term follow-up is necessary to detect the possible complication.
Contributors S-NC: substantial contributions to the conception or design of the work; final approval of the version to be published. C-CT: acquisition, analysis and interpretation of data for the work; writing the article.
Competing interests None.
Patient consent Obtained.
Ethics approval Institutional Review Board of Changhua Christian Hospital.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.