Purpose
To report 6-month outcomes of a prospective, randomized study comparing the efficacy and safety between low-fluence photodynamic therapy (PDT) and intravitreal injections of ranibizumab in the treatment of chronic central serous chorioretinopathy.
Design
Prospective, randomized, single-center pilot study.
Methods
Sixteen eyes with chronic central serous chorioretinopathy were randomized to receive either low-fluence PDT or intravitreal injections of ranibizumab: 8 eyes in the low-fluence PDT group and 8 in the ranibizumab group. Rescue treatment was considered if subretinal fluid was sustained after completion of primary treatment: low-fluence PDT for the ranibizumab group and ranibizumab injection for the low-fluence PDT group. Main outcome measures were excess foveal thickness, resolution of subretinal fluid, choroidal perfusion on indocyanine green angiography, and best-corrected visual acuity.
Results
At 3 months, the mean excess foveal thickness was reduced from 74.1 ± 56.0 μm to −35.4 ± 44.5 μm in the low-fluence PDT group ( P = .017) and from 26.3 ± 50.6 μm to −23.1 ± 56.5 μm in the ranibizumab group ( P = .058). After a single session of PDT, 6 eyes (75%) in the low-fluence PDT group achieved complete resolution of subretinal fluid and reduction of choroidal hyperpermeability, whereas 2 (25%) eyes in the ranibizumab group achieved this after consecutive ranibizumab injections. Four eyes (50%) in the ranibizumab group underwent additional low-fluence PDT and accomplished complete resolution. At 3 months, significant improvement of best-corrected visual acuity was not demonstrated in the low-fluence PDT group ( P = .075), whereas it was observed in the ranibizumab group ( P = .012). However, the tendency toward improvement of best-corrected visual acuity was not maintained.
Conclusions
In terms of anatomic outcomes, the effect of ranibizumab injections was not promising compared with that of low-fluence PDT.
Central serous chorioretinopathy (CSC) is characterized by serous detachment of the neurosensory retina. The pathophysiologic features of CSC are not certain, and various theories have been proposed, including impaired function of retinal pigment epithelium (RPE), choroidal ischemia, and choroidal hyperpermeability leading to RPE damage. Acute CSC with monofocal or paucifocal changes of RPE usually shows spontaneous resolution and has a favorable visual outcome. Chronic CSC is characterized by multifocal or diffuse decompensation of RPE associated with persistent detachment of neurosensory retina. This may lead to cystoid macular degeneration, foveal atrophy, and damage to the foveal photoreceptor layer, consequently resulting in irreversible significant visual loss.
Photodynamic therapy (PDT) was proposed for the treatment of chronic CSC based on the proposed mechanism of choroidal hypoperfusion. Over time, PDT promotes the absorption of subretinal fluid (SRF) by choroidal vascular remodeling and reduction of choroidal hyperpermeability. However, irreversible damages may result from PDT, especially conventional PDT. Several reports demonstrated the risk of complications, such as RPE changes, excessive choriocapillaris hypoperfusion, and secondary choroidal neovascularization (CNV). Recently, a few studies based on modified PDT parameters demonstrated favorable results. Reduction of verteporfin dosage, shortening of the time of laser emission, and reduction of a total light energy were suggested.
It is well known that vascular endothelial growth factor (VEGF) is related with vascular permeability. Antibodies to VEGF may reduce choroidal hyperpermeability, one of the proposed mechanisms of CSC. On the basis of this hypothesis, intravitreal administration of an anti-VEGF agent was proposed as a new treatment option in the treatment of CSC, and several reports demonstrated acceptable outcomes after intravitreal bevacizumab injection. However, the clinical results with ranibizumab have not been reported yet. The purpose of the present randomized pilot study was to compare the efficacy and safety between low-fluence PDT and intravitreal injection of ranibizumab in the treatment of chronic central serous chorioretinopathy.
Methods
This study was a prospective, randomized, pilot study performed at the Department of Ophthalmology of the Seoul National University Hospital.
Study Participants
This study enrolled patients with a history of central serous chorioretinopathy characterized by visual disturbance persisting more than 6 months or with a recurrent history of CSC. The inclusion criteria were: (1) best-corrected visual acuity (BCVA) between 0.0 and 1.0 logarithm of the minimal angle of resolution (logMAR) units, (2) presence of subfoveal fluid persisting for 3 months or more on optical coherence tomography (OCT), (3) presence of leakage and multifocal or diffuse RPE decompensation on fluorescein angiography (FA), and (4) choroidal vascular hyperpermeability and abnormal dilation of choroidal vasculature on indocyanine green angiography (ICGA). The exclusion criteria were: (1) previous treatment, such as laser photocoagulation, PDT, intravitreal injection of steroid or anti-VEGF agent; (2) evidence of CNV; (3) any other ocular diseases that could affect visual acuity; (4) systemic steroid treatment in the previous 12 months; or (5) media opacity such as cataract that could interfere with adequate acquisition of OCT, FA, and ICGA images.
Treatment Protocol
Subjects were randomized to receive either low-fluence PDT or 0.5 mg/0.05 mL intravitreal injection of ranibizumab (Lucentis; Novartis AG, Basel, Switzerland). In the low-fluence PDT group, all the patients received a 6-mg/m 2 infusion of verteporfin (Visudyne; Novartis AG) over 10 minutes followed by laser delivery at 689 nm for 83 seconds, 15 minutes after the start of the infusion. A total light energy of 25 J/cm 2 , a light dose rate of 300 mW/cm 2 , was delivered. To determine the PDT spot size, the authors measured the greatest linear dimension of the area of choroidal vascular hyperpermeability corresponding to the origin of subfoveal fluid on ICGA. If the multifocal lesions of choroidal hyperpermeability were responsible for the subfoveal fluid, separate nonconfluent laser irradiation was applied to the lesions consecutively. At the same time, if the multifocal lesions were too close to irradiate separately, a single spot was delivered that covered the entire area of multifocal choroidal hyperpermeability. In the ranibizumab group, ranibizumab was injected through the pars plana into the vitreous cavity under sterile conditions. The patients were treated with monthly intravitreal injections of ranibizumab for the first 3 months. If complete resolution of SRF was observed, although 3 monthly injections were not completed, reinjection was stopped until reaccumulation of SRF. Rescue treatment was considered if sustained or reaccumulated SRF was associated with visual acuity loss of at least 0.2 logMAR units. As rescue treatment, a single session of low-fluence PDT for the ranibizumab group and ranibizumab injection for the low-fluence PDT group were considered because gold standard treatment has not yet been established for chronic CSC.
Follow-Up Protocol
All the patients completed 6 months of follow up. At baseline and at each monthly visit, slit-lamp examination of the anterior segment, intraocular pressure assessment, dilated fundus examination, BCVA assessment, and OCT (Cirrus; Carl Zeiss Meditec, Inc, Dublin, California, USA) imaging were performed. BCVA was assessed using Early Treatment of Diabetic Retinopathy Study charts at 4 m. Scanning with Cirrus OCT was performed using the 5-line raster scan mode to document the presence of SRF and pigment epithelial detachment. Central foveal thickness was measured with the macular cube 512 × 128 mode of the Cirrus OCT (128 lines, 512 A-scans per line). To analyze the changes of SRF, the authors calculated excess foveal thickness, referring to the amount above normal. Several studies have reported the normal range of central foveal thickness obtained by Cirrus OCT to be from 257 to 300 μm. The authors calculated excess foveal thickness by subtracting normal value, 250 μm, from the measured values of central foveal thickness. FA and ICGA (Heidelberg Retina Angiography; Heidelberg Engineering, Heidelberg, Germany) were performed at baseline, month 3, and month 6. The status of RPE was evaluated, including RPE depigmentation or atrophy based on the fundus examination and FA findings.
Statistical Analysis
The primary purpose of this study was to compare the anatomic outcomes between low-fluence PDT and intravitreal ranibizumab injection, including changes of excess foveal thickness and complete resolution of SRF. The changes in BCVA, concomitant with structural changes, also were analyzed. Both ocular and systemic adverse events related to the treatments were assessed. Serial comparisons of mean BCVA and excess foveal thickness were conducted using the Wilcoxon signed-rank test. Comparison according to treatment method was carried out using the Mann–Whitney U test and Kruskal-Wallis test. Categorized variables were analyzed by the Fisher exact test. SPSS software version 15.0 (SPSS, Inc, Chicago, Illinois, USA) was used for data analysis, and a P value of < .05 was considered to be statistically significant.
Results
A total of 16 eyes of 15 patients were recruited in this study. The mean age was 48.9 ± 7.6 years (range, 35 to 65 years). Thirteen men and 2 women participated. The duration of symptoms ranged from 6 months to 6 years, with a mean of 28.9 ± 23.6 months. The mean logMAR BCVA was 0.34 ± 0.24, and the mean excess foveal thickness was 50.2 ± 57.2 μm (range, −30 to 180 μm) before treatment. At baseline, 2 eyes (12.5%) had macular pigment epithelial detachment. The study eyes were randomized into 2 treatment groups: 8 eyes in the low-fluence PDT-treated group and 8 eyes in the intravitreal injections of ranibizumab group. At baseline, no significant differences in age, gender, duration of symptoms, or logMAR BCVA were shown between the 2 treatment groups. The mean excess foveal thickness at baseline was significantly higher in the low-fluence PDT group (74.1 ± 56.0 μm) compared with the ranibizumab group (26.3 ± 50.6 μm; P = .046). The mean PDT spot size used on the low-fluence PDT group was 2687.5 ± 1254.1 μm (range, 1600 to 4500 μm). The demographic features and outline of treatment outcomes of each study eye are presented in Table 1 .
| Patient No. | Sex | Age (y) | Symptom Duration (mos) | Primary Treatment | Best-Corrected Visual Acuity (logMAR) | Excess Foveal Thickness (μm) | OCT Features: Changes in Subretinal Fluid | Period at Retreatment (mos) | Retreatment | |||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Baseline | 3 Mos | 6 Mos | Baseline | 3 Mos | 6 Mos | 3 Mos a | 6 Mos b | |||||||
| 1 | M | 47 | 12 | PDT | 0 | 0.08 | 0 | 180 | −19 | −20 | Decreased | Sustained | 5 | Ranibizumab |
| 2 | M | 52 | 12 | PDT | 0.68 | 0.72 | 0.74 | 9 | −106 | −105 | Complete c | Complete | ||
| 3 | F | 44 | 6 | PDT | 0.12 | −0.08 | −0.08 | 100 | −50 | −43 | Complete | Complete | ||
| 4 | M | 65 | 48 | PDT | 0.6 | 0.42 | 0.42 | 108 | −68 | −77 | Complete | Complete | ||
| 5 | M | 63 | 6 | PDT | 0.3 | 0.22 | 0.42 | 89 | −6 | 7 | Complete | Complete | ||
| 6 | M | 46 | 48 | PDT | 0.4 | 0.06 | −0.04 | 32 | −40 | −40 | Complete | Complete | ||
| 7 | M | 52 | 60 | PDT | 0.16 | −0.08 | −0.08 | 40 | −39 | −39 | Complete | Complete | ||
| 8 | M | 53 | 60 | PDT | 0.12 | 0.12 | 0.16 | 35 | −45 | −48 | Sustained | Decreased | 3 | Ranibizumab |
| 9 | M | 47 | 8 | Ranibizumab | 0.4 | 0.16 | 0.18 | 4 | 6 | 94 | Sustained | Increased | ||
| 10 | M | 40 | 7 | Ranibizumab | 0.12 | 0.06 | 0.04 | 98 | −4 | 62 | Decreased | Increased | ||
| 11 | M | 42 | 10 | Ranibizumab | 0.18 | 0.12 | 0.14 | −20 | −51 | −46 | Complete | Complete | ||
| 12 | F | 47 | 6 | Ranibizumab | 0.3 | 0.14 | 0.02 | 69 | 74 | −80 | Sustained | Complete | 4 | PDT |
| 13 | M | 47 | 36 | Ranibizumab | 0.34 | 0.22 | 0.08 | −30 | −104 | −105 | Complete | Complete | 5 | PDT |
| 14 | M | 52 | 48 | Ranibizumab | 0.66 | 0.36 | 0.38 | 2 | −17 | −35 | Decreased | Complete | 4 | PDT |
| 15 | M | 50 | 24 | Ranibizumab | 0.84 | 0.56 | 0.02 | −1 | −3 | −70 | Sustained | Complete | 3 | PDT |
| 16 | M | 35 | 72 | Ranibizumab | 0.22 | −0.18 | −0.12 | 88 | −86 | −84 | Complete | Complete | ||
a The change in amount of subretinal fluid at 3 months was determined compared with baseline.
b The change in amount of subretinal fluid at 6 months was determined compared with that at the 3-month follow-up visit.
Because rescue treatment was considered after completion of primary treatment, the same treatment method was maintained in each group until the 3-month follow-up. Rescue treatments were conducted in 6 (37.5%) of 16 eyes that showed reaccumulated or sustained SRF during the subsequent follow-up period: 2 eyes in the low-fluence PDT group and 4 eyes in the ranibizumab group. Also, rescue treatment was considered in 2 eyes in the ranibizumab group because of aggravation of SRF at the last follow-up visit. The proportion of eyes needing rescue treatment in the ranibizumab group (6 eyes; 75%) was higher than that in the low-fluence PDT group (2 eyes; 25%), but the difference was not statistically significant ( P = .066). In the ranibizumab group, the mean PDT spot size as rescue treatment was 2240 ± 811.2 μm (range, 1500 to 3500 μm). In 1 eye, 2 distinct lesions of choroidal hyperpermeability responsible for the subfoveal fluid were treated by separate PDT spots with widths of 2100 μm and 1500 μm.
Changes in Anatomic Findings Assessed by Optical Coherence Tomography
In the low-fluence PDT group, the mean excess foveal thickness was reduced significantly from 74.1 ± 56.0 μm at baseline to −35.4 ± 44.5 μm at 3 months ( P = .017). The SRF resolved completely in 6 eyes (75%) with significant reduction of excess foveal thickness throughout the follow-up period, whereas 2 eyes (25%) showed persistent SRF despite an additional ranibizumab injection. The excess foveal thickness of these 2 eyes also did not demonstrate significant reduction at the last follow-up visit ( P = .180).
In the ranibizumab group, the mean excess foveal thickness decreased from 26.3 ± 50.6 μm at baseline to −23.1 ± 56.5 μm at 3 months, but was not considered to be statistically significant ( P = .058). After completion of consecutive ranibizumab injections, only 2 eyes (25%) maintained complete resolution of SRF throughout the follow-up period. A total of 6 eyes revealed persistent SRF after completion of ranibizumab injections. Four eyes (50%) with persistent SRF underwent additional low-fluence PDT and achieved complete absorption of SRF by 6 months. In the remaining 2 eyes (25%) with sustained SRF, rescue treatment was not initiated because there was a tendency toward reduction of SRF. However, the accumulation of SRF was aggravated at 6 months. The changes of excess foveal thickness in each study eye are shown in Figure 1 .
Changes in Best-Corrected Visual Acuity
In the low-fluence PDT group, the mean BCVA improved from 0.30 ± 0.37 at baseline to 0.18 ± 0.27 at 3 months, but this was not statistically significant ( P = .075). Throughout the follow-up period, a significant improvement of BCVA was not observed, regardless of whether rescue treatment was performed. In the ranibizumab group, the mean BCVA improved significantly from 0.38 ± 0.25 at baseline to 0.18 ± 0.22 at 3 months ( P = .012). During the subsequent follow-up, a subanalysis in the ranibizumab group was performed according to the need for rescue treatment. At the last follow-up, there was improvement of BCVA in both subgroups: 0.13 ± 0.17 in the eyes requiring additional low-fluence PDT and 0.06 ± 0.13 in those receiving only ranibizumab injections. However, these results did not show statistical significance ( P = .062 and P = .068). There was no significant difference of BCVA between the 2 groups at 3 months ( P = .606).
Angiographic Changes
Active leakage on FA was observed in 12 eyes at baseline: 6 in the low-fluence PDT group and 6 in the ranibizumab group. The remaining 4 eyes showed multiple mottled hyperfluorescent areas without active leakage at baseline. After completion of primary treatment, 5 eyes in the ranibizumab group showed persistent active leakage on FA with only moderate reduction. At 6 months, persistent leakage was regressed in 4 eyes after an additional low-fluence PDT. All eyes in the low-fluence PDT group demonstrated complete resolution of active leakage after primary treatment, regardless of the presence of SRF.
On ICGA, 6 eyes (75%) with completely resolved SRF in the low-fluence PDT group showed narrowing of dilated choroidal vasculature and significant reduction of choroidal hyperfluorescence leakage at 3 and 6 months. In 2 eyes (25%) with sustained SRF, only a subtle reduction of choroidal hyperfluorescence was observed on ICGA at the last follow-up, despite an additional ranibizumab injection.
In the ranibizumab group, 2 eyes (25%) that maintained complete resolution of SRF after consecutive ranibizumab injections demonstrated slight reduction of choroidal hyperfluorescence on ICGA at 3 and 6 months ( Figure 2 ). In 6 eyes (75%) with sustained SRF after completion of 3 ranibizumab injections, intense choroidal hyperfluorescence with dilated choroidal vessels on ICGA was demonstrated, although a moderate reduction of leakage on FA was observed. Among them, 4 eyes received additional low-fluence PDT as rescue treatment, and there was a marked reduction in choroidal vascular hyperpermeability and a narrowing of dilated choroidal vessels after PDT ( Figure 3 ).
