Photodynamic Therapy With Verteporfin in Subfoveal Choroidal NeovascularizationSecondary to Central Serous Chorioretinopathy

Objective  To examine the efficacy and safety of photodynamic therapy with verteporfinin the treatment of subfoveal choroidal neovascularization secondary to centralserous chorioretinopathy (CSC).

Design  Prospective interventional, noncomparative case series.

Methods  After the diagnosis of a subfoveal choroidal neovascularization secondaryto CSC, 26 eyes of 24 patients were treated with photodynamic therapy withverteporfin. Patients were then followed up every 2 to 3 months, with furthertreatments performed as deemed necessary through fluorescein angiography.The mean observation was 22.2 months (range, 6-36 months; median, 24 months).

Results  There was marked visual improvement, with patients gaining a mean of1.6 lines after 1 year and a mean of 2.2 lines after 2 years. There was astatistically significant change in visual acuity from baseline to 12 and24 months (mean difference, –0.16, P = .03;and mean difference, –0.22, P = .02; respectively; t test for both). There was no correlation between patients'age or greatest linear dimension of the lesions and the final outcome (P>.10 for all). No patient experienced any adverse effects.

Conclusion  Photodynamic therapy with verteporfin resulted in a beneficial outcomein the treatment of subfoveal choroidal neovascularization secondary to CSC,without serious adverse effects in this case series.

CENTRAL SEROUS chorioretinopathy (CSC) is a disease that typically affectsmiddle-aged adults and involves the sensory retina, retinal pigment epithelium(RPE), and choroid.¹ Patients usually havemild visual loss. This generally resolves without therapy, although the diseasecan become chronic, with ensuing RPE decompensation.^2,3 Somepatients, particularly older adults,⁴ can developchoroidal neovascularization (CNV), which leads to a severe loss in visualacuity.

Photodynamic therapy (PDT) with verteporfin is a new method for thetreatment of subfoveal CNV in various diseases.^5-16 Briefly,patients with a subfoveal CNV receive an infusion with verteporfin. Thereafter,the CNV is treated with a modified diode laser (689 nm). Besides conventionalthermal laser therapy,^17-22 thisis the only method for the treatment of CNV that has a proven efficacy inlarge, multicenter, randomized clinical trials.^5-9

The fact that these clinical trials were efficacious and had few treatment-relatedadverse effects led us to examine the efficacy of PDT as a treatment for subfovealCNV secondary to CSC, as there has hitherto been no proven treatment for thisdisorder. Because most patients are young and typically active in the workforce(so-called type A personalities²³), rehabilitationof these patients is particularly important.

Twenty-six eyes of 24 patients were treated. The age range was 36 to78 years (mean ± SD, 57.0 ± 13.0 years; median, 55 years). Nineteeneyes from men and 7 from women were treated.

All patients were examined for best-corrected Snellen visual acuityand underwent a dilated fundus examination and fluorescein and indocyaninegreen (ICG) angiography. Patients with subfoveal CNV defined on a fluoresceinor ICG angiogram were treated according to treatment guidelines.^24,25 Informedconsent was received from each patient before treatment, and institutionalreview board approval was obtained.

All patients were seen at 2- to 3-month intervals, and Snellen visualacuity examination, dilated fundoscopy, and fluorescein or ICG angiographywere repeated at each visit. The diagnosis of CSC was based on the patient'shistory and typical fluorescein angiographic and ICG leakage patterns. Theexistence of drusen was ruled out to avoid a misinterpretation with age-relatedmacular degeneration.

Patients' Snellen visual acuity was converted to logMAR units for statisticalassessment. The SPSS system (SPSS for Windows version 10.0; SPSS Inc, Chicago,Ill) was used for statistical computations. The t testwas used to assess the effect of change in visual acuity from baseline, andthe Spearman rank correlation (ρ) test and a logistic regression analysis(analysis of covariance) were used to assess the effect of individual variableson the change in visual acuity and the final outcome. All P values are results of 2-tailed tests. The chosen level of statisticalsignificance was P = .05.

Patients were followed up for 6 to 36 months (mean, 22.2 months; median,24 months) after beginning PDT. All eyes were available for evaluation at6 months, 24 (92%) of 26 eyes were available for follow-up after 1 year, and19 (73%) of 26 eyes were available after 2 years. Of those patients lost tofollow-up, 1 had a visual improvement of 3 lines at the last follow-up, 5had remained stable, and 1 had a significant visual loss of 4 lines at thelast follow-up examination.

Six months after beginning PDT, there was a clear treatment benefit.Seven (27%) of 26 eyes showed a gain in visual acuity of 3 lines or more,whereas 16 (62%) of 26 remained stable (within 2 lines). Three (12%) of 26eyes lost 3 lines or more. There was a mean ± SD gain of 0.09 ±0.33 logMAR units and a mean ± SD gain of 1.00 ± 0.12 line.

After 1 year, 12 (50%) of 24 eyes had an improvement in visual acuityof 3 lines or more, and 9 (38%) of 24 remained stable. Only 3 (13%) of 24eyes lost 3 lines or more. The mean ± SD gain was 0.16 ± 0.33logMAR units, and the mean ± SD gain from baseline was 2.00 ±0.22 lines.

After 2 years, 9 (47%) of 19 eyes had an improvement of 3 lines or more.Eight (42%) of 19 had stable visual acuity (within 2 lines), and 2 (11%) of19 had lost 3 lines or more. The mean ± SD gain was 0.22 ± 0.37logMAR units, or more than 2 lines. Figure1 shows the changes in visual acuity throughout the treatment.

The greatest linear dimension (GLD) of the CNV was 2225.4 µm beforetreatment. The GLD remained virtually unchanged during treatment; the meanGLD at the last treatment was 2410 µm. However, 11 patients requiredonly 1 treatment. None of the patients showed any leakage on fluorescein angiographyat their last follow-up.

There was a statistically significant difference in the change in meanlogMAR values from baseline at 12 and 24 months (mean difference, –0.16, P = .03; and mean difference, −0.22, P = .02), that is, patients gained 1.6 and 2.2 lines, respectively,although there was no statistically relevant change at 6 months (mean difference,–0.10; P = .07). Interestingly, neither patientage nor GLD of the lesion correlated with the change (gain or loss) in visualacuity after 1, 12, and 24 months (P>.10 for all)(Table 1). However, baseline visualacuity correlated with the change in visual acuity at 12 and 24 months. Theseresults were confirmed on regression analysis. Furthermore, there was a statisticallysignificant correlation between the change at 6 months and the changes at12 and 24 months (R² = 0.87, P<.001; and R² = 0.59, P = .04; respectively).

The mean number of treatments needed per patient was 2.6 (range, 1-7;median, 2). In the first year, a median of 2 treatments was necessary; inthe second, the median number was 1. None of the patients experienced adverseeffects. No photosensitivity reaction was seen.

Figure 2A-C shows images froma typical patient with CSC and CNV. Before PDT, the visual acuity was 20/200OS. The red-free image and the fluorescein angiogram show an active CNV lesion.After 2 years, the visual acuity was 20/22 OS, and angiograms show an inactiveCNV (Figure 2D-F). The patient had3 treatments with PDT.

This study shows that PDT is a beneficial treatment for subfoveal CNVsecondary to CSC. More than three quarters of all patients have stable orbetter visual acuity 2 years after beginning treatment. Because some patientswere lost to follow-up, it is possible that there is a bias to the treatmentbenefit, with those losing visual acuity not coming back. However, the resultsindicated that most patients had at least stable visual acuity, which minimizesany possible bias, in our opinion.

Conventional laser therapy offers a proven benefit in the treatmentof CNV, particularly in lesions that are outside the center of the fovealavascular zone (extrafoveal or juxtafoveal) and in small, well-demarcatedclassic CNV lesions that are located subfoveally.^17-22 However,subfoveal treatment leads to an immediate, marked loss of visual acuity asthe overlying neurosensory retina is damaged, and extrafoveal and juxtafovealCNVs treated with laser can also have a subfoveal recurrence. Most patientstreated with PDT experience a less dramatic loss of vision, and some describea subjective improvement in visual acuity that cannot be seen on an objectivevision examination. Furthermore, as seen in the Treatment of Age-Related MacularDegeneration With Photodynamic Therapy (TAP) study,^5-7 contrastsensitivity is much better in patients treated with PDT. This makes it a viablealternative in routine clinical practice.

Central serous chorioretinopathy has a heterogeneous course. Most casesresolve spontaneously without any treatment.^4,24 However,some persons, particularly older patients, develop chronic disease, leadingto decompensation of the RPE and other complications.^2-4,25-27 Onestudy⁴ demonstrated in older patients thatCNV, persistent pigment epithelial detachment, and the accumulation of subretinalfluid are primary causes of deteriorating visual acuity, aside from the frequentrecurrences of CSC. For extrafoveal leaks, focal laser photocoagulation hasbeen advocated^28-31;however, this procedure is known to induce CNV.³²

The exact cycle of events leading to CSC is still unknown. It is generallyaccepted that CSC develops primarily as a result of choroidal vascular hyperpermeability,⁴ which can be seen on ICG angiography.^33,34 Thisleads to an increase in choroidal tissue hydrostatic pressure, which supersedesthat of the retina, thus reducing or stopping solute flow across the RPE.As a result, a serous detachment can occur, as well as a change in the RPElayer that Carvalho-Recchia et al³⁵ have describedas a "blowout" of the RPE. It is not clear under what conditions CNV can develop.Our demographic results are, however, similar to those of Spaide et al⁴ in that 16 of 24 patients were older than 50 years.

Psychopharmacologic medications, corticosteroid use, and hypertension^30,35-38 arerisk factors for CSC. Investigations in cultured pigment epithelium cellsshow that epinephrine can induce apoptosis of RPE cells, although dexamethasonehad no effect.³⁹ Stress also plays a majorrole in this disease, particularly in highly motivated, eager individualswith a type A personality.²³ Also, an increasedactivity of the sympathetic autonomic nervous system has been seen in patientswith CSC,⁴⁰ which would corroborate these findings.

Why is treatment with PDT in patients with CNV secondary to CSC successful?On the one hand, lesion size might play a role. Although no statistical significancewas found in this study, the number of patients is relatively small, whichdoes not rule out an effect in a larger cohort. All patients had a lesionsize less than 4000 µm. The effect of lesion size has been shown tobe an important factor in the success of PDT,⁴¹ withsmaller lesions in age-related macular degeneration receiving more therapeuticbenefit.

Age might be an important factor. Investigations in patients with pathologicmyopia, which albeit has a different pathophysiology, have shown that youngerpatients benefit significantly more from therapy than older patients (E.E.,unpublished data, 2003). There was no statistical correlation in the presentstudy, but it stands to reason that the occlusive effect of PDT on CNV hasa better and more long-standing effect in younger individuals. Similarly,in idiopathic CNV, favorable results have been shown with PDT with verteporfin.^16,42

Another reason for the success of PDT in patients with CNV secondaryto CSC might be because PDT may also treat the long-standing CSC. Yannuzziet al⁴³ and others (F. C. Piccolino, MD, writtencommunication, May 2002) have shown that PDT can improve visual acuity inlong-standing CSC. Therefore, our patients might have had a 2-fold advantage:treatment of the disease and the CNV. This might give the method an advantageover other procedures, such as surgery, particularly because it is less invasive.⁴⁴

Our study also shows that the results after 6 months seem to set thecourse for further success or failure: ie, patients who experience an improvementafter 6 months tend to stay improved and vice versa. This mimics results fromthe TAP trials for age-related macular degeneration,^5,6 whichhave shown that the change in the first 6 months is the decisive period forthe final outcome.

In conclusion, PDT with verteporfin is a safe and efficacious methodfor the treatment of subfoveal CNV in CSC. A main issue is the differentiationfrom other diseases, particularly CNV secondary to age-related macular degenerationand secondary to polypoidal CNV.⁴⁵ The useof ICG angiography is essential in making the diagnosis.^46,47

Corresponding author: Erdem Ergun, MD, Department of Ophthalmology,University of Vienna Medical School, Allgemeines Krankenhaus, Waehringer Guertel18-20, A-1090 Vienna, Austria (e-mail: erdem.ergun@akh-wien.ac.at).

Submitted for publication January 7, 2003; final revision received August11, 2003; accepted September 10, 2003.

Dr Stur was principal investigator in 2 studies of verteporfin (TAPand Verteporfin in Photodynamic Therapy [VIP] trials) and is principal investigatorand vice-chair in an ongoing verteporfin study (Verteporfin in Early Retreatment[VER] trial), as well as a permanent member of the Verteporfin Studies AdvisoryGroup. Dr Ergun is a coinvestigator in the VER trial.

This study was presented in part as a poster at the American Academyof Ophthalmology; November 12, 2001; New Orleans, La.