Photodynamic therapy using verteporfin for choroidal neovascularization in angioid streaks☆

Methods 

A retrospective analysis of 11 eyes of nine patients diagnosed with subfoveal and juxtafoveal choroidal neovascularization associated with angioid streaks and treated by the authors from April 2000 through October 2001 was conducted. Fluorescein angiographic criteria, using definitions previously described,12, 13 included evidence of choroidal neovascularization caused by angioid streaks that extended adjacent to or underneath the center of the foveal avascular zone. Patients ranged in age from 44 to 79 years (mean, 58.2 years). Four eyes had previously undergone conventional laser photocoagulation for extrafoveal and juxtafoveal choroidal neovascularization. All of these eyes developed subfoveal recurrences prompting treatment with verteporfin. All patients underwent color photography, fluorescein angiography, and complete ocular examinations with determination of best-corrected visual acuity (BCVA) both before treatment and at all subsequent follow-up visits. When appropriate, patients underwent further diagnostic testing for associated systemic diseases. After informed consent was obtained, patients were treated with verteporfin according to the standard protocol followed in the Treatment of Age-Related Macular Degeneration with Photodynamic Therapy Investigation (TAP).11 The follow-up period ranged from 5 to 28 months (mean, 17 months). Fluorescein angiography and funduscopic examination were performed at each follow-up visit. Fluorescein angiograms were evaluated for lesion size, conversion of the CNVM to a fibrous disciform lesion, and leakage of the choroidal neovascular membrane. Re-treatment was performed based on criteria previously established by the TAP study,9, 10, 11, 12 except in one case in which early re-treatment was performed at 6 weeks. Intervals for re-treatment ranged from 6 to 28 weeks (mean, 15 weeks).

Results 

Clinical characteristics and demographics of the patient population are presented in Table 1. Eleven eyes of nine patients with subfoveal (nine eyes) or juxtafoveal (two eyes) CNV from angioid streaks were treated with photodynamic therapy. Six of these patients were women. Predominantly classic CNV (greater than 50%) occurred in 10 eyes. Lesion size varied from 600 to 6000 μm (mean, 2900 μm). Nine of the eleven eyes displayed conversion of the CNVM into a fibrous disciform lesion by biomicroscopic examination at final follow-up. Enlargement of the CNVM was documented in seven of the 11 eyes by fluorescein angiography at final follow-up (FIGURE 1, FIGURE 2, FIGURE 3).

TABLE 1. Clinical Characteristics of the Study Population

	Case	Age/Sex/History	Eye	Duration	Lesion Size (microns)	Preop BCVA	Postop BCVA	Anatomic Outcome	Treatments(interval wks)	Follow up
	1*	79 yo, M, Paget’s	OD	11 m	2600	20/400	CF	larger, fibrous change	1	20 m
	2*	79 yo, F	OS	4 m	2400	20/200	CF	larger, fibrous change	2 (16)	18 m
	3*	65 yo, F, PXE	OD	60 m	4000	CF	20/400	stable	2 (14)	21 m
			OS	20 m	4000	CF	CF	larger, fibrous change	1	19 m
	4	45 yo, F	OS	1 w	2400	CF	CF	larger, fibrous change	2 (15)	18 m
	5†	54 yo, M, PXE	OS	1 w	2200	20/25	20/400	larger, fibrous change	3 (12, 8)	12 m
			OD	1 w	600	20/30	20/20	stable	2 (6)	5 m
	6	63 yo, F	OS	1 m	800	20/200	20/400	fibrous change	4 (15, 16, 28)	15 m
	7‡	44 yo, F, PXE	OS	37 m	2400	20/60	20/50	fibrous change	2 (12)	10 m
	8*	45 yo, F, PXE	OD	11 m	6000	20/200	5/200	larger, fibrous change	4 (12, 16, 28)	25 m
	9	50 yo, M, PXE	OS	17 m	2800	20/200	20/100	larger, fibrous change	5 (16, 18, 16, 15)	28 m

BCVA = best-corrected visual acuity; CF = counting fingers; OD = right eye; OS = left eye; PXE = Pseudoxanthoma elasticum.

* History of previous conventional laser photocoagulation for extra/juxtafoveal choroidal neovascularization with subfoveal recurrence. † Both eyes with juxtafoveal choroidal neovascularization. ‡ Predominantly occult, all others predominantly classic.

View Large Image Download to PowerPoint

FIGURE 1. Case 2. Pretreatment red-free (top left) and angiographic (bottom left) photos demonstrate subfoveal choroidal neovascular membrane adjacent to previous extrafoveal laser scar. Posttreatment red-free (top right) and angiographic photos (bottom right) at 12 months reveal marked enlargement and disciform conversion of the neovascular complex associated with hemorrhage and leakage.

View Large Image Download to PowerPoint

FIGURE 2. Case 5. Pretreatment fluorescein angiogram reveals juxtafoveal choroidal neovascularization (top left) with minimal enlargement before early re-treatment 6 weeks later (bottom left). Fellow eye demonstrates subfoveal extension of juxtafoveal neovascular complex (top right) with increased leakage and fibrous change (bottom right) at 9 months posttreatment angiogram.

View Large Image Download to PowerPoint

FIGURE 3. Case 9. Pretreatment red-free (top left) and angiographic (bottom left) photos demonstrate subfoveal choroidal neovascular membrane associated with hemorrhage. Posttreatment red-free (top right) and angiographic photos (bottom right) at 8 months reveal marked enlargement, increased leakage, and disciform conversion of the neovascular complex.

Best-corrected visual acuity before therapy ranged from 20/25 to counting fingers (CF) (mean, 20/400; median, 20/200) and BCVA at final follow-up ranged from 20/20 to CF (mean, 20/600; median, 20/400). Visual acuity of counting fingers was assigned a value of 20/1000 for statistical purposes. Seven eyes had vision of 20/200 or better preoperatively, while at final follow-up, only three eyes maintained this level of vision or better. In five eyes, the BCVA decreased, most notably from 20/25 to 20/400 in one case of a juxtafoveal choroidal neovascular membrane. Of note, in the fellow eye of this patient also with a juxtafoveal neovascular membrane BCVA remained stable at the 20/20 level with 5 months of follow-up after early re-treatment at 6 weeks (Figure 2). Final BCVA was improved in four of 11 eyes and stable in two of 11 eyes at last follow-up.

Duration from onset of symptoms inclusive of previous conventional laser therapy ranged from 1 week to 6 years (mean, 14.7 months; mode, 1 week) and did not appear to correlate with postoperative visual or anatomic outcome. Additionally, the size of the lesion also did not appear to correlate with final visual or anatomic outcome. All but two eyes were treated multiple times (range, 1–5 treatments; mean, 2.6 treatments). Both of these eyes under-went conversion to a fibrous disciform lesion after the first treatment session.

Discussion 

Loss of vision from choroidal neovascularization is the most serious ocular complication of angioid streaks. Bilateral CNV occurs in 42% to 60% of cases with an interlapsing time of approximately 18 months,4 and most patients become legally blind by the age of 50.6 Our treatment results are comparable to reported eyes treated for CNV secondary to angioid streaks, although most eyes previously described underwent conventional laser photocoagulation not involving the foveal center as a result of recommendations by the original Macular Photocoagulation Studies.4, 7, 14, 15, 16, 17, 18.

In a series of 21 eyes reported by Piro and associates,4 70% of eyes progressed to legal blindness (55% within 6 months of treatment). Initial visual acuity was 20/40 or better in over 80% of eyes. Lim and associates7 in their series of 24 eyes (CNV juxtafoveal in two eyes, subfoveal in three eyes, extrafoveal in 18 eyes, and indeterminate in one eye) with an initial mean BCVA of 20/50 noted a decreased average visual acuity of 20/120 at final follow-up.7 Brancato and associates17 and Gelisken and associates18 also reported slightly more favorable outcomes, but it should be noted that in all these series that the initial visual acuity was better and that subfoveal lesions were not included. Recurrence rates were high in both series. In our series of patients treated with photodynamic therapy, a high retreatment rate was similarly noted.

Gelisken and associates17 noted that lesions in the macula have the worst posttreatment prognosis, and Lim and associates7 suggest that the natural history of CNV associated with angioid streaks, even for extrafoveal and juxtafoveal lesions, eventually leads to subfoveal scarring. Conventional laser photocoagulation of extrafoveal choroidal neovascular lesions has been recommended because treated eyes maintain useful vision for a longer period of time than untreated eyes. Furthermore, it has been recommended that for successful treatment, eyes should be treated as early as possible, and CNV should be treated completely and adequately to not stimulate further neovascularization.17 Similarly, it is certainly possible that aggressive follow-up with shorter treatment intervals may be more effective in photodynamic therapy of subfoveal or juxtafoveal CNV secondary to angioid streaks. The parameters for follow-up and re-treatment, recommended currently at 3 months, are currently being reevaluated and may lead toward shorter treatment intervals.12 Of note, in one of our patients, the first eye with a preoperative BCVA of 20/25 was treated at 3-month intervals with progression of a juxtafoveal choroidal neovascular membrane to a fibrous disciform lesion and final BCVA of 20/400. On review of the patient’s history, he was reportedly symptomatic at less than 3 months after initial therapy and probably had recurrence of neovascularization at that time, but did not seek treatment until follow-up at which time retreatment was promptly administered. The second eye was treated more aggressively at 6 weeks, and BCVA has remained stable at 20/20 at final follow-up.

The chronicity of the condition before treatment is notable in many of our patients, most of whom were awaiting approval of photodynamic therapy. We were not able to correlate duration of neovascularization or history of previous conventional laser therapy with final visual or anatomic outcome. Additionally, size of the neovascular lesions did not appear to correlate with outcome either, although these conclusions are limited by the small size of our series. Our preliminary experience with photodynamic therapy for choroidal neovascularization associated with angioid streaks leads us to believe that it does not significantly alter the course of this disease, with most eyes undergoing enlargement and disciform transformation of the neovascular process. However, aggressive management of these patients with biomicroscopic and fluorescein angiographic examination and timely photodynamic therapy with early re-treatment when indicated may be beneficial in certain cases.