Editor, Sarcoidosis is a multi-organ disorder characterized by non-caseating granulomas. This disease has been reported to be the most common of confirmed aetiologies among all categories of uveitis (Kitamei et al. 2009). Choroidal neovascular membrane (CNVM) occurs very rarely, but is vision-threatening when it involves peripapillary locations (Gragoudas & Regan 1981; Hoogstede & Copper 1982). Reported treatments of sarcoid-related peripapillary CNVM, including systemic, oral and periocular steroids and argon laser photocoagulation, are controversial and not universally effective (Gragoudas & Regan 1981; Spalton & Sanders 1981; Hoogstede & Copper 1982; Frank & Weiss 1983). Elevated vascular endothelial growth factor (VEGF) has been implicated in systemic sarcoidosis (Matsuda et al. 2004). A recent report with short-term follow-up reported significant therapeutic response to bevacizumab in a patient with sarcoidosis-related posterior uveitis complicated by CNVM (Kurup et al. 2009). Likewise, sustained resolution was reported in a patient with inflammatory CNVM related to toxoplasmosis when treated with intravitreous bevacizumab (Guthoff & Goebel 2008). In contradiction, we report a patient with sarcoid peripapillary CNVM which was only transiently less active when treated with multiple ranibizumab and bevacizumab intravitreous injections over a longterm period. In December 2002, a 62-year-old man presented with bilateral panuveitis and a peripapillary CNVM in the left eye (Fig. 1). His best corrected visual acuity (BCVA) was 20/20 OD and 20/50 + 2 OS. The patient was diagnosed with biopsy-proven sarcoidosis and treated unsuccessfully with high-dose oral prednisone and periocular triamcinolone. There was complete resolution of the CNVM after adequate peripapillary argon laser photocoagulation (Fig. 2). (A, B) Colour and red-free fundus photographs in the left eye showing peripapillary choroidal neovascular membrane with associated subretinal fluid. (C) Early-phase angiogram with early hyperfluorescence in the peripapillary region. (D) Angiogram showing late leakage in the same distribution. (A, B) Colour and red-free fundus photographs in the left eye after peripapillary argon laser photocoagulation. (C, D) Early- and late-phase angiography revealing window defects without leakage in the area of previous peripapillary choroidal neovascular membrane. The patient was lost to follow-up and presented to our institute again in November 2006. Visual acuity was 20/20 − 2 OD with peripapillary haemorrhage and subretinal fluid along the temporal optic disc margin without an anterior or posterior chamber cellular reaction; VA OS was 20/40 and an inactive peripapillary photocoagulation scar was noted. From November 2006 to January 2007, the patient received monthly injections of intravitreous ranibizumab in the right eye. Temporary anatomical and functional improvement was noted after each injection. In March 2007, a new, second CNVM was observed along the superior peripapillary distribution with subretinal fluid. The eye was re-injected with intravitreous ranibizumab. One month later, growth and activity of both CNV lesions persisted. Because the new lesion was located along the superior edge of the optic nerve outside the maculopapular bundle, it was treated with peripapillary argon laser photocoagulation in April 2007. Although the superior CNVM resolved within 2 weeks of photocoagulation, submacular fluid from the temporal CNVM persisted (Fig. 3). Between April 2007 and December 2007, the patient received five additional intravitreous ranibizumab injections with transient improvement, but global growth of the CNVM continued. (A) Colour photograph in the right eye encompassing the peripapillary choroidal neovascular membrane complex along the temporal border of the optic nerve. (B) Red-free photograph suggests subretinal fluid in the same distribution. (C, D) Early- and late-phase angiography in the right eye showing a punctuate area of hyperfluorescence with late leakage. Also note the distribution of the superior peripapillary scar of the choroidal neovascular membrane previously treated by laser. During this time–course the patient was noted to have a worsening cataract and epiretinal membrane in the right eye. He underwent 25-gauge pars plana vitrectomy, lensectomy, posterior chamber intraocular lens implant, membrane peel and intravitreous injection of 0.1 cc of triamcinilone 40 mg/cc in December 2007. Although postoperative VA improved to 20/30 − 2, the CNVM remained active through to March 2008, despite two additional bevacizumab injections. Overall, despite nine ranibizumab and two bevacizumab injections administered over a 16-month period, subretinal fluid began to encroach closer to the fovea (Fig. 5A). The patient was finally treated with peripapillary photocoagulation along the maculopapular bundle in the right eye (Fig. 4). Four months after treatment, VA was 20/50 + 1 with pinhole to 20/25 + 1 without subretinal fluid (Fig. 5B). (A) Optical coherence tomography (OCT) scan through the maculopapular bundle in February 2008 showing subretinal fluid threatening the fovea despite nine ranibizumab and two bevacizumab intravitreous injections administered over a period of 16 months. (B) OCT scan through the same region showing resolution of subretinal fluid in May 2008 after peripapillary argon laser along the maculopapular bundle in the right eye. (A, B) Colour and red-free photographs after peripapillary argon laser photocoagulation to the choroidal neovascular membrane along the temporal border of the right optic nerve. (C, D) Window defects along this distribution without leakage are evident on angiography. In conclusion, we question recent reports of the longterm efficacy of anti-VEGF therapy in treating inflammatory CNVM. For sarcoid-related peripapillary CNVM in particular, argon laser photocoagulation may be more effective in the longterm, as it was in the left eye of our patient (prior to the anti-VEGF era) as well as in the right eye despite multiple intravitreous anti-VEGF injections.
No takes yet. Share an insight, caveat, or question.
Shah et al. (2009) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: