Anastomotic vessels remain viable after photodynamic therapy in primate models of choroidal neovascularization.

Purpose: Anastomotic vessels in exudative age-related macular degeneration (AMD) represent a serious clinical feature that reportedly does not respond well to either photocoagulation or photodynamic therapy (PDT). Anastomoses also occur in various animal models of choroidal neovascularization (CNV). In the present study, anastomotic vessels and their patency were evaluated in two primate CNV laser-trauma models after PDT, by using two novel photosensitizers.

Methods: In cynomolgus (Macaca fascicularis) and squirrel (Saimiri sciureus) monkey eyes (n = 20), matrix placement of laser photocoagulation sites elicited CNV as a component of the development of fibrovascular tissue (FVT). FVT sites received PDT according to specific drug infusion and laser light treatment parameters. FVTs and anastomoses were evaluated by fundus photography, fluorescein angiography, and histologic examination.

Results: Anastomoses averaged approximately 48% of FVT sites, with greatest occurrence in the macaque. Although PDT with each photosensitizer effectively produced FVT closure, both retinal vessels and anastomoses remained patent.

Conclusions: Although PDT is effective in closing the choroidal neovascularization in FVT, this technique was ineffective in occluding anastomotic vessels and their associated tributaries within the mid- to proximal retina. Various factors (vascular diameter, composition, blood flow, orientation) may contribute to continued anastomotic patency. By convention, such vessels would typically be defined as chorioretinal anastomoses (CRAs); however, continuing studies suggest the possibility that these neovessels constitute dual-origin hybrids. Regardless of origin, viable anastomoses provide one potential mechanism for revascularization to occur after PDT and may help to explain why CRAs are considered a poor prognostic sign in patients with AMD.