Unusual Loop-Like Vessel Formation After Intravitreal Ranibizumab Injections in AMD

Introduction

We observed an unusual closed loop-like vessel formation and its regression in the course of multiple intravitreal ranibizumab injections in an eye with exudative age-related macular degeneration. The development of such unique vessel formation might be attributed to the heterogenous treatment effect of ranibizumab on the choroidal neovascular components.

Case Report

A 76-year-old healthy woman was referred with the diagnosis of exudative age-related macular degeneration in both eyes. She had a treatment history of transpupillary thermotherapy in both eyes, two sessions of verteporfin photodynamic therapy combined with intravitreal steroid injection, an intravitreal bevacizumab injection, and an intravitreal ranibizumab (Lucentis; Genentech, Inc., San Francisco, CA) injection in only the left eye.

In October 2006, the right eye showed a large disciform scar in the macula. The left eye showed subfoveal choroidal neovascularization surrounded by subretinal fluid, subretinal hemorrhage, and lipid exudates (Figure). On biomicroscopic examination, the left eye demonstrated a large-size vessel on the surface of the central fibrotic scar. The large-size vessel originated from the grayish, active choroidal neovascularization lesion and had a bifurcation connecting to two separate sites of retinochoroidal anastomoses, seemingly functioning as a feeder choroidal vessel.¹

Figure. The Left Eye of a 76-Year-Old Woman with Bilateral Exudative Age-Related Macular Degeneration Associated with Retinochoroidal Anastomoses (RAs). (A) Color Fundus Photograph Showing Subfoveal Choroidal Neovascularization (CNV), Subretinal Fluid, Inferotemporal Subretinal Hemorrhage, and Lipid Exudates. Visual Acuity (VA) was 20/200. (B) The Magnified Color Fundus Photograph of Figure A Demonstrated a Large-Size Vessel (open Arrowhead) that Originated from the Grayish, Active CNV Component Superiorly Adjacent to the Subfoveal Fibrotic Scar. The Bifurcation (solid Arrowheads) of the Large-Size Vessel was Seen on the Surface of the Fibrotic Scar, Connecting to Two Separate Sites of RA (arrows). (C) Fluorescein Angiogram Showed Mild Staining of the Fibrotic Component of the CNV. (D) The Magnified Fluorescein Angiogram of Figure C Indicates the Retinal Arterioles (A1, A2) and the Retinal Venules (V1, V2) in the Proximity of the RA. Compared to the Color Fundus Photographs (Figs. A and B), One RA Appears to have a Connection only with the Retinal Venule (V1), Whereas the Other RA Connects with Both the Arteriole (A2) and the Venule (V2). (E) After an Intravitreal Ranibizumab Injection, the Ingrowth Site (Fig. B, Open Arrowhead) of the Presumed Feeder Choroidal Vessel had Regressed and an Unusual Closed Loop-Like Vessel (arrowheads) had Developed. One Arteriole (Fig. D, A1) Became Invisible, but Both RAs Remained (arrows). (F) One Month After an Additional Intravitreal Ranibizumab Injection, the Loop-Like Vessel and One of the Draining Retinal Venules (Fig. D, V1) Disappeared, Leaving One RA Site (arrow) and the Related Anastomotic Arteriole and Venule (Fig. D, A2 and V2). VA was 20/200.

After an intravitreal ranibizumab injection (0.5 mg) in December 2006, exudation from the active choroidal neovascularization component had markedly resolved. In addition, the ingrowth site of the presumed feeder choroidal vessel had regressed, leading to formation of an unusual closed loop-like vessel in the macula. The loop-like vessel was similar in diameter to the connecting retinal arteriole and venule. One month after an additional ranibizumab injection (0.5 mg), the loop-like vessel had disappeared, leaving one retinochoroidal anastomosis site.

Discussion

Retinochoroidal anastomoses are characteristic findings in end-stage age-related macular degeneration. The branching feeder choroidal vessel in our case might be acting as a form of compensation for the demand of the blood supply from the two separate anastomotic sites. Recently, intravitreal ranibizumab injection has been shown to be a promising therapeutic approach for the treatment of exudative age-related macular degeneration.² After one intravitreal ranibizumab injection, the active choroidal neovascularization component and the ingrowth site of the feeder choroidal vessel appeared to be selectively obliterated with the retention of two sites of retinochoroidal anastomosis. As a result, the unusual closed loop-like vessel might then have developed.

Furthermore, taking into consideration the diameter of the loop-like vessel, it was most likely that this unusual vessel was fed and drained only by the retinal circulation, without participation of the choroidal circulation. Conclusions regarding the pathophysiologic implications of this case are limited by the fact that this is a single case report and the lack of sequential high-speed indocyanine green angiograms that may have better delineated the exact contributions of the two circulatory systems.

We observed an unusual loop-like vessel formation and its regression caused by the heterogenous treatment effects of intravitreal ranibizumab injections on the choroidal neovascularization associated with multiple retinochoroidal anastomoses.