Why did a 38-year-old woman with neurofibromatosis type 1 (NF1) develop choroidal Yasunari nodules in both eyes? That’s what Danish clinicians had to figure out when the patient presented for a routine ophthalmic examination.

The patient had no known history of any systemic vasculopathy, though she did have a long history of peripheral Retinal capillary nonperfusion affecting her left eye, and years before received treatment with panretinal photocoagulation to manage this, Jakob Bjerager, MD, of Rigshospitalet in Glostrup, Denmark, and colleagues detailed in JAMA Ophthalmology.

Choroidal Yasunari nodules are generally difficult to detect on an ophthalmoscopic examination, the authors noted, but they are easily visualized using near-infrared photography.

Examination revealed no additional pathologies affecting the patient’s right eye, and her corrected visual acuity was 20/20 in both eyes. No gliomas were seen on recent MRI of the neuraxis and orbits.

When the team compared their findings with those from an assessment 3 years previously, they noted that the Central Retinal Artery and branching arcades showed signs of thinning. They also observed formation of a chorioretinal anastomosis in the superotemporal macula, which had developed at the site of a photocoagulation scar, they wrote, and “a spiral-shaped choroidal vessel interconnected with both arterial and venous vessels from the superior arcades and a vein from the inferior arcade, which perfused the arterial supply of the central retina, including the fovea.”

Bjerager and team confirmed this new retrograde arterial supply through the anastomosis using fluorescein angiography.

They noted that sequences at 0.5, 1, 5, and 20 seconds revealed “a conspicuous, rapid initial filling of the temporal macular vasculature by retrograde circulation through the laser-induced chorioretinal anastomosis (LICRA) in the superotemporal macula … which is sequentially followed by anterograde filling of the central retinal artery branches that brings perfusion to the nasal retina.”

The authors theorized that the anastomosis may have developed due to “a combination of a hydrostatic pressure gradient over the retinal pigment epithelium in the direction of the retina and vascular endothelial growth factor production associated with retinal capillary nonperfusion.”

Discussion

Data from larger cohorts have suggested that retinal vascular abnormalities affect 17% to 37% of people with NF1, depending on how these abnormalities are defined, noted Bjerager and co-authors.

Although severe retinal vascular abnormalities occur rarely in the setting of NF1, they said that peripheral or central retinal vascular occlusion, neovascular glaucoma, or occlusion of the central retinal artery or ophthalmic artery have been previously documented in at least 13 patients.

As observed in this patient, most of these cases were younger patients with impaired arterial inflow and worsening nonperfusion of the retinal capillary. That choroidal perfusion was preserved in this case suggests that vasculopathy only affected the arteries “distally to the branching of the posterior ciliary arteries from the ophthalmic artery,” the authors wrote.

Treatment with high-energy photocoagulation can cause defects in the Bruch membrane and mediate development of new blood vessels from the choroid. Under specific conditions, these can develop into a LICRA. It is this phenomenon that serves as the basis for treatment of nonischemic central retinal vein occlusion (CRVO), wrote Bjerager and co-authors.

The treatment aims to normalize venous pressure in the retina by creating auxiliary flow between the occluded high-pressure retinal venous circulation and the unobstructed low-pressure choroidal venous circulation.

In a study of 58 selected patients who received intravitreal ranibizumab injections for nonischemic CRVO, those randomly assigned to a LICRA procedure had superior visual acuity and needed fewer injections of anti-vascular endothelial growth factor compared with controls after 2 years.

“However, only two in three attempts to create an anastomosis were successful, and procedure-related complications requiring secondary surgery occurred for every third patient,” the authors wrote.

Furthermore, most efforts to create LICRA in eyes with ischemic CRVO have been unsuccessful, which Bjerager and team suggested may be related to significant damage sustained by endothelial cells due to the detrimental effects of ischemia and venous thrombosis on retinal circulation.

In the patient reported here, the authors opined that the incidental LICRA related to treatment may have saved the retina and helped maintain 20/20 visual acuity in an eye with severely occluded retinal arteries.

“Here, the mechanistic mode of action of the LICRA differed from that of eyes with CRVO, in that the chorioretinal anastomosis causes inflow of oxygenated blood to the retina rather than outflow from congested retinal veins to the choroid,” noted Bjerager and co-authors.

They cautioned that given the full year of considerable vascular remodeling that occurred in this patient, “more evidence is needed to suggest that therapeutic LICRAs can be created in patients with severe, progressive retinal arterial occlusive vasculopathy where choroidal perfusion is still intact.”