Main

Sir,

Venous Chorioretinal anastomosis has been induced using a variety of techniques in the treatment of retinal vein occlusions. This process can also occur in a number of infective, inflammatory, or traumatic processes. We describe the formation and angiographic appearance of Chorioretinal venous and arterial anastomoses following a nonpenetrating eye injury.

Case report

A 40-year-old pseudophakic white male patient presented following a blunt injury to the right eye, reportedly caused by a punch. This man had had bilateral cataract extractions using a phacoemulsification technique with intraocular lens insertion 2 years previously for idiopathic presenile cataract.

On initial examination, visual acuity was 6/18 right eye and 6/5 left eye. On the right side, a partial dislocation of his posterior chamber intraocular lens into the anterior chamber was present. The intraocular pressure was 13 mmHg on both sides by applanation tonometry. On dilated examination, a vitreous haemorrhage and an inferior Chorioretinal rupture could be seen.

The Chorioretinal rupture measured approximately five disc diameters in length and originated one and a half disc diameters inferior to the optic nerve head. The rupture passed inferotemporally, following a curvilinear course that transected the inferior temporal retinal arteriole and vein. The left eye examination was unremarkable.

Following initial treatment with topical steroid and antibiotic drops, visual acuity improved to 6/9 within 8 weeks of the injury. Anterior uveitis with secondary glaucoma continued to intermittently cause symptoms.

On review at 5 months following the injury, it was noted that although the inferior temporal retinal arteriole proximal to the chorioretinal rupture was stenosed and did not appear to have significant blood flow the arteriole proximal to the rupture was of normal calibre and appearance (Figure 1). The inferotemporal retina peripheral to the chorioretinal rupture appeared healthy with no evidence of pallor, oedema, or exudates.

Figure 1
figure 1

Colour fundus photograph of the right eye 5 months following closed globe trauma resulting in chorioretinal rupture.

Full size image

A fluorescein angiogram was carried out with digital real-time image capture using a Heidelberg Engineering scanning laser ophthalmoscope. The inferior temporal retinal arteriole distal to the chorioretinal rupture filled immediately prior to the central retinal artery, that is, from the choroidal circulation (Figure 2). A short, curved section of arteriole was visible ascending from the inferior border of the chorioretinal rupture. No flow was seen through the proximal portion of the inferior temporal arteriole.

Figure 2
figure 2

Fundus fluorescein angiogram showing the inferior branch retinal artery filling from the choroidal circulation (black arrow) and the inferior branch retinal vein emptying into the choroidal circulation (clear arrow).

Full size image

The venous circulation was also abnormal on angiography. A branch of the inferior temporal vein drained up to, but not beyond, the site of the chorioretinal rupture. The branch vein appeared to be of normal calibre with flow comparable to unaffected vessels in other segments of the fundus. As with the branch arteriole, the vessel appeared to curve deep to the retina to anastomose with the choroidal circulation. No connection was present between the proximal end of this vein and the main inferior temporal retinal vein. There was no evidence of choroidal neovascular membrane formation, retinal oedema, or exudate.

Comment

Acquired chorioretinal anastomoses have been described in disciform age-related macular degeneration, sickle cell anaemia, laser photocoagulation, severe hypertension, toxoplasmosis, toxocariasis, and trauma.1,2,3,4,5 The authors are only aware of one described case where this change has occurred following closed globe trauma.6 Anastomoses have been described affecting arterial only, venous only, or both types of vessels. In all these events, the integrity of Bruchs membrane and the retinal pigment epithelium has been lost, allowing the approximation of retinal and choroidal circulations. The pressure gradients within the eye are, from the highest to the lowest, choroidal arterial, retinal arterial, retinal venous to choroidal venous. Knowing the direction of flow from fluorescein angiography, the anastomosis described here must therefore be between choroidal artery and retinal artery, and between retinal vein and choroidal vein. This connection between unrelated large calibre retinal vessels and the choroidal circulation is postulated to be a result of their approximation by granulation tissue during the healing process.

In some cases, complications, predominately neovascular membranes, have arisen as a result of the formation of these chorioretinal anastomoses. Communications between larger calibre vessels appear to have a clinically more benign course. In this case and the small number of other comparable reported incidences, there has been no evidence of clinically significant neovascular proliferation, exudation, or oedema.6,7 It is in an effort to repeat this benign anastomotic process involving the venous drainage of the retina that both argon and YAG lasers and more recently surgery have been utilised for treatment of retinal vein occlusion.8,9