Dramatic recovery of vision in patient with acute central retinal artery occlusion treated with local intra-arterial tissue plasminogen activator

Abstract

Central retinal artery occlusion (CRAO) is a devastating disease resulting in severe visual loss in most cases. It has no proven treatment, and a variety of management options used clinically all have dismal rate of success. We report outcome of treatment with local intra-arterial thrombolysis administered 2.75 hours after devastating visual loss from incomplete CRAO.

Background

Central retinal artery occlusion (CRAO) is a devastating ocular condition that results in irrevocable visual loss in most cases. While multiple treatment modalities have been tried in treating patients with CRAO, none have proven to be effective. Intra-arterial injection of tissue plasminogen activator into an ophthalmic artery has been previously studied, but the results of a large clinical trial have concluded that this treatment is not better than observation and can cause critical side effects. Most important factor in inability of any curent therapy to reverse the devastating visual loss associated with CRAO has been late presentation of patients with this disease as by the time they typically reach an ophthalmologist many hours have passed since the onset of the visual loss causing irreversible retinal ischaemia and death. Most of the data that have been published on treating this condition have been acquired from patients who were seen many hours or even days after the onset of the visual loss and is thus difficult to extrapolate to the patients with a very acute CRAO.

Case presentation

A 72-year-old man underwent uncomplicated cataract surgery with topical anaesthesia. His medical history consisted of hypertention and hyperlipidaemia. Postoperative check 2.5 hours after surgery demonstrated 20/60 acuity, intraocular pressure (IOP) of 18 mm Hg, trace corneal oedema and centrally positioned intra-ocular lens. He returned 20 min after being discharged stating that vision in the operated eye went suddenly black. One hour after visual loss, acuity was bare light perception in the operated eye with brisk relative afferent pupillary defect (RAPD), but retina and optic nerve both appeared normal. Macular ocular coherence tomography was normal.

Investigations

An intravenous fluorescein angiography (IVFA) was performed but because of multiple eye drops administered earlier and mild corneal oedema, quality of the images was poor (figure 1). Some flow was seen in retinal vessels of the affected eye but vessel calibre in the affected eye was smaller than in the fellow eye.

Figure 1.

Top panel: red free photos demonstrating retinal arteriole attenuation in the left eye. Midpanel: colour photographs also revealing arteriolar attenuation in the left eye but absence of cherry-red spot in the macula. Lower panel: IVFA photographs 2 min after dye injection; there is decreased perfusion of retina in left eye. IVFA, intravenous fluorescein angiography.

Differential diagnosis

When initially assessing the patient with bare light perception vision in the presence of brisk RAPD but otherwise a normal ophthalmic examination, it was difficult to come up with the possible explanation for the visual loss. Given very poor visual acuity and a brisk RAPD, this presentation had to be either secondary to a diffuse retinal dysfunction or to an optic neuropathy. Absence of optic nerve head oedema left posterior ischaemic optic neuropathy (which in the absence of profound hypoperfusion is almost always secondary to giant cell arteritis (GCA)) as the only possibility. As the patient had no symptoms of GCA and the visual loss was very acute, this diagnosis was deemed very unlikely.

When considering diffuse retinal dysfunction that could have produced acute visual loss, CRAO (or potentially cilioretinal artery occlusion) was the only potential possibility. Typically, it is easy to make a diagnosis of acute CRAO because of its characteristic retinal appearancere with the presence of retinal whitening/oedema, box-carring in retinal arterioles and the cherry-red spot in the macula. Normal retinal examination made the diagnosis of CRAO difficult, but the literature search confirmed that retinal whitening and oedema as well as cherry-red spot can take hours to develop in a non-complete CRAO. Thus, by the process of elimination, CRAO was the only entity that could have produced this clinical presentation. Finding of attenuated retinal arterioles on IVFA confirmed our suspicion.

Treatment

A decision was made to proceed with administration of local intra-arterial thrombolysis (LIT) into ophthalmic artery. Interventional radiologist was available, and ophthalmic artery was cannulated 2.75 hours after onset of visual loss (figure 2).

Figure 2.

Left internal carotid angiogram demonstrating cannulated ophthalmic artery (blue arrow).

Outcome and follow-up

One minute after administration of LIT, patient stated that his vision is clearing up. Vision was at least counting fingers, and brisk RAPD seen previously had completely resolved as observed by two separate ophthalmologists who checked pupillary reaction separately. The following morning vision was 20/40 in the affected eye with the normal appearing retina. On 1-week follow-up, vision was 20/20 in each eye with no RAPD and normal appearing retinas. CT angiography of the chest, neck and brain demonstrated only mild atherosclerotic disease in the internal carotid arteries of the neck. Transthoracid echocardiogram was normal. Forty-eight hour Holter monitoring did not demonstrate abnormal heart rhytm.

Discussion

Aetiology of non-arteritic CRAO is almost always an embolus originating in the carotid arteries of the neck that travels up to the ophthalmic artery.¹ The other possibility is the embolus originating in the heart or intracranial branches of the carotids. The least likely source is the atherosclerotic disease affecting the ophthalmic artery itself. The timing of retinal tissue viability in the face of ischaemia has been debated and has been shown to be about 100 min for monkeys whose central retinal artery has been clamped with occlusions lasting 100–240 min producing variable amount of retinal ischaemia and those lasting longer than 240 min producing irreversible severe retinal damage and optic nerve ischaemia.² However, more recent data indicate that retinal survival time after complete CRAO might be much shorter, in the range of 12–15 min.³

Treatment for CRAO currently consists of variety of manoeuvres designed to lower IOP with the goal of dislodging the embolus and allowing it to travel downstream thus minimising the area of retinal ischaemia.⁴ Unfortunately, all of the proposed treatments have dismal success rate. LIT has been proposed as a revolutionary treatment for CRAO, and after several small prospective trials lauded its efficacy, it was thought to become a main treatment for patients with acute non-arteritic CRAOs.^{5 6} However, European Assessment Group for Lysis in the Eye trial that compared observation with LIT concluded there was no difference between observation and treatment groups and was stopped early because of critical side effects observed in the treatment group.⁷ There were several reasons offered on why this trial did not demonstrate success of LIT: most treated patients had a long time lapse from onset of visual loss to treatment, with the shortest time being 4.75 hours (mean of 9.5). Another proposed reason for poor performance of LIT was presumed high percentage of calcific plaques causing CRAO, which are poorly susceptible to lysis with tissue plasminogen activator (tPA).⁸ Another well-conducted randomised controlled clinical trial investigated efficacy of intravenous tPA for CRAO and enrolled16 patients who received placebo or intravenous tPA within 24 hours of symptoms onset. Two out of eight patients in the treatment group demonstrated improvement in visual acuity by three lines; however, this was no sustained at 6 months due to presumed reocclusion of central retinal artery. One patient in the treatment group suffered intracerebral haemorrhage. Both of the patients who had initial improvement of vision following treatment received intravenous tPA within 6 hours of symptoms onset and study authors hypothesised that the effective therapeutic window for administering treatment for CRAO is less than 6 hours.⁸ A meta-analysis of seven studies that included 396 patients concluded that intravenous fibrinolysis was effective if administered within 4.5 hours of symptoms onset confirming the conclusion that it is the time from symptoms onset to administration of treatment that is the most important predictor of successful outcome.⁹

In our case when the LIT was administered 2.75 hours after the onset of symptoms, there was complete restoration of retinal perfusion and function. We presume that the treatment was very successful in restoring visual acuity primary because it was administrated very quickly after the onset of symptoms and because the CRAO was not ‘complete’ (complete CRAO causes no retinal perfusion with box-carring and retinal whitening). The embolus in our case was also most likely thrombotic and not calcific in nature and thus susceptible to lysis by LIT.

Several important lessons can be learnt from this clinical observation: retina can appear normal for at least 2.5 hours after incomplete CRAO thus the diagnosis should be presumed in patients with sudden onset of devastating visual loss with RAPD and normal appearing optic nerve and retina. LIT can be very successful if administered within 3 hours of the visual loss and if the culprit is thrombotic and not calcific embolus. LIT when administered by an experienced intervential neuroradiologist is a quick and safe procedure.

Having a dedicated centre where an interventional neuroradiologist is available 24/7 for administration of LIT for patients with acute CRAOs should be our goal in setting the standard of care for treating patients with acute CARO as in select cases this treatment can be sight-saving.

Learning points.

• Patients with acute central retinal artery occlusion (CRAO) can present with entirely normal retinal examination.

• Intra-acterial tPA can be a very effective treatment for acute CRAO if administered early.

• Intra-arterial tPA is most effective for thrombotic rather than calcific emboli.

• Treatment with intra-arterial tPA for patients with acute CRAO can be vision-saving, thus identifying these patients is a true ophthalmic emergency.

• 24/7 access to a neuroradiologist who can cannulate an ophthalmic artery should be the goal in large ophthalmic care centres.