The Current Treatment of Branch Retinal Artery Occlusion and Central Retinal Artery Occlusion





Retinal artery occlusion (RAO), including central retinal artery occlusion (CRAO) and branch retinal artery occlusion (BRAO), is an ischemic stroke syndrome characterized by sudden and painless vision loss. Non-arteritic RAO is accompanied by an increased risk of systemic vascular events and death, therefore, timely secondary stroke prevention is critical . The available evidence suggests thrombolysis may be effective for non-arteritic CRAO at restoring functional vision if treatment is started within hours, whereas other treatments are not evidence-based. Care systems are evolving to decrease pre-hospital delay to presentation and in-hospital delay to diagnosis for CRAO.


Key points








  • Retinal artery occlusion is an ischemic stroke syndrome and a medical emergency.



  • Non-arteritic retinal artery occlusions are accompanied by an increased risk of systemic vascular events and timely secondary stroke prevention is critical.



  • The available evidence suggests thrombolysis may be effective for non-arteritic central retinal artery occlusion (CRAO) when treatment is started within hours of vision loss.



  • Care systems are evolving to decrease pre-hospital delay to presentation and in-hospital delay to diagnosis for CRAO.



  • The next steps in the care of CRAO include developing and validating biomarkers of retinal viability and neuroprotective treatments for CRAO to expand thrombolysis eligibility.




Introduction


Retinal artery occlusion (RAO), including central and branch retinal artery occlusion (CRAO and BRAO, respectively), is a stroke syndrome characterized by retinal ischemia. Recent years have seen an increased interest in the management of RAO and an evolution in its standard of care endorsed by leaders across multiple disciplines [ , ]. The purpose of this article is to summarize the current management of central and branch RAO, with a focus on recent advances in the treatment of non-arteritic CRAO.


Epidemiology


CRAO and BRAO are rare. The incidence of CRAO is 1.90 per 100,000 among the white US population [ ], with comparable estimates from populations in Europe and Asia [ ]. Incidence increases with age and CRAO is more common among males. Clinically diagnosed BRAO is more common than CRAO by a factor of at least 2 to 3 [ , ].


Pathophysiology and risk factors


RAO arises from sudden blockage of the central retinal artery (in the case of CRAO) or one of its branches (in the case of BRAO), which results in hypoperfusion of the downstream retinal arterioles and rapidly progressive death of neurons in the inner retina.


RAO can be arteritic (ie, associated with systemic inflammatory condition) or non-arteritic, the latter being more common. For this article, unless otherwise specified, RAO should be interpreted to refer to the non-arteritic variety.


Arteritic CRAO is most commonly due to giant cell arteritis (GCA), and occurs in 1.6% (95% confidence interval [CI], 0.4%-4.2%) of those with this disease [ ]. Arteritic CRAO in GCA typically is caused by luminal narrowing that occurs as a consequence of inflammatory vessel wall thickening. BRAO arising in patients with proven GCA is rarely described. Arteritic BRAO is more typical of other vasculitides such as Susac syndrome [ ].


The majority of non-arteritic RAOs are thromboembolic in etiology, with an embolus typically of arterial (most commonly the ipsilateral carotid artery) or cardiac origin [ ]. Common predisposing conditions include carotid artery stenosis, cardiac valvular disease, and atrial fibrillation. Much less commonly, RAO may be iatrogenic, such as from inadvertent intra-arterial injection of cosmetic fillers [ ]. CRAO has been described in association with SARS-CoV-2 infection; in such cases, the mechanism is presumably because of the pro-thrombotic state associated with the disease [ ].


Clinical presentation and natural history


CRAO is characterized by acute, profound, and painless vision loss in 1 eye. Visual acuity (VA) is no better than counting fingers in more than 80% of affected eyes [ ], although VA may be relatively preserved in the presence of a cilioretinal artery. Ipsilateral relative afferent pupillary defect and significant visual field defect are the rules.


Classic fundoscopic features in CRAO include cherry-red spot, retinal opacification, and attenuated retinal arterioles. Optic disc edema may be present in non-arteritic CRAO, however, should raise suspicion for an arteritic etiology if significant. Occasionally, an embolus may be visible in the retinal arterioles, but this is not typical [ ].


Eyes that experience CRAO occasionally experience some improvement in vision, especially within the first week; nevertheless, visual recovery from 20/200 or worse to 20/100 or better (ie, functional visual recovery) occurs in less than 1 in 5 patients [ ].


For BRAO, the degree of vision loss and fundoscopic findings tend to be more modest. VA at presentation is 20/40 or better in 74% of eyes with permanent BRAO [ ], and improvement is seen occasionally for eyes with poorer presenting VA [ ].


Risk of additional ischemic events


Although the distinction between central and branch RAO has important functional implications concerning visual outcome (and therefore, the motivation for vision-directed treatment), their risk of systemic ischemic events is comparable [ ].


Multiple population-based studies have demonstrated that patients with RAO are at increased long-term risk of subsequent stroke, myocardial infarction (MI), and death [ ]. Importantly, the greatest risk of subsequent stroke is within the first week after vision loss [ , ]. Up to 30% of eyes with acute RAO have acute cerebral ischemia on MRI [ ]. This underscores the importance of urgent referral for secondary stroke prevention and risk factor control.


Summary


A growing body of evidence, to be explored later in this article, suggests thrombolysis may improve visual outcomes in selected patients with non-arteritic CRAO. All vision care providers must recognize that RAO is an emergency because of the limited window for vision-directed treatment and the association with life-threatening conditions.


Significance


Vision-directed treatment


Secondary prevention for stroke is a critical component of the treatment of a patient with any RAO but doesn’t typically improve visual function. Vision-directed treatments aimed at improving visual outcomes have become more tenable with the advent of new directed therapies. The significant visual morbidity associated with CRAO compared with BRAO means that most of these vision-directed therapies have concentrated on CRAO and not on BRAO. In this section, the authors will discuss the challenges and limitations of these therapeutic approaches.


Retinal tolerance to ischemia


Like other neural tissue, the retina will tolerate ischemia for a finite amount of time before infarction occurs and function (vision!) is irretrievably lost. This is a critical concept with implications including the duration of the therapeutic window and motivation for neuroprotective strategies to extend that window until treatment or spontaneous recanalization occurs.


In a seminal study, Hayreh and colleagues demonstrated that, in aged rhesus monkeys, 240 minutes is the upper limit of tolerance of the retina to ischemia produced by experimental CRAO, beyond that the retina suffers massive irreversible damage [ ]. More recently, others have argued that infarction may occur in only 12 minutes to 15 minutes after complete CRAO, although incomplete CRAO is not uncommon and may benefit from a longer ischemic tolerance time [ ]. Undoubtedly, when faced with acute CRAO, time is retina.


Vision-directed treatment with thrombolysis


Background


Thrombolytic agents catalyze the breakdown of fibrin-based thrombi that commonly are implicated in embolic stroke, including RAO [ ]. Clinically, thrombolysis has well-established benefits for acute cerebral ischemia, and the Cochrane Library’s systematic review concluded thrombolytic therapy with tissue plasminogen activator (tPA) given up to 6 hours after stroke reduces the risk of death or dependency. This benefit was apparent despite an increase in symptomatic intracerebral hemorrhage (ICH), a feared complication of this therapy [ ].


Intravenous thrombolysis for central retinal artery occlusion


Intravenous thrombolysis (IVT) for CRAO has been described for a half-century, however, foundational support for this practice, in particular with modern thrombolytic agents, has more definitively emerged within the past decade [ ]. In a 2015 patient-level meta-analysis, Schrag and colleagues reported that, concerning visual recovery from 20/200 or worse to 20/100 or better, systemic thrombolysis was beneficial at 4.5 hours or earlier after vision loss compared with the natural history group (17 of 34 [50.0%] vs 70 of 396 [17.7%]; odds ratio, 4.7 [95% CI, 2.3–9.6]; P <.001). No benefit was apparent when thrombolysis was provided at after a time point from vision loss [ ]. An updated meta-analysis reported a visual recovery rate of 37.3% among patients treated with alteplase, a biosynthetic form of human tissue-type plasminogen activator, within 4.5 hours because time last known well, compared with the 17.7% recovery rate in those without treatment. A low risk of hemorrhagic complications occurs in the included studies; indeed, no case of symptomatic ICH in patients treated was noted within 4.5 hours [ ].


These data, largely based on observational studies, have served as a motivation for clinical trials comparing IV thrombolysis to placebo for CRAO. At present, there are 3 such clinical trials registered on clinicaltrials.gov underway in Europe.


Intraarterial thrombolysis for central retinal artery occlusion


Intraarterial thrombolysis (IAT) has had an important role in acute cerebral ischemia for more than 2 decades. The endovascular approach permits a high concentration of lytic agent to be delivered directly to the clot and increases lysis rates compared with IVT, while minimizing systemic exposure and hemorrhagic complications [ ].


For CRAO, the procedure involves super-selective microcatheterization of the ostium of the ophthalmic artery. Catheter-associated complications can include arterial dissection, catheter-induced vasospasm, and distal embolization of clot. Furthermore, a greater human resources requirement exists compared with IVT, because the procedure must be carried out in a catheterization laboratory with monitored anesthesia care [ ].


To date, a single randomized controlled trial (RCT) of IAT for CRAO has been reported. The European Assessment Group for Lysis in the Eye carried out a prospective, randomized, and multi-center study that compared the conservative treatment to IAT for patients with CRAO within 20 hours of vision loss. Because of similar efficacy and a higher rate of adverse reactions in the IAT group, the study was stopped after the first interim analysis [ ]. However, as no patient was treated within 4.5 hours, and only 4 of 41 were treated within 6 hours, the role of IAT at earlier time points after vision loss from CRAO is not yet defined. No active randomized trials of IAT for CRAO are currently registered on clinicaltrials.gov .


Alternative modalities of thrombolysis delivery


Intraocular endovascular delivery of thrombolytic agent during pars plana vitrectomy has been described [ , ]. In this technique, retinal arterial cannulation at the optic disc is performed with a microneedle and tPA is administered directly into the retinal circulation. A non-randomized study reported improved VA in eyes treated with this approach compared with conservative measures (ocular massage and intraocular pressure [IOP] lowering drops) although the clinical significance of VA improvement was questionable [ ]. Further study is required before such an approach should be considered. It is noteworthy that the time required for arranging a vitrectomy in the operating room would extend the delay to treatment with thrombolysis.


Non-thrombolytic vision-directed treatment


Non-thrombolytic therapies have been theorized to improve outcomes in CRAO by different mechanisms: vasodilation to increase retinal perfusion (pentoxifylline, carbogen inhalation, and sublingual isosorbide dinitrate); dislodging the embolus via rapid fluctuation in IOP (ocular massage) or laser embolysis; or increasing retinal artery perfusion pressure (IOP-lowering therapy such as topical medications or anterior chamber paracentesis). As Sharma and colleagues pointed out in their comprehensive review, these nominally conservative therapies, individually or in combination, are not supported by evidence showing benefit in improving final visual outcome or altered natural history [ ].


A meta-analysis found these conservative treatments were associated with significantly worsened VA outcomes compared with natural history, with a number needed to harm of 10.0 [95% CI, 6.8–17.4] [ ]. Although some of these therapies continue to be commonly performed, in particular ocular massage and anterior chamber paracentesis [ ], such treatments are not currently endorsed in professional guidelines [ , ].


Secondary prevention


Neurovascular secondary prevention


As detailed previously, patients with BRAO and CRAO experience an increased short- and long-term risk of vascular events including stroke, MI, as well as death [ ]. Hence, an imperative exists for both urgent etiologic workup (ie, to reduce short-term stroke risk) and long-term vascular risk factor optimization. Multidisciplinary collaboration between the vision care provider, neurologist, primary care provider, and others is essential [ ]. In this regard, the vision care provider’s obligations following diagnosis of acute RAO include referral to an emergency department (ED) or (for subacute presentations) a stroke prevention service on an urgent basis.


Ocular secondary prevention


Ocular neovascularization (NV) is detected in 16% of eyes after CRAO. Anterior segment neovascularization is most common and can produce neovascular glaucoma [ ]. The average time from CRAO to NV diagnosis is approximately 3 months, although NV may be detected as early as 1 week after CRAO [ , ]. Risk factors for ocular NV after CRAO include absence of angiographic retinal reperfusion, diabetes, history of stroke, chronic kidney disease, and increasing age [ ]. As BRAO is associated with less extensive retinal ischemia (and, therefore, reduced pro-angiogenic stimulus), NV typically is not observed following BRAO [ ]. The American Academy of Ophthalmology’s professional guidelines for retinal artery occlusions advise that patients with more extensive retinal ischemia be followed closely. Panretinal photocoagulation treatment is recommended for patients who develop NV after RAO [ ].


In addition to monitoring for neovascular sequelae, the vision-care provider (ophthalmologist or optometrist) has critical roles to play after RAO, including optimizing remaining vision and preserving the health of the fellow eye [ ].


Acute arteritic central retinal artery occlusion


Clinical suspicion and prompt empiric treatment with glucocorticoids are critical to reduce the risk of fellow-eye vision loss when CRAO is because of GCA [ ]. Our standard practice for vision-involving suspected GCA, including patients who present with possible arteritic CRAO, is to offer treatment with 3 days of intravenous methylprednisolone 1000 mg daily, followed by high dose oral prednisone while a temporal artery biopsy is arranged. Recently, it has been proposed that the interleukin-6 receptor blocker tocilizumab may have a role because acute rescue therapy in patients with glucocorticoid-resistant GCA whereby further vision loss would otherwise be imminent, but further research is needed [ ].


Present relevance & future avenues


The results of ongoing phase 3 clinical trials for IVT in CRAO are eagerly awaited and may support the conclusions of the observational data that have hitherto informed this practice. Nevertheless, other barriers (ie, beyond lack of level 1 evidence) remain to be addressed to optimize care for patients with this condition and to maximize eligibility for thrombolysis in the real-world setting.


Delay to diagnosis


Pre-hospital delay


Delay in presentation after vision loss from RAO is common. In a US study, among 484 patients with recent CRAO, Shah and colleagues reported 247 (51%) presented within 4.5 hours of vision loss, whereas 86 (17.8%) sought care after more than 24 hours [ ]. Other studies have described even less favorable times-to-presentation. In another US-based study of 91 patients with acute non-arteritic CRAO, only 20.9% presented within 4 hours of symptom onset [ ]. In a German study of 101 patients admitted for CRAO, almost 60% presented later than 4.5 hours [ ].


The first point of contact with the health care system may not be able to provide thrombolysis, and transfer to stroke center will introduce additional delay. In a report of 181 patients with CRAO referred to a tertiary care stroke center, only 62 (34%) were seen at that institution within 24 hours [ ].


As compared to the symptoms of acute cerebral ischemia, where significant efforts have been made to raise public awareness about the possibility of stroke, the urgency of acute vision loss may not be recognized by the layperson. In Ardila Jurado and colleagues’s Swiss study of 32,816 patients, the symptom-to-door time was greater for patients with CRAO than for those with cerebral stroke: 852 mins versus 300 mins on average, respectively [ ]. Revision of the widely-promoted acronym for stroke symptomatology–FAST (Face, Arms, Speech, Time) to include other stroke symptoms–Balance, Eyes to BE-FAST has been proposed as a strategy to increase public awareness [ ]. Further public education campaigns may be required.


In-hospital delay to diagnosis


Providers in acute care settings who encounter a patient with sudden vision loss may not readily consider RAO as a differential diagnosis and are not necessarily equipped to confirm the diagnosis clinically [ ]. The use of nonmydriatic fundus photography in the ED provides numerous benefits over direct ophthalmoscopy including improved detection of relevant abnormalities by ED physicians and the ability to remove review [ , ]. The latter point is particularly salient because timely in-person ophthalmologic consultation is not always available in acute care settings. Increasingly, calls are being made for ED-based fundus photography to have a central role in minimizing door-to-diagnosis time for CRAO [ , ]. Integrating optical coherence tomography (OCT) into ED workflows may further assist with making a prompt diagnosis [ ]. Collaboration between ophthalmology, neurology, and emergency medicine departments will be required for the successful implementation of this technology at the institutional level.


Biomarkers of retinal viability


Given risks associated with thrombolysis, the decision to offer treatment requires likelihood of salvageable retina. Akin to perfusion imaging after cerebral stroke to differentiate reversible (penumbra) from irreversible (infarct core) ischemic brain volume [ ], an unmet need exists for biomarkers of retinal viability to guide treatment decisions when thrombolysis is being considered. This may be particularly true in patients who present at later time points after vision loss, or when time of onset is unknown.


Fundus photography


The characteristic cherry red spot of CRAO appears after several hours of retinal ischemia; its absence on color fundus photos in the presence of other early fundus changes (segmental blood flow and retinal arteriolar attenuation) may indirectly suggest retinal viability [ ].


Optical coherence tomography


OCT provides structural information about the retina that can complement clinical assessment and fundus photography. OCT-derived metrics in recent CRAO correlate with visual outcomes [ , ]. Retinal edema detected by OCT develops in a time-dependent fashion within hours of CRAO [ ]. In a retrospective analysis of 66 patients with recent (<48 hours) CRAO, OCT retinal thickness measures identified eyes with ischemia onset less than 4.5 hours with sensitivity of 100% and specificity of 94.3% [ ]. The ability of these markers to inform treatment decision-making remains to be studied in a prospective setting.


OCT angiography (OCTA) permits non-invasive 3-dimensional visualization of retinal microvasculature and for this reason is worthy of mention in a discussion of multimodal assessment of RAO [ ]. With further technological refinements, it is plausible that OCTA will play a role in assessing retinal viability after RAO in the acute setting, but at this time such a role is not clearly defined.


Future directions


Additional techniques currently limited to research settings may have a role in retinal viability assessment in the future. These tools, which include Retinal Function Imager and Photoacoustic Imaging, have been summarized in a recent review by Mac Grory and colleagues [ ].


Neuroprotection for retinal artery occlusion


In the context of acute ischemic stroke, neuroprotection refers to strategies that maintain the viability of neural tissue that has been compromised by ischemia and, if left untreated, will go on to die. The goals of neuroprotection include limiting tissue injury and extending the treatment window for therapies that improve or restore perfusion such as thrombolysis. Translation of neuroprotective treatments for acute cerebral ischemia from bench to bedside has not yet been successful; however, this is an evolving field with numerous possible neuroprotectants for stroke emerging in recent years [ ].


Retinal neuroprotection combined with thrombolysis for CRAO is a conceptually attractive strategy to improve outcomes in this condition. Human data are lacking, but a very limited number of pharmacologic therapies have been studied in animal models of neuroprotection for CRAO and are summarized in a recent article [ ].


Hyperbaric oxygen


Hyperbaric oxygen (HBO) has been described as a primary treatment for CRAO, although it is not accepted as an evidence-based practice [ ]. Treatment protocols involve up to 24 hours of exposure to 100% oxygen at 2.0 to 2.8 atm [ ]. HBO may allow impaired inner retinal oxygenation to be supplemented by diffusion of oxygen from the choroidal circulation in the setting of supraphysiologic oxygen tension. HBO is theorized to maintain tissue viability until normal perfusion is restored, either spontaneously or because of intervention (such as thrombolysis); as such, it is a potential neuroprotective strategy. A 2018 meta-analysis of 7 RCTs of oxygen therapy for RAO (including 6 that used HBO) found a significant benefit for VA improvement for patients receiving oxygen therapy versus the control group who did not (OR, 5.61 [95% CI, 3.60–8.73]; P <.01); hyperbaric was superior to normobaric treatment. Most included studies had a low risk of bias, however, the clinical significance of VA improvement was not discussed; furthermore, many of the included studies used other interventions such as ocular massage or IOP-lowering therapies and none used thrombolysis [ ]. Centers with access to hyperbaric oxygen therapy can consider how to integrate this treatment into institutional protocols for acute treatment of RAO.


Hypothermia


Therapeutic hypothermia has proven neuroprotective effects in conditions that produce global cerebral ischemia such as cardiac arrest. In this context, systemic cooling decreases cerebral metabolic demand and reduces oxygen requirements, hence, neuronal death is reduced or delayed and superior neurologic outcomes are observed as a result [ ].


Local cooling of the eye or retina may provide a neuroprotective effect for RAO. In rats with experimental CRAO, cooling of the eye with an external icepack reduced retinal cell death in 1 study [ ]. Cooling of enucleated bovine eyes to 21°C prolonged retinal ganglion cell (RGC) ischemic tolerance compared to 37°C [ ]. Even mild degrees of local hypothermia may be beneficial in RAO: in an organ culture model, cooling to 30°C increased RGC survival after ischemia [ ]. Human data are lacking but retinal cooling may be feasible with as simple a technique as an application of an icepack to the closed eyelids as soon as suspicion for CRAO is raised.


Summary


RAO should be recognized as a medical emergency given the risk of systemic events including stroke and death, association with other high-risk conditions, and limited window for treatment for CRAO. To that end, the available evidence indicates thrombolysis is efficacious in non-arteritic CRAO, and prospective RCT data may soon be available to further guide its use. The next steps include implementing and refining protocols and pathways to ensure patients receive optimal care for RAO that will include strategies to minimize delay to diagnosis. In many centers, CRAO protocols have been designed and might serve as templates for institutions looking to do the same [ ].


In RAO, the principal role of the vision care provider is to confirm the diagnosis. In the coming years, this will likely involve participation in telemedicine as fundus photography becomes implemented into acute care workflows. Second, when the vision care provider is the patient’s first point of contact with the health care system after experiencing RAO, timely referral for stroke care falls to that provider. Although many articles (including the present one) address acute treatment of CRAO, the importance of secondary stroke prevention must be emphasized for CRAO and BRAO alike. Long-term follow-up after RAO to screen for neovascular sequelae and to optimize ocular health also falls in the domain of the patient’s ophthalmologist and optometrist. Vision science researchers are called upon to participate in research that will continue to advance the field. This includes clinical trials, as well as bench and translational science to develop and validate biomarkers of retinal viability and efficacious neuroprotection treatments, which are needed to further expand treatment eligibility for CRAO.


Clinics care points








  • The vision care provider’s obligations following diagnosis of retinal artery occlusion include referral to the emergency department or (for subacute presentations) a stroke prevention service on an urgent basis



  • Patients with acute non-arteritic CRAO should be considered for vision-directed treatment with thrombolysis



  • So-called conservative vision-directed treatments are not supported by evidence, including ocular massage and anterior chamber paracentesis



  • Patients who have experienced retinal artery occlusion require ongoing follow up with their vision care provider



  • Arteritic CRAO due to giant cell arteritis requires treatment with glucocorticoid to reduce the risk of fellow eye vision loss


Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Mar 29, 2025 | Posted by in OPHTHALMOLOGY | Comments Off on The Current Treatment of Branch Retinal Artery Occlusion and Central Retinal Artery Occlusion

Full access? Get Clinical Tree

Get Clinical Tree app for offline access