Retina-Brain Homology: The Correlation Between Ophthalmic or Retinal Artery Occlusion and Ischemic Stroke

Introduction

Ophthalmic artery occlusion (OAO) or retinal artery occlusion (RAO) is a retinal infarction results from the interruption of blood flow in the ophthalmic or retinal artery due to thromboembolism or vasospasm. This often leads to painless vision loss, visual field damage, and fundus ischemic changes like retinal pale edema.¹ The severity of symptoms correlates with the size of the obstructed blood vessels and the duration of occlusion. According to the vascular involvement, RAO can be classified into central retinal artery occlusion (CRAO) and branch retinal artery occlusion (BRAO). The most common site of occlusion for CRAO is the lamina cribrosa.² CRAO typically results in severe visual impairment, often limited to hand motion or even perception of light, and exhibit delayed direct light reflex in the ipsilateral pupil; while BRAO is associated with relatively favorable visual acuity and prognosis, with localized visual field defects corresponding to the areas of obstructed blood supply.¹ The yearly incidence rate of CRAO is 1–2 cases per 100,000 individuals, with a higher prevalence among men compared to women.³ Moreover, the incidence rate of CRAO rises with age, and individuals over 80 years old can experience rates as high as 10 cases per 100,000.⁴ OAO is relatively rare, whereas branch retinal artery occlusion (BRAO) accounts for approximately 1/3 of RAO.³ The incidence rate of transient monocular vision loss (TMVL) is not well-established due to patients often not seeking medical attention for its transient symptoms. Fundus photography is commonly relied upon for the diagnosis of OAO and RAO, as the characteristic alterations in the patient’s fundus facilitate a relatively straightforward diagnosis; while fundus fluorescence imaging is considered the gold standard for diagnosing these conditions.⁵

Stroke, like OAO or RAO, is an acute cerebrovascular disease caused by the occlusion or rupture of cerebral blood vessels, leading to tissue damage in the brain.⁶ Acute ischemic stroke (AIS) is the predominant type of stroke, representing 69.6–70.8% of all strokes in China.⁶ Data from the Global Burden of Diseases showed that stroke is the primary cause of death and disability among adults in China, with an incidence rate of 0.201% for stroke and 0.145% for AIS in 2019, making it the highest prevalence in the world.⁷ According to TOAST classification, the etiology of AIS mainly includes aortic atherosclerosis, cardiogenic embolism, arteriole occlusion and other factors, which share the same pathogenesis with OAO or RAO.⁸ Due to collateral circulation, when AIS occurs, although some neurons experience a loss of electrophysiological function, their cellular structure remains intact, and these neurons could regain normal electrophysiological function if tissue perfusion is restored.⁹ However, structural damage eventually occurs as ischemia progresses, and normal physiological function cannot be recovered even with restored tissue perfusion. These neurons are referred to as the “penumbra”.⁹ Similarly, a penumbra also exists in CRAO.¹⁰ When retinal blood flow suddenly ceases, a retinal “penumbra” consisting of hypoxia and hypoxia-tolerant cells emerges due to variations in oxygen supply from the choroidal circulation and different of hypoxia tolerance among retinal cells in different regions.^9,10 The retinal “penumbra” appears similar to normal retinal tissue in fundus images but exhibits signal attenuation in electrophysiological testing.⁹ Unlike the cerebral penumbra that ultimately transforms into an infarct core, the retinal penumbra can persist and eventually induce the formation of intraocular neovascularization.^9,11 In fact, OAO or RAO could be viewed as special types of AIS in terms of pathogenesis. There is growing support for classifying OAO or RAO as AIS and implementing comparable treatment and prevention strategies due to the rapid and severe damage caused, which would benefit secondary and tertiary prevention strategies for the disease.⁵ In 2013, a statement from the American Stroke Association and the American Heart Association defined central nervous system infarction as the permanent ischemic death of brain, spinal cord, and retinal cells, based on imaging, pathology, and clinical symptoms, thus categorizing OAO or RAO as types of stroke.¹² Relevant guidelines recommend prompt intervention and multidisciplinary consultation for assessing neurological function in patients with OAO or RAO.⁵

In this review, we firstly summarize the anatomical and functional connections between the retina and the brain. Next, we compile a list of common cardiovascular and cerebrovascular risk factors found in previous similar studies comparing OAO or RAO and AIS. Additionally, we summarize the future risk of AIS and its associated risk factors in patients with OAO or RAO. Lastly, we compare and review the recent advancements in the effectiveness and safety of intravenous and intra-arterial thrombolysis therapy in patients with OAO or RAO. Our goal is to provide inspiration and ideas for the development of a standardized multidisciplinary approach and the establishment of expedited pathways or thrombolysis guidelines for vascular interventions in the future.

Retina and Brain

The retina and optic nerve develop directly from the mesencephalon during the embryonic stage.¹³ Retinal ganglion cells are morphologically similar to cerebral neurons, composed of soma, dendrites, and axons.¹⁴ And the optic nerve, optic chiasm, and optic tract are formed by the successive axons of ganglion cells, terminating in the lateral geniculate body.¹⁴ Ganglion cells convert visual information from photoreceptors into action potentials and transmit it along their axons, which eventually reach the visual cortex in the occipital lobe by passing through the lateral geniculate body.¹⁴ Due to the neural homology and correlation between the retina and the brain, retinal degeneration has been suggested to be accompanied by cerebral neurodegeneration.^15,16

The blood perfusion of the retina and brain is closely interconnected. The ophthalmic artery is the first branch of the internal carotid artery, while the anterior and middle cerebral arteries are its terminal branches.¹⁷ Consequently, stenosis of the internal carotid artery would lead to ocular and cerebral ischemia. Under this condition, the Willis loop can compensate for the blood supply to the ophthalmic artery,¹⁸ and the ophthalmic artery can enhance cerebral vascular perfusion as collateral circulation when cerebral blood supply is insufficient.¹⁹

The retinal vessels have similar structural and physiological properties to the cerebral small vessels. Retinal vessels have a diameter of approximately 100um to 300um, which is similar to that of the cerebral small vessels.²⁰ They both exhibit low flow and high oxygen uptake, and can regulate vascular resistance and blood flow by contracting vascular smooth muscle or altering local metabolic demand to maintain tissue perfusion under hypoxia-ischemia.²¹ Similar to the blood-brain barrier, the inner blood-retinal barrier consists of pericytes and tight junctions between vascular endothelial cells, giving it the function to block macromolecules from the bloodstream from entering the tissues.²² Aqueous humor, containing abundant anti-inflammatory and immunomodulatory mediators, shares similar properties and functions with cerebrospinal fluid, thus making the eye an immune-privilege site similar to the brain.^23,24 The permeability of the blood-retinal barrier and blood-brain barrier increases under hypoxia-ischemia, ultimately leading to impaired barrier function and tissue edema.²²

Due to the shared embryonic origins and numerous similarities in anatomical structures and physiological functions, research on fundus diseases may offer insights for the onset, diagnosis, and prognosis of cerebral diseases. Additionally, research advancements in diagnosing and treating cerebral diseases are relevant for the development of diagnosis and treatment strategies for fundus diseases. Furthermore, fundus imaging enables the non-invasive visualization and quantification of the structure and function of retinal vessels and neural tissues, serving as a unique “window” for researching cerebral vascular and nervous tissues (Figure 1).²⁵

Figure 1 Retina as a “window” to brain.

The Correlation Between Ophthalmic or Retinal Artery Occlusion and Acute Ischemic Stroke

OAO or RAO shares a similar development pattern with AIS, often linked to underlying cardiovascular and cerebrovascular conditions.² Most patients with OAO or RAO present with multiple major vascular risk factors, including hypertension, diabetes mellitus, and hyperlipidemia (Table 1).^2,26–46 The blood supply to retinal and cerebral tissues primarily originates from the internal carotid artery, and multiple studies have demonstrated the presence of severe atherosclerotic stenosis (≥70% stenosis) in the internal carotid artery among patients with OAO or RAO (Table 1).^{26,32,33,35,38,43,45–48} Cardiogenic embolism is another significant cause of OAO or RAO.⁴⁶ Thus, the combination of cardiac disorders in these patients should not be neglected. Several researches have indicated that these patients are frequently detected with coronary artery disease, heart valvular disease, atrial fibrillation and other cardiac disorders (Table 1).^{29,33,36,39,40,42,43,45–50} Interestingly, several studies have revealed that some patients with OAO or RAO have a history of stroke (Table 1).^{30–33,35,38,40–43,45,47,49,51} Nonetheless, the onset of OAO or RAO still elevates the risk of AIS.^26,42

Table 1 Prevalence (%) of Cardio Cerebral Vascular Disease in Patients with OAO or RAO

Most studies have shown that patients with OAO or RAO were at an increased risk of AIS. The AIS risk was significantly higher within a few days after the onset of visual loss, and the AIS morbidity within the next three years ranged from 2% to 20% (Table 2).² Differences in the occurrence probability may be related to variations in follow-up times, patient populations, and study designs. Several clinical studies have revealed that cranial magnetic resonance imaging (MRI) of patients with RAO within 1 week of disease onset showed a high incidence (13% to 24%) of asymptomatic and multifocal of fresh cerebral infarcts caused by microarterial embolism. These infarcts were also predictive of a high incidence of AIS in the future.^53–55 One possible explanation is that when OAO or RAO occurs, it suggests the patients’ blood vessels, especially the microscopic arteries, undergo ischemic pathological changes. And due to the common vascular source between the retinal and cerebral microvasculature and their similar structural and functional characteristics,²¹ these patients have a high probability of experiencing cerebral ischemic events in microvasculature perfusion areas. They also have a higher future risk of cardiovascular and cerebrovascular events. A real-world study in America showed that patients with acute CRAO had an incidence of AIS of approximately 12.15% during hospitalization (Table 2).³³ And the risk of AIS would increase by 28-fold in the first week and 33-fold in the second week after the onset of CRAO.⁵¹ A related meta-analysis showed that the proportion of CRAO patients exhibiting AIS on MRI within 7 days was 30%, while for BRAO patients, it was 25%.⁵⁶ Additionally, a statistical analysis of a readmission database in the United States showed that the highest proportion of RAO patients who were readmitted due to AIS occurred within 150 days of RAO.³⁶ However, cerebral tissue has more collateral circulation and functional compensation than the retina, and the symptoms of cerebral neurological damage caused by arteriole spasm or occlusion are mild in early stage and progress chronically. Therefore, it is hard to draw ophthalmologists’ attention to seek help from neurologists to screen for cerebral disorders in patients with OAO or RAO.^20,26 Nearly 90% of patients who exhibited combined AIS imaging signs within 1 week of the onset of RAO had no neurologic symptoms.⁵⁴ However, the EXPRESE study and the SOS-TIA study demonstrated that early aggressive intervention in patients with TIA or minor stroke reduced the risk of recurrent AIS by 80%.^57,58 These studies emphasize the urgency and importance of early neurological screening, stroke risk assessment, and tertiary prevention strategies for patients with OAO or RAO.

Table 2 Morbidity (%) of Future AIS in Patients with OAO or RAO

The occurrence of OAO or RAO is associated with an increased incidence of distant AIS. A large national study in South Korea revealed that the morbidity rate of AIS was approximately 14.96% within 10 years after the onset of CRAO (Table 2). This represents an approximate 70-fold increase compared to the population without RAO.²⁹ Studies in China and Poland showed that the incidence of AIS in RAO patients was 19.61% within 3 years and 10.07% within 12 years, respectively (Table 2).^27,40 Therefore, it is urgent and necessary to assess the risk of stroke or other cerebral disorders and develop a treatment approach or prevention strategy for patients with OAO or RAO in clinical practice, yet it is neglected by a large proportion of ophthalmologists.⁵

Most researches have confirmed the increased risk of AIS in patients with OAO or RAO. However, there is limited exploration of risk factors or predictive models related to the future AIS risk in these patients. Nonetheless, valuable insights could still be gleaned from the existing researches. Two clinical studies in South Korea confirmed that age and atherosclerosis were the primary risk factors for future AIS in RAO patients, respectively (Table 3).^29,30 Similar conclusions were drawn by several studies in China and America, which identified additional risk factors such as male and history of stroke or TIA (Table 3).^26,33,34,42 It was noteworthy that three studies published relevant prediction models (Table 3).^29,33,42 However, the validation of these models was hindered by differences in statistical methods and the absence of reported baseline survival probabilities. Efforts are underway to establish the first clinical database of OAO or RAO patients in Shenzhen, China, with the aim of developing a prediction model for the risk of AIS in these patients [ChiCTR2300072813]. However, the limited sample size poses a challenge to the study. The establishment and validation of future prediction models should be pursued through large-sample, multicenter OAO or RAO cohort studies.

Table 3 Influence Factors for Future AIS Risk in Patients with OAO or RAO

Comparison and Advance of Thrombolysis for Ophthalmic or Retinal Artery Occlusion with Acute Ischemic Stroke

Given that OAO or RAO shares the same pathogenesis with AIS and considering the success and widespread use of thrombolysis in AIS, an increasing number of healthcare organizations are progressively exploring the development of thrombolysis for OAO or RAO. In AIS, the purpose of thrombolysis is to rescue the neurons in the penumbra. Therefore, imaging assessments of the perfusion and function of penumbra can help evaluate the necessity for thrombolytic therapy.⁹ While in CRAO, thrombolysis aims to salvage the ischemic and anoxic cells in retinal posterior pole, preventing irreversible failure of ion pumps.⁹ As for the penumbra in the retinal peripheral part, its survival status is not influenced by the intervention timing due to continuous mild oxygen supply from the choroidal circulation.⁹ Currently, thrombolysis for OAO and RAO is primarily divided into intravenous thrombolysis and intra-arterial thrombolysis. The main adverse effects include hemorrhage of intracranial, intraocular, or other organ, as well as unpredictable stroke-like neurological signs or symptoms. A network meta-analysis using the SUCRA method concluded that intravenous thrombolysis was more effective and safer than intra-arterial thrombolysis.⁶² However, it is important to consider the impact of onset-intervention time, thrombolytic duration and procedural complexity on the outcomes. The time window for thrombolysis remains uncertain. Animal experiments have demonstrated that when the retina is completely ischemic and anoxic for more than 97 minutes, retinal ganglion cells and photoreceptors start to experience irreversible damage, which gradually expands; and by 4 hours, almost the entire retina is affected.⁶³ Another study found that when retinal neurons ischemia occurs between 97 to 120 minutes, even if reperfusion is achieved, the electrophysiological activity cannot be restored.⁶⁴ Therefore, if neurons function rather than structure is used as a criterion, the time window may be shorter. However, the onset-intervention time for patients with OAO or RAO in clinical practice often exceeds 2 hours, 4 hours or even longer, and there is a lack of data from large-scale, well-designed randomized controlled trials (RCTs). Additionally, the proportion of visual function recovery after thrombolysis in patients with CRAO occurring >4 hours is comparable to the spontaneous visual recovery in the natural state,⁹ and most emboli of OAO or RAO are cholesterol or calcitonin,⁶⁵ which are not effectively treated with thrombolytic drugs. Consequently, the use of thrombolysis for vascular OAO or RAO remains a topic of controversy. In this part we will discuss intravenous thrombolysis and intra-arterial thrombolysis for OAO or RAO, respectively, and compare them with thrombolysis for AIS.

Intravenous Thrombolysis

Similar to AIS, the primary intravenous thrombolytic drug used for OAO or RAO is recombinant human tissue plasminogen activator (rtPA), with alteplase being the most frequently employed medication. The recommended dosage is 0.9 mg/kg, administered as a 10% intravenous bolus within the initial minute, followed by the remaining 90% infused over 1 hour.⁶⁶ The time window for intravenous thrombolysis in AIS is within 4.5 hours from symptom onset,⁶⁷ and the EXTEND RCT results suggest that this time window could be extended to 9 hours.⁶⁸ For OAO or RAO, the management of CRAO from American Heart Association in 2021 recommends initiating intravenous thrombolysis within 4.5 hours of disease onset.⁶⁶ Two prospective studies in Germany and America have also demonstrated that intravenous thrombolysis within 4.5 hours can lead to positive visual outcomes in patients with CRAO (Table 4).^69,70 A meta-analysis indicated that intravenous thrombolysis within 4.5 hours led to a 74.3% improvement in visual acuity of ≥0.3 logMAR, and a 39% effective rate of visual acuity ≥20/100.⁷¹ Another meta-analysis demonstrated that intravenous thrombolysis performed within 4.5 hours resulted in superior prognostic visual outcomes compared to administration after 4.5 hours.⁶² However, retinal ganglion cells are more vulnerable to ischemia and hypoxia compared to cerebral neurons,²⁶ which suggests that the time window for intravenous thrombolysis should be shorter than 4.5 hours. Currently, double-blind, double-dummy, multicenter RCTs comparing the efficacy and safety in intravenous thrombolysis for CRAO patients between alteplase and placebo are underway in France [NCT03197194, THEIA] and Germany [NCT04965038, REVISION] and these studies are anticipated to conclude in 2024 and 2026, respectively, which will contribute to the understanding of the efficacy, safety, and time window of intravenous thrombolysis in OAO or RAO.

Table 4 Efficacy and Safety of Intravenous Thrombolysis in Patients with OAO or RAO

New thrombolytic drugs teneplase, have garnered significant attention due to its rapid onset of action, prolonged half-life, ease of administration, lower injection concentration (0.25 mg/kg), shorter treatment duration, and improved efficacy and safety profile.⁷⁹ Currently, a Phase III double-blind, double-dummy, multicenter RCT on teneplase for intravenous thrombolysis in CRAO is ongoing in Norway [NCT04526951, TenCRAOS] and is anticipated to conclude in 2024. In the near future, teneplase may potentially replace alteplase as the first-line agent for intravenous thrombolysis in AIS, OAO, or RAO.

Intra-Arterial Thrombolysis

Because of the absence of evidence-based medicine, intra-arterial thrombolysis is not currently recommended as the primary treatment for AIS in the guidelines, leading to its limited use in clinical practice. There have been no large-scale multicenter RCTs on the efficacy and safety of intra-arterial thrombolysis in AIS. However, insights can still be gleaned from existing studies. According to the PROACT II study, arterial thrombolysis (9mg urokinase + heparin) significantly enhances the mRS score at 90 days for patients with AIS who undergo thrombolysis beyond the venous system.⁸⁰ Additionally, the CHOICE study showed that supplementary intra-arterial thrombolysis (0.225 mg/kg alteplase) resulted in improved mRS scores at 90 days for AIS patients who remained unvascularized after mechanical thrombolysis, without any observed adverse effects.⁸¹

In contrast to the unfavorable status of intra-arterial thrombolysis in AIS, superselective intra-arterial ophthalmic thrombolysis has been extensively researched in OAO or RAO. However, the majority of these studies are small-sample, single-center retrospective studies (Table 5). The present thrombolytic drugs of intra-arterial ophthalmic thrombolysis are urokinase or rtPA, typically administered at a consistent rate, without standardized concentration and duration of administration.⁸² In contrast to the favorable results of intra-artery thrombolysis in CRAO reported in most clinical studies,^83–88 the European EAGLE study, the sole prospective multicenter RCT on the efficacy and safety of intra-arterial thrombolysis in CRAO, concluded that intra-arterial thrombolysis was not recommended as a treatment option for acute CRAO due to a comparable visual prognosis to conservative treatment and increased adverse effects (Table 5).⁸⁹ However, in the EAGLE study, only 4 people in the intra-arterial thrombolysis group intervened within 6 hours, compared to 10 in the conservative treatment group. Considering the significant impact of the onset-intervention time on visual prognosis, the negative results may be attributed to delayed interventions during intra-arterial thrombolysis. Consequently, we should consider the impact of variations in the onset-intervention time on conclusions when literature reviewing. The optimal time window for intra-arterial thrombolysis in OAO or RAO remains uncertain, as some studies indicated that the most favorable outcomes were attained within 4 hours,^85,87 whereas others suggested that improved visual prognosis was still obtained within 6 hours or even 8 hours.^83,90 One meta-analysis indicated that even intra-arterial thrombolysis within 24 hours could result in a 60% improvement in visual acuity of ≥0.3 logMAR, and a 21.9% effective rate of visual acuity ≥20/100,⁷¹ while another meta-analysis revealed that the visual acuity prognosis following intra-arterial thrombolysis was more favorable within 6 hours, with limited improvement beyond this time frame.⁹¹

Table 5 Efficacy and Safety of Intra-Arterial Thrombolysis in Patients with OAO or RAO

Regrettably, the EAGLE study was terminated in 2009, resulting in the absence of ongoing large-scale, multicenter RCTs on the efficacy and safety of intra-arterial thrombolysis in OAO or RAO. Due to the unverified efficacy and safety, it is still not endorsed as the primary treatment option for OAO or RAO in the guidelines. In theory, thrombolytic drugs administered directly into the ophthalmic artery for thrombolysis using superselective microcatheter placement are expected to exhibit improved efficacy and safety. However, based on the existing evidence, it’s plausible that intra-artery thrombolysis may only be advantageous if the intervention occurs within 4 or 6 hours and if the embolic composition is a fibrin embolus. Consequently, the management of CRAO from American Heart Association in 2021 suggests that intra-arterial thrombolysis could be contemplated for patients with OAO or RAO who have contraindications to intravenous thrombolysis, following a comprehensive assessment of the benefits and risks.⁶⁶ Nevertheless, it’s important to acknowledge that this technique has been utilized for over two decades in accessible medical facilities.

Kadonosono et al in Japan recently developed a new technique for intra-arterial thrombolysis. They conducted vitrectomy on 13 patients with CRAO within 26–48 hours of onset, followed by a direct injection of 200ug of rtPA into retinal arteries. Postoperative funduscopic angiography revealed a 76.9% complete recanalization rate of the vessels and a 92.3% improvement in visual acuity of ≥0.3 logMAR, with a mean visual acuity improvement to 20/130. There was only one case of severe vitreous hemorrhage requiring reoperation.⁹⁹ While the reported results are promising, this technique remains highly challenging, and its efficacy and safety require further confirmation through robust RCT studies involving larger sample sizes.

Ophthalmic or Retinal Artery Occlusion in Coronavirus Disease of 2019

The Coronavirus Disease 2019 (COVID-19) has emerged as a significant global health event in recent years, claiming the lives of thousands and profoundly impacting the global economy. While in May 2023, the World Health Organization declared COVID-19 was no longer in its emergency phase, it is highly likely that coronavirus will persist alongside humanity in the long term. COVID-19, caused by a virus, can lead to severe vascular damage. It affects vascular endothelial cells through the angiotensin-converting enzyme 2 (ACE2) receptor, triggering vascular inflammation, resulting in circulatory disturbances and tissue ischemia.^100,101 Moreover, endothelial cell damage induces vasoconstriction, promotes thrombus formation, and elevates the body into a procoagulant state.^102,103 Additionally, endothelial cell impairment can activate inflammatory pathways, inciting a “cytokine storm” that leads to further organ damage.^102,103 Since ACE2 receptors are also expressed in retinal vascular endothelial cells,¹⁰⁴ theoretically, coronavirus could damage retinal vascular endothelial cells, triggering blood hypercoagulability and vascular inflammation, increasing the incidence of OAO or RAO.^105,106 A cohort study examining the prevalence of retinal artery occlusion (RAO) in the United States during the pre-COVID-19 period (January 2019 to February 2020) and the COVID-19 pandemic (March 2020 to December 2020) hinted at a slight increase in RAO incidence after the COVID-19 outbreak, but no significant difference compared to pre-outbreak levels was observed.¹⁰⁷ Another cohort study suggested that the prevalence of RAO was approximately 3 per million during six months after contracting the coronavirus, showing no significant disparity from pre-infection rates.¹⁰⁸ Although cases of COVID-19 patients developing OAO or RAO have been reported,^109,110 the low incidence of RAO, coupled with a lack of corresponding causal studies, currently does not substantiate a direct correlation between COVID-19 and OAO or RAO incidence. Further mechanistic research is needed to confirm their relationship.

Interestingly, some case reports suggest that RAO may be a complication of receiving the COVID-19 vaccine.^111,112 One possible explanation is that vaccination triggers the immune system, leading to immune-mediated retinal vascular inflammation in some individuals.¹⁰⁹ Summarizing the current published case reports, RAO often occurs after the first dose of viral vector vaccines and the second dose of mRNA vaccines.¹¹¹ However, due to the extremely low incidence of RAO post-vaccination,¹¹³ the lack of the corresponding causal studies, the association between COVID-19 vaccine administration and RAO incidence cannot currently be definitively established.

Comparison of Hyperbaric Oxygen Therapy for Ophthalmic or Retinal Artery Occlusion with Acute Ischemic Stroke

Hyperbaric oxygen therapy (HBOT) can increase the partial pressure of oxygen in plasma and the dissolved oxygen content in plasma by 23-fold.¹¹⁴ Under normoxic conditions, approximately 15% of the oxygen supply to the retina originates from the choroidal circulation.¹¹⁴ And HBOT can significantly increase the oxygen partial pressure in the choroidal circulation, allowing oxygen from the choroidal circulation to diffuse into the inner layers of the retina.^9,115 Several clinical studies have also confirmed the effectiveness of HBOT in improving visual prognosis in patients with CRAO,¹¹⁶ and suggesting that the effective treatment window should be within 24 hours of CRAO onset,^117,118 and possibly up to 72 hours,¹¹⁹ with an effective treatment duration exceeding 9 hours.¹¹⁶ It is important to note that irreversible neuronal damage has not yet occurred,¹²⁰ and the choroidal circulation is still able to provide oxygenation to the retina¹²¹ for HBOT to be effective. OAO severely affect the choroidal circulation, which is why HBOT cannot be used for OAO.^9,115 Theoretically, HBOT can diffuse oxygen to ischemic areas inaccessible to blood in AIS, thereby increasing the oxygenation level in the penumbra to prevent further infarct expansion.¹²² However, when the penumbra has fully transformed into infarction, HBOT will no longer be effective. Therefore, the application of HBOT in AIS is time-limited.¹²² Animal experiments suggest that the time window is approximately within 12 hours of AIS occurrence. However, due to the lack of high-quality clinical studies, the efficacy and time window of HBOT for AIS patients cannot be definitively determined.¹²³ Consequently, the American Heart Association/American Stroke Association’s 2019 AIS guidelines classified HBOT as a Class III recommendation, suggesting its use “only in the context of clinical trials or in patients with cerebral air embolism”.¹²⁴ Large-sample RCTs should be conducted to investigate the efficacy and safety of HBOT in patients with AIS or CRAO, and to establish a standardized treatment regimens (including time window, oxygen concentration, pressure, duration, frequency and course).

Conclusions and Future Directions

OAO or RAO share the same pathogenesis and cardio-cerebrovascular risk factors with AIS. Due to the high risk of future AIS, the central nervous system function of these patients should be promptly screened and evaluated. It is necessary to establish and validate a predictive model through a large-sample, multicenter cohort study of OAO or RAO due to the differences in conclusions among previous similar studies and the lack of reliable prediction models. Such a model would be a valuable tool for ophthalmologists to assess the risk of future AIS and implement intervention measures in these patients, as well as to raise patients’ awareness about AIS prevention and improve therapies for OAO or RAO. In addition, there is a lack of reliable evidence-based medicine on the efficacy and safety of intra-arterial thrombolysis in OAO or RAO, therefore, future large-sample, multicenter RCTs about this should be conducted, despite the challenges posed by the high demands on the onset-intervention time. And since the visual prognosis of patients with OAO or RAO depends on the onset-intervention time, it is important to develop an “retinal stroke” map network similar to that of AIS and collaborate with emergency physicians, ophthalmologists, interventionalists, and neurologists to establish a fast track for thrombolysis for these patients, aiming to increase the salvage rate and improve the visual prognosis. Lastly, the strong link between retina and brain may have implications for the development of co-treatments for retinal and cerebral disorders, and future research is worthwhile to verify whether the treatment of each has beneficial effects on the other.