Thrombolytic Therapy for Acute Central Retinal Artery Occlusion
2019; Lippincott Williams & Wilkins; Volume: 51; Issue: 2 Linguagem: Inglês
10.1161/strokeaha.119.027478
ISSN1524-4628
AutoresBrian Mac Grory, Patrick Lavin, Howard S. Kirshner, Matthew Schrag,
Tópico(s)Acute Ischemic Stroke Management
ResumoHomeStrokeVol. 51, No. 2Thrombolytic Therapy for Acute Central Retinal Artery Occlusion Free AccessReview ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissionsDownload Articles + Supplements ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toSupplementary MaterialsFree AccessReview ArticlePDF/EPUBThrombolytic Therapy for Acute Central Retinal Artery Occlusion Brian Mac Grory, MB BCh BAO, MRCP, Patrick Lavin, MB BCh BAO, MRCPI, Howard Kirshner, MD, PhD and Matthew Schrag, MD, PhD Brian Mac GroryBrian Mac Grory Correspondence to Brian Mac Grory, MB BCh BAO, MRCP, Department of Neurology, Warren Alpert Medical School of Brown University, APC 529, 593 Eddy St, Providence, RI 02903. Email E-mail Address: [email protected] From the Department of Neurology, Warren Alpert Medical School of Brown University, Providence, Rhode Island (B.M.G.) , Patrick LavinPatrick Lavin Department of Ophthalmology and Visual Sciences (P.L.), Vanderbilt University School of Medicine, Nashville, TN. Department of Neurology (P.L., H.K., M.S.), Vanderbilt University School of Medicine, Nashville, TN. , Howard KirshnerHoward Kirshner Department of Neurology (P.L., H.K., M.S.), Vanderbilt University School of Medicine, Nashville, TN. and Matthew SchragMatthew Schrag Department of Neurology (P.L., H.K., M.S.), Vanderbilt University School of Medicine, Nashville, TN. Originally published23 Dec 2019https://doi.org/10.1161/STROKEAHA.119.027478Stroke. 2020;51:687–695Other version(s) of this articleYou are viewing the most recent version of this article. Previous versions: December 23, 2019: Ahead of Print Central retinal artery occlusion (CRAO) causes an interruption of blood flow to the retina resulting in the acute onset of retinal tissue dysfunction. CRAO (along with branch retinal artery occlusion and ophthalmic artery occlusion) is a form of ischemic stroke.1 Like cerebral stroke, it can occur because of large artery atherosclerosis, embolism (from the heart, aorta, or great vessels), inflammatory vascular disease, or hypercoagulability. While many patients have some improvement in vision, only about 17% regain a functional visual acuity2 and in 50% of affected people the only remaining visual field is a small peripheral island.3 CRAO affects men and women equally (2 cases per 100 000 person years) and the incidence increases with age.4,5 It has both arteritic or nonarteritic forms, but thrombolytic therapy for treatment of the non-arteritic form (accounting for 95% of cases) will be the focus of this review.The optimal acute treatment of CRAO is controversial. A variety of therapies have been studied with the goal of protecting vision including anterior chamber paracentesis, hyperbaric oxygen, acetazolamide therapy, hemodilution, and ocular massage. None have proven efficacy, they are not recommended in professional guidelines,6,7 and may be harmful.2 More than half of academic centers in the United States will consider intravenous tPA (tissue-type plasminogen activator) to select patients who present at early time points, but it is the preferred initial treatment in only 36%.8 This approach will be discussed in greater detail in the remainder of this review.The Anatomy and Pathophysiology of Central Retinal Artery OcclusionArterial Supply to the RetinaThe ophthalmic artery (Figure 1) is the first branch of the internal carotid artery after it emerges from the cavernous sinus to pierce the dural sheath and enter the intracranial cavity.9 Its origin lies along the infero-lateral aspect of the optic nerve and gives off the central retinal artery as it courses medially to traverse the optic nerve (either superiorly or inferiorly).10 The retina has 2 vascular networks—both arising from the ophthalmic artery. The retinal arterial circulation supplies the inner retina while the choroidal network is a nexus of arterioles supplying the outer retina and choroid. Compared with a normal retina, a cherry red spot (Figure 2A) is seen in central artery occlusion because of preserved choroidal circulation underlying the fovea surrounded by a pale, edematous retina. The central retinal artery has 3 sections10: (1) The intraorbital segment (extending from the origin to the point where it pierces the dural sheath surrounding the optic nerve); (2) the intradural segment (the component of the artery lying in the space between the optic nerve and dural sheath) and; (3) the intraneural segment (the section of artery lying within the optic nerve itself). The central retinal artery travels within the optic nerve and emerges in the optic nerve head before dividing into branch retinal arteries where is supplies the inner retina.11 In 26% of people,12 a portion of the retina is supplied by the cilioretinal artery—a branch arising from the posterior ciliary circulation.13 The diameter of the central retinal artery is ≈160 µm at the optic nerve head.14,15 The narrowest section of the CRA is about 2 mm before the visible section where it passes through the lamina cribrosa16; there is an additional narrow section even more proximally where it pierces the dura near where it enters the substance of the optic nerve.10 Likely, these are the primary sites where emboli lodge. The ophthalmic artery has a rich anastomotic network with anastomoses arising through a deep network incorporating the internal maxillary artery and anterior temporal artery as well as a superficial network incorporating the dorsal nasal artery and the facial artery.11 Hayreh and Weingeist17,18 found that 89% of eyes had residual retinal circulation on fluorescein angiography after occlusion of the central retinal artery. This may be because of cilioretinal capillary anastomoses via the microcirculation of the optic disc, from the pial anastomoses originating proximal to the site of occlusion11,19 or from partial recanalization of the central retinal artery.Download figureDownload PowerPointFigure 1. Medial to lateral view of the arterial supply to the optic nerve and retina.Download figureDownload PowerPointFigure 2. A, Funduscopic images of healthy (left) and ischemic (right) retina from the same patient. The ischemic retina illustrates retinal whitening, a cherry red spot, boxcarring of the branch retinal arterioles and a small strip of viable tissue immediately temporal to the optic disc as a result of a small cilioretinal artery. B, Calcified embolus (left) and branch retinal artery occlusion (right). C, Fundoscopic image (left) of a patient presenting with sudden, painless, monocular visual loss and found to have a vitreo-retinal hemorrhage partially obscuring visualization of the optic disc. Computed tomography of the brain without contrast (right) of a patient with a vitreo-retinal hemorrhage in the right eye evident as a hyperdensity in the posterior aspect of the globe. The authors are grateful to Dr Lory Snady-McCoy for permission to reproduce this image.Retinal Ischemic ToleranceRetinal ganglion cells and their axons (the most distal portion of the second cranial/optic nerve) are components of the central nervous system. The retina is extremely metabolically active20 and dependent on the constant availability of substrates for anerobic glycolysis. Retinal ganglion cell death sets in motion a process of axonal degeneration with optic atrophy ensuing over a period of several weeks. Ischemia to the retina has been modeled in ex vivo culture, rat, feline, and porcine models as well as in nonhuman primates. The thin inner retina is layered on the outer retina and choroid, and there is diffusion of oxygen from the choroid to the inner retina during central retinal artery occlusion.20 Thus, the development of a valid model is dependent on selectively interrupting central retinal artery blood flow while preserving the choroidal circulation. By analogy to ischemic stroke of the cerebrum, there is a retinal ischemic penumbra with observed anoxic, hypoxic, and normoxic compartments produced during central retinal artery occlusion.21 In nonhuman primates, occlusion of the central retinal artery proximal to its entry into the body of the optic nerve longer than 105 minutes was sufficient to produce severe retinal tissue injury.17,18,22 The degree of cell death in the retina correlated with the duration of occlusion of the central retinal artery. In older animals with experimentally induced chronic hypertension and an atherogenic diet, irreversible tissue loss did not occur until 240 minutes after occlusion.23 This suggests more retinal tolerance in older animals with vascular risk factors. It is not known whether or not the retinal ischemic tolerance is increased in humans, where the lifespan is much longer.Characteristics of Particles Obstructing the Retinal VasculatureTraditional teaching in the ophthalmological literature argues that most emboli in CRAO are cholesterol or calcium (Figure 2B) and therefore not lysable.19,24 However, for all major causes of CRAO—carotid disease, valvular disease, and atrial fibrillation—emboli are associated with fibrin, even if containing calcium or cholesterol crystals:The studies suggesting the majority of emboli in CRAO are not fibrin-based on the appearance of retinal emboli on fundoscopy,24,25 but emboli visualized on fundoscopy are in branch retinal arterioles (Figure 2C), and because the narrowest points in the course of the CRA are proximal to the globe, the overwhelming majority of obstructing emboli lodge in retrobulbar sites and are not visible on fundoscopy.Acute ischemic cerebral stroke (either symptomatic or asymptomatic) occurs in 30% or more of patients with CRAO (Table I in the online-only Data Supplement); moreover, patients with CRAO are at increased risk of subsequent ischemic stroke26,27 arguing for commonality between the lesions obstructing the retinal arterial circulation and the cerebral circulation.Atrial fibrillation is found in approximately 10.6% to 21.4% of patients with CRAO28,29 and the rate of CRAO is higher in AF than in age-matched controls.30 Emboli due to AF are like thrombotic and therefore potentially lysable.Embolic material captured during carotid artery stenting31 and after transcatheter aortic valve replacement32 is usuallycomposed—at least in part—of thrombus. This suggests that even if an obstructing particle has crystalline elements, it still may be amenable to thrombolysis.Thrombolytic Therapy for Central Retinal Artery OcclusionIntravenous ThrombolysisIntravenous thrombolysis reduces morbidity from acute ischemic stroke33,34 when given within 4.5 hours of the time a person was last free of symptoms. Currently, the American Heart Association's guidelines for the treatment of acute ischemic stroke do not specifically address CRAO.35,36 However, CRAO causing retinal ischemia conforms to the definition of acute ischemic stroke (along with cerebral and spinal ischemia).1 Likely, the lack of recommendation for treatment is because there is no high-quality randomized data demonstrating benefit for thrombolysis in CRAO. The only randomized trial performed to-date37 of intravenous thrombolysis for CRAO was underpowered and enrolled patients from 4.5 to 24 hours after the onset of symptoms. The lack of efficacy in this trial is likely explained by the enrollment of patients beyond the time limits of retinal ischemic tolerance, as the animal experiments published by Hayreh et al23 suggest that irreversible retinal tissue loss occurs in every case of retina ischemia of 240 minutes or more.Schrag et al2 conducted a meta-analysis of studies concerning the use of intravenous thrombolysis for CRAO. Patient-level data were obtained for 9 studies and a total of 147 patients accounting for 80% of the cases in the published literature at the time. Of the studies included, 2 used alteplase, 4 streptokinase, and 3 urokinase. There was a higher rate of improvement in visual outcome in the patients treated with thrombolysis compared with studies reporting on the natural history of CRAO or compared with studies reporting alternative therapies used for CRAO. Thirty four patients received IV thrombolysis within 4.5 hours of symptom onset and 50% (17 out of 34) had recovery of functional visual acuity. This represented a 32.3% absolute risk reduction with a number needed to treat of 4 (2.6–6.6, 95% CI). Crucially, there was no significant difference in the recovery rate of patients treated with IV tPA after 4.5 hours from last known well compared with the natural history of CRAO suggesting late treatment is probably futile. There were 5 serious hemorrhagic events (including 4 fatalities) out of 147 people identified who were treated with an IV thrombolytic agent, all attributed to the use of the older agents urokinase and streptokinase. Eight out of 13 patients treated with alteplase within 4.5 hours had a favorable visual outcome compared with only 4 out of 23 patients treated between 4.5 and 12 hours.Since the analysis of Schrag et al, there have been 3 more case series published concerning the use of IV tPA for CRAO:Nedelmann et al38 treated 46 patients with CRAO, all of whom were evaluated for a retrobulbar spot sign,39 which is felt to be indicative of a calcium-based embolus, and thus less amenable to thrombolytic therapy compared with a fibrin-based clot. Intravenous thrombolysis was given in 11/46 of these patients—4/11 without a spot sign had a significant visual improvement whereas 7/11 with a spot sign had persistent visual impairment.Préterre et al40 treated 30 patients over a 10-year period with IV tPA at their institution. The mean time to treatment was 273 minutes and 55.2% had a significant visual improvement at 1 month post-diagnosis.Schultheiss et al41 studied 20 consecutive patients treated with intravenous thrombolysis within 4.5 hours of onset. This study was a single arm study; however, when compared with the medical therapy arm of the EAGLE (European Assessment Group for Lysis in the Eye) Study,42 the enrolled patients had a superior outcome.Only 1 patient in these 3 recent series had an intracerebral hemorrhage and this occurred in the setting of immediate anticoagulation. One patient in the randomized, placebocontrolled trial published in 201137 suffered a symptomatic intracranial hemorrhage, however, was found to have cerebral amyloid angiopathy which increased their liability to intracranial hemorrhage. Two patients in this trial developed neovascularization of the retina necessitating panretinal photocoagulation. The incidence of ocular neovascularization of the iris is up to 18.2%43 after central retinal artery occlusion. This is associated with a risk of neovascular glaucoma. Reperfusion reduces the risk of developing neovascularization.44An ongoing clinical trial in France (NCT03197194) is randomizing patients, aged 18 to 80, with an initial visual acuity of <1/20 (20/400) to receive either intravenous alteplase at a dose of 0.9 mg/kg or aspirin 300 mg within 4.5 hours of symptom onset (anticipated enrollment of 70 patients and a planned completion date of September 2020).Intraarterial ThrombolysisIntraarterial thrombolysis (also referred to as local intraarterial thrombolysis in the published literature) is given via cannulation of the femoral artery, introduction of a catheter into the internal carotid artery then the proximal ophthalmic artery at which point thrombolysis is administered. Thus, a precise dose of thrombolytic can be tailored to the individual patient in real time. This requires an expert in the ophthalmological examination to be present for the duration of the procedure (to perform fundoscopy and test visual acuity after each dose of tPA). However, intraarterial therapy is labor intensive to deploy, requires a functioning interventional suite available on an emergency basis, poses technical challenges (including a risk of ophthalmic artery catheter-induced spasm), and is associated with a risk of periprocedural stroke (up to 8%45). The EAGLE trial (European Assessment Group for Lysis in the Eye)42 was prematurely terminated for futility after enrollment of 84 patients presenting within 20 hours of symptom onset of whom 44 were treated with intraarterial thrombolysis and 40 with medical therapy. The primary end point was best corrected visual acuity 30 days after enrollment. There was no difference in the rate of clinically significant visual improvement between the 2 groups, and there was a higher rate of adverse events in the group receiving intraarterial thrombolysis (37.1% versus 3.4%). The major problem with this trial is that no patients were treated with intraarterial thrombolysis within 4.5 hours (the mean time between symptom onset and treatment was 13 hours). Because intraarterial delivery of thrombolysis is technically complex, this timeline probably does not represent a procedural failure of the trial, but rather the reality of how difficult it is to rapidly deploy this treatment modality.Rationale for Further StudyWhile thrombolytic therapy for CRAO is currently not considered standard of care, it is offered at more than half of academic medical centers in the United States.8 We think that there is clinical equipoise and that further study is warranted.Devastating Visual OutcomeOnly 17% of people with CRAO regain a functional visual acuity in the affected eye without treatment.2 Monocular visual loss impairs quality of life because of reduced field of vision, impairments in depth perception and increased vulnerability to blindness with vision-threatening disorders in the opposite eye. This has been demonstrated in acquired monocular visual loss from surgical enucleation46 as well as in patients with CRAO specifically47 who have long-term quality of life impairments in domains as diverse as overall social functioning, mental health and level of dependency on others. This justifies exposure to thrombolysis and its attendant risks in selected patients. This is particularly true given that the risk of symptomatic intracranial hemorrhage is likely similar to that seen in stroke mimics—≈1%.48 For those patients with CRAO and concurrent cerebral ischemia (without measurable deficits on clinical examination), the risk of thrombolysis is likely higher and would equate to that seen thrombolysis for minor stroke. In the PRISMS trial, symptomatic intracranial hemorrhage occurred in 3.2% of patients (1.3% when using the SITS-MOST definition of ICH).49,50 In making decisions regarding thrombolysis for acute stroke, physicians assign the most weight to motor and speech deficits51 and may underweight other domains of neurological impairment. However, qualitative research suggests that patients are strongly averse to even mild residual deficits.52 Monocular visual loss is likely associated with a different emotional valence in different patient based on their premorbid functional baseline and thus the risk of intracranial hemorrhage associated with thrombolysis should be interpreted in this light.Positive Indicators From Previous StudiesSchrag et al2 demonstrated a benefit to treatment with intravenous thrombolysis when given within 4.5 hours. Hayreh et al17 demonstrated that the longer the period of retinal arterial occlusion, the more damage takes place. Earlier treatment of CRAO improves visual outcome53 in the same way as it improves neurological outcome in cerebral stroke.54 Treatment beyond the duration of retinal tolerance time likely will be ineffective, even if there is recanalization of the implicated artery, because irreversible tissue damage has occurred.Mechanistic PlausibilityThe recanalization rate in response to thrombolytic therapy improves as a vessel narrows55 (Figure 3). The diameter of the CRA is optimal for tPA success—≈160 µm at the optic nerve head14,15 and may be smaller in the presence of vascular risk factors.56Fluorescein angiography demonstrates that in cases of amaurosis fugax, microemboli fragment quickly.57 Thrombolysis may accelerate this process or lead to resolution of retinal artery occlusion in cases where irreversible tissue loss may have otherwise occurred.Most occluding lesions likely are composed—at least in part—of fibrin (as discussed above) and therefore there is a potential benefit to thrombolysis even in those cases of CRAO that arise from the aorta or carotid circulation.Download figureDownload PowerPointFigure 3. Relative diameter of major vessels of the anterior circulation. ICA indicates internal carotid artery; M1, M2, M3, subdivisions of the middle cerebral artery; and OA, ophthalmic artery.Opportunities and Challenges in Designing a Clinical TrialThe greatest opportunity in designing a clinical trial for thrombolysis in CRAO lies in the ability to deploy existing infrastructure and care protocols designed for the treatment of acute cerebral stroke. We think that the window for enrollment should be within 4.5 hours. This is important (1) to ensure that treatment occurs within the time frame that would permit salvage of retinal tissue, (2) to harmonize the treatment protocol with that for acute ischemic stroke, and (3) to limit the risk of hemorrhagic brain injury given the non-trivial rate of concurrent brain ischemia among patients with CRAO.CRAO may efficaciously be treated with the more modern thrombolytic agent, tenecteplase (TNK). The EXTEND-IA TNK58 study demonstrated that intravenous TNK led to a superior 90-day functional outcome when compared with alteplase as measured by the modified Rankin Scale score (63% of those treated with TNK had an modified Rankin Scale score of 0 to 2 at 90 days compared with only 50% of those treated with alteplase). TNK is an attractive option for the treatment of CRAO as it can be given as a single bolus (owing to its longer half-life of 22 minutes compared with a 3.5 minute half-life for tPA),59 has a 15-fold higher fibrin specificity,59 potentially a lower risk of systemic hemorrhage and a higher rate of arterial recanalization than alteplase.60,61Attempting a clinical trial that enrolls patients within 4.5 hours of the onset of visual symptoms is a significant challenge for 3 key reasons:Achieving an effective enrollment rate of patients will require multiple centers, cooperation with community practitioners, public outreach (analogous to the brain attack campaign), and an adjustment of existing emergency department triage protocols to activate an acute stroke code.Two important steps add to the complexity and resources needed to administer a trial of thrombolysis in CRAO: (1) examination by an experienced ophthalmologist for confirmatory funduscopic examination findings (Table I in the online-only Data Supplement) and to aid in the exclusion of other conditions that also present with painless monocular visual loss, such as nonarteritic ischemic optic neuropathy, intraocular hemorrhage (Figure 2C), retinal detachment, and functional visual loss and (2) the erythrocyte sedimentation rate and CRP (C-reactive protein) must be checked on all patients, 50 years and older, to aid with the exclusion of giant cell arteritis (a proposed management protocol is included in Figure 4).The selection of a readily discernible, objective, and patient-centered outcome is challenging. Assuming that only patients with a visual acuity of 20/200 or worse at onset are enrolled, one reasonable approach would be to select functional recovery in the affected eye at 90 days to a visual acuity of 20/100—this is one line above the threshold of legal blindness on the Snellen chart—as the primary outcome. The evaluation of this outcome should be performed by an experienced ophthalmologist to ensure that perceived visual recovery is not merely a function of eccentric fixation (a compensatory mechanism whereby patients with absent central vision but a preserved paracentral window of vision can adapt to fixate on this location, which can mimic central visual recovery62). Secondary outcomes that directly reflect central retinal artery recanalization (independent of visual acuity) include optical coherence tomography angiography of the macula and Doppler ultrasonography of the central retinal artery at 90 days. These results would provide important context to a primary outcome of functional visual recovery.Download figureDownload PowerPointFigure 4. Management algorithm of patients presenting with sudden, painless, monocular visual loss and a suspicion for central retinal artery occlusion. CRP indicates C-reactive protein.ConclusionsCurrent evidence dictates that central retinal artery occlusion should be treated in an emergent fashion in the same manner as acute cerebral ischemic stroke. It is not only an ophthalmological problem, it is a disorder of the cerebral arterial tree that affects the retina. There is a need for further randomized, controlled trials examining the different agents currently available for intravenous thrombolysis within the 4.5 hours window for treatment of this devastating condition.AcknowledgmentsWe express their gratitude to Dr Lory Snady-McCoy of Brown University/Rhode Island Hospital for providing the funduscopic image used in Figure 2C.DisclosuresNone of the authors have conflicts of interest related to this data. This manuscript is not under review at any other journal. There are no redundant publications based on this dataset. All co-authors meet the ICMJE requirements for authorship.FootnotesThe online-only Data Supplement is available with this article at https://www.ahajournals.org/doi/suppl/10.1161/STROKEAHA.119.027478.Correspondence to Brian Mac Grory, MB BCh BAO, MRCP, Department of Neurology, Warren Alpert Medical School of Brown University, APC 529, 593 Eddy St, Providence, RI 02903. Email [email protected]eduReferences1. Sacco RL, Kasner SE, Broderick JP, Caplan LR, Connors JJ, Culebras A, et al; American Heart Association Stroke Council, Council on Cardiovascular Surgery and Anesthesia; Council on Cardiovascular Radiology and Intervention; Council on Cardiovascular and Stroke Nursing; Council on Epidemiology and Prevention; Council on Peripheral Vascular Disease; Council on Nutrition, Physical Activity and Metabolism. An updated definition of stroke for the 21st century: a statement for healthcare professionals from the American Heart Association/American Stroke Association.Stroke. 2013; 44:2064–2089. doi: 10.1161/STR.0b013e318296aecaLinkGoogle Scholar2. Schrag M, Youn T, Schindler J, Kirshner H, Greer D. Intravenous fibrinolytic therapy in central retinal artery occlusion: a patient-level meta-analysis.JAMA Neurol. 2015; 72:1148–1154. doi: 10.1001/jamaneurol.2015.1578Google Scholar3. Hayreh SS, Zimmerman MB. Central retinal artery occlusion: visual outcome.Am J Ophthalmol. 2005; 140:376–391. doi: 10.1016/j.ajo.2005.03.038CrossrefMedlineGoogle Scholar4. Park SJ, Choi NK, Seo KH, Park KH, Woo SJ. Nationwide incidence of clinically diagnosed central retinal artery occlusion in Korea, 2008 to 2011.Ophthalmology. 2014; 121:1933–1938. doi: 10.1016/j.ophtha.2014.04.029Google Scholar5. Leavitt JA, Larson TA, Hodge DO, Gullerud RE. The incidence of central retinal artery occlusion in olmsted county, minnesota.Am J Ophthalmol. 2011; 152:820–3.e2. doi: 10.1016/j.ajo.2011.05.005Google Scholar6. Arnold AC. Urgent evaluation of the patient with acute central retinal artery occlusion.Am J Ophthalmol. 2018; 196:xvi–xvii. doi: 10.1016/j.ajo.2018.08.033Google Scholar7. Olsen TW, Pulido JS, Folk JC, Hyman L, Flaxel CJ, Adelman RA. Retinal and ophthalmic artery occlusions preferred practice pattern®.Ophthalmology. 2017; 124:P120–P143. doi: 10.1016/j.ophtha.2016.09.024Google Scholar8. Youn TS, Lavin P, Patrylo M, Schindler J, Kirshner H, Greer DM, et al. Current treatment of central retinal artery occlusion: a national survey.J Neurol. 2018; 265:330–335. doi: 10.1007/s00415-017-8702-xCrossrefMedlineGoogle Scholar9. Gibo H, Lenkey C, Rhoton ALMicrosurgical anatomy of the supraclinoid portion of the internal carotid artery.J Neurosurg. 1981; 55:560–574. doi: 10.3171/jns.1981.55.4.0560CrossrefMedlineGoogle Scholar10. SINGH S, DASS R. The central artery of the retina. I. origin and course.Br J Ophthalmol. 1960; 44:193–212. doi: 10.1136/bjo.44.4.193Google Scholar11. Michalinos A, Zogana S, Kotsiomitis E, Mazarakis A, Troupis T. Anatomy of the ophthalmic artery: a review concerning its modern surgical and clinical applications.Anat Res Int. 2015; 2015:591961. doi: 10.1155/2015/591961Google Scholar12. Brown GC, Shields JA. Cilioretinal arteries and retinal arterial occlusion.Arch Ophthalmol. 1979; 97:84–92. doi: 10.1001/archopht.1979.01020010024006Google Scholar13. Hayreh SS. The cilio-retinal arteries.Br J Ophthalmol. 1963; 47:71–89. doi: 10.1136/bjo.47.2.71Google Scholar14. Dorner GT, Polska E, Garhöfer G, Zawinka C, Frank B, Schmetterer L. Calculation of the diameter of the central retinal artery from noninvasive measurements in humans.Curr Eye Res. 2002; 25:341–345. doi: 10.1076/ceyr.25.6.341.14231Google Scholar15. Lee KE, Klein BE, Klein R, Meuer SM. Association of retinal vessel caliber to optic disc and cup diameters.Invest Ophthalmol Vis Sci. 2007; 48:63–67. doi: 10.1167/iovs.05-1203Google Scholar16. Hertel . Beitrag zur kenntnis der angiosklerose der zentralgafasse des auges.1901; 52:191–200.Google Scholar17. Hayreh SS, Weingeist TA. Experimental occlusion of the central artery of the retina. I. ophthalmoscopic and fluorescein fundus angiographic studies.Br J Ophthalmol. 1980; 64:896–912. doi: 10.1136/bjo.64.12.896Google Scholar18. Hayreh SS, Weingeist TA. Experimental occlusion of the central artery of the retina. IV: retinal tolerance time to acute ischaemia.Br J Ophthalmol. 1980; 64:818–825. doi: 10.1136/bjo.64.11.818Google Scholar19. Hayreh SS. Prevalent misconceptions about acute retinal vascular occlusive disorders.Prog Retin Eye Res. 2005; 24:493–519. doi: 10.1016/j.preteyeres.2004.12.001CrossrefMedlineGoogle Scholar20. Wangsa-Wirawan ND,
Referência(s)