Subjects

This study was a retrospective observational case series that included consecutive nonarteritic cases of RAO between June 2005 and December 2012 that were evaluated at the Department of Ophthalmology and Neurology of Yonsei University Medical Center. Among the 53 subjects with diagnoses of RAO who underwent brain MRI, the 33 subjects who underwent diffusion-weighted imaging within 7 days of the onset of symptoms were enrolled. The University of Yonsei Institutional Review Board approved this retrospective research, and the research adhered to the tenets of the Declaration of Helsinki and good clinical practices (IRB No. 4-2013-0142).

The diagnoses of RAO were based on the classic clinical findings on fundus photography and retinal fluorescein angiography. The demographic data, initial neurologic symptoms and signs, visual acuities, medical histories (including hypertension, diabetes mellitus, smoking habits, and renal disease); histories of previous cerebrovascular accidents or transient ischemic attacks; and laboratory findings (including lipid profiles, erythrocyte sedimentation rates and C-reactive protein levels) of the subjects were reviewed. Detailed neurologic examinations were performed by experienced neurologists. The results of the following evaluations that were performed for etiologic work-up were also reviewed: magnetic resonance angiography, catheter angiography, 12-lead electrocardiography, transthoracic echocardiography, transesophageal echocardiography, and hematology for hypercoagulable states.

Etiological Classification

Causes were classified based on the Trial of Org 10172 in Acute Stroke Treatment (TOAST) subtype classification system, which is the most widely used etiologic classification system for patients after acute stroke. The TOAST system categorizes causes into the following 5 subgroups: 1) large-artery atherosclerosis; 2) cardioembolism; 3) small-artery occlusion; 4) stroke of other determined cause; and 5) stroke of undetermined cause. The classifications are based on clinical symptoms and the results of etiologic evaluations that include brain imaging, cardiac evaluation, angiography, and other laboratory tests that evaluate prothrombotic conditions. Large-artery atherosclerosis is diagnosed when brain imaging reveals significant (>50%) stenosis or occlusion of a relevant artery (the carotid artery in cases of RAO) that is most likely due to atherosclerosis. Potential sources of cardioembolism should be excluded. Cardioembolism is diagnosed in patients with cardiac sources of emboli without large-artery atherosclerotic sources. Small-artery occlusion is diagnosed when patients present with traditional lacunar syndromes or have brain stem or subcortical lesions of <15 mm without evidence of large-artery atherosclerosis or cardioembolism. The diagnosis of stroke of other determined causes includes patients with rare stroke causes that include hypercoagulable states, hematologic diseases, or nonatherosclerotic vasculopathies with no evidence of large-artery atherosclerosis or cardioembolism. Acute strokes due to iatrogenic causes are classified as stroke of other determined cause. Stroke of undetermined cause is further divided into the following 3 subgroups: 1) 2 or more causes identified; 2) negative evaluation; and 3) incomplete evaluation. The negative evaluation subgroup includes patients who have no recognized potential causes despite extensive evaluations. The incomplete evaluation subgroup includes patients with no identified cause and insufficient evaluation.

In this study, the subjects were divided into the following 4 diagnostic subgroups: large-artery atherosclerosis, cardioembolism, stroke of other determined cause, and stroke of undetermined cause. No subject met the criteria for 2 or more identified causes or incomplete evaluation; therefore, all subjects included in the category of stroke of undetermined cause were categorized into the negative evaluation subgroup. Large-artery atherosclerosis and cardioembolism, the 2 most common causes of stroke, were considered to be the major causes, and the other causes were considered to be minor causes.

Diffusion-weighted MRI

Subjects who underwent diffusion-weighted imaging within 7 days of the onset of symptoms were included in this study. Brain MRIs were performed with a 1.5-Tesla system (Intera or Achieva; Philips Medical Systems, Best, The Netherlands; or Signa Horizon, GE Medical Systems, Milwaukee, WI, USA) or a 3.0-Tesla system (Intera or Achieva, Philips Medical Systems). Lesions that were hyperintense on diffusion-weighted images were regarded as acute ischemic lesions. Acute ischemic lesions remain hyperintense for only 7-14 days. The sizes of the lesions were defined by the maximum diameter of the hyperintense areas as measured across the axial slices. Diffusion-weighted imaging lesions were classified as lesions in 1 vascular territory (single or multiple lesions in the unilateral anterior circulation or in the posterior circulation) or as lesions in multiple vascular territories (multiple lesions in the bilateral anterior circulations or in both the anterior and posterior circulations).

Statistical Analyses

Chi-square tests or the Fisher exact test (for categorical variables) and t tests (for continuous variables) were used to compare the baseline demographic characteristics, risk factors and etiologies between the central retinal artery occlusion (CRAO) and branch retinal artery occlusion (BRAO) groups. Chi-squared tests or the Fisher exact test were used to compare the underlying etiologies between the diffusion-weighted imaging lesion-positive and lesion-negative groups. A P value <0.05 was considered statistically significant. Statistical analyses were performed with SPSS software (v 18.0, SPSS, Chicago, IL, USA).