Central retinal artery occlusion and cerebral stroke

Sir,

Aspects of the review by Varma and colleagues¹ conspire to amplify the confusion that surrounds central retinal artery occlusion (CRAO). First, they report that a foveal cherry-red spot is present in 90% of eyes examined within 1 week of CRAO onset, whereas only 58% show concomitant macular opacification. Surely these ‘classic’ CRAO signs must co-exist?

Second, Varma and colleagues discuss ‘transient’ CRAO, a small subgroup of eyes in the Iowa CRAO classification characterised by CRA reopening by the time of initial presentation and fluorescein angiography.^{2, 3} They state that ‘transient’ CRAO is ‘analogous to a transient ischaemic attack affecting the eye’ since ‘restoration of blood flow … results in symptom resolution’.¹ This is not the case. The Iowa classification dispenses with terminological convention by labelling events as ‘transient’ even though complete reversal of symptoms and signs is precluded by a duration of ischaemic anoxia exceeding 4 h (inner retina’s maximum survival time).⁴ They also state that ‘transient’ CRAO presents ‘greatly varied fundus findings’.¹ This is not the case. If the duration of ‘transient’ CRAO exceeds 4 h, the ‘classic’ signs will persist for at least a week. Alternative presentations in the form of ‘scattered patches of retinal opacity’^{2, 3} usually signify ‘partial’, not ‘transient’, CRAO.⁵

Third, Varma and colleagues believe ‘only a rare subgroup of individuals have viable tissue’ in the inner retina by 24 h from CRAO onset.¹ This is not the case. Although the survival time of anoxic tissue in the posterior pole has been exceeded, this ‘core’ of infarction will be surrounded by a zone of critically hypoxic but viable tissue (the ‘ischaemic penumbra’), just as in cerebral stroke. Unlike penumbral cerebral cortex, however, penumbral retina can persist indefinitely.⁵ Reperfusion of this tissue explains the recovery of the ERG b-wave after CRA unclamping following 4 h of CRAO.⁴ It can also account for the visual improvement frequently attributed to ‘natural history’ in patients,^{1, 2, 3} provided neural connectivity is maintained between reoxygenated penumbra and optic nerve. However, if the CRA does not reopen and cilioretinal collaterals are not formed, the enduring penumbral state of the tissue will stimulate intraocular neovascularisation (developing in 15–20% of CRAO eyes).⁵