To the Editor:

We congratulate Fallico et al. on their clear collation and presentation of data on the incidence of acute cerebral ischaemia detected by magnetic resonance imaging (MRI) following acute central retinal artery occlusion (CRAO) and branch retinal artery occlusion (BRAO) [1]. Their article reports 25–30% of these patients have evidence of cerebral ischaemia.

We would like to make three points. Firstly, while many of the diffusion-weighted lesions identified were asymptomatic, their presence is important. MRI in a truly asymptomatic population has been reported to yield no diffusion-positive lesions [2]. Therefore, identifying diffusion-positive lesions in the population of patients with CRAO or BRAO implies that these patients are at risk of clinical ischaemic events many of which may be preventable with early diagnostic evaluation and treatment.

Secondly, in the data presented, up to 45% of acute lesions on MRI were not ipsilateral to the affected eye. This emphasises how the use of MRI can highlight the need to look for a more proximal, probably cardiac, source of emboli. These patients may benefit from echocardiography and prolonged cardiac monitoring for atrial fibrillation in particular.

Thirdly, we fully support the close collaboration of neurological and ophthalmological vascular services. However, we think hospital admission or emergency department attendance could be avoided for most patients who could be dealt with very effectively through rapid access TIA clinics. In practical terms, however, this approach is likely to include some delay. Therefore, we would encourage ophthalmologists to institute vascular secondary prevention measures immediately on diagnosis of CRAO or BRAO before referring on for further neurological evaluation.