The effects of daily low-dose aspirin on white matter hyperintensity lesions and retinal vascular calibre in healthy older adults: the ENVIS-ion exploratory neuroimaging substudy of the ASPREE randomised clinical trial.

BACKGROUND Cerebral small vessel disease and alterations in retinal vascular calibre (RVC) are recognised precursors of stroke, dementia, and cognitive decline. We aimed to assess the effect of low-dose aspirin on white matter hyperintensity (WMH), a marker of cerebral small vessel disease, and changes in RVC.

METHODS We conducted a prospectively planned exploratory neurovascular substudy (ENVIS-ion) of the Aspirin in Reducing Events in the Elderly (ASPREE) double-blinded randomised clinical trial of 19 114 older adults (aged ≥70 years), who had no previous cardiovascular disease, stroke, or cognitive impairment at baseline. Participants were allocated to daily enteric-coated aspirin 100 mg or matching placebo using computer-generated randomisation and underwent MRI of the brain and fundus photography at two clinical trials sites in Australia at baseline and after 3 years. WMH and RVC measures were assessed by graders blinded to study treatment allocation. The effects of aspirin on total and regional WMH volumes (as a percentage of total brain volume) and RVC over time were analysed using linear models. ASPREE is registered with ClinicalTrials.gov, NCT01038583, and the International Standard Randomised Controlled Trial Number Registry, ISRCTN83772183.

FINDINGS Between April, 2010, and April, 2012, 610 participants from the eligible ASPREE cohort of 2346 individuals were enrolled in the ENVIS-ion substudy. Of the 610 participants (mean age 75·2 years [SD 4·1], 289 [47%] male and 321 [53%] female), 312 were assigned to receive aspirin and 298 to placebo. Over 3 years, the aspirin group had a greater increase in the percentage of deep WMH (β 0·14 [95% CI 0·01 to 0·27]) but there was no difference between aspirin and placebo groups in changes from baseline to 3 years in total brain WMH (0·05 [-0·02 to 0·11]) or periventricular WMH (0·03 [-0·03 to 0·09]). There was no evidence of an aspirin effect on RVC. The rate of major haemorrhage was higher in the aspirin arm for the ASPREE study (hazard ratio 1·38, 95% CI 1·18 to 1·62).

INTERPRETATION In this exploratory study, there was no evidence that low-dose aspirin in healthy older adults had any effect on RVC or attenuated the progression of WMH during a 3-year period.

FUNDING National Health and Medical Research Council of Australia.