BACKGROUND

Hypertension is the leading risk factor for cerebral small vessel disease. We aimed to determine whether antihypertensive drug classes differentially affect microvascular function in people with small vessel disease.

METHODS

We did a multicentre, open-label, randomised crossover trial with blinded endpoint assessment at five specialist centres in Europe. We included participants aged 18 years or older with symptomatic sporadic small vessel disease or cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) and an indication for antihypertensive treatment. Participants were randomly assigned (1:1:1) to one of three sequences of antihypertensive treatment using a computer-generated multiblock randomisation, stratified by study site and patient group. A 2-week washout period was followed by three 4-week periods of oral monotherapy with amlodipine, losartan, or atenolol at approved doses. The primary endpoint was change in cerebrovascular reactivity (CVR) determined by blood oxygen level-dependent MRI response to hypercapnic challenge in normal-appearing white matter from the end of washout to the end of each treatment period. Efficacy analyses were done by intention-to-treat principles in all randomly assigned participants who had at least one valid assessment for the primary endpoint, and analyses were done separately for participants with sporadic small vessel disease and CADASIL. This trial is registered at ClinicalTrials.gov, NCT03082014, and EudraCT, 2016-002920-10, and is terminated.

FINDINGS

Between Feb 22, 2018, and April 28, 2022, 75 participants with sporadic small vessel disease (mean age 64·9 years [SD 9·9]) and 26 with CADASIL (53·1 years [7·0]) were enrolled and randomly assigned to treatment. 79 participants (62 with sporadic small vessel disease and 17 with CADASIL) entered the primary efficacy analysis. Change in CVR did not differ between study drugs in participants with sporadic small vessel disease (mean change in CVR 1·8 × 10%/mm Hg [SE 20·1; 95% CI -37·6 to 41·2] for amlodipine; 16·7 × 10%/mm Hg [20·0; -22·3 to 55·8] for losartan; -7·1 × 10%/mm Hg [19·6; -45·5 to 31·1] for atenolol; p=0·39) but did differ in patients with CADASIL (15·7 × 10%/mm Hg [SE 27·5; 95% CI -38·3 to 69·7] for amlodipine; 19·4 × 10%/mm Hg [27·9; -35·3 to 74·2] for losartan; -23·9 × 10%/mm Hg [27·5; -77·7 to 30·0] for atenolol; p=0·019). In patients with CADASIL, pairwise comparisons showed that CVR improved with amlodipine compared with atenolol (-39·6 × 10%/mm Hg [95% CI -72·5 to -6·6; p=0·019) and with losartan compared with atenolol (-43·3 × 10%/mm Hg [-74·3 to -12·3]; p=0·0061). No deaths occurred. Two serious adverse events were recorded, one while taking amlodipine (diarrhoea with dehydration) and one while taking atenolol (fall with fracture), neither of which was related to study drug intake.

INTERPRETATION

4 weeks of treatment with amlodipine, losartan, or atenolol did not differ in their effects on cerebrovascular reactivity in people with sporadic small vessel disease but did result in differential treatment effects in patients with CADASIL. Whether antihypertensive drug classes differentially affect clinical outcomes in people with small vessel diseases requires further research.

FUNDING

EU Horizon 2020 programme.