Genetics
Genetic Lp(a) raises large-artery stroke risk 23% but not small vessel stroke, Mendelian randomisation of 343,681 (J Am Heart Assoc 2026)
Original title: Genetically Predicted Levels of Lipoprotein(a) and Risk of Cerebrovascular Disease
Mendelian randomisation study using genetic associations with Lp(a) (n=343,681, two LPA variants explaining 36% of variance) and 12 cerebrovascular phenotypes, including ischaemic stroke subtypes (up to 62,100 cases) and intracranial haemorrhage subtypes (up to 15,400 cases), estimating effects per 100 nmol/L genetically predicted Lp(a) increase. Genetically predicted Lp(a) was associated with increased risk of all-cause ischaemic stroke (OR 1.04, 95% CI 1.02-1.07) and, most strongly, large artery atherosclerotic stroke (OR 1.23, 95% CI 1.14-1.33; early-onset OR 1.37, 95% CI 1.15-1.64); a nominal association appeared with cardioembolic stroke (OR 1.07, 95% CI 1.01-1.13) and none with small vessel stroke (OR 0.98, 95% CI 0.91-1.06). Secondary analyses showed significant associations with carotid plaque and atrial fibrillation, nominal associations with lobar haemorrhage and microinfarcts, and null associations with cerebral small vessel disease phenotypes. The authors conclude elevated Lp(a) drives ischaemic stroke primarily through large artery atherosclerosis, supporting prioritising these patients in Lp(a)-lowering stroke-prevention trials.
Original abstract
Background: Lipoprotein(a) (Lp(a)) is a highly atherogenic lipoprotein and the target of investigational therapies. Using a Mendelian randomization study design, we aimed to clarify associations between genetically predicted Lp(a) levels and cerebrovascular disease outcomes and related phenotypes.
Methods: We obtained genetic associations with Lp(a) levels (n=343 681), ischemic stroke subtypes (≤62 100 cases), intracranial hemorrhage subtypes (≤15 400 cases), and 12 related cerebrovascular phenotypes. Lp(a) was proxied using 2 LPA genetic variants (rs10455872 and rs3798220) that explain 36% of the variance in Lp(a) levels. We performed Mendelian randomization analyses to estimate the association of a genetically predicted 100 nmol/L increase in Lp(a) levels on each outcome.
Results: Genetically predicted Lp(a) levels associated with significantly increased risk of all-cause ischemic stroke (odds ratio [OR], 1.04 [95% CI, 1.02-1.07], P=2.05×10-4) and large artery atherosclerotic stroke (OR, 1.23 [95% CI, 1.14-1.33], P=3.54×10-7). There was a nominal association with cardioembolic stroke (OR, 1.07 [95% CI, 1.01-1.13], P=0.02), and no evidence for association with small vessel stroke (OR, 0.98 [95% CI, 0.91-1.06], P=0.60). Associations with early-onset stroke were similar, though with a greater magnitude of association for large artery atherosclerotic stroke (OR, 1.37 [95% CI, 1.15-1.64], P=5.58×10-4). Analyses of secondary outcomes paralleled these findings, including significant associations of genetically predicted Lp(a) with carotid plaque and atrial fibrillation, nominal associations with lobar hemorrhage and autopsy-confirmed microinfarcts, and null associations with cerebral small vessel disease phenotypes.
Conclusions: Elevated Lp(a) is primarily associated with ischemic stroke due to large artery atherosclerosis, while showing no link to cerebral small vessel disease. These findings support prioritization of patients with atherosclerotic cerebrovascular disease in Lp(a)-lowering stroke prevention trials.
Summary written by lp-a.org from the published abstract; figures as published. Page updated 17 August 2026. Methods.