Transl Stroke Res. 2026 Oct 8;17(5):109. doi: 10.1007/s12975-026-01499-6.
ABSTRACT
Lipid traits are established risk factors for atherosclerotic disease; however, their relationship with intracranial aneurysms and subarachnoid haemorrhage (SAH) remains unclear. Recent Mendelian randomisation (MR) studies have reported inconsistent findings. This study aimed to review the literature and assess the association between lipid traits and SAH using a one-sample MR analysis accounting for statin use and nonlinearity. We performed a systematic review of MR studies investigating lipid traits and intracranial aneurysms/SAH. A meta-analysis was not undertaken because studies used overlapping datasets. In parallel, we conducted a one-sample MR analysis of 486,603 UK Biobank participants. Genetic instruments for low-density lipoprotein (LDL), high-density lipoprotein (HDL), total cholesterol (TC), and triglycerides (TG) were derived from published genome-wide association studies. Associations with SAH were adjusted for lipid-lowering therapy. Nonlinear relationships were explored using cubic splines and piecewise MR. While eight MR studies met the inclusion criteria, seven relied on the same outcome cohort. Despite this, these studies showed no consistent direction of effect across lipid traits, suggesting that the estimates were unstable. In the UK Biobank, there was no evidence of an association between genetically predicted lipid traits and SAH for LDL (HR 1.06 [0.64-1.73]), HDL (HR 2.57 [0.42-15.74]), TC (HR 1.06 [0.71-1.60]), or TG (HR 1.05 [0.69-1.59). Results were consistent across different MR methods and sensitivity analyses, including adjustment for lipid-lowering therapy. Current evidence does not support a major association between lipid traits and SAH, with the heterogeneity observed across previous MR studies likely reflecting methodological limitations rather than a consistent biological effect. However, small effects and lipid-independent effects of lipid-lowering therapies cannot be excluded.
PMID:42848158 | DOI:10.1007/s12975-026-01499-6

