Lipid-Lowering Therapy and Hemorrhagic Stroke Risk: Comparative Meta-Analysis of Statins and PCSK9 Inhibitors

Statins were shown to increase hemorrhagic stroke (HS) in patients with a first cerebrovascular event in 2006 (SPARCL), likely due to off-target antithrombotic effects, but continued to sometimes be used in patients with elevated HS risk due to absence of alternative medications. Recently, the PCSK9Is (proprotein convertase subtilisin kexin 9 inhibitors) have become available as a potent lipid-lowering class with potentially less hemorrhagic propensity.

We performed a systematic comparative meta-analysis assessing HS rates across all completed statin and PCSK9I randomized clinical trials with treatment >3 months, following PRISMA guidelines. In addition to HS rates across all trials, causal relation was probed by evaluating for dose-response relationships by medication (low versus high medication dose/potency) and by presence and type of preceding brain vascular events at inception (none versus ischemic stroke/transient ischemic attack versus HS).

The systematic review identified 36 statin randomized clinical trials (204 918 patients) and 5 PCSK9I randomized clinical trials (76 140 patients). Across all patient types and all medication doses/potencies, statins were associated with increased HS: relative risk 1.15, P=0.04; PCSK9Is were not (P=0.77). In the medication dose/potency analysis, higher dose/potency statins (7 trials, 62 204 patients) were associated with magnified HS risk: relative risk, 1.53; P=0.002; higher dose/potency PCSK9Is (1 trial, 27 564 patients) were not (P=0.99). In the type of index brain vascular injury analysis for statins (5 trials, 9772 patients), prior ischemic stroke/transient ischemic attack was associated with a magnified risk of HS: relative risk, 1.43; P=0.04; and index intracerebral hemorrhage was associated with an extremely high effect estimate of risk of recurrent HS: hazard ratio, 4.06. For PCSK9Is, prior ischemic stroke/transient ischemic attack (1 trial, 5337 patients) was not associated with increased HS risk (P=0.97).

Statins increase the risk of HS in a medication dose- and type of index brain vascular injury-dependent manner; PCSK9Is do not increase HS risk. PCSK9Is may be a preferred lipid-lowering medication class in patients with elevated HS risk, including patients with prior HS.