Abstract T P144: Association Between Lipoprotein (a) Levels and Vascular Events in the Stenting and Aggressive Medical Management for Preventing Recurrent Stroke in Intracranial Stenosis (SAMMPRIS) trial

Abstract

Background: Lipoprotein (a) (Lp(a)) has been implicated in atherothrombogenesis and prior studies have suggested its pathogenic role in the development of intracranial atherosclerosis (ICAS). Objective: We sought to determine the association between baseline Lp(a) levels and risk of vascular events during follow-up in patients randomized to the medical arm of the SAMMPRIS trial. Methods: Baseline Lp(a) data was available for 226 of 227 patients randomized to the medical arm in SAMMPRIS. High Lp(a) was defined as > 65 mg/dL in Blacks and > 35 mg/dL in Whites and Others. The primary outcome was any ischemic stroke and the secondary outcome was a composite of vascular events (ischemic stroke, MI and vascular death). Kaplan-Meier curves for the time to the outcomes were compared between patients with high and low Lp(a) using the log-rank test. Using Lp(a) as a continuous variable, Cox proportional hazards model was used to determine the association between Lp(a) and time to outcomes of interest. Results: Overall, 17/92 (18.5%) patients with high Lp(a) levels had an ischemic stroke during follow-up versus 24/134 (17.9%) with normal Lp(a) levels, with no difference between the Kaplan-Meier curves of the groups (p=0.94). In a Cox proportional hazards model relating Lp(a) to time to ischemic stroke, Lp(a) was not statistically significant (p = 0.82). Similarly, 23/92 (25%) patients with high Lp(a) levels had a vascular event during follow-up versus 29/134 (21.6%) with normal Lp(a) levels, with no difference between the Kaplan-Meier curves (p=0.58). In a Cox proportional hazards model relating Lp(a) to time to ischemic stroke, MI, or vascular death, Lp(a) was not statistically significant (p = 0.46). Conclusion: Patients with high Lp(a) levels in the SAMMPRIS trial did not have an increased risk of ischemic strokes or vascular events during follow-up. Although prior studies have suggested a pathogenic contribution of high Lp(a) to the development of ICAS, our results do not find an association between high Lp(a) and increased risk of vascular events.