Lp(a) above 31.2 mg/dL raises poor 3-month stroke outcome risk 79%, cohort of 175 (Brain Behav 2026)

Original title: Clinical Study on the Association Between Lipoprotein(a) Levels and Prognosis in Patients With Acute Ischemic Stroke

Brain Behav · 1 May 2026 · 6

Chen X, Zhang Y, Tang Y, Hu J, He X, Wu J

Retrospective cohort of 175 patients with acute ischaemic stroke, testing Lp(a) as a predictor of poor 3-month functional outcome (modified Rankin Scale above 2). Elevated Lp(a) was an independent risk factor for poor outcome (OR 1.09 per unit, p=0.004), with a significant threshold effect at 31.2 mg/dL, above which the risk of poor outcome rose 79% (p=0.0398). Combining Lp(a) with admission NIHSS score gave higher discriminative accuracy (AUC 0.6820) and specificity (0.9208) than either marker alone, with sensitivity analyses confirming robustness. The authors conclude Lp(a) is a stable, independent predictor of poor 3-month stroke outcome with a clear threshold at 31.2 mg/dL, supporting its use alongside NIHSS for early risk stratification.

Original abstract

Background: Lipoprotein(a) (Lp(a)) is a known cardiovascular risk factor, but its role in acute ischemic stroke (AIS) prognosis remains unclear. This study investigated the association between Lp(a) levels and 3-month functional outcomes in AIS patients.

Methods: We conducted a retrospective cohort study of 175 AIS patients. Prognosis was assessed using the 3-month modified Rankin Scale (mRS > 2 defined as poor). Logistic regression and threshold effect models analyzed the association, while ROC curves compared the predictive performance of Lp(a) and NIHSS scores.

Results: Elevated Lp(a) was an independent risk factor for poor 3-month prognosis (OR = 1.09, p = 0.004). A significant threshold effect was identified at 31.2 mg/dL, above which the risk of poor outcomes increased by 79% (p = 0.0398). Combining Lp(a) with admission NIHSS scores yielded a significantly higher AUC (0.6820) and specificity (0.9208) than either indicator alone. Sensitivity analyses confirmed the robustness of these findings.

Conclusion: Lp(a) is a stable and independent predictor of poor 3-month outcomes in AIS patients, exhibiting a clear threshold effect at 31.2 mg/dL. The combined Lp(a)-NIHSS model enhances prognostic accuracy, supporting Lp(a) as a valuable biomarker for early risk stratification.

Summary written by lp-a.org from the published abstract; figures as published. Page updated 17 August 2026. Methods.