Elevated Lipoprotein(a) is independently associated with symptomatic intracranial atherosclerotic stenosis in acute ischemic stroke
Evangelos Panagiotopoulos, Maria-Ioanna Stefanou, Vasileios Giannopapas, Eleni Bakola, Georgia Papagiannopoulou, Aikaterini Theodorou, Vaia Lambadiari, Michele Romoli, Ignatios Ikonomidis, Evangelos Liberopoulos, Sotirios Giannopoulos, Odysseas Kargiotis, Klearchos Psychogios, Apostolos Safouris, Mira Katan, Georgios P. Paraskevas, Annerose Mengel, Lina Palaiodimou, Georgios TsivgoulisBackground
Symptomatic intracranial atherosclerotic stenosis (sICAS) represents a frequent cause of recurrent acute ischemic stroke (AIS). Although lipoprotein(a) [Lp(a)] is an established risk factor for atherosclerosis, its association with sICAS and large-artery atherosclerotic (LAA) stroke has been scarcely investigated in Caucasians.
Objectives
To investigate the association between serum Lp(a) levels and sICAS, and secondarily evaluate their association with LAA stroke.
Design
Cross-sectional, single-center observational study.
Methods
We enrolled consecutive patients with AIS. Multivariable logistic regression assessed the association between serum Lp(a) levels and sICAS. Discriminative performance was assessed using receiver operating characteristic (ROC) analysis; associations between Lp(a) and LAA stroke were evaluated in secondary models.
Results
Among 407 consecutive AIS patients, 43 (10.6%) had sICAS. Higher Lp(a) levels were independently associated with sICAS after multivariable adjustment. Each 1-unit increase in log 10 -transformed Lp(a) was associated with approximately 3-fold increased odds of sICAS (adjusted odds ratio [aOR]: 3.34; 95% CI: 1.78-6.25; p<0.001), and each 10 nmol/L increase in Lp(a) conferred a 5% increase in sICAS odds (aOR: 1.05; 95% CI: 1.02-1.08; p=0.003). Lp(a) levels ≥75 nmol/L were associated with approximately 3-fold higher odds of sICAS (aOR: 3.16; 95% CI: 1.56-6.41; p=0.001). In ROC analysis, an Lp(a) cutoff of 53.8 nmol/L demonstrated a high negative predictive value (93%), but a low positive predictive value (19%) for sICAS. In secondary analyses, higher Lp(a) levels were also independently associated with LAA stroke.
Conclusion
Elevated Lp(a) levels were independently associated with sICAS and LAA stroke in our cross-sectional study. The cutoff of 53.8 nmol/L was associated with a low likelihood of sICAS, warranting external validation. Lp(a) may serve as a routinely available biomarker for etiologic risk stratification in AIS, rather than as a stand-alone diagnostic marker.