Increased Plasma Lipoprotein(a) Found in Large-Artery Atherosclerotic, but Not Small-Artery Occlusive, Stroke

We previously reported that plasma lipoprotein(a) [Lp(a)]1 is an independent risk factor for coronary artery disease, abdominal aortic aneurysm, peripheral arterial disease, and ischemic stroke (1). Although the exact pathophysiological role of Lp(a) has not been definitively established, both atherogenic and thrombogenic mechanisms have been proposed. In this study, we measured plasma Lp(a) in ischemic stroke patients who had been subdivided into etiologic subtypes with the TOAST (Trial of ORG 10172 in Acute Stroke Treatment) classification to determine if stroke subtype is a confounder of the Lp(a) association. Ischemic stroke patients (n = 245) and arterial disease–free and stroke-free controls (n = 435) were recruited from the Otago region of New Zealand as previously reported (1). Stroke patients were placed into TOAST classification subtypes by means of a computer algorithm developed by Goldstein and colleagues(2). The TOAST subgroups were large-artery atherosclerosis (LAA), cardioembolism, small-artery occlusion (SAO), undetermined due to 2 or more causes (multiple etiologies) (ME), or undetermined etiology due to negative or incomplete evaluation. Inclusion criteria for controls were an age >52 years and no history of ischemic heart disease or stroke, including transient ischemic attack [Questionnaire for Verifying Stroke-Free Status (QVSFS) score = 0]. Plasma lipoprotein and demographic risk factors were assessed as previously described(1). The study was approved by the local human ethics committee, and all study participants gave written informed consent before being recruited into this study. The χ2 test and ANOVA were used to assess nominal and normally distributed variables, respectively. We used the Mann–Whitney U-test to assess Lp(a) because of the nongaussian distribution of this variable. Logistic regression models were used to stratify Lp(a) by control population quartiles. Odds ratios were calculated for the fourth quartile (>50 nmol/L) vs the first quartile (<4.5 nmol/L). Multiple logistic regression was used to identify statistically significant interactive effects of variables on susceptibility to ischemic stroke. Demographic factors associated with ischemic stroke are shown in Table 1 ; statistically significant confounders were included in subsequent multivariable logistic regression models. Plasma Lp(a) concentrations were significantly higher in the total-stroke group than in vascular disease–free controls (P < 0.03, Table 1). Lp(a) was differentially associated with TOAST stroke subtypes. Most notably, SAO strokes were not associated with Lp(a), and the association with the all-strokes group was largely attributable to increased concentrations in the “atherothrombotic” stroke subtypes (LAA and ME). Overall, 86% of ME patients had LAA as one of their identified pathologies. The ME cases with LAA were grouped with the LAA TOAST subgroup to form a patient population designated as having an “atherothrombotic” stroke. Compared with stroke-free controls, atherothrombotic cases had an odds ratio of 3.47 (95% CI, 1.26–9.60; P < 0.02) and an adjusted (sex, age, hypertension, dyslipidemia, smoking history, diabetes, and history of coronary heart disease) odds ratio of 3.36 (95% CI, 1.04–10.89; P < 0.05). Lp(a) concentration was also significantly higher in this group than in the SAO stroke subgroup (P < 0.02). Demographic profiles of the ischemic stroke populations.1 Results are expressed as the mean (SD) or as the median (interquartile range). Odds ratios include the 95% CI and the P value. The adjusted model includes sex, age, hypertension, dyslipidemia, smoking history, diabetes, and history of coronary heart disease. Note the differential risk association between Lp(a) and TOAST stroke subtypes. CE, cardioembolism; UE, undetermined due to negative or incomplete evaluation; BMI, body mass index; HDL-C, HDL cholesterol. P < 0.05, vs controls. P < 0.01, vs controls. P < 0.0001, vs controls. P < 0.002, vs controls. P < 0.05, vs ME. P < 0.05, vs SAO. Demographic profiles of the ischemic stroke populations.1 Results are expressed as the mean (SD) or as the median (interquartile range). Odds ratios include the 95% CI and the P value. The adjusted model includes sex, age, hypertension, dyslipidemia, smoking history, diabetes, and history of coronary heart disease. Note the differential risk association between Lp(a) and TOAST stroke subtypes. CE, cardioembolism; UE, undetermined due to negative or incomplete evaluation; BMI, body mass index; HDL-C, HDL cholesterol. P < 0.05, vs controls. P < 0.01, vs controls. P < 0.0001, vs controls. P < 0.002, vs controls. P < 0.05, vs ME. P < 0.05, vs SAO. This study therefore suggests that our previously reported association of Lp(a) with ischemic stroke was principally due to the presence of the atherothrombotic (LAA and ME) stroke subgroups. In contrast, the plasma Lp(a) concentration in the SAO stroke group was indistinguishable from that in stroke-free elderly control individuals. This observation is consistent with other studies that have suggested that plasma Lp(a) concentrations in SAO patients are approximately half those of LAA patients (3); however, the present report is the first to simultaneously examine all stroke subtypes and to measure Lp(a) with an apolipoprotein A isoform–insensitive method. We therefore concluded that plasma Lp(a) is associated with atherothrombotic cardiovascular disease phenotypes and appears to show no association with nonatherosclerotic pathologies such as SAO, which is typically associated with lipohyalinosis. Because a large proportion (86%) of ME cases had LAA as one of their associated pathologies, we considered it reasonable to combine these ME cases with LAA cases in the LAA group to further increase the power of the study. Of note, however, is that Lp(a) has been implicated as being both atherogenic and thrombogenic; therefore, high concentrations of this plasma lipoprotein may be associated with a compounded risk of ME stroke. We recommend that this possibility be investigated in future, larger-sized studies of stroke populations. Such differential risk associations have also been reported for other stroke-related biomarkers (e.g., brain natriuretic peptide and d-dimer) as independent predictors of cardioembolic stroke, but not of other stroke subtypes(4). In conclusion, this study suggests that plasma Lp(a) is differentially associated with ischemic stroke subtypes and highlights the importance of etiologic subclassification in assessing stroke-susceptibility biomarkers. Author Contributions:All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article. Authors’ Disclosures of Potential Conflicts of Interest:No authors declared any potential conflicts of interest. Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, or preparation or approval of manuscript. Acknowledgments: We thank Vicky Phillips, Jennifer Cole, Terry Biltoft, and the Otago Vascular Assessment Laboratory for their technical assistance and recruitment of patients. We thank Santica M. Marcovina, Northwest Lipid Research Laboratories, University of Washington, Seattle, for allowing us to use her Lp(a) assay.