15289-s-bos

5 75 | Lp(a) and AoVC in FH of plasma containing 30 ng apo(a) protein. Plasma samples were mixed with SDS gel- loading buffer and heated at 98oC for 5 minutes. Gel electrophoresis was performed under a reducing condition with SDS and a 1.75% agarose gel. Proteins were then transferred to a PVDF membrane (Millipore) using a semi-dry blotting system (Millipore Graphite Electroblotter II) at 13V for 1 hour. Themembrane was subsequently incubated with blocking buffer containing 1% Bovine Serum Albumin (BSA) for 1 hour at 38oC to reduce non-specific binding. Primary antibody incubation was performed using monoclonal antibody 1A2 against apo(a) KIV [22], followed by incubation with goat anti-mouse IgG antibody coupled to Horseradish peroxidase (HRP) (Pierce, 1:3000). Enhanced chemiluminescence (ECL) and visualization on film were used for detection. A mixture of human plasma samples with 5 isoforms of known number of apo(a) KIV repeats was used as reference material. Statistical analysis Data are presented as mean + standard deviation or median (Interquartile range, IQR) for continuous variables and as number (%) for categorical variables. Mann-Whitney U and Fisher’s exact tests were performed to analyze the differences between 2 continuous and categorical parameters, respectively. The presence of AVC or CAC was defined as AVC or CAC score of more than 0 AU. The severity of AVC or CAC was expressed by the AU score as continuous variables. Apo(a) phenotypes were categorized into 2 groups: low molecular weight (LMW) apo(a) (KIV < 22 repeats) and high molecular weight (HMW) apo(a) (KIV > 22 repeats). When 2 apo(a) isoforms were detected in the immunoblot, the smaller isoform was used for categorization as discussed recently [6]. Hypertension was defined as a systolic blood pressure (SBP) above 140 mm Hg or a diastolic blood pressure (DBP) above 90 mm Hg or those who were receiving antihypertensive therapy at the time of inclusion. Cholesterol-year score, a measurement of life-long cholesterol burden, was calculated using the formula (untreated total cholesterol x years without statins) + (statin-treated total cholesterol x years with statins) [23]. Correlations between Lp(a) concentrations and other variables were determined using Spearman correlation test. In order to account for different levels of CAC score, regressionmodelswere fitted on Lp(a) concentration, the presence and the severity of AVC separately, using CAC score as an independent variable. Subsequently, the residuals were used to calculate Spearman correlation coefficient (r). Significant predictors for the presence of AVC (AVC score > 0) were identified using univariate logistic regression analysis. The variables entered to the model were classical risk factors for CVD or parameters previously suggested [24].