15289-s-bos

80 | Chapter 5 patients. Our results are in line with previous findings showing that Lp(a) concentrations might have a casual role in AVC development [13], which may result in symptomatic end-stage aortic valve disease [12, 15, 25, 26]. Statin treatment has substantially improved the prognosis of patients with heterozygous FH [27]. Nonetheless, we observed impressive CAD in asymptomatic, long-term statin-treated FH patients [28]. It might be that additional risk factors that are independent of LDL levels determine the residual CVD risk in treated FH. In the Jupiter trial, statin negligibly influenced the average Lp(a) concentrations, whereas Lp(a) levels determined the residual risk [29]. In our treated FH patients, we found that Lp(a) levels were associated with AVC but not with CAC. It is, therefore, tempting to speculate that Lp(a) contributed to the residual CVD risk of treated FH through AVC and subsequent aortic valve stenosis. It is well-established that plasma Lp(a) concentrations are largely genetically determined by the variation in the number of KIV repeats that are based on genetic variants at the LPA locus. However, we cannot exclude that Lp(a) is solely an associated factor for AVC in the present study. In the future, a large genetic study, i.e. a Mendelian randomization approach, is required to investigate the causality. Interestingly, LMW apo(a) phenotypes were not significantly associated with AVC, despite the strong correlation between apo(a) phenotypes and Lp(a) concentration. Moreover, the frequency of LMW apo(a) is similar in patients with and without AVC. We cannot exclude that this is a chance finding, because 14 of the 29 patients who expressed an LMW isoform also expressed an HMW isoform. The large group of mixed isoforms reduces the power for detecting an association between the specific LMW isoform and AVC. To study this association in FH a larger cohort is required. Alternatively, the LMW apo(a) phenotype was previously found to be a strong predictor for CVD outcomes [8, 30, 31], apo(a) KIV repeat number was not significantly associated with aortic valve stenosis in the general population [12]. As CAC and AVC have several pathological similarities and share risk factors [32-34], and CAC had a strong predictive value towards AVC [35, 36], we tested whether Lp(a) concentration was generally associated with calcification or specifically associated with calcification at the aortic valve. Interestingly, Lp(a) was associated only with AVC but not CAC in our patients. Moreover, the association between Lp(a) concentration and AVC was not influenced by the adjustment for CAC. These results suggest that high Lp(a) levels might point at FH patients specifically at risk for AVC. How Lp(a) is involved in the mechanism of AVC development remains to be established.