Oxidized Phospholipids and Risk of Calcific Aortic Valve Disease: The Copenhagen General Population Study

Abstract

Objective: Lipoprotein(a) is causally associated with calcific aortic valve disease (CAVD). Lipoprotein(a) carries proinflammatory and procalcific oxidized phospholipids (OxPL). We tested whether the CAVD risk is mediated by the content of OxPL on lipoprotein(a).

Approach and results: A case-control study was performed within the Copenhagen General Population Study (n=87 980), including 725 CAVD cases (1977-2013) and 1413 controls free of cardiovascular disease. OxPL carried by apoB (apolipoprotein B-100; OxPL-apoB) or apolipoprotein(a) (OxPL-apo(a)) containing lipoproteins, lipoprotein(a) levels, LPA kringle IV type 2 repeat, and rs10455872 genetic variants were measured. OxPL-apoB and OxPL-apo(a) levels correlated with lipoprotein(a) levels among cases (r=0.75 and r=0.95; both P<0.001) and controls (r=0.65 and r=0.93; both P<0.001). OxPL-apoB levels associated with risk of CAVD with odds ratios of 1.2 (95% confidence interval [CI]:1.0-1.6) for 34th to 66th percentile levels, 1.6 (95% CI, 1.2-2.1) for 67th to 90th percentile levels, 2.0 (95% CI, 1.3-3.0) for 91st to 95th percentile levels, and 3.4 (95% CI, 2.1-5.5) for levels >95th percentile, versus levels <34th percentile (trend, P<0.001). Corresponding odds ratios for OxPL-apo(a) were 1.2 (95% CI, 1.0-1.5), 1.2(95% CI, 0.9-1.6), 2.1(95% CI, 1.4-3.1), and 2.9(95% CI, 1.9-4.5; trend, P<0.001) and were similar for lipoprotein(a). LPA genotypes associated with OxPL-apoB, OxPL-apo(a), and lipoprotein(a) levels and explained 34%, 46%, and 39%, respectively, of the total variation in levels. LPA genotypes associated with risk of CAVD; a doubling in genetically determined OxPL-apoB, OxPL-apo(a), and lipoprotein(a) levels associated with odds ratio of CAVD of 1.18 (95% CI, 1.10-1.27), 1.09 (95% CI, 1.05-1.13), and 1.09 (95% CI, 1.05-1.14), respectively, comparable to the corresponding observational estimates of 1.27 (95% CI, 1.16-1.39), 1.13 (95% CI, 1.08-1.18), and 1.11 (95% CI, 1.06-1.17).

Conclusions: OxPL-apoB and OxPL-apo(a) are novel genetic and potentially causal risk factors for CAVD and may explain the association of lipoprotein(a) with CAVD.

Keywords: apolipoproteins B; case-control study; genotype; lipoprotein(a); phospholipids.

Figure 1

Concentration distributions of OxPL-apoB and lipoprotein(a) in controls and cases. Note, one lipoprotein(a) measurement of 327 mg/dL among cases was not included in the depicted lipoprotein(a) data (lower right panel).

Figure 2

Association of OxPL-apoB with lipoprotein(a) levels in controls and cases. The best fit linear regression line is shown in grey. One lipoprotein(a) measurement of 327 mg/dL among cases was not included in the depicted data (lower panel); the OxPL-apoB value in this individual was 22.0 nmol/L.

Figure 3

Risk of calcific aortic valve disease as a function of OxPL-apoB or lipoprotein(a) levels. Analyses were adjusted for age and sex or multivariable adjusted. Analyses excluded cases without matched controls, and vice versa (N=17). Abbreviations: OR, odds ratio; CI, confidence interval.

Figure 4

OxPL-apoB and lipoprotein(a) levels as a function of LPA genotypes. Boxes show median and interquartile range (also given in numbers) and error bars depict the 10th and 90th percentiles.

Figure 5

Risk of calcific aortic valve disease per doubling in OxPL-apoB or lipoprotein(a) levels in observational and genetic (instrumental variable) analyses. Abbreviations: OR, odds ratio; CI, confidence interval; RR, relative risk.