Abstract 1092: Genetically Low HDL Cholesterol without Increased Risk of Ischemic Heart Disease
In epidemiological studies, reduced levels of high-density lipoprotein (HDL) cholesterol are inversely related to increased cardiovascular risk. Whether this is a causal effect is unclear. We tested this hypothesis using Mendelian randomization. We examined the ability of mutations associated with lifelong reductions in plasma levels of HDL cholesterol per se to predict risk of ischemic heart disease (IHD) in 9190 white individuals from the general population, The Copenhagen City Heart Study, followed prospectively for 28 years. Heterozygosity for five nonsynonymous mutations in ATP-Binding-Cassette transporter A1 (ABCA1) was associated with substantial reductions in HDL cholesterol per se (mean percentile 15th, 95 percent confidence interval (CI): 8th to 22nd; P<0.00001 versus noncarriers), and an absolute reduction in HDL cholesterol of 20 mg per deciliter (0.52 mmol per liter; P<0.00001 versus noncarriers). The predicted hazard ratio for IHD for this HDL decrease, assuming the well-known inverse relationship between plasma HDL cholesterol and risk, was 1.7 (95 percent CI, 1.6 to 1.8). However, the observed hazard ratio was 0.7 (95 percent CI, 0.3 to 1.6) in ABCA1 heterozygotes versus noncarriers, a result which was confirmed in an independent case-control study including 2497 cases with IHD and 7722 controls (odds ratio 0.4, 95 percent CI, 0.2 to 1.2). The clinical implications are that individuals with an isolated low HDL cholesterol and no other risk factors should not be treated pharmacologically or otherwise in an attempt to increase plasma levels of HDL cholesterol, since they are probably not at an increased risk of cardiovascular disease. This is also supported by recent reports that torcetrapib, a cholesteryl ester transfer inhibitor which increases plasma levels of HDL cholesterol, failed to protect against progression of atherosclerosis. In conclusion, a genetically lifelong reduction in plasma levels of HDL cholesterol per se does not predict an increased risk of IHD. This suggests that the inverse relationship observed in epidemiological studies between decreased levels of HDL cholesterol and increased cardiovascular risk may not be causal.