Plasma HDL cholesterol and risk of myocardial infarction: a mendelian randomisation study.
Level 3 - non-randomized controlled study
Mendelian randomization meta-analysis using genetic instrumental variables in observational cohorts and case-control studies.
PubMed 22607825 · doi:10.1016/S0140-6736(12)60312-2
What was done
The authors performed two Mendelian randomisation analyses to test whether the association between plasma HDL cholesterol and myocardial infarction (MI) is causal. First, they analyzed a single nucleotide polymorphism (SNP) in the endothelial lipase gene (LIPG Asn396Ser) across 20 studies comprising 20,913 MI cases and 95,407 controls. Second, they evaluated a genetic score of 14 common SNPs exclusively associated with HDL cholesterol in up to 12,482 MI cases and 41,331 controls. A genetic score of 13 common SNPs exclusively associated with LDL cholesterol served as a positive control.
What was found
Carriers of the LIPG 396Ser allele had higher HDL cholesterol (0.14 mmol/L higher, p=8x10^-13) without differences in other risk factors, which observationally expected a 13% reduction in MI risk (OR 0.87, 95% CI 0.84-0.91). However, the 396Ser allele showed no association with MI (OR 0.99, 95% CI 0.88-1.11, p=0.85). Observationally, a 1 SD increase in HDL cholesterol was associated with lower MI risk (OR 0.62, 95% CI 0.58-0.66), whereas a 1 SD increase genetically instrumented by the 14-SNP score was not associated with MI (OR 0.93, 95% CI 0.68-1.26, p=0.63). In contrast, the LDL cholesterol genetic score was strongly associated with MI risk (OR 2.13 per 1 SD increase, 95% CI 1.69-2.69, p=2x10^-10), consistent with observational epidemiology (OR 1.54, 95% CI 1.45-1.63).
Why it matters
These findings challenge the assumption that raising plasma HDL cholesterol will uniformly reduce myocardial infarction risk, demonstrating that epidemiological correlations between HDL-C and cardiovascular disease do not necessarily reflect a causal relationship.
Limits
The study evaluated specific genetic variants and mechanisms (LIPG and a 14-SNP genetic score) and may not generalize to all biological pathways that influence HDL function or other HDL particle characteristics not captured by plasma HDL-C concentrations.
Cited by
- supports Genetic variants associated with high LDL are associated with increased heart disease risk, whereas genetic variants associated with HDL cholesterol levels are almost never associated with heart disease risk.