Genetically elevated lipoprotein(a) and increased risk of myocardial infarction
Pia R Kamstrup, Anne Tybjaerg-Hansen, Rolf Steffensen, Borge G Nordestgaard - JAMA, 2009
In Danish general-population cohorts, both observational and genetic analyses linked higher lipoprotein(a) to higher myocardial-infarction risk. Using the LPA KIV-2 size polymorphism as a genetic instrument (Mendelian randomization), genetically elevated Lp(a) was associated with a hazard ratio of 1.22 per doubling, an estimate largely free of confounding and reverse causation. The authors concluded the data are "consistent with a causal association," cautious wording rather than a claim of definitive proof.
Key findings
- Genetically elevated lipoprotein(a), instrumented by the LPA KIV-2 repeat polymorphism, was associated with a hazard ratio of 1.22 (95% CI 1.09-1.37) for myocardial infarction per doubling of Lp(a) - a Mendelian-randomization estimate robust to confounding and reverse causation.
- Observationally, multivariable-adjusted MI risk rose with Lp(a) level, reaching a hazard ratio of 2.6 (95% CI 1.6-4.1) for the highest category (>95th percentile).
- A moderate-elevation category gave a non-significant hazard ratio of 1.2 (95% CI 0.9-1.6), the confidence interval crossing 1.0.
- The authors concluded the data are "consistent with a causal association" between elevated Lp(a) and MI - cautious wording, not a claim of definitive proof.
Effect measures
- Hazard Ratio: 1.22 per doubling of genetically elevated lipoprotein(a)95% CI 1.09-1.37
- Hazard Ratio: 2.695% CI 1.6-4.1
- Hazard Ratio: 1.295% CI 0.9-1.6
Why this evidence tier (High)
- Risk of bias:
- Mendelian randomization using the LPA KIV-2 size polymorphism as a genetic instrument, largely robust to confounding and reverse causation. A single-locus, candidate-gene instrument is the main internal-validity caveat.
- Precision:
- The genetic instrumental-variable estimate is statistically significant with a moderate, reasonably tight interval (HR 1.22, 95% CI 1.09-1.37).
- Directness:
- Hard clinical endpoint (myocardial infarction) and an Lp(a)-specific exposure; directly addresses whether Lp(a) is causal.
- Consistency:
- Concordant with PROCARDIS (Clarke 2009) and the broader Lp(a) Mendelian-randomization literature.
- Funding / COI:
- Funding and author financial disclosures are not shown on the PubMed record and were not retrievable from the primary source this session; the JAMA full text carries them.
High certainty that genetically elevated Lp(a) is causally associated with higher myocardial-infarction risk, with the authors framing the result as "consistent with" causality rather than definitive proof.
- Population:
- Three Danish general-population studies (the Copenhagen City Heart Study, the Copenhagen General Population Study, and the Copenhagen Ischemic Heart Disease Study). Plasma lipoprotein(a), the LPA kringle IV type 2 (KIV-2) size polymorphism, and incident/prevalent myocardial infarction were assessed, with KIV-2 repeat number used as a genetic instrument (Mendelian randomization / instrumental-variable analysis).
- Conflicts of interest:
- Not displayed on the PubMed abstract; funding source and author financial disclosures were not captured from the primary source this session. The JAMA full text carries the full disclosure and funding statements.
- Funding:
- Not captured from the PubMed record; the funding statement appears in the JAMA full text and was not retrievable this session.
Limitations
- Single genetic instrument (the LPA KIV-2 size polymorphism); a candidate-gene Mendelian-randomization design rather than a genome-wide instrument.
- Conducted in three Danish general-population cohorts, so generalizability beyond a European-ancestry population is uncertain.
- The causal claim is framed cautiously ("consistent with a causal association"), not as definitive proof.
- Funding and per-author conflict-of-interest disclosures were not retrievable from the PubMed record this session.